PUBLISHED
UNITED STATES COURT OF APPEALS
FOR THE FOURTH CIRCUIT
No. 10-5296
UNITED STATES OF AMERICA,
Plaintiff – Appellee,
v.
THOMAS ROYAL,
Defendant – Appellant.
Appeal from the United States District Court for the District of
Maryland, at Baltimore. Richard D. Bennett, District Judge.
(1:09-cr-00439-RDB-1)
Argued: October 26, 2012 Decided: October 1, 2013
Before TRAXLER, Chief Judge, DIAZ, Circuit Judge, and Catherine
C. EAGLES, United States District Judge for the Middle District
of North Carolina, sitting by designation.
Affirmed in part, vacated in part, and remanded by published
opinion. Judge Diaz wrote the opinion, in which Chief Judge
Traxler and Judge Eagles joined.
ARGUED: James Christopher Fraser, VENABLE, LLP, Baltimore,
Maryland, for Appellant. John Walter Sippel, Jr., OFFICE OF THE
UNITED STATES ATTORNEY, Baltimore, Maryland, for Appellee. ON
BRIEF: James Wyda, Federal Public Defender, Baltimore, Maryland,
Paresh S. Patel, OFFICE OF THE FEDERAL PUBLIC DEFENDER,
Greenbelt, Maryland, for Appellant. Rod J. Rosenstein, United
States Attorney, Baltimore, Maryland, for Appellee.
DIAZ, Circuit Judge:
Thomas Royal was convicted by a jury of unlawfully
possessing ammunition after being previously convicted of a
crime punishable by imprisonment for a term exceeding one year,
in violation of 18 U.S.C. § 922(g)(1). At sentencing, the
district court determined that Royal was an armed career
criminal under the Armed Career Criminal Act (“ACCA”), 18 U.S.C.
§ 924(e), triggering a fifteen-year mandatory minimum sentence.
The court sentenced him to 188 months’ imprisonment.
Royal advances three arguments on appeal. First, he
contends the government failed to present sufficient evidence to
convict him of knowingly possessing “ammunition,” arguing that
since the rounds he possessed were loaded in an antique firearm,
the government had a burden to show that the rounds were
actually designed for use in a non-antique firearm. Second,
Royal asserts that the district court committed plain error when
it instructed the jury that the phrase “knowingly possessed
ammunition” meant that Royal knew the rounds were “ammunition as
we commonly use the word.” Finally, Royal argues that the
district court reversibly erred by using the modified
categorical approach to determine that his prior conviction
under Maryland’s second-degree assault statute constituted a
predicate conviction under the ACCA. We conclude that
sufficient evidence supported the jury’s verdict and that the
2
district court did not err in its jury instructions. However,
in light of the Supreme Court’s recent holding in Descamps v.
United States, 133 S. Ct. 2276 (2013), we sustain Royal’s
challenge to the district court’s application of the modified
categorical approach and its imposition of the ACCA sentencing
enhancement. Accordingly, we affirm in part, vacate in part,
and remand for resentencing.
I.
On appeal from a criminal conviction, we view the evidence
in the light most favorable to the government. United States v.
Herder, 594 F.3d 352, 358 (4th Cir. 2010).
A.
On January 8, 2009, Sergeant Jones and Detective Rayam of
the Baltimore, Maryland Police Department stopped Thomas Royal
for driving with an expired registration plate. During the
stop, Detective Rayam observed Royal place his hand in his front
left jacket pocket, where Rayam noticed a bulge. Suspecting
criminal activity, the officers asked if they could search the
car. After initially consenting to the search, Royal grew
agitated and attempted to push his way past Detective Rayam.
Sergeant Jones used his Taser device to subdue him, and a
subsequent search of Royal’s person revealed an antique Iver
Johnson revolver loaded with five .32 caliber rounds.
3
Since Royal had previously been convicted of second-degree
assault, “a crime punishable by imprisonment for a term
exceeding one year,” the federal Gun Control Act (“GCA”)
prohibited him from knowingly possessing “any firearm or
ammunition” that has traveled in interstate commerce. 18
U.S.C. § 922(g)(1). “Firearm” and “ammunition” are defined
terms under the GCA. As is relevant here, the term “firearm”
explicitly “does not include an antique firearm,” which is any
firearm “manufactured in or before 1898.” Id. § 921(a)(3),
(a)(16)(A). “The term ‘ammunition’ means
ammunition . . . designed for use in any firearm.” Id.
§ 921(a)(17)(A).
Royal was charged in a one-count indictment for possession
of ammunition by a prohibited person, in violation of GCA
§ 922(g)(1). At trial, the government called Special Agent
David Collier of the Bureau of Alcohol, Tobacco, Firearms, and
Explosives (“ATF”) as “an expert in identification of firearms
and ammunition” and the “interstate nexus of firearms and
ammunition.”
Collier testified that Royal’s revolver had been
manufactured in 1895 and therefore, for the purposes of the GCA,
was an “antique firearm,” which is not a “firearm.” Collier
also testified that the rounds had been manufactured outside
Maryland and had traveled in interstate commerce. He gave no
4
testimony about whether Royal’s rounds had been “designed for
use in any firearm” such that they technically fell within the
statute’s definition of “ammunition,” but conceded on cross-
examination that he did not know the dates of the rounds’
manufacture.
The evidence showed that the rounds were .32 caliber, and
manufactured by the arms companies Remington and Winchester.
The rounds themselves were also entered into evidence and
published to the jury. The government presented no specific
evidence as to the rounds’ design. Nor, aside from cross-
examining Collier about the rounds’ manufacture dates, did Royal
himself raise any issues concerning the design of the
ammunition.
At the close of the government’s case, Royal moved under
Fed. R. Crim. P. 29 for a judgment of acquittal, arguing that
the government had presented insufficient evidence to support a
guilty verdict. The district court denied the motion. Royal
advanced no defense but renewed his Rule 29 motion, which the
court again denied.
After closing arguments, the district court instructed the
jury that the government needed to prove that Royal knowingly
possessed one or more pieces of ammunition. It read them the
GCA’s statutory definition of ammunition: “‘Ammunition’ is
ammunition or cartridge cases, primers, bullets, or propellant
5
powder designed for use in any firearm,” J.A. 148. With respect
to the mens rea component, the district court instructed the
jury, without objection from Royal, that “whether the defendant
acted knowingly . . . means [whether] he knew that the
ammunition was ammunition as we commonly use the word.” The
jury returned a guilty verdict on the sole count.
A conviction for being a felon in possession of a firearm
or ammunition normally carries a maximum penalty of ten years in
prison. 18 U.S.C. § 924(a)(2). However, at Royal’s sentencing,
the government argued that he was subject to a fifteen-year
mandatory minimum sentence because he had three prior
convictions “for a violent felony or serious drug offense,” id.
§ 924(e)(1), including a 2007 guilty plea to Maryland second-
degree assault. Royal argued that under the categorical
approach, this conviction did not qualify as a “violent felony”
because “violent force” is not categorically required for a
conviction under Maryland’s second-degree assault statute. The
district court rejected this argument and, following our
decision in United States v. Alston, 611 F.3d 219 (4th Cir.
2010), applied the so-called modified categorical approach to
conclude that, based on facts admitted in his plea colloquy,
Royal’s 2007 second-degree assault conviction was indeed a
violent felony. The district court thus applied the ACCA
§ 924(e)(1) enhancement and sentenced Royal to fifteen years and
6
eight months in prison. Royal timely appealed. Following oral
argument, we placed the case in abeyance pending the Supreme
Court’s decision in Descamps, which issued in June 2013.
II.
The issues before us are (1) whether Royal was entitled to
a Rule 29 judgment of acquittal due to insufficient evidence;
(2) whether the district court committed plain error when it
instructed the jury that the phrase “knowingly possessed
ammunition” meant “[Royal] knew the ammunition was ammunition as
we commonly use the word”; and (3) whether the district court
erred in applying the modified categorical approach to determine
that Royal’s 2007 Maryland second-degree assault conviction
qualified as a predicate “violent felony” under the ACCA.
A.
We first consider Royal’s argument that he was entitled to
a judgment of acquittal because the government offered
insufficient evidence to support his conviction. We review de
novo the district court’s decision to deny a defendant’s Rule 29
motion for judgment of acquittal. United States v. Smith, 451
F.3d 209, 216 (4th Cir. 2006). On an appeal challenging the
sufficiency of evidence, we assess the evidence in the light
most favorable to the government, and the jury’s verdict must
stand unless we determine that no rational trier of fact could
7
have found the essential elements of the crime beyond a
reasonable doubt. United States v. Young, 609 F.3d 348, 355
(4th Cir. 2010).
In order to prove that Royal possessed ammunition in
violation of the GCA, the government was required to prove
beyond a reasonable doubt that (1) Royal was a convicted felon;
(2) he knowingly possessed ammunition; and (3) the ammunition
had traveled in interstate commerce. See United States v. Moye,
454 F.3d 390, 395 (4th Cir. 2006). Royal challenges the
sufficiency of the government’s evidence as to only the second
prong, insisting that the government failed to show that the
five rounds found in his revolver were “ammunition” within the
statute’s definition. Specifically, he argues that the
government never met its burden of proving beyond a reasonable
doubt that the rounds were “designed for use in any firearm”;
that is, the government never showed that the rounds were
designed for use in any non-antique firearm. One of our sister
circuits has helpfully articulated this distinction:
Bullets are “ammunition” if they are “designed for use
in any firearm.” 18 U.S.C. § 921(a)(17)([A])
(emphasis added). If these bullets had been designed
exclusively for use in [defendant’s antique] revolver,
they would not be “ammunition” because by definition
this antique revolver is not a “firearm.” On the
other hand, if the bullets were designed for use, not
just in this antique revolver, but in other guns
manufactured after 1898, then it would appear, given
the literal language of the definition, that they are
8
“ammunition” because they would be designed for any
firearm.
United States v. Mixon, 457 F.3d 615, 618 (7th Cir. 2006).
The government does not contest this literal reading of the
statute. Rather, it posits that its burden of proof did not
entail a need to demonstrate that the rounds were designed for
use in a non-antique firearm. Instead, it argues that whether
the ammunition was designed exclusively for use in an antique
firearm is an affirmative defense that must be raised by the
defendant and supported by evidence before the government must
disprove its application.
To support this theory, the government cites numerous cases
holding that the “antique firearm exception” in 18
U.S.C. § 921(a)(3) is an affirmative defense that the defendant
bears the burden of raising. Because the antique firearm
exception is an affirmative defense to a § 922(g) firearm
charge, the government insists, it follows that a claim that
ammunition was designed for use in a non-antique firearm is also
an affirmative defense, which Royal failed to raise.
We accept Royal’s literal reading of the statute and agree
that rounds designed for use exclusively in antique firearms do
not meet the GCA’s definition of “ammunition.” However, we also
agree with the government that the antique firearms exception is
an affirmative defense, which Royal failed to raise.
9
It is well established that the antique firearm exception
is an affirmative defense to a firearm charge under § 922(g).
See, e.g., United States v. McMillan, 346 F. App’x 945, 947 (4th
Cir. 2009) (unpublished); United States v. Lawrence, 349 F.3d
109, 122 (3d Cir. 2003) (“Every circuit court of appeals that
has considered this issue has agreed that [the antique firearm
exception] is an affirmative defense that must initially be
raised by sufficient evidence to justify shifting a burden of
proof to the government.”); United States v. Mayo, 705 F.2d 62,
75 (2d Cir. 1983) (“We find no indication in the language of the
statute that Congress intended the government to prove in all
criminal prosecutions under 18 U.S.C. § 922 that the illegal
firearms transactions involved weapons that were not
antiques.”). This owes to the longstanding principle that “an
indictment or other pleading founded on a general provision
defining the elements of an offense . . . need not negative the
matter of an exception made by a proviso or other distinct
clause . . . . [I]t is incumbent on one who relies on such an
exception to set it up and establish it.” McKelvey v. United
States, 260 U.S. 353, 357 (1922). Accordingly, since
§ 921(a)(3) clearly sets apart the antique firearm exception as
a distinct proviso to the general definition of “firearm,”
courts have not hesitated to place the burden on defendants to
10
raise it as an affirmative defense. We agree with this broadly
held view.
Unlike § 921(a)(3)’s antique firearm exception, which
stands alone as a separate sentence untethered to the general
definition of “firearm,” the “designed for use in any firearm”
language of § 921(a)(17)(A)’s definition of “ammunition” is part
and parcel of the definitional sentence. Consequently, it is
the government’s initial burden to prove as an element of the
offense that the rounds were “designed for use in any firearm.”
We are not persuaded, however, that that burden somehow
incorporates the antique firearms exception, thereby requiring
the government to initially come forth with proof that the
ammunition was not designed exclusively for use in antique
firearms. Instead, if a defendant seeks the shelter of the
antique firearms exception as it relates to § 921(a)(17)(A)’s
“designed for use” clause, it remains incumbent on him to raise
the exception as an affirmative defense at trial.
This Royal failed to do. Although he is correct that a
defendant need only produce “more than a scintilla of evidence”
to raise an affirmative defense, United States v. Sligh, 142
F.3d 761, 762 (4th Cir. 1998), and that an affirmative defense
may be raised by the testimony of the government’s own
witnesses, see Sherman v. United States, 356 U.S. 369, 373
(1958), Royal fails to satisfy even this minimal burden.
11
Special Agent Collier’s testimony that the rounds were
found in an antique revolver and that he did not know the dates
of their manufacture constituted the full extent of the evidence
on this issue. This evidence, we conclude, was too attenuated
to sufficiently raise the defense. Although the ammunition was
found loaded in an antique revolver, the mere fact that the
ammunition happened to fit in an antique firearm does not mean
it was designed for antique firearms. Cf. Mixon, 457 F.3d at
618 (“It is true that the bullets were in the cylinder [of an
antique revolver], but that simple fact hardly establishes as a
matter of law that they were designed for, and could be safely
used, in this weapon.”). Royal, meanwhile, offered no testimony
or evidence on the rounds’ design and never asserted at trial
that the ammunition was not designed for use in a modern
firearm. Accordingly, because the jury ultimately did not hear
even a scintilla of evidence to suggest that the rounds were
designed for use exclusively in an antique firearm, Royal was
not entitled to have it consider the issue as an affirmative
defense.
Meanwhile, although the government did not present evidence
specifically going to the ammunition’s design, the evidence was
still sufficient to establish that the ammunition was “designed
for use in any firearm.” The evidence showed that the
ammunition was .32 caliber, manufactured by Remington and
12
Winchester, and the ammunition itself was shown to the jury.
Since most people are familiar with the appearance of modern
ammunition, we find that the jury could reasonably have
concluded that .32 caliber rounds manufactured by well-known
firearms companies were “designed for use in any firearm.”
As the district court correctly noted, the government’s
evidence satisfied all three elements of the § 922(g)(1)
violation. We therefore hold that sufficient evidence supported
Royal’s conviction.
B.
We next address Royal’s challenge to the district court’s
jury instructions. Since Royal did not object to the
instructions, we review for plain error. United States v.
Robinson, 627 F.3d 941, 953-54 (4th Cir. 2010). Under this
standard, Royal “must establish that the district court erred,
that the error was plain, and that it affected his substantial
rights.” Id. at 954 (internal quotation marks omitted).
Royal asserts that the district court plainly erred when it
instructed the jury that the phrase “knowingly possessed
ammunition” meant that Royal “knew the ammunition was ammunition
as we commonly use the word.” He relies on United States v.
Tomlinson, where we held that to sustain a conviction, a
district court must instruct the jury that the defendant must
have had “knowledge of those facts that bring the firearm within
13
[the] legal definition” prohibited by the GCA. 67 F.3d 508, 514
(4th Cir. 1995). Here, Royal insists the district court ran
afoul of Tomlinson when it failed to instruct the jury that the
government needed to prove that Royal knew the rounds were
“ammunition” within the GCA’s definition, and specifically that
Royal knew the rounds were designed for use in a non-antique
firearm.
We disagree. In light of our first holding, we find no
error, plain or otherwise, in the district court’s instructions
on the “knowing possession of ammunition” element of the
offense. The district court properly instructed the jury as to
the statutory definition of the word ammunition. And we are
satisfied that the instructions adequately informed the jury
that, to sustain a conviction, Royal needed to have knowledge of
those facts that brought the rounds in this case within that
legal definition. See United States v. Frazier-El, 204 F.3d
553, 561 (4th Cir. 2000) (“The conventional mens rea of criminal
statutes . . . requires not that a defendant know that his
conduct was illegal, but only that he ‘know the facts that make
his conduct illegal.’” (quoting Staples v. United States, 511
U.S. 600, 605 (1994))). As we have already explained, evidence
that the rounds were designed exclusively for use in an antique
firearm is not required to prove an element of the offense, but
rather provides an affirmative defense. Because the matter of
14
the rounds’ design was not in issue on the facts presented, the
district court did not err, plainly or otherwise, by failing to
mention it in its jury instructions.
C.
Finally, we address Royal’s challenge to the district
court’s use of the modified categorical approach and its
determination that his 2007 Maryland second-degree assault
conviction qualified as a violent felony under the ACCA, 18
U.S.C. § 924(e)(1). We review this determination de novo.
United States v. Harcum, 587 F.3d 219, 222 (4th Cir. 2009).
In Descamps, the Supreme Court recently clarified whether
courts may apply the modified categorical approach to assess,
for ACCA sentencing enhancement purposes, the violent nature of
a defendant’s prior conviction under an indivisible criminal
statute (i.e., one that does not set out elements of the offense
in the alternative, but which may nevertheless broadly
criminalize qualitatively different categories of conduct).
Answering that question in the negative, the Court explained
that the modified categorical approach “serves a limited
function: It helps effectuate the categorical analysis when a
divisible statute, listing potential offense elements in the
alternative, renders opaque which element played a part in the
defendant's conviction.” Descamps, 133 S. Ct. at 2283.
15
In this case, Royal’s 2007 Maryland second-degree assault
conviction is predicated on a facially indivisible statute. 1
Nevertheless, the government has argued elsewhere that
authoritative judicial decisions have, in effect, converted
Maryland’s second-degree assault statute from indivisible to
divisible, because the Maryland courts have held that the
completed battery form of second-degree assault may consist of
either “offensive physical contact” or infliction of “physical
harm.” Supp. Br. of Appellee at 18-19, United States v.
Barillas, No. 11-5141 (4th Cir. Aug. 28, 2013), ECF No. 42
(quoting Nicolas v. State, 44 A.3d 396, 402 (Md. 2012)).
Because an assault involving “physical harm” would qualify as a
violent felony for sentencing purposes, the argument goes,
courts may continue to apply the modified categorical approach
to Maryland second-degree assault convictions, in order to
determine whether the defendant’s conviction was in fact for the
“physical harm” variety of the offense. Id. at 21.
In addressing this argument, we acknowledge that the
Supreme Court in Descamps “reserve[d] the question whether, in
determining a crime’s elements, a sentencing court should take
1
Maryland’s statute prohibiting second-degree assault
provides simply that “[a] person may not commit an assault.”
Md. Code, Crim. Law § 3-203(a). "Assault" encompasses “the
crimes of assault, battery, and assault and battery, which
retain their judicially determined meanings.” Id. § 3-201(b).
16
account not only of the relevant statute’s text, but of judicial
rulings interpreting it.” 133 S. Ct. at 2291. We need not
resolve that question here, however, because regardless of
whether judicial decisions might in theory turn an indivisible
statute into a divisible one, that is simply not what the
Maryland courts have done with respect to the completed battery
form of second-degree assault.
As the Court explained in Descamps, offenses are divisible
when they consist of alternative elements through which the
offense may be proved. Id. at 2283. By “elements,” the Court
meant factual circumstances of the offense that the jury must
find “unanimously and beyond a reasonable doubt.” Id. at 2288
(citing Richardson v. United States, 526 U.S. 813, 817 (1999)).
Thus, to decide whether “offensive physical contact” and
“physical harm” are alternative elements of the completed
battery form of second-degree assault, we consider how Maryland
courts generally instruct juries with respect to that offense.
To convict a defendant of an assault of the battery
variety under Maryland law, “the State must prove that: (1) the
defendant caused offensive physical contact with, or harm to,
the victim; (2) the contact was the result of an intentional or
reckless act of the defendant and was not accidental; and (3)
the contact was not consented to by the victim or was not
legally justified.” Nicolas, 44 A.3d at 407 (quoting,
17
favorably, trial court jury instructions). Maryland juries are
not instructed that they must agree “unanimously and beyond a
reasonable doubt” on whether the defendant caused “offensive
physical contact” or “physical harm” to the victim; rather, it
is enough that each juror agree only that one of the two
occurred, without settling on which. See also Robinson v.
State, 58 A.3d 514, 528, 531 (Md. Ct. Spec. App. 2012) (quoting
instruction requiring jury to find, among other elements, “that
the defendant caused offensive physical contact with or physical
harm to [the victim],” and describing that instruction as
“mirror[ing] the pattern jury instruction for second degree
assault”).
Rather than alternative elements, then, “offensive physical
contact” and “physical harm” are merely alternative means of
satisfying a single element of the Maryland offense.
Consequently, because “[t]he dispute here does not concern any
list of alternative elements,” the modified approach “has no
role to play.” Descamps, 133 S. Ct. at 2285.
Instead, we must apply the traditional categorical
approach, under which we look “only to the statutory definition
of the state crime and the fact of conviction to determine
whether the conduct criminalized by the statute, including the
most innocent conduct, qualifies as a ‘crime of violence.’”
United States v. Torres–Miguel, 701 F.3d 165, 167 (4th Cir.
18
2012) (internal quotation marks omitted). And because, “as we
have repeatedly observed,” Maryland’s second-degree assault
statute reaches any unlawful touching, whether violent or
nonviolent and no matter how slight, “convictions under the
statute, including [Royal’s], cannot categorically be crimes of
violence.” Karimi v. Holder, 715 F.3d 561, 568 (4th Cir. 2013).
Accordingly, Royal’s 2007 Maryland second-degree assault
conviction does not constitute a predicate “violent felony”
supporting a sentencing enhancement under ACCA § 924(e)(1). Cf.
Johnson v. United States, 559 U.S. 133, 140-42 (2010) (holding
that a violent felony under the ACCA necessarily involves the
use of “violent force”).
III.
For the foregoing reasons, we conclude that sufficient
evidence supported Royal’s conviction and that the district
court did not err in its jury instructions. However, in light
of Descamps, the district court’s application of the modified
categorical approach to support Royal’s ACCA sentencing
19
enhancement was in error. Accordingly, we affirm Royal’s
conviction, vacate his sentence, and remand for resentencing. 2
AFFIRMED IN PART,
VACATED IN PART,
AND REMANDED
2
We deny Royal’s motion for leave to file a supplemental
brief.
20