FILED
                                                                        United States Court of Appeals
                                        PUBLISH                                 Tenth Circuit

                      UNITED STATES COURT OF APPEALS                            July 19, 2016

                                                                            Elisabeth A. Shumaker
                             FOR THE TENTH CIRCUIT                              Clerk of Court
                         _________________________________

UNITED STATES OF AMERICA,

      Plaintiff - Appellee,

v.                                                            No. 15-2019

CODY LITTLE,

      Defendant - Appellant.
                      _________________________________

                     Appeal from the United States District Court
                           for the District of New Mexico
                          (D.C. No. 5:12-CR-02654-RB-1)
                       _________________________________

Andre Poissant, Assistant Federal Public Defender, Office of the Federal Public
Defender, Las Cruces, New Mexico, for Defendant-Appellant.

Shaheen P. Torgoley, Assistant United States Attorney (Damon P. Martinez, United
States Attorney, Office of the United States Attorney, Albuquerque, New Mexico with
her on the brief), for Plaintiff-Appellee.
                          _________________________________

Before TYMKOVICH, Chief Judge, LUCERO and HOLMES, Circuit Judges.
                 _________________________________

LUCERO, Circuit Judge.
                    _________________________________


       After police discovered stolen weapons in Cody Little’s residence, a jury

convicted him of being a felon in possession of a firearm and of possessing a stolen

firearm. Little appeals, challenging several jury instructions. We agree with Little that
constructive possession requires proof of intent to exercise dominion and control over an

object following the Supreme Court’s opinion in Henderson v. United States, 

(2015). However, because the evidence presented at trial compels the conclusion

that Little intended to exercise control over the weapons, we hold the district court’s error

in omitting the intent element from its jury instruction was harmless. We further

conclude that the district court permissibly issued instructions regarding aiding and

abetting and possible guilt of others, and that a deliberate indifference instruction was

harmless. Finally, we hold that the district court erroneously relied on the Sentencing

Guidelines’ residual clause in calculating Little’s offense level. Thus, exercising

jurisdiction under 28 U.S.C. § 1291 and 18 U.S.C. § 3742(a)(2), we affirm Little’s

convictions but vacate his sentence and remand for resentencing.

                                              I

       On the night of October 24, 2011, seven firearms were stolen from Southwest

Arms in Lovington, New Mexico. The store owner received an anonymous tip that

Little was the burglar. She passed the tip on to investigating authorities. Law

enforcement learned that Little was residing on the property of Lacosta Blythe, not

far from Southwest Arms.

       On November 1, 2011, several law enforcement officials, including Agent

Anthony Budrow, visited Blythe’s property. Blythe, her boyfriend, and two minor

children lived in a residential trailer on her lot. Little resided nearby in a six-by-eight

foot “well house” he rented from Blythe. Several storage sheds were also located on

the property.

                                              2
      Upon arriving at the scene, Budrow activated a recording device. The officers

spent several minutes speaking to two or three individuals on the porch of Blythe’s

trailer. Seven and a half minutes after arriving, Budrow made eye contact with Little

as Little was exiting the well house. Two officers pursued Little down a walkway

and behind the trailer. Budrow remained at the front of the property and observed

that no one else entered or exited the well house. One of the officers who pursued

Little escorted him back to the front of the trailer. The other officer stayed behind,

where he observed assault rifle hand guards inside of an open shed. Upon seeing the

guards, the officers sought a search warrant. They cleared the premises and ensured

no one entered any building on the property while they waited for the warrant to

issue, although Little indicated he wanted to go back into the well house.

      The officers received and executed a warrant later that day. Upon entering the

well house, they saw two plainly visible shotgun shells on a shelf below eye level.

They also recovered two firearms matching guns stolen in the burglary. One of the

weapons—a loaded assault rifle—was found inside a duffel bag under or inside a

sleeping bag on a bed. The other weapon—a shotgun—was found under the bed.

Officers described the well house as cramped. The bed, which ran almost the entire

length of the well house, was the only place to sit.

      No other firearms were discovered on the premises. However, officers found a

small bag of methamphetamine in the trailer. They agreed that the trailer’s residents

would not be charged for possession of the drugs if they cooperated with the

officers—particularly by recovering any other stolen firearms. One day after the

                                           3
search, Blythe turned over a third gun found between two sheds on the property. The

gun was missing its hand guard, and matched the guard found inside the shed.

      Little was indicted for being a felon in possession of a firearm and ammunition

in violation of 18 U.S.C. §§ 922(g)(1) and 924(a)(2). Following his first trial, the

jury deadlocked and a mistrial was declared. The government then filed a

superseding indictment adding a count for possession of stolen firearms, §§ 922(j)

and 924(a)(2). The superseding indictment also charged aiding and abetting as to

both counts. Little again proceeded to trial.

      At the second trial, the government presented testimony from several law

enforcement officers describing the foregoing search and seizure. Blythe testified

that the last time she had entered the well house was in mid-September 2011, a week

after Little moved in. She saw Little access the well house on a daily basis, but never

saw anyone else enter. Blythe further testified that Little installed a lock on the well

house door about a week before the search—around the time Southwest Arms was

burgled—but did not give her a key. However, officers did not recall seeing the lock,

and testified that the well house was not locked when they performed their search.

Blythe also averred that when she showed Little the police inventory report after the

search, he responded, “[t]hey only found two?” The jury also heard a recorded phone

call Little placed from jail during which he described another burglary suspect as a

“snitch,” and angrily accused him of “snitch[ing] me out.”




                                            4
      The jury convicted Little on both counts.1 The district court calculated a total

offense level of 28, based in part on Little’s prior convictions for crimes of violence.

It imposed a within-Guidelines sentence of 150 months’ imprisonment. Little timely

appealed.

                                           II

      We review de novo whether jury instructions, as a whole, correctly state the

law and provide the jury with an understanding of the issues. United States v.

Wittgenstein, 

, 1168 (1998). We will disturb a judgment only if we

have “substantial doubt that the jury was fairly guided.” United States v. Smith, 

, 1424 (10th Cir. 1994).

                                           A

      Little challenges the instruction on constructive possession. Conviction for

possession of a firearm pursuant to § 922 requires proof of “knowing possession.”

See United States v. Heckard, 

, 1228 (10th Cir. 2001). Such

possession may be either actual or constructive. See United States v. Mills, 

, 549 (10th Cir. 1994). The district court used the unmodified Tenth Circuit

Criminal Pattern Jury Instruction § 1.31 (2011), which provides that constructive

possession exists when a person not in actual possession “knowingly has the power at

a given time to exercise dominion or control over an object.” Little argues that the




      1
        Little was convicted for possession of the items recovered from the well
house, not the firearm recovered by Blythe after the initial search.
                                            5
court should have instead instructed that constructive possession exists when a

person knowingly has the power and the intention to exercise dominion or control.

      In United States v. Colonna, 

(10th Cir. 2004), we held that for

purposes of constructive possession “[i]t is not necessary to show that the defendant

intended to exercise . . . dominion or control.” 

We have subsequently

reaffirmed that holding. See, e.g., United States v. Jameson, 

, 1211 n.2

(10th Cir. 2007). As we explained in United States v. Ledford, 

(10th

Cir. 2005), a panel may not overrule Colonna “[a]bsent intervening Supreme Court or

en banc authority to the contrary.” 

      Little argues that Henderson v. United States, 

(2015), provides

the intervening Supreme Court authority lacking in Ledford.2 In Henderson, the

Court held that an individual convicted of being a felon in possession of a firearm

may direct the transfer of a seized firearm to a third party, provided that the third

party would not grant the individual access to the weapon. 135 S. Ct at 1784-85. It

rejected the government’s argument that allowing such a transfer would place the

felon in constructive possession of the weapon. 

Crucially, for our

purposes, the Court observed that “[c]onstructive possession is established when a

person, though lacking physical custody, still has the power and intent to exercise

control over the object.” 

(emphasis added) (citing Black’s Law


      2
        Although Henderson was decided after Little’s sentencing, when the
Supreme Court decides a new rule, “the integrity of judicial review requires that we
apply that rule to all similar cases pending on direct review.” Griffith v. Kentucky,

, 323 (1987).
                                            6
Dictionary 1047 (5th ed. 1979) and 2A K. O’Malley, J. Grenig, & W. Lee, Federal

Jury Practice and Instructions, Criminal § 39.12, p. 55 (6th ed. 2009)). The Court

further observed that “the very hallmark of possession” is that a defendant have

“broad[] command over the gun’s location and use.” 

n.3.

      We agree that Henderson changes the law of constructive possession in our

circuit. In Henderson, the Court squarely held that constructive possession requires

both power to control an object and intent to exercise that control. 

.

Because Colonna’s disavowal of an intent requirement is incompatible with the

Supreme Court’s decision in Henderson, we overrule that point of law. See United

States v. White, 

, 1123 n.2 (10th Cir. 2015) (“Although typically, one

panel of this court cannot overrule the judgment of another panel, we may do so if an

intervening decision from the Supreme Court invalidates our previous analysis.”).

We thus hold that constructive possession exists when a person not in actual

possession knowingly has the power and intent at a given time to exercise dominion

or control over an object. Accord 

.3

      This holding aligns our circuit not only with Supreme Court precedent but with

every other circuit but one that has considered the issue. See United States v.

Introcaso, 

, 270 (3d Cir. 2007); United States v. Bustamante, 

, 889 (7th Cir. 2007); United States v. Piwowar, 

, 955 (8th Cir.

      3
        In United States v. Ibarra-Diaz, 

(10th Cir. 2015), an opinion
issued after Henderson was decided, we stated in dicta that constructive possession
requires that a “defendant knowingly has the power to exercise control or dominion
over the item.” 

(quotation omitted). We did not
consider Henderson’s impact on our circuit precedent in Ibarra-Diaz.
                                           7
2007); United States v. McFarlane, 

, 59 (1st Cir. 2007); United States v.

Gardner, 

, 713 (6th Cir. 2007); United States v. Jones, 

,

788 (5th Cir. 2007); United States v. Ruiz, 

, 1089-90 (9th Cir. 2006);

United States v. Paulino, 

, 222 (2d Cir. 2006); United States v. Greer,

, 1271 (11th Cir. 2006); United States v. Scott, 

, 435-36

(4th Cir. 2005); see also United States v. Johnson, 

, 1993 U.S. App.

LEXIS 26474, at *5 (D.C. Cir. 1993) (unpublished). But see United States v.

Littlejohn, 

, 1338 (D.C. Cir. 2007) (“To establish constructive

possession, the government must show that the defendant knew of, and was in a

position to exercise dominion and control over, the contraband.” (quotation

omitted)).

      Having decided that constructive possession requires intent to exercise control,

we must consider whether the district court’s jury instruction constitutes reversible

error. The government contends that there was no error because a different

instruction defined the word “knowingly” as “voluntarily and intentionally.” Thus,

the government suggests, the district court instruction that constructive possession

required knowledge and control effectively required intent. But reading these

instructions together, the jury was informed that constructive possession requires that

a defendant voluntarily and intentionally have the power to exercise dominion or

control over an object. Intentionally having the power to exercise dominion is not

the same as intending to exercise dominion. For example, a felon who knows his

neighbor keeps a gun in his bedroom and who is given a key to watch his neighbor’s

                                           8
house would intentionally have the power to exercise control over the weapon, but

would not be guilty of constructive possession without the intent to actually exercise

control. See Benjamin C. McMurray, Hands Off the Gun! A Critique of United

States v. Jameson and Constructive Possession Law in the Tenth Circuit, 85 Denv. U.

L. Rev. 531, 561 (2008) (listing hypotheticals). We thus conclude that the

“voluntarily and intentionally” instruction does not remedy the deficiency.

      However, “[e]ven when the district court fails to include an element of the

crime in the instruction (including a mens rea element), we still apply the harmless

error rule, asking whether it appears beyond a reasonable doubt that the error

complained of did not contribute to the verdict obtained.” United States v. Sorensen,

, 1229 (10th Cir. 2015). We hold that the district court’s error was

harmless because a reasonable jury would be compelled to conclude that Little

intended to exercise control over the weapons. The government submitted substantial

evidence demonstrating that Little had exclusive control over the well house,

including his placement of a lock on the door around the time the firearms were

stolen. And when Blythe reported to Little that police seized two guns, he responded

“[t]hey only found two?” Little argues that Blythe was an unreliable witness, but

even setting her testimony aside, there is no dispute that Little spent at least seven

and a half minutes in the six-by-eight foot well house when officers arrived on the

day of the search. Law enforcement found plainly visible ammunition in the well

house and weapons on and under Little’s bed, which spanned nearly the entire length

of the house and was the only place to sit. Given his exclusive control of the tiny

                                            9
well house, and the seven and a half minutes he indisputably spent in it, it would be

unreasonable to conclude that Little did not know about the presence of the weapons.

See United States v. Hishaw, 

, 571 (10th Cir. 2000) (“In most cases, the

defendant’s dominion, control, and knowledge may be inferred if he has exclusive

possession of the premises on which the object was found.”). The same is true as to

Little’s intent to exercise command over their location. See United States v. Griffin,

, 695 (3d Cir. 2012) (“Exclusive control over the premises allows the

jury to infer the knowledge and intent to control objects within those premises . . .

.”); United States v. Alanis, 

, 592 (7th Cir. 2001) (“When a gun is found

in a defendant’s bedroom . . . it would not be improper for the jury to infer that the

defendant had both knowledge of the firearm and an intent to exercise dominion and

control over it merely from its presence in the bedroom . . . .” (quotation and

alteration omitted)).

      We acknowledge that the foregoing cases stand for the proposition that a jury

may infer constructive possession under such circumstances, not that a jury must

make that inference. But given the particular facts of this case, we conclude there is

no reasonable possibility that the jury would have found that Little had knowledge of

the weapons at issue but lacked intent to exercise control over them.4 We thus hold

that the constructive possession instruction constituted harmless error.


      4
        The dissent correctly observes that we generally undertake harmless error
analysis sua sponte only if the harmlessness of the error is certain. See Mollett v.
Mullin, 

, 920 (10th Cir. 2003) (quotation omitted). The “certainty”
burden is necessarily more stringent than the “reasonable doubt” test we would apply
                                           10
                                           B

      Little challenges the district court’s instruction on aiding and abetting. He

argues that inclusion of the instruction was inconsistent with the government’s theory

that Little was the principal. He further argues that there was insufficient evidence to

support the instruction.

      We have repeatedly held that a district court may provide an aiding and

abetting instruction even if the government argues a defendant is guilty as a

principal. See United States v. Cooper, 

, 1050-51 (10th Cir. 2004);

United States v. Cueto, 

, 1275-76 (10th Cir. 1980). So long as there is

record evidence from which the jury could find that the defendant “was either a

principal, or an aider and abettor,” the government may submit both theories to the

jury. 

(quotation omitted).

      We thus turn to Little’s sufficiency argument. Generally, a conviction for

aiding and abetting requires that the defendant: “(1) willfully associate[s] himself

with the criminal venture, and (2) seek[s] to make the venture succeed through some

action of his own.” United States v. Bowen, 

, 1078 (10th Cir. 2008).

Little suggests that a conviction for aiding and abetting a § 922(g) offense requires




had the government thoroughly developed its harmlessness argument. See United
States v. Serawop, 

, 669 (10th Cir. 2005). But Little indisputably spent
seven and a half minutes in the six-by-eight foot well house with plainly visible
ammunition and two stolen firearms while police questioned his landlord and
acquaintances outside his door. Contrary to the dissent, we are left with no doubt
that Little intended to exercise control over the weapons.
                                          11
evidence that another specific felon possessed the weapons and that the aider and

abettor knew or had cause to know of this other person’s status as a felon.5

      Even assuming Little is correct regarding the evidence necessary to support an

aiding and abetting instruction,6 we conclude the government met its burden. Little

argued that other individuals may have possessed the stolen firearms and

ammunition—in particular, Blythe. During cross-examination, Blythe admitted she

was a felon. She also testified that she and Little had known each other since they

were children. The jury could infer that if Blythe possessed the weapons, Little

intentionally sought to make that possession succeed by allowing her to keep the

weapons in the well house over which he had exclusive control, and that Little knew

      5
         Little does not challenge the language of the instruction on appeal, and any
such challenge is therefore waived. Adler v. Wal-Mart Stores, Inc., 

,
679 (10th Cir. 1998) (“Arguments inadequately briefed in the opening brief are
waived.”). Similarly, Little argued to the district court that an aiding and abetting
instruction should not be given because the evidence did not support such an
instruction, but he did not object to the particular language of the instruction or
proffer alternative language. See United States v. Zapata, 

, 1190 (10th
Cir. 2008) (“[A] generalized objection to an instruction is insufficient to preserve a
specific objection on appeal.”).
      6
         The circuits are split as to whether a defendant charged with aiding and
abetting a felon in possession must know the principal was a felon. Compare United
States v. Xavier, 

, 1286 (3d Cir. 1993) (“[T]here can be no criminal
liability for aiding and abetting a violation of § 922(g)(1) without knowledge or
having cause to believe the possessor’s status as a felon.”), and United States v.
Gardner, 

, 715 (6th Cir. 2007) (“[I]n order for aiding-and-abetting
liability to attach under § 922(g), the government must show that the defendant knew
or had cause to know that the principal was a convicted felon.”), with United States
v. Canon, 

, 1442 (9th Cir. 1993) (government is not required to show
defendant knew of principal’s status as a felon). But see United States v. Graves, 

, 1188 (9th Cir. 1998) (questioning the Ninth Circuit’s position, and
suggesting that it would be “logical” to require that an aider abettor under § 922(g)
have knowledge of the principal’s status).
                                          12
of Blythe’s status as a felon. We accordingly conclude that there existed sufficient

evidence to support the aiding and abetting instruction.

                                            C

      Little challenges the “deliberate ignorance” instruction for insufficient

evidence. The jury was informed it could infer knowledge if Little “was aware of a

high probability of the existence of the charged firearms or ammunition where he was

living, unless the defendant did not actually believe the charged firearms or

ammunition were present where he was living.”

       “[A] deliberate ignorance instruction is proper only when evidence has been

presented showing the defendant purposely contrived to avoid learning the truth.”

United States v. Bornfield, 

, 1129 (10th Cir. 1998) (quotation omitted).

In making this determination, we view the evidence in the light most favorable to the

government. 

“alerts the jury that [an] act of avoidance . . .

motivated by sufficient guilty knowledge [may] satisfy the knowing element of the

crime.” 

alteration omitted). However, the risk of providing this

instruction when there is “evidence of direct knowledge but no evidence of avoidance

of knowledge is that the jury could still convict a defendant who merely should have

known about the criminal venture.” United States v. Manriquez Arbizo, 

, 249 (10th Cir. 1987).

      The government argues evidence that Little should have known about the

firearms is sufficient to show that he deliberately avoided knowledge. And the

district court overruled Little’s objection to this instruction based on the substantial

                                           13
evidence suggesting he knew or should have known that the firearms were in the well

house. But Manriquez Arbizo rejected this rationale. 

(“[I]f the

evidence against the defendant points solely to direct knowledge of the criminal

venture, it would be error to give the instruction.”). Allowing a deliberate ignorance

instruction premised on evidence of constructive knowledge reduces the standard for

conviction from knowledge to recklessness or negligence. See United States v.

Hilliard, 

, 1516 (10th Cir. 1994) (deliberate ignorance instruction is

improper when rationale for instruction is “too close to premising criminal liability

upon a reckless disregard for the truth or a negligent failure to inquire”). We are not

directed to any evidence in the record suggesting that Little deliberately avoided

knowledge of the firearms, and thus agree that the instruction was improper.

      Nevertheless, we conclude that the instruction was harmless. When

overwhelming evidence supports a finding of actual knowledge, “the tendering of a

‘willful blindness’ instruction is harmless beyond a reasonable doubt even when the

government does not introduce evidence to support such a theory.” United States v.

Sasser, 

, 1553 (10th Cir. 1992). As discussed in Section 

,

the evidence showing Little had actual knowledge of the weapons was substantial.

Because we are convinced beyond a reasonable doubt that the deliberate ignorance

instruction played no part in the jury’s verdict, we hold that the instruction was

harmless.




                                           14
                                              D

       Little argues that the district court erred in giving a “possible guilt of others”

instruction. The court provided the Tenth Circuit Criminal Pattern Jury Instruction

§ 1.19 (2011), explaining that “[t]he fact that another person also may be guilty is no

defense to a criminal charge” and that “[t]he question of the possible guilt of others

should not enter your thinking as you decide whether this defendant has been proved

guilty of the crime charged.”

       Little argues that the use notes to the pattern instruction suggest it should be

used only if a defendant requests an instruction that a person other than the defendant

could be guilty of the offense. Little did not request such an instruction. But the use

notes do not state that a guilt of others instruction is appropriate only in response to a

defendant’s request. See Tenth Circuit Criminal Pattern Jury Instruction § 1.19, use

note (2011). At trial, Little argued that someone else was involved in storing the

firearms on Blythe’s property. The district court properly instructed the jury that

involvement by third parties would not constitute an absolute defense. Moreover,

Little’s suggestion that this instruction amounted to directing the jury to ignore the

defense’s argument that someone else committed the crime is foreclosed by

precedent. See United States v. Oberle, 

, 1423 (10th Cir. 1998).7


       7
         Little also contends that even if the four instructions 

were not
individually misleading, they were misleading as a whole because they would allow a
guilty verdict even if someone else placed the firearms in Little’s home and Little had
only constructive knowledge of the weapons or ammunition. We agree that the
instructions erroneously failed to require intent to exercise control and that the deliberate
ignorance instruction was not warranted. However, as 

, these errors were
                                             15
                                            III

       The district court determined that Little’s total offense level was 28, due in

part to two prior convictions for crimes of violence. See U.S.S.G. § 4B1.2(a)

(defining “crime of violence”). Little had multiple prior convictions for “battery

upon a peace officer” in New Mexico. For a conviction under this charge, battery

must result in actual injury to the officer, represent a threat to the officer’s safety, or

present a meaningful challenge to the officer’s authority. State v. Padilla, 

, 493 (N.M. 1997).

       The district court stated that a New Mexico conviction for battery upon a

police officer meets the requirement for a crime of violence “in one of two ways.”

First, the court suggested that such convictions qualify under U.S.S.G. § 4B1.2(a)(1)

because the “actual injury” requirement means that the crime has as an element the

use of force. “In parallel,” the court observed that the “‘actual threat to safety, or

meaningful challenge to authority’ aligns with the Sentencing Guideline’s residual

clause.” See U.S.S.G. § 4B1.2(a)(2) (defining “crime of violence” as including

offenses that “otherwise involve[] conduct that presents a serious potential risk of

physical injury to another”). Thus, the district court concluded, any conviction for

the New Mexico offense would qualify as a crime of violence.

       Following Little’s sentencing, we held that under Johnson v. United States,

(2015), the Guidelines’ residual clause is unconstitutionally vague.

harmless given the overwhelming evidence that Little had actual knowledge and intent to
exercise control over the firearms. For the same reason, we reject his challenge to the
instructions as a whole.
                                            16
United States v. Madrid, 

, 1210 (10th Cir. 2015). This holding applies

to the matter before us, which was “pending on direct review” when Johnson and

Madrid were decided. 

. As the government concedes,

reliance on the residual clause constitutes plain error. We thus remand for

resentencing.

                                         IV

      For the foregoing reasons, Little’s convictions are AFFIRMED. We

VACATE Little’s sentence and REMAND for resentencing.




                                          17
No. 15-2019, United States v. Little

HOLMES, Circuit Judge, dissenting.

      I write separately because I dissent from the majority’s ultimate resolution

of Mr. Little’s challenge to the constructive-possession jury instruction. Though I

agree with the majority’s conclusion that this instruction is legally flawed

because it omits the element of intent to exercise control, I respectfully object to

the majority’s determination that the district court’s error in giving this

instruction was harmless. Consequently, while the majority would affirm the

district court’s judgment, I would reverse based on the constructive-possession

instructional error and remand for further proceedings.

      More specifically, I would decline to conduct a harmless-error analysis

because the government has not made a specific harmless-error argument

regarding the constructive-possession instructional error, and the conditions that

would ordinarily support sua sponte harmless-error review are not satisfied here.

Consequently, I would hold that the district court’s error stemming from its

allowance of the constructive-possession instruction is reversible error and that

the court’s judgment therefore cannot stand. I would not definitively opine on the

other issues in this case.

      Stated with greater particularity, I agree with the majority’s conclusion that

Henderson v. United States, --- U.S. ----, 

(2015), constitutes

intervening Supreme Court precedent that alters the elements of constructive

possession under 18 U.S.C. § 922(g) found in our circuit’s precedent: that is,
contrary to the Tenth Circuit decisions that preceded it, Henderson makes clear

that the government must establish that the defendant had an intent to exercise

control over the firearm (or other statutorily proscribed items), not just the power

to do so. See 

(“Constructive possession is

established when a person, though lacking . . . physical custody, still has the

power and intent to exercise control over the object.” (emphasis added)).

Because the constructive-possession instruction in this case did not include an

intent-to-exercise-control component, as Henderson requires, I agree with the

majority’s conclusion that the district court erred in submitting this instruction to

the jury and this error was of constitutional magnitude. See, e.g., Neder v. United

States, 

, 8 (1999) (noting that a district court errs when it gives “a jury

instruction that omits an element of the offense”); 

(“The Government

argues, correctly we think, that the absence of a ‘complete verdict’ on every

element of the offense . . . violates the Sixth Amendment’s jury trial guarantee.”).

      Having concluded that the constructive-possession instruction was fatally

defective, our court would customarily turn—as the majority does—to the

question of whether the error was harmless. See Fed. R. Crim. P. 52(a)

(providing that an error “that does not affect substantial rights must be

disregarded”); United States v. Wittgenstein, 

, 1169 (10th Cir.

1998) (“Having found that the district court misstated the arrest element of an 8

U.S.C. § 1326 offense, we must assess whether the instruction nevertheless

                                          2
constitutes harmless error.”); see also United States v. Blechman, 

,

1067 (10th Cir. 2011) (“We will not reverse a defendant’s conviction on the basis

of the district court’s erroneous admission of hearsay evidence if the error was

harmless to the defendant.”). However, ordinarily, “[t]he burden of proving the

error was harmless is on the Government.” United States v. Serawop, 

, 669 (10th Cir. 2005); see, e.g., United States v. Lott, 

, 723 (10th

Cir. 2006) (“Under harmless error review, the government has the burden of

proving beyond a reasonable doubt that the constitutional violation did not

contribute to the judgment.”); United States v. Summers, 

, 1303

(10th Cir. 2005) (“Under a harmless error analysis, the government bears the

burden of demonstrating that the error was harmless beyond a reasonable

doubt.”). But the government has failed to carry its harmless-error burden here.

      The government makes no specific harmless-error argument with respect to

the constructive-possession instructional error. I recognize that the government

offers some general and cursory assertions that appear to advance the contention

that our court should deem errors in any of the challenged instructions to be

harmless. See, e.g., Aplee.’s Br. at 15 (noting in its “Summary of the Argument”

section that, “read together, each of these [challenged] instructions, in the context

of the record as a whole, was not misleading, and even if unsupported resulted in

only harmless error”). However, these assertions are conclusory and skeletal

when applied to the specific circumstances of the jury’s consideration of the

                                          3
constructive-possession instruction; in my view, they do not constitute a

cognizable harmless-error argument regarding that instruction. See, e.g., United

States v. Faust, 

, 1248 n.4 (10th Cir. 2015) (“Mr. Faust, however,

does not elaborate on this bare assertion; thus, we may rightly deem any argument

that could be based on it to be waived.”); United States v. Pursley, 

,

1231 n.17, 1232 (10th Cir. 2009) (“Under our precedent, this skeletal reference is

insufficient to raise the ex parte/disclosure concern as a discrete appellate

issue.”); see also Bronson v. Swensen, 

, 1105 (10th Cir. 2007)

(“[T]hese cursory statements, without supporting analysis and case law, fail to

constitute the kind of briefing that is necessary to avoid application of the

forfeiture doctrine.”). Indeed, one could reasonably assert that the government

had no intention of presenting a harmless-error argument concerning the

constructive-possession instructional error because the government demonstrated

at other points in its brief that it knew how to articulate a harmless-error argument

when it wanted to do so, see, e.g., Aplee.’s Br. at 32 (devoting a section of its

brief to the harmlessness of any error associated with the willful-blindness

instruction), but it did not do so regarding the constructive-possession error.

      Thus, absent a harmless-error argument from the government regarding the

erroneous constructive-possession instruction, the question becomes whether it is

appropriate for our court to conduct a harmless-error analysis sua sponte. I

answer that question in the negative—viz., it is not appropriate. “[T]his court

                                           4
may in its discretion ‘initiate harmless error review in an appropriate case.’”

United States v. Samaniego, 

, 1224 (10th Cir. 1999) (emphasis

added) (quoting United States v. Torrez–Ortega, 

, 1136 (10th Cir.

1999)); see also United States v. Vanderwerff, 

, 1279 (10th Cir.

2015) (noting that “we have on occasion engaged in harmless-error

analysis—either at the request of a party or, ‘in an appropriate case,’ sua sponte”

(quoting 

)). However, we “should . . . be hesitant to

engage in an ‘unsolicited, unassisted, and undirected harmless error review.’”

United States v. Holly, 

, 1308 (10th Cir. 2007) (quoting 

); cf. United States v. Doe, 

, 1184 (10th Cir. 2009)

(Holmes, J., dissenting) (“The government has failed to argue that the inadequacy

of the information [i.e., the charging document] was harmless error. While we

may, in certain limited circumstances, conduct a sua sponte harmless error

analysis, it would not be appropriate to do so here.”); United States v. Pryce, 

, 1347 (D.C. Cir. 1991) (Williams, J., announcing judgment of the

panel) (“Where a court analyzes the harmless error issue wholly on its own

initiative, it assumes burdens normally shouldered by government and defense

counsel. This drain on judicial resources inevitably causes delay for parties in

other cases.”).

      “In considering whether to do so, we have ‘cited with approval three

factors suggested by the Seventh Circuit in determining whether an appellate

                                          5
court should address harmlessness when the government has failed to do

so . . . .’” Mollett v. Mullin, 

, 920 (10th Cir. 2003) (quoting

); see United States v. Giovannetti, 

,

227 (7th Cir. 1991). That is, we look to “(1) the length and complexity of the

record; (2) whether the harmlessness of the errors is certain or debatable; and

(3) whether a reversal would result in protracted, costly, and futile proceedings.”

(quoting 

). “However,

‘confusion about what the third factor contributes to the analysis has caused this

and other courts to merely reference [the third factor] but not apply it.’” 

original) (quoting 

n.2).

      The most important of these factors is the second one—i.e., the certainty of

harmlessness. See 

(noting that “it may be appropriate to

[apply harmless-error review] where the certainty of the harmlessness is readily

apparent”). This factor can be dispositive under certain circumstances. See

United States v. Gonzalez-Flores, 

, 1101 (9th Cir. 2005) (“[T]he

second factor—the court’s certainty as to the harmlessness of the error—is of

particular importance.”); 

(Holmes, J. dissenting)

(analyzing only the second factor, i.e., the certainty of harmlessness, because

“[t]he most important of the[ ] factors is the certainty of the result”); see also

United States v. Rodriguez-Preciado, 

, 1143 (9th Cir. 2005)

(Berzon, J., dissenting in part) (“[T]he touchstone of whether courts should reach

                                           6
harmless error sua sponte is the extent to which the harmlessness of the error is

open to question.”); cf. Gover v. Perry, 

, 301 (6th Cir. 2012) (noting

that “the adoption of [the Seventh Circuit’s test] does not mandate that a circuit

court review the record before it for harmlessness in every instance, only that it

has discretion to do so when the error is harmless upon review of a clear record”

(emphasis added)). “Evaluation of the certainty of the harmlessness necessarily

requires this court to review the record to some extent, though not to the same

degree as would be required pursuant to a full harmless error review.” 

.

      Having reviewed the record, I do not believe that the conditions for

conducting a harmless-error review sua sponte are satisfied here. In particular,

the harmlessness of the error stemming from the constructive-possession

instruction is far from certain or readily apparent; at a minimum, harmlessness

could be vigorously debated here. See 

; 

. In other words, the most important condition—the second one—is not

satisfied. And, at least under the circumstances here—where I discern no other

variables that could materially alter the discretionary calculus, cf. United States v.

Rose, 

, 1415 (1st Cir. 1997) (“While we find helpful the reasoning

of the Seventh Circuit, we do not restrict ourselves to [its] test. The exercise of

discretion involves the balancing of many elements.” (citation omitted))—that

conclusion dooms the argument for sua sponte harmless-error review.

                                           7
      More specifically, there is little evidence in the record that addresses Mr.

Little’s intent to exercise control over the firearms and ammunition at issue. As I

see it, the only evidence that meaningfully suggests Mr. Little’s intent to exercise

control was Ms. Blythe’s testimony that Mr. Little put a lock on the door to the

well house about a week before the police arrived to conduct the search—that is,

shortly after the firearms were taken from Southwest Arms—and that he did not

give her a key to the lock. See R., Vol. VII, at 189 (Test. of Ms. Blythe)

(answering “Cody” Little to the question “Who put the lock on that [well house]

door?,” and stating that he did so “[a]bout a week before the cops showed up”);

(responding “No” when asked “Did you have a key?”); cf. United States

v. Piwowar, 

, 955 (8th Cir. 2007) (discerning an intent to exercise

control when there was “evidence [defendant] owned the building where the

refrigerator [containing firearms] was located and possessed the sole key to the

refrigerator”).

      However, the testimony of two law-enforcement officers failed to solidly

confirm that the lock was on the door at the time of the search. See R., Vol. VII,

at 62 (Test. of Agent Budrow) (averring “I don’t totally recall” a lock on the well

house at the time of the search, but “I believe there was”); 

(Test. of Lt.

Brackeen) (“I don’t recall if that lock was there at that time [of the search] or

not.”). More importantly, even if there was a lock on the door at the time of the

search, the jury heard testimony indicating that Mr. Little did not use it at all

                                           8
times. Specifically, law-enforcement testimony indicated that the well house door

was unlocked when the officers executed the search. See 

(Test. of Agent

Budrow) (averring that the door to the well house “was not locked” at the time of

the search); 

(Test. of Sgt. Clark) (testifying that the door to the well

house “was unlocked” when the firearms were found). Such testimony could have

diluted in the eyes of a reasonable jury the probative force of the evidence

regarding Mr. Little putting a lock on the door because it could have suggested to

the jury that Mr. Little was not especially concerned about others—instead of

himself—exercising control over the firearms and ammunition.

      At bottom, what is clear is that there was scant evidence before the jury

that addressed Mr. Little’s alleged intent to exercise control over the firearms and

ammunition at issue. Of course, that is not really surprising since the government

had no reason to believe that Mr. Little’s intent to exercise control would be

relevant to this case. And, the relevant evidence that was presented (e.g., the lock

evidence), hardly put the question of Mr. Little’s intent to exercise control beyond

debate. Consequently, I cannot conclude that the harmlessness of the

constructive-possession instructional error here was certain or readily apparent;

therefore, the important second factor does not support sua sponte harmless-error

review. And, at least under the circumstances here, this conclusion is dispositive.

That is, I need not go beyond this second factor. I conclude that it is

inappropriate to conduct harmless-error review sua sponte of the constructive-

                                           9
possession instructional error.

      To be sure, after seemingly overlooking the government’s failure to argue

harmless error, the majority concludes that the constructive-possession

instructional error was harmless, primarily by relying on Mr. Little’s supposed

exclusive control of the well house. However, the majority’s harmless-error

analysis does not alter my view that the resolution of the harmlessness question is

at least debatable and far from certain or readily apparent.

      As the majority candidly acknowledges, the cases on which it relies only

hold that constructive possession may be inferred from a defendant’s exclusive

control of the premises in which the firearms are found; the cases do not indicate

that a defendant’s exclusive control of the premises necessitates an inference of

constructive possession. See United States v. Hishaw, 

, 571 (10th

Cir. 2000) (holding that “the defendant’s dominion, control, and knowledge may

be inferred if he had exclusive possession of the premises on which the object was

found” (emphasis added)); see also United States v. Griffin, 

, 695

(7th Cir. 2012) (noting that “if the government demonstrates that the defendant

had ‘exclusive control’ over the property where the contraband was discovered, a

jury may reasonably infer that he constructively possessed the items, including

the contraband, found on that property” (emphasis added)); United States v.

Alanis, 

, 592 (7th Cir. 2001) (holding that “constructive possession



                                         10
may be established by a showing that the firearm was seized at the defendant’s

residence” (emphasis added) (quoting United States v. Walls, 

, 867

(7th Cir. 2000))). Moreover, though the majority relies on the “particular facts of

this case” to support its conclusion that “there is no reasonable possibility that the

jury would have found that Little had knowledge of the weapons at issue but

lacked intent to exercise control over them,” Maj. Op. at 10, the majority does not

identify those particular facts. And such facts are not readily apparent to me.

      Indeed, the principal pillar of the majority’s harmlessness conclusion—Mr.

Little’s purported exclusive control of the well house—is far from rock solid.

Though Ms. Blythe indicated that she had not been in the well house since “about

a week” after Mr. Little’s mid-September move-in date, R., Vol. VII, at 184, 190,

she acknowledged that she still stored an “ice chest and buckets” there, 

Moreover, though Ms. Blythe testified that Mr. Little put a lock on the door about

a week before the search (that is, after the robbery), and did not give her a key, as

noted above, there are two weaknesses associated with this testimony: first, there

was no evidence to solidly confirm it; and, second, even if her testimony was

true, there was evidence indicating that Mr. Little did not use that lock at all

times. Furthermore, Ms. Blythe could not rule out the possibility that “people

came over to [her] home” when she was away. 

Accordingly, the

evidence of Mr. Little’s alleged exclusive control over the well house is not

sufficiently firm for me to conclude that it is certain or readily apparent that a

                                          11
reasonable jury would have inferred from it that Mr. Little possessed an intent to

exercise control over the firearms and ammunition in the well house.

      I need not (and thus do not) definitively opine on whether the government

could have carried its burden of proving that the district court’s constructive-

possession instructional error was harmless. The government did not even get out

of the starting blocks on the issue; in other words, it did not even attempt to carry

its harmless-error burden. Guided by our precedent—including Samaniego,

Holly, and Mollett—I have inquired whether it is appropriate for our court to

conduct harmless-error review sua sponte. It is not. That is because the

harmlessness of the constructive-possession instructional error is at least

debatable, and far from certain or readily apparent. Therefore, I would conclude

that the district court’s omission from its constructive-possession instruction of

the element of intent to exercise control is reversible error. Accordingly, I would

reverse and remand the case for further proceedings. 1




      1
             As noted, because I would reverse based on the constructive-
possession instructional error, I have no need to definitively opine on Mr. Little’s
other appellate challenges. However, I pause to observe that, as I read its
opinion, the majority is remanding based on Johnson v. United States, 

(2015), because—like the government—the majority cannot “uncouple the
ruling,” Aplee.’s Br. at 38, of the district court and determine whether the district
court relied on the residual clause because it “did not specify whether [Mr.
Little’s convictions] were [crimes of violence] because of the elements of the
crime, or because of the residual clause,”