BERNARD FLEMING v. UNITED STATES

District of Columbia Court of Appeals No. 14-CF-1074 NOV 10 2016 BERNARD FLEMING, Appellant,

v. CF1-1328-14

UNITED STATES, Appellee.

On Appeal from the Superior Court of the District of Columbia Criminal Division

BEFORE: GLICKMAN, FISHER, and EASTERLY, Associate Judges.

JUDGMENT

This case came to be heard on the transcript of record and the briefs filed, and was argued by counsel. On consideration whereof, and as set forth in the opinion filed this date, it is now hereby

ORDERED and ADJUDGED that appellant’s convictions for second- degree murder and other crimes are affirmed. The case is remanded for the trial court to vacate two of appellant’s three convictions of possession of a firearm during a crime of violence (“PFCV”) on merger grounds.

For the Court: Dated: November 10, 2016. Opinion by Associate Judge Stephen H. Glickman.

Opinion by Associate Judge Catharine Easterly, concurring in the judgment and joining in Parts I, III, and IV of the opinion for the court.

Notice: This opinion is subject to formal revision before publication in the Atlantic and Maryland Reporters. Users are requested to notify the Clerk of the Court of any formal errors so that corrections may be made before the bound volumes go to press.

DISTRICT OF COLUMBIA COURT OF APPEALS

No. 14-CF-1074 11/10/16

BERNARD FLEMING, APPELLANT,

V.

UNITED STATES, APPELLEE.

Appeal from the Superior Court of the District of Columbia (CF1-1328-14)

(Hon. Robert E. Morin, Trial Judge) (Argued April 5, 2016 Decided November 10, 2016)

Peter H. Meyers for appellant.

Nicholas P. Coleman, Assistant United States Attorney, with whom Channing D. Phillips, United States Attorney, and Elizabeth Trosman, Suzanne Grealy Curt, S. Vinet Bryant, and Kathryn L. Rakoczy, Assistant United States Attorneys, were on the brief, for appellee.

Before GLICKMAN, FISHER, and EASTERLY, Associate Judges.

Opinion for the court by Associate Judge GLICKMAN.

Opinion by Associate Judge EASTERLY, concurring in the judgment and joining in Parts I, III, and IV of the opinion for the court, at page 21.

GLICKMAN, Associate Judge: In Roy v. United States,1 this court approved a jury instruction in a murder prosecution on the “gun battle” (or “urban gun battle”) theory of causation. The instruction permitted the jury to find that the defendant, by engaging in a gun battle in a public space, was responsible for causing the death of an innocent bystander killed by a stray bullet even if it was not the defendant who fired the fatal round. For the jury to come to that conclusion, the approved instruction required it to find, inter alia, that the bystander’s death was a “reasonably foreseeable” consequence of the defendant’s participation in the gun battle.2

The trial court gave a similar gun battle causation instruction in the present case, and the jury found appellant guilty of second-degree murder while armed.3 But whereas the victim in Roy was an innocent bystander, in this case it was one of

1 871 A.2d 498 (D.C. 2005). 2 Id. at 506-7 n.8; see also id. at 509 (explaining that “[w]hile the evidence was unclear” as to which combatant fired the fatal shot, “such a determination is unnecessary if both men prepared for and undertook to participate in the gun battle where it was clearly foreseeable that others would be endangered”). 3 The jury also convicted appellant of two counts of assault with intent to kill while armed, one count of carrying a pistol outside the home or place of business, and three counts of possession of a firearm during a crime of violence (“PFCV”). All those charges arose from the same incident as the second-degree murder charge, but the gun battle causation instruction did not implicate them.

3 appellant’s antagonists in the shootout who was killed. The prosecution requested the instruction because evidence suggested that the fatal shot may have been fired not by appellant or one of his confederates, but by a confederate of the decedent. Appellant, who objected at trial, argues that a gun battle causation instruction is improper where the decedent was a participant in the battle rather than, as in Roy and other cases in which this court previously has encountered the instruction, 4 an innocent bystander. We disagree, however, and conclude that the trial court did not err in giving the gun battle instruction in this case.

Appellant also seeks a new trial on grounds of prosecutorial misconduct, but we conclude that the record does not support his claims of impropriety. We therefore uphold appellant’s convictions. However, as the government concedes, his three PFCV convictions merge, so we direct that two of those counts be vacated on remand.5

4 See McCray v. United States, 133 A.3d 205, 223-26 (D.C. 2016); Blaine v. United States, 18 A.3d 766, 768 (D.C. 2011). 5 See Matthews v. United States, 892 A.2d 1100, 1106 (D.C. 2006) (“[M]ultiple PFCV convictions will merge . . . if they arise out of a defendant’s uninterrupted possession of a single weapon during a single act of violence.”) (citing Nixon v. United States, 730 A.2d 145, 153 (D.C. 1999)).

I.

Appellant was tried for the murder of Michael Jones in a gunfight that erupted on the night of July 7, 2012. The shooting was the culmination of events that began with a hostile confrontation earlier that evening between appellant and Michael Jones’s brother, Maurice Jones.

Maurice testified that at around 7:00 p.m. on July 7, he left his apartment at Eighth and R Street Northwest to walk to a nearby store. On the way there, Maurice encountered appellant, who was with two other men. Appellant taunted Maurice and struck him on the chin. Being outnumbered, Maurice retreated to his apartment.

About half an hour later, Maurice and his girlfriend, Kendra Wingate, heard banging on his front door and several voices outside. They ignored the banging and did not open the door. After the banging stopped, Maurice looked out and saw appellant waiting with two companions named Joseph Peoples and Rakeem McMillan. In order to confront them on even terms, Maurice phoned his brother Michael and a friend named Eric Cunningham and asked them to come to his apartment. While he waited for them, Maurice looked outside from time to time

5 and observed appellant, Peoples, and McMillan gesture for him to come out. After a while, appellant and his companions departed.

Not long afterward, Michael Jones and Eric Cunningham arrived at Maurice’s apartment, together with a friend of Michael’s named James Hamlin. The four men then left on foot to look for appellant. About a block away, Maurice spotted Joseph Peoples rapidly descending an exterior stairway on the apartment building at 1730 Seventh Street known as Lincoln Tower. Ignoring Maurice’s hail, Peoples crossed Seventh Street to join Rakeem McMillan in front of a church. Moments later, according to Maurice, Peoples turned and began shooting at him and his three companions as they arrived at Lincoln Tower.

Michael drew a gun and fired back at Peoples before he was killed by a bullet in the head. Although Maurice testified that no one else in his group was armed, the parties at trial stipulated that Hamlin also fired a gun in response to the attack. The stipulation was corroborated by shell casings from two different weapons found in the vicinity of Michael’s body. This, in conjunction with inconclusive forensic evidence regarding Michael’s wound and the bullet recovered from his body, raised the possibility that Michael was killed by “friendly fire” from Hamlin.

Other shell casings recovered by police at the scene indicated that shots also were fired from a second-floor balcony of Lincoln Tower overlooking Seventh Street. Video surveillance footage obtained by the police from inside Lincoln Tower showed appellant on that balcony during the shooting.6 The surveillance footage also appeared to show appellant retrieving what could have been a weapon from inside the building and bringing it to the balcony just before the shooting started. Footage from just after the shooting showed appellant hurrying to the sixth floor and rendezvousing with Peoples, and Peoples receiving something from appellant that he then stashed in a stairwell. The police recovered firearms from that location. Relying on this evidence, the government contended at trial that appellant armed himself after seeing Maurice and company arrive at Lincoln Tower, shot at them from the balcony, and then, after Michael was down and the battle ended, handed his gun to Peoples, who hid it in the stairwell.

Both appellant and Peoples were arrested later that night. They were indicted on one count of first-degree murder while armed, three counts of assault

6 Maurice, who fled after sustaining a gunshot wound in his chest, did not see appellant during the gunfight.

7 with intent to kill while armed, and related charges. The two were tried together.7 At the close of the prosecution’s case-in-chief, the trial court granted the government’s motion to dismiss the first-degree murder counts and to proceed instead on the lesser-included offense of second-degree murder while armed. While the jury found appellant guilty of that offense (along with two counts of assault with intent to kill while armed and the related weapons offenses), it acquitted Peoples of the homicide.8

II.

Appellant argues that the trial court erred as a matter of law by giving a gun battle causation instruction where the victim was not a mere bystander but rather was an active participant in the battle. Our review of this contention is de novo, requiring us to determine whether the instruction is a correct and adequate statement of the law.9

7 Hamlin also was named in the indictment, but his case was severed before trial. 8 Peoples was found guilty only of carrying a pistol outside the home or business and of tampering with physical evidence. 9 See Brown v. United States, 139 A.3d 870, 875 (D.C. 2016) (“While we review for abuse of discretion a trial court’s assessment of whether a jury instruction is supported by the evidence, see Wheeler v. United States, 930 A.2d (continued…)

The gun battle causation instruction this court approved in Roy reflected the fact that the murder victim in that case was a bystander; it required the jury to determine, inter alia, whether “it was reasonably foreseeable that death or serious bodily injury to innocent bystanders could occur as a result of the defendant’s conduct” in engaging in a gun battle.10 In this case, where the victim was not a bystander, the trial court omitted the reference to innocent bystanders and required the jury to determine whether “it was reasonably foreseeable that death or serious bodily injury [] could occur as a result of the defendant’s conduct during the gun battle.”11 The instruction was given to clarify that appellant could be liable for

(continued…) 232, 238 (D.C. 2007), we review de novo the content of the instructions actually given, see Wilson-Bey v. United States, 903 A.2d 818, 827 (D.C. 2006) (en banc)[.]”); see also Roy, 871 A.2d at 507. 10 Roy, 871 A.2d at 507 n.8 (emphasis added). 11 The full gun battle causation instruction in this case was as follows:

An element of second degree murder is that the defendant caused the death of Michael Jones. A person causes the death of another person if his conduct is a substantial factor in bringing about the death and if it was reasonably foreseeable that the death or serious bodily injury could result from such conduct. It is not necessary for the government to prove that the defendant personally fired the fatal round in this case. Rather, if the government proves beyond a reasonable doubt: Number 1, the (continued…)

9 killing Michael Jones even if, as evidence suggested, the fatal shot might have been fired by Hamlin rather than by appellant or his accomplice.

We conclude that the trial court’s modified gun battle causation instruction was legally sound. As stated in Roy, “[i]n this jurisdiction we have held findings of homicide liability permissible where: (1) a defendant’s actions contribute substantially to or are a substantial factor in a fatal injury . . . and (2) the death is a reasonably foreseeable consequence of the defendant’s actions.”12 The gun battle instruction applies these principles of proximate causation to the realities of urban warfare, in which the antagonists are collectively responsible for creating a zone of great danger – including a substantial likelihood that errant bullets will hit

(continued…) defendant was armed and prepared to engage in a gun battle. Number 2, he did, in fact, engage in a gun battle on July 7th, 2012 at approximately 10:30 p.m. Number 3, the defendant’s conduct . . . was a substantial factor in the death of Michael Jones. Number 4, it was reasonably foreseeable that death or serious bodily injury . . . could occur as a result of the defendant’s conduct during the gun battle. And, Number 5, the defendant did not act in self-defense. If these circumstances are proved beyond a reasonable doubt, the defendant is deemed to have caused the death of Michael Jones. 12 Roy, 871 A.2d at 507-08 (internal citations omitted).

10 unintended targets.13 It is true that Roy focused on the “highly increased risk to noncombatants” in particular and reasoned that “[i]t is this increased risk to innocent bystanders which justifies the application of proximate cause liability to those participants who willfully choose to engage in these battles.”14 But the combatants themselves are obviously at great risk as well, and it surely is no less foreseeable that a stray bullet fired in the heat of a gun battle will unintentionally hit an ally or confederate in the fight rather than someone outside it. For the purposes of applying principles of proximate causation, it thus makes no meaningful difference whether the reasonably foreseeable victim of a shootout was a participant in the battle or a bystander. In either case, all those who intentionally carried on the gun battle and thereby shared in creating the danger may be found to have substantially contributed to the ensuing fatality.

The evidence at trial in this case permitted the jury to find that appellant proximately caused the death of Michael Jones, either by shooting Jones himself or

13 See id. at 507; see also id. at 511 (Glickman, J., concurring in part and dissenting in part) (“Where two or more persons ‘voluntarily and jointly created a zone of danger,’ it is fair to hold each one ‘responsible for his own acts and the acts of the others’ that ensued.”) (quoting People v. Russell, 693 N.E.2d 193, 195 (N.Y. 1998)). 14 Id. at 507.

11 by instigating and participating in a shootout in which one of the other combatants shot Jones, intentionally or otherwise. Accordingly, we hold that the trial court did not err in giving the gun battle causation instruction.

III.

Appellant also claims that he was denied a fair trial as a result of “repeated prosecutorial improprieties” to which he made timely objection. He identifies the improprieties as (1) asking a witness if she was scared to testify; (2) vouching for Maurice Jones’s credibility and otherwise expressing personal opinions in closing and rebuttal argument; and (3) submitting a new exhibit after the trial concluded. Appellant argues that even though the trial court took corrective action in response to the prosecutor’s questions and comments, the court’s remedies were inadequate to dispel the cumulative prejudice to his defense.

In reviewing allegations of prosecutorial misconduct at trial, we begin by determining whether the challenged conduct actually was improper.15 If it was not,

15 See, e.g., Washington v. United States, 884 A.2d 1080, 1088 (D.C. 2005).

12 that ends our inquiry, and we have no occasion to go on to evaluate the trial court’s response and whether there was reversible error.16

In the present case, we do not reach the second step in the analysis, because we reject appellant’s claims of impropriety.

A. “Are you scared to be here today?”

The first alleged impropriety occurred during the direct examination of Kendra Wingate, who was with Maurice Jones in his apartment on the evening of July 7. Appellant charges that “without explanation or factual basis, the prosecutor asked Wingate, ‘Are you scared to be here today?’”17 Appellant argues that this question was improper and highly prejudicial, and that although the trial court

16 If we find that the challenged conduct was improper and that timely objection to it was made, we must determine whether the trial court erred in overruling or otherwise responding to the objection and, if so, whether the error was harmless. “In making that determination, we consider the gravity of the impropriety, its relationship to the issue of guilt, the effect of any corrective action by the trial judge, and the strength of the government’s case.” Id. (internal quotation marks omitted). If there was no timely objection, on the other hand, then our review is only for plain error. Id. 17 Br. for Appellant at 15.

13 sustained his objection, the damage had been done – “government counsel had succeeded in raising the concern about the witness’s fear of testifying.”18

We have recognized that questions to witnesses about their fear of testifying may be prejudicial because “(1) they suggest to the jury a decision based on guilt by association; and (2) the evidence plays on the passions and fear of the jury, by suggesting that a threat exists against the witnesses.”19 “If the trial court admits evidence of threats solely to go to the general credibility or bias of the witness, such admission has been held to be an abuse of discretion.”20 But we have recognized that there are situations in which such inquiries are probative and permissible. One such situation is where the inquiry may help “to explain the specific behavior of a witness, such as inconsistent statements, delay in testifying, or unusual courtroom demeanor.”21 We have held that inquiries into a witness’s fear are not improper where they help to explain prior inconsistent testimony, the

witness’s lack of memory about prior testimony, or her general reluctance to testify 18 Id. 19 Ebron v. United States, 838 A.2d 1140, 1149 (D.C. 2003) (citation and internal quotation marks omitted); see also Gordon v. United States, 783 A.2d 575, 586 (D.C. 2001). 20 Mercer v. United States, 724 A.2d 1176, 1184 (D.C. 1999). 21 Gordon, 783 A.2d at 586 (quoting Mercer, 724 A.2d at 1184).

14 – especially where there are not less prejudicial means by which to explain the witness’s behavior.22 In this case, Wingate’s testimony and demeanor on the stand justified the inquiry.

Before asking Wingate whether she was scared to testify, the prosecutor asked her a series of questions about the events of the evening of July 7. When asked about the banging and voices she heard at the door, Wingate repeatedly claimed that she could not remember details such as how loud the voices were or whether they were male or female. The prosecutor followed up, referencing her professed inability to remember such details during prior interviews with investigators and asking whether it was “true that [Wingate] did not remember . . . or was it that [she] did not want to be here testifying?” Wingate first responded that she did not remember, but then changed course and stated that she “did not want to testify” but refused to say why not. Only after repeated attempts to cajole Wingate into explaining her contradictory testimony and her unwillingness to testify did government counsel ask whether she was “scared.” The court sustained a defense objection and appellant sought no additional remedy.

22 See Mercer, 724 A.2d at 1187-90.

The government’s question was asked to ascertain the reason for the witness’s reluctance and refusal to give truthful, relevant testimony. Wingate had provided inconsistent explanations for her inability to recall salient facts about the evening of July 7, and even when pressed, she was unwilling to explain why she did not wish to testify. There is no indication that there existed any less prejudicial means of explaining Wingate’s recalcitrance. Because the government’s question was “meant to explain specific behavior of the witness while testifying,” we conclude that, while the court did not abuse its discretion in sustaining the objection to it, the inquiry was not improper.23

B. Vouching for a Witness’s Credibility and Expressing Personal Opinions as to the Facts.

Appellant further contends that in closing and rebuttal argument, the prosecutor improperly vouched for the veracity of the government’s key witness, Maurice Jones, and expressed personal opinions regarding the facts of the case. In particular, appellant complains that the prosecutor asked rhetorical questions

23 See Mercer, 724 A.2d at 1190. We note also that appellant did not request further relief, such as a specific curative or limiting instruction, after the trial court sustained his objection to the prosecutor’s question. The trial court did not plainly err by not taking additional corrective action sua sponte to preclude any possibility of prejudice. See Allen v. United States, 649 A.2d 548, 555 (D.C. 1994).

16 implying belief in Jones’s credibility – e.g., “[d]o we really think he’s naming the wrong folks?” and “[d]o we really think he’s just making names up?” – and asserted that “[w]hat he’s telling you is what he can absolutely remember. He doesn’t embellish it.” Appellant also complains that, in discussing the evidence at trial, the prosecutor conveyed her personal opinions by asserting what “we know” and what was (or was not) “true” – as, for example, when the prosecutor stated that “[w]e know someone was firing” from the second floor balcony.24

Prosecutorial vouching and other expressions of personal opinion are improper because they

can convey the impression that evidence not presented to the jury, but known to the prosecutor, supports the charges against the defendant and can thus jeopardize the defendant’s right to be tried solely on the basis of the evidence presented to the jury; and the prosecutor’s opinion carries with it the imprimatur of the Government

24 Although the trial court viewed the prosecutor’s remarks as “rhetorical flourishes” rather than statements of personal opinion, it granted appellant’s request for a curative instruction. After the prosecutor’s closing and rebuttal, the court admonished the jurors to disregard anything they perceived to be the expression of personal opinion by counsel and to decide the case on the evidence alone.

and may induce the jury to trust the Government’s judgment rather than its own view of the evidence.[25]

Given this rationale, the condemnation of personal opinion requires us to focus carefully on the nature of the prosecutor’s remarks. Like other advocates, prosecutors may and should marshal the evidence properly admitted at trial and fairly argue its legitimate significance.26 That goes for argument, grounded in the evidence, that the jury should (or should not) credit a witness’s testimony. “[T]he key inquiry is whether the attorney is commenting on the evidence, which he may do, or expressing a personal opinion, which is taboo. A comment will be within the acceptable range as long as it is in the general nature of argument, and not an outright expression of opinion.”27

Our examination of the record in this case satisfies us that government counsel’s challenged remarks were fair comments on the evidence admitted at trial, not improper expressions of personal opinion. Her rhetorical questions and

25 United States v. Young, 470 U.S. 1, 18 (1985); see also Mathis v. United States, 513 A.2d 1344, 1348 (D.C. 1986). 26 See Dixon v. United States, 565 A.2d 72, 77 (D.C. 1989) (“The proper exercise of closing argument is to review the evidence and to explicate those inferences which may reasonably be drawn from the evidence.”) (quoting Bertolotti v. State, 476 So.2d 130, 134 (Fla. 1985)). 27 Irick v. United States, 565 A.2d 26, 35 (D.C. 1989) (italics in the original).

18 remarks regarding Maurice Jones’s testimony did not constitute personal vouching for his credibility; they were asked in the context of argument drawing reasonable inferences as to why Maurice, whose credibility the defense had attacked, might not have remembered certain aspects of the gun battle. The comments did not suggest that the prosecutor had knowledge outside the evidence admitted at trial or invite the jury to trust the government’s judgment.28 Similarly, whenever the prosecutor told the jury what “we know” or what was “true,” she referred to the evidence that made it so. For instance, when the prosecutor stated that “[w]e know someone was firing” from the second floor balcony, she completed the sentence by explaining that it was “because there were shell casings left behind on that balcony – not one, not two, but three right on the balcony – and several more that matched it down below which is evidence of the fact that someone was standing there firing.” There was never “any suggestion that the prosecutor was in effect adding [her] own unsworn testimony to the evidence in the record.”29 We conclude that the comments challenged by appellant were not improper.

28 Cf. Mitchell v. United States, 569 A.2d 177, 184 (D.C. 1990) (holding that prosecutor’s characterization of two government witnesses as “‘truthtelling individual[s]’ [was] very close to the personal expression that this court has condemned,” but “was a fair response” with respect to the witness whose credibility had been attacked by defense counsel). 29 Irick, 565 A.2d at 36 (footnote omitted). Likewise, when government counsel argued what was “true” or not, she did so by reminding the jury of what (continued…)

C. The Tardy Offer of a New Government Exhibit.

Lastly, appellant argues that the government improperly waited until after closing arguments were concluded to move an important new exhibit into evidence. This exhibit was a compilation on one disc of the several surveillance video clips that individually had been admitted into evidence during the trial. The clips in the compilation were not altered and nothing new was added, but appellant objected that he was unfairly surprised and should have been given the opportunity to challenge the compilation. The trial court, finding nothing unfair in the admission of a single disc that merely compiled the unaltered video evidence already admitted, overruled the objection.

We agree with the trial court that there was no unfair surprise. While the compilation disc should have been moved into evidence earlier – like other exhibits, prior to the close of evidence, as is the normal procedure – it is an exaggeration to characterize the government’s tardy motion as misconduct. There is no indication that the prosecutor deliberately withheld the exhibit to gain a

(continued…) evidence it had (or had not) seen, in order to rebut parts of the defense closing arguments.

20 tactical advantage. We discern nothing improper or prejudicial about the late admission of a disc merely compiling in a convenient format the video evidence that was already properly before the jury. Appellant had a full and fair opportunity to challenge the admission of each video clip when it was offered during trial, to explore and elucidate its probative worth, and to assess and interpret it in closing argument. Even if appellant is right that the chronologically ordered compilation of the separate videos on one disc “provide[d] a narrative supportive of the government’s case,”30 there was nothing misleading or otherwise unfair in that. As the trial court observed, the compilation was akin to a notebook of exhibits arranged in sequence, a completely appropriate practice. We reject appellant’s claim of prosecutorial misconduct.

IV.

For the foregoing reasons, we affirm appellant’s convictions for second-degree murder while armed and other crimes, and we remand the case for the trial court to vacate two of appellant’s three PFCV convictions on merger grounds.

So ordered.

30 Br. for Appellant at 16.

EASTERLY, Associate Judge, concurring in the judgment and joining in Parts I, III, and IV: At the time Michael Jones was shot and killed, there were people around him shooting guns at each other.1 Bernard J. Fleming may have been one of the individuals shooting at Mr. Jones and his companions. No witness ever saw and no videotape ever showed Mr. Fleming shooting a gun, and he was never found in possession of a weapon. He was, however, seen with another man up on a second-floor balcony where police found bullet casings and a bullet; he was also seen before the shooting bringing an object—possibly a gun—to the balcony and removing an object when the shooting was over; and he could have hidden two guns later discovered by the police under a stairwell in the building, one of which might have been used in the shooting. But even if Mr. Fleming possessed a gun and was one of the persons shooting at Mr. Jones and his companions, no evidence connected any shot he might have fired to the shot that killed Mr. Jones. The bullet retrieved from Mr. Jones’ body was so fragmented that it could not provide

1 The record indicates that three people were seen with guns that evening: the decedent, Michael Jones; his friend, James Hamlin; and Mr. Fleming’s co- defendant, Joseph Peoples. Forensic analysis of all the gun casings and bullets showed that three distinct weapons were used in the exchange of gunfire (a .40 caliber Glock-type firearm belonging to Mr. Hamlin; a “Hi Point” pistol belonging to the decedent; and another .40 caliber Glock-type firearm fired from the sidewalk, though it may have been fired from the balcony).

22 an evidentiary link to the shooter. Moreover, evidence showed that Mr. Jones was shot in the back of the head, making it possible that he was shot by Mr. Hamlin.

Nonetheless, Mr. Fleming was held accountable for Mr. Jones’ death.2 He was convicted of second-degree murder while armed based on the “urban gun battle” theory of causation that this court announced in Roy v. United States, 871 A.2d 498 (D.C. 2005), i.e., the proposition that a defendant who exchanges gunfire with another individual “proximate[ly] cause[s]” any death that results, whether or not the defendant fired the fatal shot, id. at 506. I assume the division is bound by this court’s decision in Roy to uphold Mr. Fleming’s conviction.3 But I cannot vote to affirm without commenting on the dangerous incoherence of Roy’s “causation” analysis. In truth, it dispenses with causation altogether, and, by thus reducing the government’s burden of proof for murder, captures in its net individuals who can only be proved to have committed far less serious crimes. Roy was wrongly decided. Moreover, I cannot sign on to the division’s opinion applying Roy, which

effectively expands its illogic. I write separately to explain the flaws with Roy’s 2 Mr. Peoples was also charged with second-degree murder but was found not guilty. Mr. Hamlin was only charged with and convicted of weapons offenses. 3 See M.A.P. v. Ryan, 285 A.2d 310, 312 (D.C. 1971). Roy may have been in conflict with other decisions of this court at the time it was decided. See infra note 9. But given subsequent decisions applying the urban gun battle theory, see infra note 24, I am unsure a division of this court could overturn Roy.

23 causation reasoning and to call for en banc review to reject Roy’s urban gun battle theory of guilt for murder.

I. The Causation Problem in Roy

It is a “basic postulate of our criminal law” that we impose liability when “a free agent [is] confronted with a choice between doing right and doing wrong and choos[es] freely to do wrong.” Carter v. United States, 252 F.2d 608, 616 (D.C. Cir. 1957).4 We thus rely on principles of causation and complicity to define criminal culpability: we punish people criminally for the harms they themselves have caused5 or the crimes they have conspired with or helped others to commit.6

4 Our conception in criminal law of the individual as a rational actor, in control of his own decisions, is apparent in our law of defenses. See, e.g., Barrett v. United States, 377 A.2d 62, 63–64 (D.C. 1977) (holding that “temporary insanity created by voluntary use of intoxicants will not relieve a defendant of criminal responsibility”). 5 Cf. Prezzi v. United States, 62 A.2d 196, 198 (D.C. 1948) (holding that a defendant charged with negligent homicide “is not responsible for what the other [participants in the accident] did or for the negligence of the other [participants],” and that a jury may not find a defendant guilty if “the accident was caused by the negligence of [other participants]”). 6 See, e.g., Wilson-Bey v. United States, 903 A.2d 818, 841 (D.C. 2006) (en banc) (describing “doctrines that permit conviction” on the basis of complicity, i.e., Pinkerton conspiracy liability and aiding and abetting liability).

24 See Sanford H. Kadish,7 Complicity, Cause and Blame: A Study in the Interpretation of Doctrine, 73 Cal. L. Rev. 323, 327 (1985) (discussing the “separate bodies of doctrine” that dictate when an individual may be held responsible for a crime: causation, “where results of a person’s action happen in the physical world,” and complicity, “where results take the form of another person’s voluntary action” and the defendant is somehow affiliated with that person); id. at 332–33 (explaining that these doctrines are premised on our foundational conception that humans are individual, volitional actors who choose whether or not to comply with the law). This court in Roy, however, disregarded our criminal law’s foundational principles of individual culpability and authorized a determination of guilt for murder in the absence of either actual causation or complicity.

In Roy, an innocent bystander was killed by a stray bullet fired either by Mr. Roy or his codefendant Mr. Settles. Roy, 871 A.2d at 501–02. The government prosecuted both men for murder, on the theory, approved by the trial court, that “it did not matter which [man] started the gun battle or which bullet actually killed”

7 Sanford Kadish has been described as “America’s foremost scholar of the criminal law.” Michael S. Moore, Retirement of Sanford Kadish, 79 Cal. L. Rev. 1401 (1991).

25 the bystander; it was enough that their actions had “turn[ed] city streets into an urban battle ground.” Id. at 506. The trial court instructed the jury that the only acts the government had to prove were that Mr. Roy and Mr. Settles (1) had each “armed and prepared” himself “to engage in a gun battle” and (2) “did, in fact, engage in a gun battle.” Id. at 507 n.8. The trial court then instructed the jury that it could find each man guilty of murder if “defendant’s conduct . . . was a substantial factor” in the bystander’s death, and it was “reasonably foreseeable” that a bystander could be killed or suffer serious bodily injury. Id. at 506–07 n.8. The trial court declined to instruct the jury, as requested by Mr. Roy and Mr. Settles, that the government had to prove that the two men had an “agreement to engage in mutual combat involving firearms.” Id. at 506.

On appeal, Mr. Roy and Mr. Settles both argued that the court’s jury instructions regarding causation improperly allowed each man to be found guilty of murder in the absence of evidence that either man directly caused the bystander’s death or that they were confederates. This court correctly disclaimed complicity as a basis for Mr. Roy’s and Mr. Settles’ culpability for murder,8 but it then

8 The court in Roy employed terminology that misleadingly suggests organized affiliations and allegiances that might support theories of complicity, e.g., “combatants,” “battles,” “pocket wars,” and “shoot out[s]” at “High Noon,” 871 A.2d at 507–08, but in fact the court rejected a requirement that the (continued…)

26 incorrectly held that complicity was unnecessary so long as the two men “prepared for and undertook to participate in [a] gun battle where it was clearly foreseeable that others would be endangered” and had “contribute[d] substantially to or [were] a substantial factor in [the] fatal injury.” Id. at 507–09. In reaching this determination, this court asserted that it was simply applying principles of “proximate cause.” Id. at 507.

The court in Roy misunderstood causation doctrine. Causation “consist[s] of two constituent parts: actual cause and legal [proximate] cause.” Burrage v. United States, 134 S. Ct. 881, 887 (2014). Actual causality “requires proof that the harm would not have occurred in the absence of—that is, but for—the defendant’s

(continued…) participants “reach a meeting of the minds to do battle,” id. at 508. To establish complicity, the government would have had to have shown that these adversaries either conspired with each other, necessitating an agreement to engage in criminal activity, Wilson-Bey, 903 A.2d at 840–41, or that one man aided and abetted the other “in the commission of the specific crime . . . charged,” id. at 831 (emphasis in original) (quoting Roy (Nakia) v. United States, 652 A.2d 1098, 1104 (D.C. 1995)). But the government could not show that Mr. Roy and Mr. Settles, as adversaries, had any sort of agreement; nor could it show that either man wanted the other to succeed in the charged crime—murder—where he himself was the target. See, e.g., Rivers v. Commonwealth, 464 S.E.2d 549, 551–52 (Va. Ct. App. 1995) (holding that “no existing common law theory,” including accomplice liability, supported a conviction for a bystander’s death in a gun battle because the appellant and the shooter “were not co-felons who acted in concert; instead they acted in opposition to each other and did not share the same criminal goal”).

27 conduct.” Id. at 887–88; see also McKinnon v. United States, 550 A.2d 915, 917 (D.C. 1988) (“In every criminal case, the government has the burden of showing that the defendant’s conduct . . . was a cause in fact of the harm for which he . . . is charged . . . .”). “Thus if A shoots at B intending to kill but misses, but at that moment B drops dead of some cause wholly unconnected with the shooting, A is not liable for the murder of B, in spite of the simultaneous existence of the two required ingredients, A’s intentional conduct and the fatal result.” Wayne R. LaFave, Substantive Criminal Law § 6.4 (2d ed.). As Professor LaFave explains, “[w]hat is missing is the necessary causal connection between the conduct and the result of conduct; and causal connection requires something more than mere coincidence as to time and place.” Id. (footnote omitted).

Only if actual or but-for cause is established do we ask whether the defendant’s actions are the legal or proximate cause of the harm. The core inquiry is whether there is some reason—e.g., an intervening event or attenuation—that precludes holding defendant criminally culpable for the ultimate injury. See Butts v. United States, 822 A.2d 407, 418 (D.C. 2003) (limiting liability on the basis of proximate cause when an “intervening cause . . . was so unforeseeable that the actor’s negligent conduct, though still a substantial causative factor, should not result in the actor’s liability”); Paroline v. United States, 134 S. Ct. 1710, 1720

28 (2014) (“[P]roximate cause is more restrictive than a requirement of factual cause alone.”). In other words, we consider proximate cause after actual, but-for cause has been established, and it acts to limit causation, not to extend it.

The causation analysis in Roy is flawed both as to actual cause, which it failed to consider, and as to proximate cause, which it defined incorrectly.

The court in Roy ignored the actual or but-for cause requirement in our second-degree murder statute, which obligates the government to prove that the defendant “with malice aforethought . . . kill[ed] another”9 person. D.C. Code § 22-2103 (2013 Repl.) (emphasis added). Obviously someone had fired the shot that killed the bystander, but the government’s investigation failed to establish

9 The clear meaning of this language was recognized by this court long before Roy. Examining identical language in the first degree murder statute, D.C. Code § 22-2101, this court explained that it requires the government to prove that the defendant “inflicted injury on the decedent from which he died.” Waller v. United States, 389 A.2d 801, 807 (D.C. 1978); id. at 807–08 (noting that “[t]he type of conduct subject to its sanctions is clearly identified in words of common understanding, with little room for misinterpretation or conjecture”); see also Christian v. United States, 394 A.2d 1, 48 (D.C. 1978) (“By its terms . . . the first- degree murder statute imposes . . . liability solely on the person who does the killing. Other participants in the felony are exposed to . . . liability only by virtue of the aiding and abetting statute.”); Ruffin v. United States, 76 A.3d 845, 854 (D.C. 2013) (acknowledging the “presumption that identical words used in different parts of the same act are intended to have the same meaning”).

29 whether that person was Mr. Roy or Mr. Settles.10 To the extent that the court connected the bystander’s death to Mr. Roy’s or Mr. Settles’s acts of “prepar[ing] for and undert[a]k[ing] to participate in [a] gun battle” (i.e., his acts of possessing and firing a weapon), this evidence did not establish actual causation: simple possession of a gun does not cause any injury, let alone a fatal one, and even firing that gun cannot be said to cause a gunshot death if the fired bullet missed any possible target.11

Disregarding this fundamental failure of proof, Roy jumped to proximate cause analysis, noting that this court has “long recognized proximate causation as a valid theory of second-degree murder liability.” Roy, 871 A.2d at 508 n.11.12 It is

10 Mr. Settles fired first and fired three shots; Mr. Roy then fired back. Roy, 871 A.2d at 502–03. 11 To slightly modify Professor LaFave’s hypothetical, if A shoots at B intending to kill but misses, B shoots back and misses, and C is killed by one of their bullets, unless the government can show that A fired the fatal shot, A cannot be found criminally responsible for C’s murder, “in spite of the simultaneous existence of the two required ingredients[:] A’s intentional conduct and the fatal result.” See LaFave, supra, at § 6.4. 12 Roy cited, inter alia, to Comber v. United States, 584 A.2d 26, 38–39 (D.C. 1990) (en banc), for this proposition. But this passage in Comber contains no discussion of this principle, much less support for the court’s “proximate cause” analysis; instead, the passage cited by the court in Roy is a discussion of the meaning of the “malice aforethought” mens rea requirement for second-degree murder. To the extent Comber discusses principles of proximate causation at all, it (continued…)

30 correct that the District has long held that if a defendant actually causes an injury that ultimately leads to death, the defendant may be convicted for murder. Accord, Hopkins v. United States, 4 App. D.C. 430, 439 (D.C. Cir. 1894) (reaffirming the centuries-old rule “that if one give wounds to another, who neglects the cure of them, or is disorderly and doth not keep that rule which a person wounded should do, yet if he die it is murder or manslaughter, according as the case is; because if the wound had not been, the man had not died, and, therefore, neglect or disorder in the person who received the wounds shall not excuse the person who gave them”). But “proximate” cause does not relieve the government of proving that the defendant in fact acted in a way that caused an initial injury.13

(continued…) does so in the context of limiting liability for misdemeanor manslaughter, “mindful of the danger” of using misdemeanor manslaughter to “cast too wide a net” of homicide liability. Id. at 50. 13 Thus, in each of the proximate cause cases cited by the court in Roy, the defendant had actually caused some injury to the decedent, and the issue was whether an intervening event broke the causal chain such that the defendant could not be convicted of manslaughter or murder. See Butts, 822 A.2d at 417–18 (holding that possible negligence by the victim of a vehicular homicide when crossing the street was not an intervening event that negated drunk driver’s criminal liability); McKinnon, 550 A.2d at 917–18 (holding that, where the defendant slashed the decedent’s throat, requiring immediate surgery, an infection was a reasonably foreseeable consequence and not an intervening cause of death); Baylor v. United States, 407 A.2d 664, 669–70 (D.C. 1979) (upholding trial court’s refusal to instruct the jury on intervening negligence where defendant’s action— stabbing his wife with great force—substantially contributed to her death, even (continued…)

As noted above, proximate cause is a limiting concept. See Paroline, 134 S. Ct. at 1720. Had this limiting concept been properly applied in Roy, it would have precluded the jury from finding Mr. Roy criminally culpable for murder due to a shot fired by Mr. Settles, or vice versa—not enabled it. The criminal law does not accept an understanding of causation in which one person, A, “causes” another, B, to voluntarily shoot and kill a third party. Instead, absent some connective theory of complicity between A and B (which Roy disclaimed, see supra note 8 and accompanying text), B is the only one responsible for his voluntary actions and their fatal consequences.14 See H.L.A. Hart & A.M. Honoré, Causation in the Law 326 (2d ed. 1985) (“The free, deliberate, and informed intervention of a second person, who intends to exploit the situation created by the first, but is not acting in concert with him, is normally held to relieve the first actor of criminal responsibility.”); Kadish, supra, at 391 (Under the common law, volitional, criminal actions “are seen not as caused happenings, but as the product of the

(continued…) where subsequent surgery necessitated by defendant’s actions may also have injured her). 14 There are limited exceptions where courts trace causation through B to A and only hold only A criminally culpable: if A uses B as an innocent instrumentality, Kadish, supra, at 370, or if B’s conduct is legally justified, id. at 395. But neither of these scenarios is contemplated by the urban gun battle narrative.

32 actor’s self-determined choices, so that it is the actor who is the cause of what he does, not [the individual] who set the stage for his action.”). Put another way, even assuming arming oneself with a gun and firing it could satisfy the direct causation requirement, the volitional, felonious act of someone else then shooting and killing the decedent is an “intervening cause” that breaks this chain of criminal causation.15

Implicit in Roy’s theory of proximate causation for urban gun battles, however, is an assumption that another person’s act of firing a fatal shot can be a “foreseeable” event that does not break the causal chain between the defendant’s actions—possessing and firing a gun—and a resulting death. 871 A.2d at 507–09. Though foreseeability is certainly part of proximate cause analysis, see McKinnon, 550 A.2d at 917–18, the voluntary choices of other individuals (again, absent some

15 But it is not even clear that the government must prove this temporal sequence under an urban gun battle theory, i.e., that the defendant “engaged” in a gun battle first and then a person died. Rather, Roy suggests that, as long as the government proves that two events occurred—the defendant fired his weapon, and a person was shot and killed—liability for murder is established regardless of the timing. In Roy, the government could not prove who fired the fatal shot, but the evidence was undisputed that Mr. Settles was the first to fire his gun and that he fired three shots before Mr. Roy ever fired back. 871 A.2d at 502–03. It was thus possible that Mr. Settles immediately shot and killed the bystander before Mr. Roy ever “engaged” in the gun battle. By implication, Mr. Roy’s engagement was sufficient to make his actions a “substantial factor” in the bystander’s death, even if it was subsequent to that event. Id. at 507 n.8.

33 connective theory of complicity) are not the type of “foreseeable” consequences that fall within the boundaries of criminal proximate cause. Foreseeability relates to consequences in the physical, natural world, e.g., lighting a match next to a flammable substance;16 but, because humans are individual, volitional actors, in criminal law, we do not conceive of their responses to the actions of others as “foreseeable” and thus serving as the basis for criminal culpability.17 “[W]hatever

16 See, e.g., Williams v. United States, 20 F.2d 269, 270 (D.C. Cir. 1927) (rejecting the argument that decedent’s death was not caused by or foreseeable to the defendant where she struck the decedent with a lighted lamp, “resulting in the igniting of [his] clothes . . . and thereby causing his death”); see also Kadish, supra, at 332 (“If I light a match in an area containing explosive vapors that ignite, starting a fire that burns down a building, I may be blamed for the burning of the building because I can be said to have caused it. I started a chain of events that led to the burning of the building through cause and effect relationships governed by laws of nature.” (emphasis added)). 17 As Professor Kadish explains, if someone else lit the match, the criminal law would affix blame to the defendant only if there was complicity:

I may have persuaded another responsible person to light the match or helped him by giving him a match for the purpose. The other person then caused the burning of the building. But whether I am to be blamed for the other person’s action would not be assessed by asking whether I caused his action in the same sense that his lighting the match caused the fire. Rather, my responsibility would be determined by asking whether my persuasion or help made me accountable for the other person’s actions and what they caused.

(continued…)

34 the relation of one person’s acts to those of another, it cannot be described in terms of that sense of cause and effect appropriate to the occurrence of natural events without doing violence to our conception of a human action as freely chosen.”18 Kadish, supra, at 335.

This conception of causation and the bar on holding an individual criminally liable for the volitional actions of others with whom the individual is not complicit is long-standing. Indeed, the Supreme Court of Massachusetts addressed this precise question in Commonwealth v. Campbell, 89 Mass. (7 Allen) 541, 543 (1863). In Campbell, the government asked the court to instruct the jury that if it the defendant participated in “a riotous assembly,” he could be found guilty of manslaughter “although the evidence may fail to show whether the shot which killed the deceased was fired by the rioters with whom the [defendant] was acting in concert, or by the soldiers who . . . resist[ed] the attack made . . . by the

(continued…) Kadish, supra, at 332–33 (emphasis added). In short, “[t]he doctrine of causation deals with fixing blame for natural events. The doctrine of complicity deals with fixing blame for the criminal action of another person.” Id. at 333. 18 See also H.L.A. Hart & A.M. Honoré, Causation in the Law 41 (1st ed. 1959) (“A deliberate human act is therefore most often a barrier . . . in tracing back causes in such inquiries: it is something through which we do not trace the cause of a later event and something to which we do trace the cause through intervening causes of other kinds.”).

35 rioters . . . .” Id. at 543. Considering “whether the doctrine in question ha[d] any just foundation in the recognized principles of law by which criminal responsibility for the acts of others is regulated and governed,” the Supreme Court of Massachusetts determined it did not. Id. In statements almost tailor-made to rebut Roy’s urban gun battle theory, the court explained:

No person can be held guilty of homicide unless the act is either actually or constructively his, and it cannot be his act in either sense unless committed by his own hand or by some one [sic] acting in concert with him or in furtherance of a common object or purpose.

Id. at 544. The court then continued:

Certainly that cannot be said to be an act of a party in any just sense, or on any sound legal principle, which is not only not done by him, . . . but is committed by a person who is his direct and immediate adversary, and who is, at the moment when the alleged criminal act is done, actually engaged in opposing and resisting him and his confederates and abettors in the accomplishment of the unlawful object for which they are united.

Campbell, 89 Mass. at 544–45. Based on these longstanding principles of causation, Roy was wrongly decided.

II. The Causation Problem in Fleming

As noted above, Roy’s urban gun battle theory is predicated on an assumption that one person can “cause” another to shoot and kill a third party. The majority opinion does not press this faulty conception of causation. Instead, it shifts focus to hold that an individual who exchanges gunfire with another “proximately causes” any death that results because both individuals are “collectively responsible for creating a zone of great danger—including a substantial likelihood that errant bullets will hit unintended targets.” Ante, at 9. Repackaging the culpable criminal conduct supporting a murder conviction as the act of creating a “zone of danger,” however, does not solve the causation problems inherent in Roy’s urban gun battle theory.

Until now, in the criminal law context, a “zone of danger” has been a mens rea—not a causation—concept19: if a defendant, with specific intent, shoots and

19 The concept of a “zone of danger” is separately employed in tort cases to determine whether a plaintiff has a viable claim of negligent infliction of emotional distress. See, e.g., Jane W. v. President & Dirs. of Georgetown Coll., 863 A.2d 821, 826–28 (D.C. 2004) (“To establish a prima facie case of negligent infliction of emotional distress, [plaintiff] must show that she was in the zone of physical danger created by [the negligent actor’s] conduct”). But that tort law term of art has no application in a murder prosecution.

37 kills or injures his target as well as an unintended bystander, his malicious intent to kill anyone within the zone of danger he created by firing “a hail of bullets” is inferred. Ruffin v. United States, 642 A.2d 1288, 1298 (D.C. 1994) (adopting a criminal law mens rea theory of concurrent intent denied by Ford v. State, 625 A.2d 984 (Md. 1993), and later adopted by Henry v. State, 19 A.3d 944 (Md. 2011)); see also, e.g., Washington v. United States, 111 A.3d 16 (D.C. 2015); Mobley v. United States, 101 A.3d 406 (D.C. 2014); Castillo-Campos v. United States, 987 A.2d 476 (D.C. 2010); Brown v. United States, 934 A.2d 930 (D.C. 2007); Walls v. United States, 773 A.2d 424 (D.C. 2001). A defendant’s inferred concurrent intent to shoot and kill or injure everyone in a particular “zone of danger,” however, does not address the causation gap that exists where there is no proof that the defendant hit anyone or acted in complicity with someone who did.20

20 That said, the urban gun battle theory also impedes meaningful examination of a defendant’s intent. To be guilty of murder a defendant must act with malice. Comber, 584 A.2d at 38–40 (describing the four types of murder encompassed by malice, including wanton and willful disregard of an unreasonable human risk). But under the urban gun battle theory, the fact finder is never asked to examine whether the defendant acted with malice before the fatal shot is fired. Again, the defendant need not fire the fatal shot, and the malice requirement applies only generally to the defendant’s “engagement” in the urban gun battle, without regard to when the defendant fired a gun. See supra note 15. This is in sharp contrast to the theory of concurrent intent, which examines a defendant’s intent at the time when he fired a gun and killed multiple individuals (or when his co-conspirators or accomplices did so). Moreover, casting the blanket of an “urban gun battle” over a shooting incident virtually guarantees that the fact finder will conclude the defendant’s actions were dangerously reckless, obscuring (continued…)

But the more fundamental problem with reframing Roy’s causation theory to require that the government prove nothing more than that the defendant, contemporaneously with others, created a “zone of danger” is that this reframing actually changes the substantive nature of the crime. Under D.C. Code § 22-2103, a person is guilty of second-degree murder when he “kills another,” id. (emphasis added); see also supra note 9, with the requisite state of mind, not when he creates a zone of danger in which another person somehow gets killed. The actions of an individual who creates a “zone of danger” do not themselves kill or lead to the death of anyone: death comes from the action of the shooter who fires the fatal shot. This court cannot “appl[y] principles of proximate causation,” ante, at 10, to bridge this causation gap, for reasons already discussed. And we are not empowered to effectively rewrite the murder statute and to dispense altogether with the direct causal connection between a defendant’s actions and a death. See Santos v. District of Columbia, 940 A.2d 113, 116 (D.C. 2007) (“In general, the ‘definition of the elements of a criminal offense is entrusted to the legislature.’”); cf. United States v. Heinlein, 490 F.2d 725, 736 (D.C. Cir. 1973) (refusing to

(continued…) examination of the individual motivations of each actor as events unfold in real time.

39 extend the felony-murder statute “beyond its common law origins” and holding that “[f]urther action by Congress would be necessary to that end”).21

• • * *

Courts may feel “pressure” to somehow find criminal liability in cases “where the culpable consequence of [the] defendant’s action is some unintended but voluntary action of another,” but Professor Kadish warns that “[t]here is no way to extend liability in these cases . . . in a way that does not require a significant departure from doctrinal premises.” Kadish, supra, at 399–400, 402–03. This court’s urban gun battle theory is a case in point. With no real causation

21 The Council could create a new “zone of danger” murder crime. Accord, Rivers, 464 S.E.2d at 554 (rejecting an urban gun battle type theory of guilt as inconsistent with the common law and concluding that “the legislature, not this Court, is the appropriate forum in which to amend [the law] to provide for criminal liability [for opponents in a gun battle]”). But it is unclear why the Council would feel pressed to do so. We already have serious felonies that prohibit actions that cause danger in the community, e.g., assault with intent to kill, D.C. Code § 22-401 (2013 Repl.) (punishable by up to fifteen years imprisonment), and assault with intent to kill while armed, D.C. Code § 22-4502 (2013 Repl.) (punishable by up to thirty years imprisonment).

40 requirement (and no complicity requirement), this theory is unmoored from established common law principles of criminal culpability.22

In effect, the court in Roy made a policy choice to create a new crime that allows an individual to be convicted of murder without proof that he caused (or was complicit in) a murder. If the objective was to ensure that certain homicides beget at least one murder conviction, it is antithetical to another “cardinal principle of Anglo-American jurisprudence that, in Blackstone’s immortal words, better ten guilty persons should go free than one innocent person be convicted.” United States v. Greer, 538 F.2d 437 (D.C. Cir. 1976) (citing 4 William Blackstone, Commentaries on the Laws of England 352 (1769)). Indeed, the creation of this new crime leads to a perverse result: in contrast to cases where the government shoulders its burden to prove the identity of the killer, in urban gun battle cases, the government can successfully prosecute more people for murder based on far less

22 We should not be reassured that a few other courts have similarly misapplied causation principles to recognize an “urban gun battle” theory of guilt for murder. See Roy, 871 A.2d at 507 n.10. We may look to other jurisdictions to see how they have analyzed and applied common law doctrines, but we have an obligation to scrutinize other courts’ analyses and refuse to follow suit where their analyses are flawed.

41 proof.23 But the biggest problem with this court’s decision in Roy is that we exceeded our judicial role. It is not our job to create new crimes. We do violence to the law when we purport to apply the common law doctrine but in fact act as a legislature. Before more damage is done,24 we need a course correction. Roy should be overturned by this court en banc, not revised and effectively extended by the panel in this case.

23 Cf. Wilson-Bey, 903 A.2d at 838 (holding that aiders and abettors may only be held accountable for crimes committed by a principal when the aider and abettor has the same mens rea as the principal, because “it is particularly inappropriate to permit the conviction of an aider and abettor upon a lesser [evidentiary] showing . . . than is required vis-à-vis a principal when the defendants are being prosecuted for a homicide”). 24 Roy’s problematic “proximate cause” theory of causation has been applied in a handful of cases, though never challenged on these grounds. See, e.g., Bryant v. United States, No. 14-CF-268 (D.C. Nov. 3, 2016); McCray v. United States, 133 A.3d 205 (D.C. 2016); Blaize v. United States, 21 A.3d 78 (D.C. 2011); Blaine v. United States, 18 A.3d 766 (D.C. 2011). Notably, the division’s decision in this case is at least in tension with this court’s recent decision in Bryant: this division now holds that the urban gun battle theory of liability can apply to the shooting death of non-bystanders, because “it makes no meaningful difference whether the reasonably foreseeable victim of a shootout was a participant in the battle or a bystander.” Ante, at 10. But in Bryant the court determined that the urban gun battle theory could be used in first-degree murder cases because it is the “increased risk to innocent bystanders which justifies the application of proximate cause liability to those participants who willfully choose to engage in these battles.” Bryant v. United States, No. 14-CF-268, slip op. at 18 (D.C. Nov. 3, 2016). Such conflicting statements about the boundaries of the urban gun battle theory of liability are inevitable when the court is applying a judicially-legislated, ad hoc theory of culpability that has no foundation in common law doctrine.