268 A.3d 833

D.C.

2022

TIMOTHY D. CALLAHAM v. UNITED STATES

No. 19-CF-167

DISTRICT OF COLUMBIA COURT OF APPEALS

February 3, 2022

Nоtice: 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.

Appeal from the Superior Court of the District of Columbia (CF3-8133-18)

(Hon. Kimberly S. Knowles, Trial Judge)

(Argued September 30, 2020 Decided February 3, 2022)

Vincent A. Jankoski for appellant.

David P. Saybolt, Assistant United States Attorney, with whom Jessie K. Liu, United States Attorney at the time the brief was filed, and Elizabeth Trosman, Assistant United States Attorney, were on the brief, for appellee.

Before BLACKBURNE-RIGSBY, Chief Judge, and THOMPSON* and EASTERLY, Associate Judges.

EASTERLY, Associate Judge: Timothy Callaham was convicted of robbery after a jury found that he and a companion, who was carrying what appeared to be a gun, approached a man in a striped shirt at the back of a market and walked out with two items that the man in the striped shirt tossed onto the floor. Although the putative complainant did not testify at trial, the incident was captured on surveillance video. The government introduced the video footage as evidence and called as witnesses two detectives who had reviewed the footage, but had not been present for the actual incident. In his brief to this court, Mr. Callaham makes several arguments related to the admission of the video footage and testimony about what the footage showed; he also argues that the jury‘s verdict was coerced.

Based on the particular facts of this case where (1) the body of evidence was limited and the case boiled down to whether the government had proved that a robbery occurred; (2) the jury struggled with that question and sent the court multiple notes seeking guidance as well as two notes indicating deadlock, prompting a Winters anti-deadlock instruction; (3) the Winterized jury announced a verdict which broke down (i.e., was revealed to be non-unanimous) in polling; and (4) the court gave the jury poll breakdown instruction which, although it contained the language endorsed by this court in Crowder v. United States, 383 A.2d 336, 342 n.11 (D.C. 1978) (recommending language intended to alleviate pressure on a jury to reach a verdict), still echoed the more coercive elements of the Winters anti-deadlock instruction, we conclude that there was a substantial risk that the jury‘s verdict was coerced. We reverse on this ground, but address to the extent necessary the other issues raised by Mr. Callaham in the event of a retrial.

I. Facts and Procedural History

One early morning in May 2018, Metropolitan Police Department officers responded to a call about a man with a gun at Mellon Market, on the corner of Mellon Street and Martin Luther King Jr. Avenue, SE. Later that morning, the market‘s owner called MPD to report thаt someone had entered the store with a gun the night before. MPD Detective¹ Taylor Volpe responded to Mellon Market to review the footage captured by the market‘s thirty-two surveillance cameras. After watching the video, Detective Volpe identified and approached a man in a striped shirt outside the store who appeared to be the same person who had been standing in the back of the market and had tossed items on to the floor. Detective Volpe testified that this person, David Garvin, “didn‘t have a whole lot to say” about the incident.

Later, MPD officers retrieved the surveillance footage from Mellon Market, as well as closed-circuit television footage from an MPD camera located on the same corner as the market. Another detective assigned to the case, Kоnrad Olszak, identified Mr. Callaham from the Mellon Market footage. Mr. Callaham was arrested for an unrelated offense at his girlfriend‘s apartment. On the theory that the Mellon Market surveillance video captured an armed robbery of Mr. Garvin, Mr. Callaham was charged with both that offense,² and possession of a firearm during the commission of a crime of violence.³ Detective Olszak executed a search warrant at the apartment where Mr. Callaham was arrested and seized three items of evidence: “[a] pair of [white] jeans with multiple cuts on them“; “[a] pair of Nike Air Max 95s“; and “[a] box ... [for a] paint gun.” The government also recovered a recording of a jail call made from Mr. Callaham‘s account, in which he stated that the police would “see ... [him] getting [in and] out [of] the car [seen leaving the scene on surveillance footage] ... [and] say that‘s aiding and abetting.”

Because there were no eyewitnesses,⁴ the government‘s case hinged on the Mellon Market and CCTV surveillance footage. Detective Volpe testified that he reviewed surveillance video at Mellon Market the morning after the incident. He told the jury it showed “a robbery” by two men, one holding what looked like an assault rifle and another who he identified as Mr. Callaham, and detailed what he had seen in the footage, over an objection from defense that Detective Volpe was being “ask[ed] ... to testify about his memory of something he has no personal knowledge of.” Detective Volpe testified that the two men had approached a third man, whom he identified as Mr. Garvin, in the store; that “Mr. Callaham” picked up a bag and wallet that “Mr. Garvin” dropped; and that “Mr. Callaham” and his companion then left. The government later introduсed an approximately seven-and-a-half-minute compilation of clips of the relevant portions of the surveillance footage, again over Mr. Callaham‘s objection, during Detective Olszak‘s testimony. The court tried to cabin the detective‘s testimony somewhat, ruling that he could not “give his own interpretation of things” he saw in the video, but could “point out things to bring things to the jury‘s attention.” Detective Olszak then testified about what he saw in the footage (which had no audio), as it played to the jury.

The day after Detective Olszak‘s testimony began, the court expressed its concern about the government witnesses “testifying or narrating” what they saw in the video. The court indicated that, based on its understanding of the law, these witnesses “were not supposed to be” doing the latter, but that they could point “out certain things on the video so that the jury could be oriented.” Even so, the court wanted to make sure the jurors understood that they “decide[] if it‘s Mr. Callaham . . . if in fact he had a gun, if in fact he was walking at this pace . . . The detective is pointing it out to them, but they have to see it and be convinced by it.” The court thus proposed giving an instruction “explaining that to [the jury].”

The government opposed the instruction, noting that in questioning Detective Olszak, the prosecutor had made clear that the detective was testifying about “his opinion [of what] was in the video,” and that this was “really no different than showing photographs or objects to any expert—or I shouldn‘t even say ‘expert‘—any witness. The witness testifies about what they view the thing as, what it appears to them.” The court noted, however, that “usually” when the government shows a witness a photograph, they are a percipient witness: “they have seen the knife on the ground, they have seen the gun, or they were there.” The court distinguished identification testimony from other testimony and reiterated its concern that the evidence was “just coming out in a very weird way“: in a way that might obscure for the jurors that it was “their decision about what actually happened.”

Defense counsel also objected to the court‘s proposal to give an instruction to the jury, but on the ground that it was too little, too late. Arguing that the government had “chose[n] to present the video in the way in which they chose to present it” and that “I don‘t think that this [damage] can be undone,” counsel moved for a mistrial. The court denied the motion for a mistrial, and instructed the jury that morning—before the government resumed its direct examination of Detective Olszak—that the detective‘s testimony was to “assist you in the identification of suspects in the video as well as to direct you to portions of the exhibits. It is for you to decide who is depicted in the videos, what happened, and how to interpret the video. You are the finders of the facts.”

After not quite three days of trial, the jury began deliberating the morning of Tuesday, December 4, 2018.⁵ At 3:34 p.m., it sent a note containing three questions, followed by another note with a fourth question at 4:34 p.m. All of these questions related to whether the government had proved that a robbery had in fact occurred under the legal framework the trial court had provided: (1) “[The] [i]nstructions refer to the property of a ‘complainant’ being taken. (Page 14, after #8): If there is no ‘complainant’ can there be an (‘armed‘) robbery?“; (2) “What or who is a ‘complainant’ [?]“; (3) “If some or all of the property taken belonged to one of ‘robbers’ is it robbery to take it from the ‘victim‘[?]“; (4) “The ‘victim’ hаs not to our knowledge ‘complained.’ Is there a robbery? Must the victim complain?” With both parties’ approval, the court provided responses to these questions.

The jury returned on Monday, December 10, 2018 (the court had been closed for various reasons the preceding Wednesday through Friday). At 10:00 a.m., the jury sent a note indicating deadlock: “Your Honor: We are at an impasse. We have extensively reviewed the instructions and evidence and remain deadlocked.” The court denied defense counsel‘s motion for mistrial, or alternatively for a Winters anti-deadlock instruction,⁶ and instead granted the government‘s request to give the “initial instruction to [a] jury that indicates it cannot agree” set forth in Criminal Jury Instructions for the District of Columbia, No. 2.601(I) (5th ed. 2018). The court instructed the jury, in part:

[Y]our note indicates that you‘ve been unable to reach a unanimous decision at this time. My best judgmеnt is that you‘ve been . . . deliberating for a total of approximately five hours, which is not unusual in cases such as this. As a result, I am going to ask you to deliberate further in this case and that you keep an open mind about the case with a view to listening to others and expressing your own point of view to see whether you can reach a unanimous decision. Please resume your deliberations at this time.

Subsequently, at 10:52 a.m., the jury sent a note asking three more questions, again all related to whether the government had proved that a robbery had occurred under the legal framework provided by the court. The jury asked: “What is [p]roof of taking without right? How do we know [he] didn‘t have a right to property? What is the basis of knowing whether or not the defend[ant] had right to property or not?” The court responded to the note and asked the jury to continue deliberаting. An hour later, at 11:55 a.m., the jury sent another note asking to see the parties’ “stipulations of fact that we were told cannot be disputed.” With the parties’ agreement, the jury was given the single stipulation in evidence (which addressed only the authenticity of a jail call made from Mr. Callaham‘s account).

An hour later, at 12:50 p.m. (just before lunch), the jury sent a second note indicating a deadlock: “After re-watching and reviewing the evidence, and after further intense discussions, we are still at an impasse. We carefully considered and re-considered each of the elements of the allegations [and] ... [w]e are unable to reach a unanimous verdict.”⁷

The court denied defense counsel‘s renewed motion for a mistrial, and granted both parties’ requests to give a Winters instruction. When the jury returned from lunch, the court instructed it, in part:

Although the verdict must be the verdict of each juror and not merely giving in to the views of other jurors, nevertheless, you should examine the questions submitted to you honestly and with proper regard and respect for the opinions of each other.

You should consider that it is desirable that the case be decided. . . . With this in mind, it is your duty to decide the case if you can conscientiously do so. You should listen to each other‘s arguments with a willingness to be convinced. Thus, where there is disagreement, jurors who are for acquittal should seriously consider whether their doubt is a reasonable one if it makes no impression upon the minds of other jurors who are equally honest and equally intelligent as themselves and who have heard the same evidence with the same attention with an equal desire to evaluate the evidence and decide the case and under the sanction of the same oath.

And, on the other hand, jurors who are for conviction ought seriously ask themselves whether they might not reasonably doubt the correctness of a judgment which is not shared by other jurors on the jury and whether they should trust the sufficiency of that evidence which fails to carry conviction in the minds of fellow jurors, who, again, are equally honest and equally intelligent as themselves and who have heard the same evidence with the same attention and equal desire to evaluate the evidence and decide the case and under the sanction of the same oath.

With those thoughts in mind, I am going to ask that you resume your deliberations.

At 4:45 p.m., the jury sent a note indicating it had reached a verdict on count one, robbery while armed. After the foreperson announced in open court that the jury had found Mr. Callaham guilty of robbery ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍while armed, the court begаn polling the jury. When asked if they agreed with the announced verdict, the second juror said “No.” The court immediately stopped the poll and dismissed the jury for the day.

The parties returned to court the following morning, Tuesday, December 11. Defense counsel, for the third time during deliberations, moved for a mistrial. After discussion with the parties, and over defense counsel‘s objection, the court gave the jury the model Criminal Jury Instruction 2.603 from the Criminal Jury Instructions for the District of Columbia (5th ed. 2018) including the optional language discussed by this court in Crowder v. United States, 383 A.2d 336, 342 n.11 (D.C. 1978).⁸ The court instructed the jury, in part:

. . . I can tell you in the polling of the jury it has become apparent that you may not have reached a unanimous verdict. For this reason, I‘m asking you to return to the jury room for further consideration of your verdict. If you are unanimous, your foreperson should send me a note indicating that. And I will poll you again. If you are not unanimous, please resume deliberations and see if you can reach a unanimous verdict.

It is your duty as jurors to consult with one another and to deliberate with a view to reaching an agreement if you can do so without violence to individual judgment. Each of you must decide the case for yourself but do so only after an impartial consideration of the evidence with your fellow jurors.

In the course of your deliberations, do not hesitate to re-examine your own views and change your opinion if convinced it is erroneous. But do not surrender your honest conviction as to the weight or effect of evidence solely because of the opinion of your follow [sic] jurors or for the mere purpose of reaching a verdict.

With that, I‘m going to send you back to the jury room, ladies and gentlemen. Thank you.

It is not clear from the record when precisely on the morning of Tuesday, December 11, the court delivered this instruction. However, the prior evening, the court had instructed the parties to return to court at 9:15 a.m., and that morning, the court had (1) discussed the Crowder instruction with the parties; (2) taken a recess so the court could read relevant case law and so the government could consult with the appellate division; and (3) continued to discuss the Crowder instruction after the recess, before the court called the jury in to actually give that instruction. It thus appears the jury deliberated for less than two hours—possibly substantially less—before it sent a note to the court, at 11:42 a.m., stating that it had again reached a verdict. It found Mr. Callaham not guilty on count one, robbery while armed, but guilty of the lesser included robbery charge, and not guilty on count two, possession of a firearm during the commission of a crime of violеnce. This timely appeal followed.

II. Whether There is a Substantial Risk that the Jury‘s Verdict was Coerced

When a defendant exercises their right to a jury trial, the jury‘s verdict will have legitimacy only if it is the product of unanimous decision making, devoid of coercion. Green, 740 A.2d at 25. When a jury struggles to reach a verdict, the court unquestionably “has a right and duty to urge [the] jury to work diligently to reach a fair and freely arrived at verdict if possible.” Id. at 26. There comes a point, however, when a defendant is entitled to the benefit of a jury that deliberates thoughtfully but does not reach a verdict; such an inconclusive outcome is itself an assessment of the evidence that must be respected. See Winters, 317 A.2d at 539 n.10 (Gallagher, J., concurring) (“It is fundamental that a defendant is entitled to the benefit of a disagreement by the jury.“); accord Smith v. United States, 542 A.2d 823, 825 (D.C. 1988) (“In our legal system, the minority in a jury deserves respect and credеnce.“).

Mr. Callaham argues this point was reached in his case when the jury, having already received a Winters anti-deadlock instruction, announced a guilty verdict that broke down in polling. He asserts the subsequent delivery of a Crowder instruction—including its reiteration of the “[jury‘s] duty to deliberate and reach agreement” language from the Winters instruction—coerced the jury‘s verdict. “[O]ur evaluation of jury coercion focuses on probabilities, not certainties.” Davis v. United States, 669 A.2d 680, 685 (D.C. 1995). Thus our task is not to determine whether the jury‘s verdict finding Mr. Callaham guilty of robbery was in fact coerced, but only whether the record “reveals a substantial risk of a coerced verdict.” Coley v. United States, 196 A.3d 414, 420 (D.C. 2018) (internal quotation marks omitted). We assess the risk of juror coercion “from the jurors’ perspective,” inquiring into both “the inherent coercive potential of the entire situation before the trial court and the ameliorative or exacerbating impact of the judgе‘s actions in response to that situation.” Hankins v. United States, 3 A.3d 356, 361 (D.C. 2010) (internal quotation marks omitted). If we cannot “say with assurance that the jury arrived at its verdict freely and fairly,” we must reverse. Id. at 362 (internal quotation marks omitted). Here, we conclude there was a substantial risk that the court‘s delivery of the Crowder instruction coerced the jury‘s verdict.

A. The Coercive Potential of the Circumstances

As our case law requires, we must first consider the coercive potential of the circumstances at the time the court gave the challenged instruction. This necessitates that we look beyond the jury poll breakdown that immediately preceded the instruction (which is what the government focuses on in its brief) and examine the totality of the jury‘s deliberations. We begin by acknowledging the limited amount of evidence in the case, and the jury‘s evident difficulty in reaching any conclusions based upon it. Because the putative complainant declined to testify, the case turned on the jury‘s interpretation of about seven-and-a-half minutes of silent surveillance footage from in and around the Mellon Market convenience store—footage that was ambiguous as to the relationship between the man identified as Mr. Callaham and the two other men in the back of the Mellon Market, and as to why the man identified as Mr. Garvin dropped some items on the floor before the man identified as Mr. Callaham picked them up. The central question (assuming the jury credited the identification of Mr. Callaham as one of the men depicted) was whether this footage showed his participation in a robbery. The jury immediately honed in on this question and they struggled with it. In an effort to gain clarity, they sent two notes containing multiple questions seeking guidance in answering it, then said that they were deadlocked despite having “extensively reviewed the instruсtions and evidence.” After being told to continue their deliberations, they sent two more notes with questions revealing their efforts to make sense of the evidence, followed by another deadlock note. Despite “re-watching and reviewing the evidence” and “further intense discussions” they told the court they were “at an impasse” and “unable to reach a unanimous verdict.” In short, well before the delivery of the jury poll breakdown instruction, there were signs that this might be a case where the jury justifiably would not be able to reach a verdict. See United States v. Thomas, 449 F.2d 1177, 1183 (D.C. Cir. 1971) (“Equivocal evidence can raise problems for conscientious jurors, and increase their susceptibility to judicial prodding for a verdict they seem otherwise unable to reach.“); see also Fortune v. United States, 65 A.3d 75, 86 (D.C. 2013) (“[Where] the jury had deliberated for more than eight hours over the course of two days (with an intervening weekend) and sent the judge three notes stating unequivocally, and with increasing emphasis, that it was at an impasse . . . the prospect that any further pressure on this jury to reach a unanimous verdict would induce jurors to abandon their honest convictions in order to do so[ ]was elevated.“).

Second, we consider the delivery of a Winters instruction to the jury in response to the second deadlock note. A Winters instruction is one of three anti-deadlock charges listed in the Criminal Jury Instructions for the District of Columbia which a trial court may issue at its discretion “when jurors cannot agree,” Criminal Jury Instructions for the District of Columbia, Comment to No. 2.601 (5th ed. 2018) (internal quotation marks omitted), but the Winters instruction represents “the highwater mark for an anti-deadlock charge.” 317 A.2d at 534. As Winters itself acknowledged, the charge has a “sting“: it stresses the “desirab[ility]” that the case be decided and the jury‘s “duty” to reach a decision, and it directs the jurors to reconsider their own views. Id. at 533-34; see also Hankins, 3 A.3d at 360 n.3 (acknowledging the “emphatic language” of the Winters instruction). Moreover, it does not caution the jurors against “surrender[ing] [their] honest convictions as to the weight of the evidence or effect of the evidence solely because of the opinion of your fellow jurors, or only for the purpose of returning a verdict,” like the Gallagher anti-deadlock instruction. Winters, 317 A.2d at Appendix to Concurring Opinion (Gallagher, J., concurring); see also Hankins, 3 A.3d at 363 (explaining the Gallagher instruction is less coercive than the Winters instruction for this reason).

Mindful of its coercive potential, we have warned that the Winters instruction “should not be given routinely, but only after careful consideration by the trial judge of the nature of the case and the length of the deliberations.”⁹ Smith, 542 A.2d at 825. Here, we have reason to be especially concerned ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍about the coercive effect of a Winters instruction because of what happened after the jury was so instructed: it returned an ostensibly unanimous verdict that fell apart during polling. In other words, the record suggests that the Winters instruction coerced a temporary verdict and contributed to the coercion of the ultimate verdict.

Third, we consider the jury poll breakdown on its own terms. The object of polling the jury after a guilty verdict is “to ascertain for a certainty that each of the jurors approves of the verdict as returned; that no one has been coerced or induced to sign a verdict to which [they] do[] not fully assent.” Green, 740 A.2d at 25; see also id. (“The jury poll . . . has long been regarded as a useful and necessary device for preserving the defendant‘s right to a unanimous verdict.“). To be sure, not every jury poll breakdown signifies the presence of coercion, see id. at 26, but based on the facts detailed above and the absence of any alternative explanation—e.g., there is no indication that this jury, which sent the court multiple notes with sophisticated questions, was confused or simply impatient with the collaborative process—the inference of coercion is strong. Moreover, a jury poll breakdown creates an additional risk of coercion as a result of revealing, in open court, one or more dissenting jurors, which is heightened when their dissent follows an anti-deadlock instruction. See Harris v. United States, 622 A.2d 697, 705 (D.C. 1993) (explaining that when it becomes apparent that a jury is split, “factors that help to establish the existence or degree of inherent coercive potential include (but are not limited to): . . . whether the identity of a dissenting juror (or jurors) is revealed in open court, . . . whether the judge knows the identity of a dissenting juror (or jurors) and whether the juror is aware of the judge‘s knowledge, . . . and whether an ‘anti-deadlock’ instruction has been given and, if so, whether this has occurred under circumstances where the potential for coercion is high.“).¹⁰ If the trial court thereafter instructs the

jury to keep deliberating, an “obvious danger” exists that the dissenting juror (and any affiliated minority) will interpret the court‘s instruction as “requiring further deliberations in order to eliminate [their] dissent.” Crowder, 383 A.2d at 342 n.11.

Taking into account this “entire situation,” see Hankins, 3 A.3d at 361, we conclude that the coercive potential was high at the time the trial court opted to deliver a Crowder instruction. “Even in a situation with a high degree of inherent coercive potential,” however, “we have said that coercion may be averted where a court acts with appropriate precaution.” Id. at 363. Thus, we turn next to the actual delivery of the Crowder instruction and consider whether it sufficiently mitigated—or instead exacerbated—the coercive elements operating upon the jury.

B. The Impact of the Crowder Instruction in this Case

As noted above, the Crowder instruction is intended for situations where a jury poll reveals a lack of unanimity. It is not meant to serve as an anti-deadlock instruction. But given that our analysis must be “from the jurors’ perspective,” Hankins, 3 A.3d at 361, the ostensible purpose of the instruction does not matter if

the message received by the jury is otherwise. We conclude that here the court‘s delivery of the Crowder instruction did not substantially reduce the risk of coercion and instead operated like a second anti-deadlock instruction.

Looking to the text of the Crowder instruction, it is in substance quite similar to a Winters instruction. Although the Crowder instruction reminds jurors not to “surrender [their] honest conviction[s],” it also encourages them not to “hesitate to reexamine [their] own views,” 383 A.2d at 343 n.11, much as Winters directs jurors to “seriously” consider if their views are reasonable. 317 A.2d at 534. Moreover, just as the Winters instruction emphasizes the desirability “that the case be decided,” id., so too the Crowder charge emphasizes jurors’ “duty . . . to consult with one another and to deliberate with a view to reaching an agreement.”¹¹ 383 A.2d at 343 n.11. Indeed, because of the similarity between the two instructions, as well as our

concern that the coercive portions could be reinforced, we have held that a Winters instruction should not follow a Crowder instruction. See Benlamine v. United States, 692 A.2d 1359, 1364–65 (D.C. 1997) (reversing where trial court gave Crowder instruction and then Winters instruction after ninth juror dissented in jury poll); see also Davis, 669 A.2d at 685 (reversing where trial court gave Winters instruction after third juror dissented in jury poll).

Apart from these coercive elements, considered in context, the court‘s instruction was an unequivocal directive to the jury that it was not in fact done: it would have to continue to try to reach a verdict. At this point, the jury had been deliberating for nearly two days, focusing on one question—the meaning of the seven-and-a-half-minute compilation of video footage—and had already received both Criminal Jury Instruction 2.601 and the Winters instruction in response to two jury notes reporting deadlock. Cf. Harris, 622 A.2d at 703 (“The factual context of a case can give rise to a situation where an ‘anti-deadlock’ instruction becomes coercive.“). Under the circumstances, there is a significant risk that the jurors understood the Crowder charge to “resume deliberations and see if you can reach a unanimous verdict” as a directive that the court would not let them go until they reached a unanimous verdict. In other words, the Crowder instruction effectively functioned as a second, impermissibly coercive, anti-deadlock instruction. Fortune, 65 A.3d at 85 (“A jury instruction is impermissibly coercive if it would objectively appear to force a juror to abandon his honest conviction as a pure accommodation to the majority of jurors or the court.” (internal quotation marks omitted)).¹²

Because a jury which reports that it has reached an impasse more than once “is particularly vulnerable to pressure to reach a verdict,” see Grant v. United States, 85 A.3d 90, 98 (D.C. 2014), this court has “repeatedly held” that giving a second anti-deadlock instruction is error. Id. at 100; see also id. at 98 (explaining that any instruction following an anti-deadlock charge “must be carefully drawn to ensure that it does not contain the key components of an anti-deadlock instruction“); Epperson, 495 A.2d at 1175–76 (“[R]epeatedly giving ‘hung jury’ instructions to a ‘hung jury’ risks affecting adversely the integrity of a verdict.“); Criminal Jury Instructions for the District of Columbia, Comment to No. 2.601 (5th ed. 2018) (“[A]n anti-deadlock charge generally should only be delivered to a hung jury once.“).

The jurors in this case invested three days of their time to determine if a seven-and-a-half-minute compilation of surveillance video could support a conviction for robbery beyond a reasonable doubt. They ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍“extensively reviewed” the law as explained to them in the court‘s instructions. They asked insightful follow-up questions. They re-watched the video footage. They had “intense discussions.” They were Winterized. And yet they could not reach a unanimous verdict. It was only after their continued internal disagreement was laid bare in a jury poll breakdown and the court gave them a Crowder instruction directing them to continue deliberating that they announced they had found Mr. Callaham guilty of a lesser included offense. Because we conclude there was a substantial risk that one or more jurors felt undue pressure “to abandon [their] honest conviction[s] as a pure accommodation to the majority of jurors or the court,” Fortune, 65 A.3d at 85 (internal quotation marks omitted), we must reverse and remand for a new trial.

III. Admission of Video Footage and Related Testimony

Because we reverse and remand for a new trial, we address to the extent necessary Mr. Callaham‘s arguments related to the admission clips of video footage introduced at trial and testimony about that footage. Specifically, Mr. Callaham argues: (1) the government‘s failure to call the prosecutor who compiled the clips violated Mr. Callaham‘s Sixth Amendment right to confront witnesses against him, (2) Detective Volpe‘s summary of the events depicted in the footage violated the best evidence rule, and (3) the detectives, as non-percipient lay witnesses, should not have been permitted to narrate the events captured on video.

A. Confrontation Clause

Mr. Callaham argues his Sixth Amendment Confrontation Clause right was violated when he was not permitted to cross-examine the prosecutor who compiled the government‘s video exhibits. We review this claim de novo. Carrington v. District of Columbia, 77 A.3d 999, 1003 (D.C. 2013).

The Confrontation Clause protects against the introduction of a “testimonial” statement of an absent witness, Young v. United States, 63 A.3d 1033, 1039 (D.C. 2013) (citing Crawford v. Washington, 541 U.S. 36, 53–54 (2004)), defined as one “‘made for the purpose of establishing or proving some fact’ relevant to investigating or prosecuting a crime.” Carrington, 77 A.3d at 1003–04 (quoting Crawford, 541 U.S. at 51). Mr. Callaham has not demonstrated that the prosecutor who compiled the video exhibits made any such statement. Rather, the prosecutor simply took all the clips from the relevant timefrаme in the authenticated surveillance footage;¹³ combined them sequentially without enhancing, rearranging, or editing them; and repeatedly signaled that they had been stitched together by asking Detective Olszak to explain the varying camera angles.¹⁴ The prosecutor could have moved the non-compiled footage into evidence in the same sequence, as fifteen separate exhibits, but that would have done little other than slow down the presentation of evidence.¹⁵

United States v. Smith, 640 F.3d 358 (D.C. Cir. 2011), to which Mr. Callaham cites, is not to the contrary. Id. at 362–64. Unlike the court clerk in Smith, who wrote a letter certifying that the defendant had a felony conviction, the prosecutor here did not “create a record” for trial. See id. at 363 (quoting Melendez-Diaz v. Mass., 557 U.S. 305, 323 (2009)). Instead, as explained, the prosecutor simply organized the preexisting footage so that all the clips showing the alleged crime (a number of which showed the same time period frоm different vantage points) were played in chronological order.

Accordingly, we conclude that the admission of the compiled videos did not violate Mr. Callaham‘s right to confrontation under the Sixth Amendment.

B. Best Evidence Rule

We are no more persuaded by Mr. Callaham‘s argument that Detective Volpe‘s testimony from memory about the surveillance video‘s content violated the best evidence rule because the government never introduced the “original” surveillance footage. Even assuming this argument was preserved, it is meritless. Pursuant to the best evidence rule, “[a]n original writing, recording, or photograph¹⁶ is [generally] required in order to prove its content.” Fed. R. Evid. 1002; see also Gassaway v. Gassaway, 489 A.2d 1073, 1075, n.4 (D.C. 1985) (explaining that the District‘s practice aligns with the federal rule). But here, the government did introduce the surveillance footage about which Detective Volpe testified, in the form of several clips assembled in the manner described аbove. Mr. Callaham does not allege that the compilations were inaccurate reproductions of the unedited footage and has not shown that the government‘s exhibits were insufficient duplicates under Fed. R. Evid. 1003 (“A duplicate is admissible to the same extent as the original unless a genuine question is raised about the original‘s authenticity or the circumstances make it unfair to admit the duplicate.“). See also Fed. R. Evid. 1001(e) (defining duplicate). Mr. Callaham thus does not have a best evidence argument vis-a-vis Detective Volpe‘s testimony.

C. Non-Percipient Lay Witness Narration of Video Footage

Lastly, Mr. Callaham argues that Detectives Olszak and Volpe improperly “narrated the contents of [the] video[-]recordings” despite not having “witnessed the events depicted” in the videos in real time. We need not address the particulars of their testimony because, while the government would doubtless seek to present the video footage evidence at any retrial, whether and how it might seek to elicit testimony about the videos could change, for example, if Mr. Garvin opted to testify.¹⁷ Accordingly, we address this issue only to make three points:

First, we reaffirm the requirement that lay witness testimony generally must be based on personal knowledge,¹⁸ whether it is proffered as fact¹⁹ or opinion.²⁰

Second, we reject the government‘s argument that the detectives “witnessed” the events in question—and thereby obtained personal knowledge of them—solely by watching recorded surveillance footage.²¹

Third and finally, we endorse the trial court‘s efforts to ensure that the jurors in this case understood that they were the finders of fact and it was for them to decide what the video footage showed.

IV. Conclusion

For the reasons set forth above, we reverse and remand.

So ordered.

Notes

At the time of the incident, Detective Volpe had not yet been promoted from his position as an investigating officer. To avoid confusion, we refer to him as Detective Volpe throughout.

D.C. Code §§ 22-2801, -4502 (2012 Repl. & 2021 Supp.).

D.C. Code § 22-4504(b) (2012 Repl. & 2021 Supp.).

The government called two Mellon Market employees to the stand, but both stated that they left the immediate area before the alleged robbery occurred, one upon seeing a man holding what he believed to be a gun, and the other upon hearing cries that someone had a gun.

The exact timing is unclear, but it was likely аlmost lunchtime, as the government had given its rebuttal and the court had read the jury instructions that same morning.

(en banc); see also Criminal Jury Instructions for the District of Columbia, No. 2.601(III) (5th ed. 2018).

The penultimate line of the note, crossed out but still legible, said “one or more of us still have a reasonable doubt.”

The instruction is intended to address “a jury ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍that, in polling, simply reveals a split.” . The first paragraph is the standard language. The second and third paragraphs, which appear in brackets in the model instruction, “have been recommended [by this court in Crowder, 383 A.2d at 342 n.11] for use in cases where there is a particularly high likelihood of juror coercion.” Comment to Criminal Jury Instructions for the District of Columbia, No. 2.603 (5th ed. 2018). We refer to this entire instruction as the Crowder instruction.

The government argues that the delivery of a Winters instruction in response to the jury‘s second deadlock note was not an abuse of discretion because it was “either neutral or servеd to alleviate pressure on the jury.” Whether the court abused its discretion in giving a Winters instruction is not the issue before us and we do not so hold. But the fact that the court could have reasonably chosen to give the Winters instruction does not mean that choice was a “neutral” action, nor does it preclude us from concluding based on subsequent events that the instruction had a coercive effect.

Because the trial court appropriately stopped the poll after the second juror dissented from the announced guilty verdict, the court did not learn the precise numerical split, nor is it clear that the second juror was the only dissenter. See (also identifying as factors to consider “the degree of isolation of a dissenting juror (or jurors)” and “whether the exact numerical division of the jury is revealed.“). But we evaluate the “risk [of coercion] on a continuum” and all the factors listed in Harris need not be present for a court to conclude that a jury poll break down contributed to a substantial risk of coercion. .

Notably, the “coercion reducing elements” of the bracketed language in the Crowder poll breakdown instruction are actually derived from an American Bar Association charge which we have concluded “was fashioned to be, and is nationally regarded by the courts and scholars, as being an anti-deadlock instruction.” ; see also (quoting ABA Project on Minimum Standards for Criminal Justice, Trial by Jury, Commentary, § 5.4(a) (Approved Draft, 1968)); . What is more, the language in the bracketed portion of the Crowder instruction also appears as an alternative anti-deadlock instruction to the Winters instruction in Criminal Jury Instruction 2.601. See Criminal Jury Instructions for the District of Columbia, No. 2.601(III.A) (5th ed. 2018) (“Thomas instruction“).

Our assessment that the Crowder instruction did not ameliorate the situation is reinforced by what happened next. The jury‘s swift return of a unanimous verdict after receiving this instruction is some evidence that the jurors settled their disagreements about whether the surveillance footage showed a robbery simply so they could put an end to the process of deliberation. Compare (finding no substantial risk of a coerced verdict where jury deliberated for several hours before returning a unanimous verdict after receiving the Winters instruction), with (observing that the jury‘s ninety-minute deliberation before returning a unanimous verdict after receiving an anti-deadlock instruction “tend[ed] to show coercive effect” of the instruction). And the fact that the jury ultimately acquitted Mr. Callaham of armed robbery—the charge it had earlier indicated Mr. Callaham was guilty of—and possession of a firearm during the commissiоn of a crime of violence, and announced a verdict on the lesser included offense of robbery, suggests a compromise verdict born of coercion not persuasion. The government‘s theory of the case was that Mr. Callaham had acted in concert with the man with the gun to rob Mr. Garvin. The defense‘s theory of the case was that the video did not prove that a robbery, armed or otherwise, had in fact occurred. Finding that the video depicted Mr. Callaham committing a robbery but not an armed robbery seems less plausible than either theory. Cf. (where trial counsel asserted a misidentification defense, reasoning that unanimous verdict acquitting appellant of first-degree premeditated murder and convicting of lesser-included offense of second-degree murder “may well have been a compromise,” and thus “tend[ed] to substantiate appellant‘s contention that the verdict was coerced“).

The surveillance videos were authenticated, over defense objection, by the market owner and an MPD employee, and Detective Olszak testified that the government‘s exhibits consisted of the same surveillance videos he reviewed in his investigation. Mr. Callaham has not renewed his objection, on appeal, to the authentication of the videos.

While Mr. Callaham—who received the unedited videos in discovery—argued in closing that the government pieced the videos together “to tell the story they wanted to tell,” he did not contend that material portions of the videos were missing or altered.

It also bears noting that Mr. Callaham did not ask for the full videos to be admitted into evidence, for example under the rule of completeness. See, e.g., .

Fed. R. Evid. 1001(c) defines a photograph as “a photographic image or its equivalent stored in any form.”

For this reason too, we need not address the government‘s argument that Mr. Callaham only partially preserved this claim.

But see, e.g., Fed. R. Evid. 1006 (authorizing the use of summary witnesses in cases where the evidence is voluminous).

Fed. R. Evid. 602 (“A witness may testify to a matter only if evidence is introduced sufficient to support a finding that the witness has personal knowledge of the matter.“); (endorsing the personal knowledge requirement in Fed. R. Evid. 602 for lay witnesses); see also (“The testimony of witnesses is admissible if predicated upon concrete facts within their own senses, as distinguished from their opinions or conclusions drawn from such facts.” (internal quotation marks omitted)).

Fed. R. Evid. 701 (“If a witness is not testifying as an expert, testimony in the form of an opinion is limited to one that is: (a) rationally based on the witness‘s perception; (b) helpful to clearly understanding the witness‘s testimony or to determining a fact in issue; and (c) nоt based on scientific, technical, or other specialized knowledge within the scope of Rule 702 [governing testimony of expert witnesses]“); (endorsing the personal knowledge requirement of Fed. R. Evid. 701 for lay opinion testimony); (quoting Fed. R. Evid. commentators’ admonition that the rule permitting lay opinion testimony “is not designed to encourage speculation on the part of the witnesses concerning events that they have not perceived“); see also (holding that lay witness opinion testimony identifying the defendant as the person depicted in surveillance footage may be admitted only if (1) the testimony is rationally based on the perception of a witness who has sufficient familiarity with the defendant‘s appearance because of their “substantial contact with the defendant,” (2) the testimony is “helpful to the factfinder in the determination of a fact in issue,” and (3) the court is “reasonably satisfied that . . . the lay witness is more likely to accurately identify the defendant thаn is the factfinder“).

See, e.g., (concluding witnesses lacked personal knowledge of events captured in video when they “did not perceive [the events] in real time“); cf. (affirming conviction where trial court expressly limited witness’ narration of surveillance video to the portions showing events to which he was an eyewitness).

Holmes v. United States, 92 A.3d 328 (D.C. 2014), cited by the government, does not hold otherwise. The only issue before the court in Holmes was whether a police officer‘s testimony about his observation of a live ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍feed from a surveillance camera in a clothing store constituted hearsay. Id at 331. We concluded that a surveillance system is “not a person” but “a tool,” and a tool cannot “make[] an ‘assertion to the [testifying] witness‘“; thus the surveillance system did not generate an out of court statement offered for the truth of the matter asserted. Id. Holmes did not address, much less qualify, the requirement that a lay witness testify based on their own personal knowledge.

Judge Thompson was an Associate Judge of the court at the time of argument. Although her term expired on Septembеr 4, 2021, she will continue to serve as an Associate Judge until her successor is confirmed. See D.C. Code § 11-1502 (2012 Repl.). She was qualified and appointed on October 4, 2021, to perform judicial duties as a Senior Judge and will begin her service as a Senior Judge on a date to be determined after her successor is appointed and qualifies.

Read the detailed case summary