Walker v. Commonwealth

PRESENT: Goodwyn, C.J., Powell, Kelsey, McCullough, Chafin, and Mann, JJ., and Koontz, S.J. JACQUES LAMAR WALKER OPINION BY v. Record No. 220378 JUSTICE STEPHEN R. McCULLOUGH June 1, 2023 COMMONWEALTH OF VIRGINIA FROM THE COURT OF APPEALS OF VIRGINIA This appeal presents two primary issues. First, we consider whether the Due Process Clause of the United States Constitution requires a court to pre-screen eyewitness identification testimony before the eyewitness can be permitted to make an identification of the defendant for the first time in open court. We also examine whether the jury was properly instructed that the defendant was subject to the second or subsequent offense provisions of Code § 18.2-53.1 for using a firearm in the commission of robbery. With regard to the first issue, in accord with the majority of courts that have considered this question, we conclude that the Due Process Clause does not require a court to pre-screen an eyewitness identification made for the first time in court. Second, we hold that the jury was properly instructed on the second or subsequent offense provisions of Code § 18.2-53.1. BACKGROUND I. A WELLS FARGO BANK IS ROBBED AT GUNPOINT BY A MASKED INDIVIDUAL. On the afternoon of May 23, 2016, a man wearing a mask, a hooded sweatshirt, and a distinctive yellow work vest “barge[d]” into the lobby of a Wells Fargo bank branch. Edlin Cottrell, one of the tellers, heard a scream, which another teller described as a “blood curdling scream.” Cottrell saw the masked man and ducked behind her desk. The robber struck a customer and knocked him down. The robber then went behind the teller line to where the bank employees were. He pointed a gun at the tellers and told them in a “demanding,” and “rushed” voice to put money in a bag. His demeanor was “forceful.” The bag, a black bag with handles, appeared to be the kind of plastic bag used at liquor stores. The teller complied, as did other bank employees. A third teller mostly hid under her desk. Once the money was placed in the bag, the robber then ran out the front door. All told, the robber ran off with over $15,000. The mask covered the robber’s nose and mouth. His eyes and skin tone were visible. Based on skin tone, he was described as black. A teller described the robber as medium build and maybe 5’8” or 5’9”. A person in the parking lot observed the robber exiting the bank and getting into the passenger side of a white Acura automobile. Another person was driving. II. THE INVESTIGATION AND WALKER’S ARREST. The Wells Fargo bank branch that was robbed is located at Tackett’s Mill Shopping Center in Prince William County. Approximately four hours before the robbery of the Wells Fargo bank, an employee of a different bank, a branch of BB&T, located a few miles from the Wells Fargo bank that was robbed, heard a report of suspicious activity. Acting on this report, she followed a suspicious vehicle with two men in it. The vehicle was a late model white Acura. She wrote down the license plate and reported it to the police. She also was able to observe the two men when they stepped out of the car. One of the men was “shorter and . . . stockier” and the other was “taller and thinner.” The thinner, taller man was in the passenger seat. The thinner of the two was wearing a “yellow shirt.” She saw their faces. She identified Walker at 2 trial as one of the men in the white Acura. Police later learned that an Acura with the reported license plate was registered to a Jacques Walker. Some of the money the Wells Fargo tellers placed in the bag was “strapped money,” meaning a stack of bills wrapped with a band. The straps have a Wells Fargo Bank stamp on them, which has a date and a number for a particular branch. These straps also bear the initials of a service manager and a particular teller. These straps help ensure the money is accounted for and help the tellers identify certain bundled amounts, such as $500 or $5,000. A police officer who responded to the 9-1-1 call from the bank robbery observed a yellow vest on the ground, by the road. Police collected the vest, as well as additional items of clothing, including a black ski mask and a pair of jeans, further down the road from the Wells Fargo bank that was robbed. After the robbery, one of the bank customers, the man struck by the robber, noticed a pistol on a mulch pile in the bed of his pickup truck. He called the police, who retrieved the gun. The gun turned out to be a BB gun. Two days after the robbery, on May 25, 2016, police arrested Walker in Maryland, after a traffic stop. The white Acura he was traveling in was being driven 91 miles per hour in a 55 mile per hour zone. The license plate matched the license plate for the white Acura that the BB&T bank employee had noted. Walker was the passenger in the vehicle. Walker had about $2,600 in cash in his pocket. There was also a suitcase in the back seat. It contained a backpack, and inside that backpack was a black plastic bag with $9,060 in cash. The cash from the suitcase was wrapped in the distinctive bands from the Wells Fargo bank. Walker told the police in Maryland that they were headed to a casino. Police also recovered two cell phones. 3 When questioned by the police about the Wells Fargo robbery, Walker stated that he was working for an employment service called Labor Ready that day, on a moving job in Warrenton. He said he worked from 8:00 a.m. to 3:00 p.m. A work ticket from Labor Ready, however, showed that Walker had worked four hours that day. The evidence established that four hours is the minimum amount of pay a worker receives; the actual hours worked could have been less. A manager for Labor Ready confirmed that Walker worked on May 23, 2016, for a moving job. She identified the yellow vest worn by the robber as “the same type” of vest the company provides to persons who work for them. She identified Walker as the bank robber from a still photo. When asked how she could identify Walker as the robber, when the robber was wearing a mask, she responded: Well, he started working for us in the colder months so a lot of the workers come in very bundled up so you learn to recognize them when they’re coming in. So and I had worked with him for four months, he’d been coming in every day just about asking for work so after a while you really get to know your workers and so I recognized him. App. 236. A search of the phones recovered from the Acura in Maryland revealed that someone used one of the phones to pull up an article titled A Man in a Yellow Vest Robs Tackett’s Mill Wells Fargo. A forensic analysis of the search history of the phone revealed a query for “do sweat have DNA.” DNA testing on the mask recovered by the side of the road revealed a mixture. Walker could not be eliminated as the major contributor to the mixture. No DNA was recovered from the yellow vest found by the road. The pair of jeans found near the road also contained a DNA mixture that was a match for Walker. 4 Walker offered alibi evidence from a “good friend” and five-time convicted felon, who testified that Walker worked on the moving job all day, from around noon to about 6:00 p.m. Walker himself testified to that effect. However, this evidence differed in a number of ways from the alibi he provided to a detective just days after the robbery. Walker also offered evidence from another convicted felon that a man named “Mike” may have committed the robbery. Walker testified that he had no idea there was money in the car when it was pulled over in Maryland. He blamed the robbery on his brother. III. MOTION TO EXCLUDE EYEWITNESS TESTIMONY AND TRIAL. Walker was indicted on four counts of robbery and four counts of use of a firearm in the commission of a felony, as well as one count of abduction for pecuniary benefit. Walker filed a pro se, handwritten motion in limine in which he argued, among other things, that the court should “exclude any in court witness identification of the defendant, unless the witness has already been vetted in their ability to identify the suspect in a blind 6-man photo line up with only [their] eyes showing, because the suspect wore a ski mask.” He contended that “[t]he [r]isk is to[o] high of suggestive identification because the defendant is here in the courtroom.” He also argued that such an identification would be more prejudicial than probative. After hearing argument on the motion by Walker, who was assisted by standby counsel, the court denied the motion. At a jury trial, one of the bank employees, Irene Caison, testified that she saw the robber come in the bank lobby and she noticed he was holding a gun. She initially dropped down behind the counter. She soon became worried that the robber would kill her if she did not get up, so she did. When the robber demanded money, she complied, filling the bag he provided with cash from different drawers. She testified she was “looking at him,” and “staring at his eyes, 5 because [that was] all [she] could see.” She tried to keep the robber calm by saying “we’re going to give you the money.” Caison was asked, “[d]o you see anyone in the courtroom today from those events on May 23, 2016?” She answered in the affirmative. She was then asked “[w]ho do you recognize?” to which she answered, identifying Walker, “[i]t’s the gentleman right there.” She testified that she recognized his eyes. She said that the robber was “[v]ery close,” within arm’s reach, that she could have touched him. She was not asked to review a photo lineup before trial. When Walker, representing himself, asked Caison if she saw any other part of the individual’s face, she said “just your eyes, your eyebrows, and your skin complexion.” She said she had also recognized him at the preliminary hearing. In its instructions to the jury, the circuit court included an instruction on eyewitness testimony, based on a model jury instruction. See Virginia Model Jury Instructions – Criminal, No. 2.800, Note on Eyewitness Identification (2018-19 repl. ed.). The jury found Walker guilty of abduction for pecuniary benefit, four counts of robbery, and four counts of using a firearm in the commission of robbery. The circuit court also instructed the jury to fix Walker’s punishment at five years for three of the four use of a firearm charges, as “second or subsequent offense[s]” under Code § 18.2-53.1. He was sentenced to serve an active sentence of 45 years in the penitentiary. IV. POST-TRIAL MOTION AND APPEAL. Walker filed a post-trial motion, again raising the due process issue based on Caison’s in-court identification. The circuit court denied the motion. 6 On appeal, a divided panel of the Court of Appeals affirmed. Walker v. Commonwealth, 74 Va. App. 475 (2022). We granted Walker an appeal based on the following assignments of error: I. The Court of Appeals erred by ruling that the trial court did not err in admitting Ms. Caison’s in-court identification of Mr. Walker as the perpetrator or in refusing Mr. Walker’s request to implement protective procedures. II. A. The Court of Appeals erred by ruling that the trial court did not err in instructing the jury at sentencing that the second, third, and fourth guilty findings were to be treated as second or subsequent convictions subject to the enhanced sentencing provision of Code § 18.2-53.1. B. If the Court of Appeals is correct that Ansell v. Commonwealth, 219 Va. 759 (1979) is controlling (and not distinguishable), Ansell should be modified, overturned, or reversed. ANALYSIS I. THE DUE PROCESS CLAUSE DOES NOT REQUIRE JUDICIAL PRE-SCREENING OF EYEWITNESS IDENTIFICATION BEFORE A WITNESS MAKES AN IDENTIFICATION FOR THE FIRST TIME IN COURT. The Due Process Clauses of the Fifth and Fourteenth Amendments to the United States Constitution provide that no person shall be deprived “of life, liberty, or property, without due process of law.” 1 The phrase “due process of law” traces its origin to “[a] series of statutes enacted during Edward III’s reign.” Andrew T. Bodoh, The Road to “Due Process”: Evolving Constitutional Language from 1776 to 1789, 40 T. Jefferson L. Rev. 103, 113 (2018). These statutes “specified and updated the interests protected and the nature of the protections afforded under the Magna Carta’s law of the land and denial of justice clauses. These statutes are the original link between due process language and the Magna Carta.” Id. One of these statutes, 1 Article I, Section 11 of the Virginia Constitution likewise provides “[t]hat no person shall be deprived of his life, liberty, or property without due process of law.” The parties do not ask us to construe that Clause. 7 from 1363, “interpreted the Magna Carta as prohibiting taking or imprisoning a man, or putting him out of his freehold, ‘without process of the law.’” Id. at 113-14. Relying on a series of United States Supreme Court decisions construing the Due Process Clause, Walker argues that an identification that is made for the first time in court is the kind of suggestive, improper identification procedure that the Due Process Clause regulates. He contends that an identification by an eyewitness that is made for the first time in court should be subject to judicial pre-screening, at least in some instances. 2 “[Q]uestions of constitutional interpretation . . . are questions of law subject to de novo review” on appeal. Ashland, LLC v. Virginia-American Water Co., 301 Va. ___, ___, 878 S.E.2d 378, 381 (2022). There is no denying that an in-court identification is suggestive. The defendant is seated at counsel table, in the courtroom. A witness then points, sometimes with dramatic flair, to the defendant as the culprit. The question is whether this practice is improperly or unnecessarily suggestive within the intendment of the Due Process Clause. First, it is common ground that the United States Supreme Court has not construed the Due Process Clause to require pre-screening by a trial court of in-court identifications. The Court’s most recent decision to address eyewitness identifications is Perry v. New Hampshire, 565 U.S. 228 (2012). Perry, authored by Justice Ginsburg, explained that, as a general proposition, “[t]he Constitution . . . protects a defendant against a conviction based on evidence of questionable reliability, not by prohibiting introduction of the evidence, but by affording the 2 The Commonwealth argues that Walker’s due process argument is procedurally defaulted. Although Walker’s pro se motion could have been crafted with greater precision, we conclude that it was sufficient to alert the court that Walker was raising a due process challenge to the admissibility of eyewitness testimony. Therefore, we proceed to resolve the issue on the merits. 8 defendant means to persuade the jury that the evidence should be discounted as unworthy of credit.” Id. at 237. The Court then addressed a subset of cases involving improperly suggestive, out of court police procedures, such as Neil v. Biggers, 409 U.S. 188 (1972) and Manson v. Brathwaite, 432 U.S. 98 (1977). The Court held that these cases set forth “the approach appropriately used to determine whether the Due Process Clause requires suppression of an eyewitness identification tainted by police arrangement.” 565 U.S. at 238. The Court noted that “[a] primary aim of excluding identification evidence obtained under unnecessarily suggestive circumstances . . . is to deter law enforcement use of improper lineups, showups, and photo arrays in the first place.” Id. at 241. The Court went on to say that “the Due Process Clause does not require a preliminary judicial inquiry into the reliability of” eyewitness identifications that are “not procured under unnecessarily suggestive circumstances arranged by law enforcement.” Id. at 248. Neither Perry, nor the cases that precede it, support the argument that the Due Process Clause applies to in-court identifications. Instead, they manifest a concern for out-of-court, improperly suggestive police procedures. In United States v. Domina, the United States Court of Appeals for the Ninth Circuit explained the different dynamic of an in-court identification versus an out of court identification: The concern with in-court identification, where there has been suggestive pretrial identification, is that the witness later identifies the person in court, not from his or her recollection of observations at the time of the crime charged, but from the suggestive pretrial identification. Because the jurors are not present to observe the pretrial identification, they are not able to observe the witness making that initial identification. The certainty or hesitation of the witness when making the identification, the witness’s facial expressions, voice inflection, body language, and the other normal observations one makes in everyday life when judging the reliability of a person’s statements, are not available to the jury during this pretrial proceeding. There 9 is a danger that the identification in court may only be a confirmation of the earlier identification, with much greater certainty expressed in court than initially. When the initial identification is in court, there are different considerations. The jury can observe the witness during the identification process and is able to evaluate the reliability of the initial identification. 784 F.2d 1361, 1368 (9th Cir. 1986) (citation omitted). Second, the Supreme Court’s decision in Perry points to a fundamental concern of our Constitution: the mechanism for determining the reliability of evidence is generally the trial itself, not a pretrial weeding out of evidence by judges. Perry is not an isolated case. In Kansas v. Ventris, 556 U.S. 586, 594 n.* (2009), for example, the United States Supreme Court declined to fashion a broad exclusionary rule for statements made by jailhouse informants on the basis that such statements are “inherently unreliable.” The Court observed that “[o]ur legal system . . . is built on the premise that it is the province of the jury to weigh the credibility of competing witnesses.” Id. Additionally, addressing a separate constitutional right, the right to confrontation, the Supreme Court had this to say: Admitting statements [merely because they are] deemed reliable by a judge is fundamentally at odds with the right of confrontation. To be sure, the Clause’s ultimate goal is to ensure reliability of evidence, but it is a procedural rather than a substantive guarantee. It commands, not that evidence be reliable, but that reliability be assessed in a particular manner: by testing in the crucible of cross- examination. The Clause thus reflects a judgment, not only about the desirability of reliable evidence (a point on which there could be little dissent), but about how reliability can best be determined. Crawford v. Washington, 541 U.S. 36, 61 (2004). Criticizing its prior test, which called for courts to assess the reliability of evidence, the Supreme Court observed that judicial determinations of reliability improperly “replace[d] the constitutionally prescribed method of assessing reliability with a wholly foreign one.” Id. at 62. 10 Developing rules to address conduct outside the courtroom that produces tainted evidence in the courtroom is one thing. It is another thing to extend these rules, designed to address flawed police practices prior to trial, to the presentation of evidence in the courtroom when there has been no allegation of impropriety prior to trial. Extending due process this far would supplant the truth-finding function of a trial. The trial is the crucible that is designed to determine the reliability of this evidence.3 Third, in construing the Due Process Clause, we are engaged in an interpretive exercise, not a policy-making one. In ascertaining the meaning of constitutional provisions, courts have often examined the historical understanding of words and phrases as well as longstanding practices. See, e.g., United States v. Jones, 565 U.S. 400, 411 (2012) (applying what the Court characterized as “an 18th-century guarantee against unreasonable searches” to conclude that the installation of a GPS tracking device was a search); In re Winship, 397 U.S. 358, 362 (1970) (the standard of proof beyond a reasonable doubt is a “historically grounded right[]” (quoting Brinegar v. United States, 338 U.S. 160, 174 (1949)); Yoder v. Commonwealth, 107 Va. 823, 827-28 (1907) (drawing from the common law to conclude that summary petty contempt does not require trial by jury); Cullen v. Commonwealth, 65 Va. (24 Gratt.) 624, 628 (1873) (relying on the common law to expansively construe the right to avoid compelled self-incrimination). The practice of in-court identifications has existed continuously for centuries. This long and 3 It is not clear why, conceptually, a due process requirement to judicially pre-screen evidence would be limited to in-court identifications. The same logic, if adopted, would lead to pre-trial screening of statements by jailhouse informants, undercover operatives used to purchase narcotics, witnesses with undeniable biases, and more. It is also unclear whether this judicial pre-screening would be limited to evidence offered by the prosecution, or whether it would extend to evidence offered by the defense. 11 uninterrupted historical practice is a powerful indicator that the Due Process Clause does not require a judicial pre-screening of identifications made for the first time in court. Fourth, experts who have studied eyewitness testimony have criticized the test adopted by the United States Supreme Court to address the reliability of eyewitness testimony. The five factors the Court articulated in Neil, 409 U.S. at 199-200 and Manson, 432 U.S. at 114, “were drawn from earlier judicial rulings and not from scientific research.” National Research Council, Identifying the Culprit: Assessing Eyewitness Identification 32 (2014). “As critics have pointed out, the Manson v. Brathwaite test includes factors that are not diagnostic of reliability.” Id. at 6. Extending to new horizons a test that authorities on eyewitness testimony universally view as flawed hardly seems like the right solution. Fifth, we note that, although a minority of courts have concluded that the Due Process Clause regulates eyewitness identifications made for the first time in court, 4 a strong majority of courts have rejected such an interpretation of the Clause. 5 4 United States v. Greene, 704 F.3d 298, 305-07 (4th Cir. 2013); United States v. Rogers, 126 F.3d 655, 658-59 (5th Cir. 1997); United States v. Archibald, 734 F.2d 938, 941-43 (2d Cir. 1984); State v. Dickson, 141 A.3d 810, 820-27 (Conn. 2016). Two of these federal decisions pre-date Perry. Additionally, the Supreme Court of Massachusetts has relied on “common law principles of fairness,” as opposed to the Due Process Clause, to impose a “good reason” requirement for identifications that are made for the first time in court. Commonwealth v. Crayton, 21 N.E.3d 157, 169 & n.16 (Mass. 2014) (cleaned up). 5 See, e.g., United States v. Thomas, 849 F.3d 906, 910-11 (10th Cir. 2017); Lee v. Foster, 750 F.3d 687, 691-92 (7th Cir. 2014); United States v. Hughes, 562 F. App’x 393, 398 (6th Cir. 2014); United States v. Whatley, 719 F.3d 1206, 1214-17 (11th Cir. 2013); Domina, 784 F.2d at 1368-69; Garner v. People, 436 P.3d 1107, 1119-20 (Colo. 2019) (en banc); Byrd v. State, 25 A.3d 761, 767 (Del. 2011); State v. Doolin, 942 N.W.2d 500, 511-15 (Iowa 2020); Fairley v. Commonwealth, 527 S.W.3d 792, 799-800 (Ky. 2017); State v. King, 934 A.2d 556, 561 (N.H. 2007); State v. Ramirez, 409 P.3d 902, 912-13 (N.M. 2017); People v. Brazeau, 759 N.Y.S.2d 268, 271 (App. Div. 2003); State v. Hickman, 330 P.3d 551, 572 (Or. 2014) (en banc), modified on reconsideration, 343 P.3d 634 (Or. 2015) (en banc) (per curiam); State v. Lewis, 609 S.E.2d 515, 518 (S.C. 2005). 12 For all of these reasons, we conclude that the Due Process Clause does not require a court to conduct a pre-screening of an eyewitness’s testimony before that witness is permitted to identify the defendant for the first time in court. In concluding that the Due Process Clause does not require the pre-screening of in-court identifications, we are not blind to perils associated with eyewitness testimony. Over 50 years ago, the United States Supreme Court observed that “[t]he vagaries of eyewitness identification are well-known” and that “the annals of criminal law are rife with instances of mistaken identification.” United States v. Wade, 388 U.S. 218, 228 (1967). Since then, a large body of research has shaped our understanding of the way human memory works and how it degrades over time, the heightened risks associated with identifying a person of another race, the impact of conducting different types of lineups, and many other facets of eyewitness testimony. 6 There are ways to address the problem of flawed eyewitness testimony, however, without tearing down our long-accepted understanding of the Due Process Clause to rebuild it on the 6 The impact of flawed eyewitness testimony in the context of actual cases, as opposed to social science research, is difficult to determine. There have been many exonerations, based on DNA evidence, in which erroneous eyewitness testimony played a central or a contributing role. The number of these DNA-based exonerations is tiny in comparison to the overall volume of criminal convictions. One difficulty in determining the real-world impact of flawed eyewitness testimony is that the overwhelming majority of criminal cases, more than ninety percent, are resolved by guilty pleas. See Virginia Criminal Sentencing Commission, Annual Report 25 (2022) (92% of criminal cases in Virginia in 2022 were resolved by a guilty plea). Most of the cases that do proceed to trial, as this case exemplifies, do not rely exclusively on eyewitness testimony. Furthermore, of the subset of criminal cases that do proceed to trial, a small number are jury trials. Id. (reflecting 1% of criminal cases were tried to a jury in 2022). Judges often sift evidence differently than juries. Finally, eyewitness testimony is generated in a wide variety of ways. Some witnesses will experience protracted exposure to a defendant over a period of hours or even days. Other witnesses will have only very brief encounters. Some witnesses will have better eyesight, observational skills, and memory than others. Some identifications will be made by persons who have known a defendant very well for a long period of time and others from witnesses who have never before seen the defendant. Therefore, although the DNA exonerations no doubt understate the number of convictions associated with flawed eyewitness testimony, it is difficult to assess to what extent such testimony results in convictions of innocent persons. 13 shifting sands of social science. Of course, a defendant can make use of the traditional safeguards of the right to counsel, the right to present evidence, and cross-examination to expose mistaken eyewitness testimony. See, e.g., Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 596 (1993) (concluding that “[t]he[ ] conventional devices” of “[v]igorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence. . . . rather than [its] wholesale exclusion.”). Additionally, law enforcement officers can be (and many have been) trained on using best practices when conducting lineups. That training can be modified as our understanding of the issue progresses. Judges and lawyers can also be (and many have been) educated on the subject. Legislatures also have a role to play. For example, the General Assembly has required law enforcement agencies in Virginia to “establish a written policy and procedure for conducting in-person and photographic lineups.” Code § 19.2-390.02. Finally, a jury can receive an instruction on the subject of eyewitness testimony. Whether to grant such an instruction is a matter of discretion for the trial judge. Daniels v. Commonwealth, 275 Va. 460, 466 (2008). In this case, the circuit court did just that, using an instruction developed by the Model Jury Instructions Committee. See Model Jury Instruction No. 2.800, Note on Eyewitness Identification. For all of these reasons, we hold that the Due Process Clause did not compel the circuit court to pre-screen the identification Caison made of the defendant, when that identification was made for the first time in court. II. THE CIRCUIT COURT DID NOT ABUSE ITS DISCRETION IN RULING THAT CAISON’S IDENTIFICATION WAS MORE PROBATIVE THAN PREJUDICIAL. Walker also argues that the circuit court erred in concluding that the evidence was more probative than prejudicial. Caison’s identification of Walker as the robber was certainly 14 relevant. “Relevant evidence may be excluded if . . . the probative value of the evidence is substantially outweighed by . . . the danger of unfair prejudice.” Va. R. Evid. 2:403(a)(i). It is well-settled that “[t]he responsibility for balancing the competing considerations of probative value and prejudice rests in the sound discretion of the trial court.” Ortiz v. Commonwealth, 276 Va. 705, 715 (2008) (alteration in original) (quoting Spencer v. Commonwealth, 240 Va. 78, 90 (1990)). When balancing these considerations, it is of course true that “all probative direct evidence generally has a prejudicial effect to the opposing party.” Lee v. Spoden, 290 Va. 235, 251 (2015). “‘[U]nfair prejudice’ refers to the tendency of some proof to inflame the passions of the trier of fact, or to invite decision based upon a factor unrelated to the elements of the claims and defenses in the pending case.” Id. The term “speaks to the capacity of some concededly relevant evidence to lure the factfinder into declaring guilt [or liability] on a ground different from proof specific to the [case elements].” Id. (alterations in original) (quoting Old Chief v. United States, 519 U.S. 172, 180 (1997)). “‘Unfair prejudice’ within its context means an undue tendency to suggest decision on an improper basis, commonly, though not necessarily, an emotional one.” Id. at 251-52 (quoting Old Chief, 519 U.S. at 180 (construing the federal analogue to Va. R. Evid. 2:403)). Given the deference we accord to trial judges in balancing probative value with undue prejudice, we conclude that the circuit court did not abuse its discretion in weighing the probative nature of the in-court identification versus any risk of undue prejudice. Caison’s identification of Walker as the robber was certainly probative. And Walker’s argument of “undue prejudice” turns on the claim that Caison’s identification was not reliable. As noted above, however, the reliability of the identification is to be tested by cross-examination and by 15 assessing the identification in the context of the overall evidence. Accordingly, we affirm the judgment of the Court of Appeals on this point. III. WALKER WAS PROPERLY CONVICTED OF A SECOND OR SUBSEQUENT OFFENSE OF USE OF A FIREARM IN THE COMMISSION OF A FELONY. Code § 18.2-53.1 prohibits the use or display of a firearm in the commission of certain felonies, including robbery. The statute imposes “a mandatory minimum term of five years for a second or subsequent conviction under the provisions of this section.” Id. Walker challenges the imposition of a heightened punishment for a second or subsequent offense, noting that, here, all of the offenses occurred in a single event, the robbery of the Wells Fargo bank on May 23, 2016. We review a circuit court’s interpretation of statutes de novo. Jones v. Williams, 280 Va. 635, 638 (2010). A penal statute, such as the one at issue, “must be strictly construed and any ambiguity or reasonable doubt as to its meaning must be resolved in [the defendant’s] favor. This does not mean, however, that [the defendant] is entitled to a favorable result based upon an unreasonably restrictive interpretation of the statute.” Ansell v. Commonwealth, 219 Va. 759, 761 (1979). In Ansell we construed the second or subsequent offense provisions of this statute to apply when a defendant entered a simultaneous guilty plea to multiple charges. In that case, the defendant pled guilty to two charges of robbery, one charge of attempted robbery, and three charges of use of a firearm in the commission of a felony. Id. at 760-61. All of the offenses occurred in a period of 45 minutes. Id. at 761. The circuit court sentenced the defendant to a recidivist punishment for a second or subsequent offense on two of the firearm counts. Id. The defendant objected to the enhanced punishment. Id. We affirmed the circuit court judgment, reasoning that “[t]o give the statute the construction sought by Ansell would enable an offender who used a firearm in the commission of a series of serious felonies over an extended period of 16 time prior to his apprehension to enjoy the status of a first offender as to each violation.” Id. at 763. We concluded that such a “result would negate the legislative intent and would require an unreasonably restrictive interpretation of the statute.” Id. We also noted that the purpose of the statute was to deter violent criminals rather than to reform them. Id. The United States Supreme Court examined a comparably worded statute in Deal v. United States, 508 U.S. 129 (1993), superseded by statute, First Step Act of 2018, Pub. L. No. 115-391, § 403(a), 132 Stat. 5194, 5221-22 (2018), as recognized in United States v. Davis, __ U.S. __, __, 139 S. Ct. 2319, 2324 n.1 (2019). That statute provided: Whoever, during and in relation to any crime of violence . . . uses or carries a firearm, shall, in addition to the punishment provided for such crime of violence . . . , be sentenced to imprisonment for five years . . . . In the case of his second or subsequent conviction under this subsection, such person shall be sentenced to imprisonment for twenty years. Id. at 130 (omissions in original) (quoting 18 U.S.C. § 924(c)(1) (1988 ed., Supp. III)). The phrase “second or subsequent conviction” at issue in Deal is the same phrasing used in Code § 18.2-53.1. The Supreme Court explained that although “[a] judgment of conviction includes both the adjudication of guilt and the sentence,” in this context, “‘conviction’ refers to the finding of guilt by a judge or jury that necessarily precedes the entry of a final judgment of conviction.” Id. at 132. The Court reasoned that “if ‘conviction’ in § 924(c)(1) meant ‘judgment of conviction,’ the provision would be incoherent, prescribing that a sentence which has already been imposed (the defendant’s second or subsequent ‘conviction’) shall be 5 or 20 years longer than it was.” Id. We find this reasoning persuasive. In 2018, Congress changed the federal statutory law so that, going forward, only a second § 924(c) violation committed “after a prior [§ 924(c)] conviction . . . has become final” triggers the 25-year minimum. First Step Act of 2018, Pub. L. No. 115-391, § 403(a), 132 Stat. 17 5194, 5221-22 (2018). In other words, Congress responded to the Supreme Court decision in Deal by altering the statute. In contrast, the General Assembly has not elected to amend the recidivist provisions of Code § 18.2-53.1 that we construed in Ansell. Walker seeks to analogize his case to Batts v. Commonwealth, 30 Va. App. 1 (1999), but we find the comparison inapposite. In that case, the defendant had been convicted of a firearm offense in one proceeding, but the sentencing had been continued. Id. at 5-6. In a separate trial for robbery and use of a firearm, the Commonwealth sought to use the conviction in the first proceeding to enhance the punishment for the firearm charge in the second proceeding. Id. The Court of Appeals held that the jury should not have been instructed on the enhanced punishment provision of Code § 18.2-53.1 because – although a jury in a separate proceeding had returned a verdict of conviction – no final order had been entered. Id. at 12. The Court observed that “[a] final sentencing order was a necessary predicate to this action, and [the judge] had not entered one on the earlier firearm offense.” Id. The Court concluded that “[t]he jury’s verdict in that case was not a final conviction without the entry of the sentencing order and, therefore, could not be used to establish the predicate first offense.” Id. (citing Ramdass v. Commonwealth, 248 Va. 518, 520 (1994)). The Court of Appeals in Batts did not distinguish Ansell. The two cases, however, are not in conflict. As the Court of Appeals explained below in the appeal of the present case, there exists “a key distinction between Batts and this case — Batts involved charges contested in separate prosecutions while all of the findings of guilt in this case arise from a single prosecution.” Walker, 74 Va. App. at 497. When a defendant faces separate prosecutions, there is a risk that the first conviction, which served as a predicate for an enhanced penalty in a second prosecution, might be set aside. There would then be no first conviction. Indeed, that actually 18 occurred in Batts, when the judge in the first proceeding set aside the jury’s verdict on the first firearm conviction. Batts then stood convicted of a second conviction even though there was no first conviction. Batts, 30 Va. App. at 9. This risk, of a second offense without a predicate first offense, is not present when multiple charges of use of a firearm in the commission of a felony are tried in a single prosecution. Finally, Walker, responding to our statement in Ansell that this statute serves a deterrent purpose, argues that social science research demonstrates that recidivist statutes do not, in fact, serve a deterrent function. Criminal punishment serves a number of purposes, including incapacitation, deterrence, and retribution. Regardless of whether a recidivist statute actually serves a deterrent purpose, it remains the case that a person who robs more than one person is more blameworthy than a person who robs a single individual. Each person Walker robbed experienced the terror in that moment of being robbed at the point of a gun, as well as whatever sequelae may flow from that event. Whatever purpose or purposes the legislature may have had when it enacted the recidivism provisions of this statute — the statute does not say — the language of Code § 18.2-53.1, combined with the weight of our precedent, requires affirmance. Accordingly, Walker was properly convicted of use of a firearm as a second or subsequent offense. CONCLUSION For the foregoing reasons, the judgment of the Court of Appeals of Virginia will be affirmed. Affirmed. 19