Attorneys: Sean D. Reyes, Att'y Gen., Kris C. Leonard, Asst. Solic. Gen., Clint T. Heiner, Salt Lake City, for petitioner Nathalie S. Skibine, Lisa J. Remal, Salt Lake City, for respondent
This opinion is subject to revision before final publication in the Pacific Reporter
On Certiorari to the Utah Court of Appeals Third District, Salt Lake County
The Honorable Randall N. Skanchy
No. 121910892 Attorneys: Sean D. Reyes, Att'y Gen., Kris C. Leonard, Asst. Solic. Gen., Clint T. Heiner, Salt Lake City, for petitioner Nathalie S. Skibine, Lisa J. Remal, Salt Lake City, for respondent ASSOCIATE CHIEF JUSTICE LEE authored the opinion of the Court, in which CHIEF JUSTICE DURRANT, JUSTICE HIMONAS, JUSTICE PETERSEN, and JUDGE BROWN joined. Having recused himself, JUSTICE PEARCE does not participate herein; DISTRICT JUDGE JENNIFER A. BROWN sat. ASSOCIATE CHIEF JUSTICE LEE, opinion of the Court:
¶1 Manuel Antonio Lujan was convicted of aggravated robbery based on eyewitness identification testimony and other evidence admitted at trial. The court of appeals reversed the conviction under the "reliability" factors set forth in State v. Ramirez, 817 P.2d 774 (Utah 1991). State v. Lujan, 2015 UT App 199, 357 P.3d 20. Ramirez identified five factors for courts to consider in assessing the reliability (and hence admissibility) of eyewitness identification testimony under the due process clause of the Utah Constitution: (1) the "opportunity" of the eyewitness to view the suspect; (2) the degree of attention paid to the suspect; (3) the witness's capacity to observe the event; (4) the degree of "spontane[it]y" and "consisten[cy]" of the eyewitness testimony; and (5) "the nature of the event being observed." 817 P.2d at 781 (citation omitted). Applying these factors, the court of appeals concluded that the testimony in question was "legally insufficient . . . to warrant a preliminary finding of reliability and, therefore, admissibility." Lujan, 2015 UT App 199, ¶ 15 (quoting Ramirez, 817 P.2d at 784). And it reversed on the ground that the State had not carried its burden of establishing that "the improperly admitted eyewitness identifications were harmless beyond a reasonable doubt." Id. ¶ 16.
¶2 In so doing, however, the court of appeals also raised concerns about the viability of the standard set forth in Ramirez. In light of developments in "scientific and legal research regarding the reliability of eyewitness identification" testimony since our decision in Ramirez, the majority indicated that it had "every reason to believe" that the Ramirez framework "must be revisited" by this court. Id. ¶ 10 n.1. Then-Judge Pearce dissented but echoed the view "that the time may have arrived for the Utah Supreme Court to revisit its holding" in Ramirez. Id. ¶ 21 (Pearce, J., dissenting).
¶3 We granted certiorari in light of the court of appeals' open call for our reconsideration of Ramirez. And in the course of our consideration of this case a number of developments have ensued. We asked for supplemental briefing on the question of whether and to what extent the Ramirez factors set a freestanding guarantee of evidentiary reliability rooted in the Utah Constitution. We then reheard the case after a member of the court retired while the matter was under advisement. And in the meantime our court considered and promulgated a new rule of evidence governing the admissibility of eyewitness identification testimony. See UTAH R. EVID. 617 (effective November 1, 2019).
¶4 These developments have informed our consideration of the important questions presented in this case. In light of them we now take up the court of appeals' request that we revisit the factors set forth in our decision in Ramirez. And we do so first by specifying the "order of operations" in assessing the reliability and admissibility of eyewitness identification testimony. We clarify that the threshold step in this assessment is a matter for our rules of evidence. We hold that those rules, including (in cases going forward) new rule 617 of the Utah Rules of Evidence, prescribe the factors that trial courts should consider in judging the reliability and admissibility of eyewitness identification evidence. And we note that our established rulemaking process lends itself nicely to adaptation over time in response to developments in scientific and legal scholarship in this important field.
¶5 We also contrast our adaptive rulemaking process with our settled method of constitutional interpretation. Our recent cases have clarified our carefully circumscribed role in interpreting the constitution. We have emphasized that the provisions of this charter document are not a license for common-law policymaking but instead a fixed set of limits on the operation of our government. Such limits are interpreted in accordance with the public understanding of the constitution when it was originally established. And these premises highlight a key limitation on the factors set forth in our Ramirez decision—the fact that the Ramirez court spoke vaguely of advancing constitutional "due process" interests but nowhere rooted the factors we adopted in the text or original understanding of the Utah Constitution. There is some tension and confusion in our case law on the question whether the Ramirez factors are mandated as a matter of state constitutional law. But it is clear that we have never identified a basis for these factors in the interpretive methodology that governs our approach to questions of state constitutional law—the original public meaning of the due process clause of the Utah Constitution. And that shortcoming is sufficient for us now to reinforce a point we alluded to in our decision in State v. Hubbard, which is that the Ramirez factors themselves are not rooted in constitutional soil. See 2002 UT 45, ¶ 27, 48 P.3d 953 (the Ramirez factors "provide guidance" but are not "exhaustive or exclusive" considerations in determining whether identifications are "violative of due process").
See In re Steffensen, 2016 UT 18, ¶ 7, 373 P.3d 186 (explaining that the due process clause of the Utah Constitution is not a "free-wheeling constitutional license" for this court to "assure fairness on a case-by-case basis," but a guarantee of procedural rights "measured by reference to traditional notions of fair play and substantial justice" (citation and internal quotation marks omitted)).
See South Salt Lake City v. Maese, 2019 UT 58, ¶ 18, 450 P.3d 1092 (noting that "[w]hen we interpret constitutional language, we start with the meaning of the text as understood when it was adopted"); Zimmerman v. Univ. of Utah, 2018 UT 1, ¶ 25, 417 P.3d 78 (explaining that we interpret the Utah Constitution by examining its "text . . . as understood when it was adopted in the late nineteenth century"); Neese v. Utah Bd. of Pardons & Parole, 2017 UT 89, ¶ 96, 416 P.3d 663 (clarifying that we "interpret the Constitution according to how the words of the document would have been understood by a competent and reasonable speaker of the language at the time of the document's enactment" (alteration in original) (citation omitted)).
Compare State v. Ramirez, 817 P.2d 774, 778, 780-81 (Utah 1991) (articulating specific factors for judicial assessment of the "required constitutional admissibility analysis"), with State v. Hubbard, 2002 UT 45, ¶ 27, 48 P.3d 953 (explaining that the Ramirez factors are "not an exhaustive or exclusive list" for examining due process concerns).
¶6 We revisit and clarify Ramirez on this basis. We endorse the need for revising and updating the factors set forth in that opinion. But we emphasize that the revising and updating is done as a matter of our revisions to the Utah Rules of Evidence, and not by treating the Utah Constitution as a vessel for judicial policymaking.
¶7 This is not to say that there is no role for the due process clause in a case like this one. Our decisions in Ramirez and Hubbard also reinforced a premise established under the federal Due Process Clause in binding precedent of the United States Supreme Court. That premise is that eyewitness identification evidence may be excluded if it is produced as a result of suggestive police activity and the taint of suggestive police procedures creates a "substantial likelihood of misidentification." Neil v. Biggers, 409 U.S. 188, 201 (1972); see also Ramirez, 817 P.2d at 784 (considering reliability in the context of a "blatant[ly] suggestive" showup); Hubbard, 2002 UT 45, ¶¶ 23, 26 (examining whether "procedural actions taken by [police]" are so "impermissibly suggestive" as to create a "substantial likelihood of irreparable misidentification" under both the federal and state due process clauses). We endorse and reaffirm that principle here, which controls as a matter of stare decisis. But we clarify that in the face of suggestive police activity the due process standard is still only a constitutional backstop to the threshold inquiry into reliability and admissibility under our rules of evidence. And we emphasize that that threshold inquiry under our rules may render the constitutional inquiry unnecessary in many cases.
¶8 We thus reverse the court of appeals on the ground that the legal framework it established (and, indeed, the one it invited us to reconsider) is no longer viable. Because we have substantially reformed the law in this field, we might, in an ordinary case, be inclined to remand to the district court to allow it to apply our new standards to the facts of this case in the first instance. We see no need to do so here, however. Instead we reinstate the jury verdict on an alternative basis advanced by the State—on the ground that any arguable error in admitting the eyewitness identification evidence in this case was harmless in light of the other evidence in the record establishing Manuel Lujan's guilt.
¶9 Early in the morning before sunrise on November 25, 2012, a man went out to get his car ready for an upcoming annual inspection. When he sat down in the driver's seat he noticed that the car's dome light was on and that someone had opened the rear driver's side door. Then he saw a man who was later identified as Manuel Lujan, who closed the rear door and opened the driver's door. Lujan squatted next to the driver's seat with his face about eight to nine inches from the man's face and asked, "why you following me?" Lujan then stood up, opened his jacket, and reached near his waist for what the man thought was a knife or gun. The man decided to retreat to the house, fearing he might be stabbed or shot.
¶10 At that point the man stood up, placing himself at eye level with Lujan and within such close proximity that the men were "almost touching." The man then slowly moved towards the house, maintaining visual contact with Lujan the entire time. Lujan followed the man, moving into the light of the car's headlights. Once inside, the man turned on the floodlights, locked his door, and woke up his younger brother. The two stepped outside the house in time to see Lujan drive the car off the property. And the man's brother quickly called the police, who responded shortly thereafter.
¶11 The man's encounter with Lujan occurred before sunrise. But there were streetlights on across the street, the man's porch light was on, and the car's headlights and dome light remained on throughout the encounter.
¶12 When the police arrived at the man's house, the man gave a description of the robber. He stated that the robber was about 5'10", 180 pounds, "Spanish," wearing a black jacket, and had "'longish hair' [that] poked out of [a] beanie to 'mid-ear length.'" While the man was giving his statement, the police officer noticed a trail of liquid on the ground from where the car had been parked and leading north out of the driveway. The officer cut the interview short to follow the trail of liquid. He found the car abandoned a few blocks away near an elementary school. The officer then had additional officers set up a containment area and called for a K-9 unit to help him search for the culprit.
¶13 One of the officers participating in the safety sweep around the school heard a noise sounding like someone climbing a chain link fence near the classrooms. The officer followed the noise and found Lujan "curled into a ball" next to a fenced-off air conditioning unit outside one of the classrooms. When the officer asked Lujan why he was hiding, Lujan responded "somebody is following me." The officer described Lujan as a Hispanic man with closely shaven hair and a goatee, wearing a black beanie and a black jacket.
¶14 Approximately thirty minutes after the robbery, an officer transported the man to the site where the police had found Lujan. Lujan was the only non-police officer in the area, was in handcuffs, and was illuminated with police spotlights. The police asked the man if he could identify the robber. And he identified Lujan as the same person who had accosted him and stolen his car.
¶15 Lujan was charged with first-degree-felony aggravated robbery. A few months later, the defense requested a lineup identification procedure. The man identified two possible perpetrators in the lineup—Lujan and another person who "looked familiar." The man again identified Lujan as the robber at the preliminary hearing. After the preliminary hearing, Lujan moved to suppress the in-court identification as well as evidence that the man had identified Lujan in the showup. The court denied his motion and permitted the man to identify Lujan at trial as the person who robbed him. At trial, the defense called an expert to testify about the reliability of eyewitness identification. But the jury ultimately convicted Lujan for aggravated robbery.
¶16 Lujan appealed. The court of appeals reversed and vacated Lujan's conviction, remanding for a new trial. It concluded that the eyewitness identification testimony was not reliable, and thus inadmissible, under the reliability factors laid out in State v. Ramirez, 817 P.2d 774 (Utah 1991). State v. Lujan, 2015 UT App 199, ¶ 19, 357 P.3d 20. Yet it also openly called on this court to revisit the Ramirez factors, asserting that Ramirez was an "unreliable tool" and stating that the "scientific and legal research regarding the reliability of eyewitness identifications has progressed significantly in the last twenty-four years." Id. ¶ 10 n.1.
¶17 Then-Judge Pearce dissented. He "agree[d] with the majority that the time may have arrived for the Utah Supreme Court to revisit its holding" in Ramirez. Id. ¶ 21 (Pearce, J., dissenting). Yet he also noted that the court of appeals was "duty-bound to apply" the Ramirez standard so long as it stands as precedent. Id. And he proceeded to disagree with the majority's conclusion in applying Ramirez to the facts of this case. Id. Judge Pearce concluded that "the showup involving Defendant in this case was substantially less troublesome than that the Ramirez court approved." Id. And he accordingly indicated that he would have affirmed Lujan's conviction. Id. ¶ 31.
¶18 We granted certiorari. We review the court of appeals' decision de novo, affording no deference to its analysis of the important legal questions presented.
¶19 The court of appeals has invited us to reconsider and revise the standards of evidentiary reliability set forth in our decision in Ramirez. And understandably so, as the scientific and legal scholarship in this field has evolved substantially since the time of our decision in that case—in a manner raising serious questions about the continuing viability of the factors we prescribed in the Ramirez decision. See State v. Clopten, 2015 UT 82, ¶ 53, 362 P.3d 1216 (noting that "eyewitness memory science" relied upon in prior precedent had "already been called into question by subsequent research"). We granted certiorari to take up the court of appeals' challenge. And in the course of the briefing and argument we also uncovered a key point of conflict or imprecision in our case law in this field. The imprecision concerns the legal basis for the factors prescribed in our Ramirez opinion.
¶20 On one hand, we have sometimes suggested that the Ramirez factors are rooted in constitutional soil. See, e.g., State v. Ramirez, 817 P.2d 774, 778 (Utah 1991); State v. Guzman, 2006 UT 12, ¶ 21, 133 P.3d 363; State v. Hubbard, 2002 UT 45, ¶ 25, 48 P.3d 953. In the Ramirez case itself, for example, we spoke of a "required constitutional admissibility analysis" for eyewitness identification testimony. 817 P.2d at 778. And when we articulated factors for judging reliability and admissibility of such testimony, we spoke of the matter as presenting a "constitutional" question. Id. at 780-81.
¶21 Lujan has advanced this view of the Ramirez factors in his briefing in this case. Citing Ramirez and other decisions, Lujan has urged us to reinforce the factors that we prescribed in Ramirez. And he asks us to treat them as establishing a threshold standard of evidentiary reliability that a court should use in "fulfilling its charge as [a] gatekeeper . . . [for] eyewitness identification." Lujan asserts that this standard is guaranteed by the due process clause of our Utah Constitution.
¶22 The State urges a contrary view. It asks us to "free the [Ramirez] factors of a constitutional foundation" and instead establish that the threshold standard of reliability and admissibility of evidence is a matter for our rules of evidence. It maintains that keeping up with "current research" is best "accomplished through evidentiary means" given the "fast-changing dynamics" of the field.
¶23 This position also finds some support in our case law—and, as the State notes, in recent authority from the United States Supreme Court. In Hubbard we stated that the factors set forth in Ramirez "provide guidance" on the reliability of eyewitness identification testimony but suggested that the list is not "an exhaustive or exclusive list of factors that may be considered in determining whether an identification is reliable." 2002 UT 45, ¶ 27. Our cases, moreover, have never proffered the Ramirez factors as a freestanding standard of evidentiary admissibility. Instead, as the State notes in its briefing, almost every case in which we have applied the Ramirez factors has involved a threshold showing of state action in the form of suggestive police activity. This implicates a key question on which we requested supplemental briefing in this case: whether the Utah due process clause establishes a freestanding guarantee of evidentiary reliability or instead just comes into play in the face of suggestive police activity.
Hubbard is the one case in which we applied the Ramirez factors in the absence of suggestive police activity. 2002 UT 45, ¶ 26. But as noted above, we made clear in that case that we did not consider their application constitutionally required. Id. ¶ 27; see also infra ¶¶ 26-28.
See Ramirez, 817 P.2d at 777-84 (addressing admissibility of identification following a one-person showup arranged by police); see also State v. Hollen, 2002 UT 35, ¶¶ 9-11, 29-64, 44 P.3d 794 (addressing admissibility of identification following a photo array arranged by police); State v. Hoffhine, 2001 UT 4, ¶¶ 7, 13-19, 20 P.3d 265 (addressing admissibility of identification following a two-person showup arranged by police); State v. Decorso, 1999 UT 57, ¶¶ 7, 41-47, 993 P.2d 837, abrogated on other grounds by State v. Thornton, 2017 UT 9, 391 P.3d 1016 (addressing admissibility of identification following a police lineup); State v. Willett, 909 P.2d 218, 224 (Utah 1995) (addressing admissibility of identification following a photo array arranged by police).
¶24 The State points us to recent authority from the United States Supreme Court on this question—Perry v. New Hampshire, 565 U.S. 228 (2012). The Perry Court held that the threshold inquiry for admissibility of eyewitness identification testimony is a matter for "statutes and rules . . . govern[ing] the admissibility of evidence." Id. at 237. It also stated that "due process concerns arise only when law enforcement officers use an identification procedure that is both suggestive and unnecessary." Id. at 238-39. "Even when the police use such a procedure," the Perry Court held that "suppression of the resulting identification is not the inevitable consequence." Id. at 239. "Instead of mandating a per se exclusionary rule, . . . the Due Process Clause requires courts to assess, on a case-by-case basis, whether improper police conduct created a substantial likelihood of misidentification." Id. (citation and internal quotation marks omitted). "The due process check for reliability" of eyewitness identification testimony, in other words, "comes into play only after the defendant establishes improper police conduct." Id. at 241.
¶25 We need not and do not decide whether to endorse the Perry framework as a matter of the law of due process under the Utah Constitution. In this case the State has not challenged the allegation of suggestive police activity. And for that reason we need not decide the broader question whether the Utah due process clause establishes a freestanding guarantee of the reliability of eyewitness identification testimony that would attach in the absence of state action in the form of suggestive police activity.
¶26 That still leaves the question whether, in a case involving suggestive police activity, the Ramirez factors themselves are mandated by the Utah Constitution. We hereby hold that they are not so mandated. We suggested as much in our decision in Hubbard. 2002 UT 45, ¶ 27. And, importantly, Lujan has proffered no basis for a determination that the Ramirez factors would have been understood by the public as an element of the guarantee of "due process of law" at the time of the ratification of the Utah Constitution. This is a fatal deficiency under our case law. We have repeatedly reinforced the notion that the Utah Constitution is to be interpreted in accordance with the original public meaning of its terms at the time of its ratification. See South Salt Lake City v. Maese, 2019 UT 58, ¶ 18, 450 P.3d 1092; Zimmerman v. Univ. of Utah, 2018 UT 1, ¶ 25, 417 P.3d 78; Neese v. Utah Bd. of Pardons & Parole, 2017 UT 89, ¶ 96, 416 P.3d 663. And we have emphasized that it is this mode of analysis that controls. The due process clause of the Utah Constitution is not a "free-wheeling constitutional license" for this court to "assure fairness on a case-by-case basis." In re Steffensen, 2016 UT 18, ¶ 7, 373 P.3d 186. It is a guarantee of procedural rights "measured by reference to traditional notions of fair play and substantial justice." Id. (citation and internal quotation marks omitted). And Lujan has not established a basis for the conclusion that the "traditional notions" of due process rooted in the original public meaning of the Utah Constitution would sustain a decision to chisel the Ramirez factors into constitutional stone.
¶27 Nor can we find any such basis in our opinion in Ramirez. The Ramirez opinion looked only to evolving social science in its articulation of the reliability factors that it identified. It based the factors on "well-respected and essentially unchallenged empirical studies" as laid out in State v. Long, 721 P.2d 483 (Utah 1986), even while conceding that the holding in Long "was not squarely based on the state constitution." Ramirez, 817 P.2d at 780. The opinion established this "more empirically based approach" solely because the court "judge[d] this to be a more appropriate approach." Id.
¶28 These sorts of considerations—rooted in evolving social science and legal scholarship—may be appropriate grounds for our provision of "guidance" on the reliability of eyewitness identification testimony. See Hubbard, 2002 UT 45, ¶ 27. But such evolving grounds are not a basis for establishing fixed principles of constitutional law. And our decision in Ramirez nowhere offered an originalist basis for constitutionalizing the reliability factors set forth in that opinion.
¶29 We reinforce that understanding of Ramirez here. In so doing we reverse the decision of the court of appeals, which applied the factors set forth in the Ramirez opinion as the threshold basis for assessing the admissibility of the eyewitness identification testimony in this case. We thus clarify that the threshold legal framework for judging the admissibility of eyewitness identification testimony is established under our rules of evidence. Those rules, importantly, have the virtue of being subject to nimble reformulation and revision in response to changes in prevailing scientific and legal scholarship of relevance to the reliability of eyewitness identification testimony. See Clopten, 2015 UT 82, ¶ 80 (Lee, A.C.J., concurring) (explaining that concerns about our eyewitness identification testimony law "can and should be dealt with by an amendment to our rules of evidence" and noting the "many virtues of that system" in contrast to having judges "make rules of evidence on the fly based on evolving social science").
This is, of course, no knock on the court of appeals. That court was bound to follow our case law as it stood in place at the time of the decision before us on certiorari review. And our precedent was plausibly read as establishing the Ramirez factors as a freestanding, constitutionally mandated standard for judging the admissibility of eyewitness identification testimony. So although we reverse the court of appeals, the judges of that court are to be applauded for a decision that helpfully cued up an important question for our review—and that prompted a reformulation of precedent that is only ours to make.
¶30 And we note that our rulemaking process has in fact fulfilled this task. Even while this case was under advisement, our advisory committee on the Utah Rules of Evidence considered and proposed a new rule of evidence governing the admissibility of eyewitness identification testimony—rule 617. This rule draws on recent scholarship in social science journals and law journals of relevance to the reliability of eyewitness identification testimony. And it establishes factors and standards for a trial court to employ in judging the admissibility of eyewitness testimony.
¶31 This new rule, of course, was not in place at the time of the trial in this proceeding. So it could not have been applied in the disposition of this case. But other rules of evidence were in place at the time of the trial court proceedings, and those rules could and should have been applied in assessing Lujan's challenge to the admissibility of the eyewitness identification testimony in this case.
¶32 The governing rules of evidence were not the basis for the lower court decisions before us on certiorari in this case. And in the absence of a lower court decision applying the governing law as now clarified by this court, we might ordinarily remand the case for reconsideration in light of the newly revised standard. We see no need to do so here, however, because we affirm the verdict in the trial court on an alternative basis advanced by the State. We hold that any arguable error in the admission of the eyewitness testimony in this case was harmless beyond a reasonable doubt in light of the substantial evidence connecting Lujan to the crime even absent the testimony of the eyewitness.
¶33 In the paragraphs below we first highlight the standards in our rules of evidence (both at the time of trial and under newly adopted rule 617) that should have formed the basis for the threshold inquiry into the admissibility of the eyewitness testimony in this case. Second, we clarify the remaining role for due process in a case involving suggestive police activity—a constitutional backstop for the reliability standards set forth in our rules of evidence. And finally, we conclude by explaining the basis for our decision to affirm the trial verdict without any need for a remand under the framework as clarified in this decision.
¶34 Our rules of evidence include several tools for assessing the reliability and admissibility of eyewitness identification testimony. And importantly, these tools leave room for consideration of, and adaptation based on, developments in scholarly literature in this field.
¶35 A key question for admissibility under the rules in place at the time of the trial in this case arises under rule 403. That rule provides that "[t]he court may exclude relevant evidence if its probative value is substantially outweighed by a danger of" (among other things) "unfair prejudice." UTAH R. EVID. 403.
¶36 The threshold inquiry under rule 403 goes to the probative value of the testimony of a given eyewitness. Where there is an eyewitness to a crime, his testimony typically will have at least some probative value. Yet developments in the scholarly literature have uncovered some grounds for questioning the strength or reliability of such evidence. Important research has identified both "estimator variables" and "system variables" that may tend to undermine the reliability of a given eyewitness account. See MASSACHUSETTS SUPREME JUDICIAL COURT STUDY GROUP ON EYEWITNESS EVIDENCE, REPORT AND RECOMMENDATIONS TO THE JUSTICES (2013); State v. Henderson, 27 A.3d 872 (N.J. 2011). And these variables may be considered in assessing both the probative value of a given piece of eyewitness identification testimony and the possibility of it producing unfair prejudice.
¶37 Estimator variables are factors connected to the event, witness, or perpetrator—items over which the justice system has no control. See Henderson, 27 A.3d at 895. These are factors that may affect the reliability of an eyewitness account. They include (among others) the viewing conditions at the time of the event (distance, lighting, etc.), the amount of stress (or duress) the witness was under, whether there was a weapon that the witness focused on, witness characteristics (age, impairment, etc.), perpetrator characteristics (like age and race, given that witnesses are better at identifying persons of their own age and race), and factors affecting memory decay. Id. at 904-10.
¶38 System variables consist of factors controlled by the court or law enforcement. See id. at 895-96. Examples of system variables that may affect the reliability of an eyewitness account are the use of double-blind identification procedures, the quality of pre-identification instructions, and the use of proper lineup construction. See id. at 896-903.
¶39 These lists are exemplary. The cited sources list other estimator and system variables. And research in this field is ongoing—over time we are understanding better and better how human perception and recall work when it comes to eyewitness identification. But the core point is this: The perception and memory of an eyewitness account may be affected by a wide range of factors not intuitively obvious to a juror (or even a judge), and careful consideration of these factors is important in assessing the reliability or probative value of eyewitness testimony. See Nicholas A. Kanh-Fogel, The Promises and Pitfalls of State Eyewitness Identification Reforms, 104 KY. L.J. 99, 124 (2016).
¶40 Our rulemaking process is set up in a manner that enables us to react nimbly to further developments in this field. And as our understanding of the factors that affect the reliability of eyewitness testimony develops, our application and understanding of our rules of evidence can likewise evolve.
¶41 The above-listed factors are crucial to the assessment of unfair prejudice. Eyewitness testimony is common. And it is sometimes viewed as the gold standard. It is tempting to say that "there is almost nothing more convincing . . . than a live human being who takes the stand, points a finger at the defendant, and says 'That's the one!'" Henderson, 27 A.3d at 889 (citation and emphasis omitted). But some eyewitness accounts are fool's gold. An eyewitness who is affected by significant estimator or system variables may appear to present a highly probative account of the crime; but false appearance of probity may ultimately translate into unfair prejudice. Cf. State v. Maestas, 1999 UT 32, ¶ 26, 984 P.2d 376 ("[B]ecause jurors do not appreciate the fallibility of [eyewitness] identifications, they often give eyewitness testimony undue weight.").
¶42 The above-noted estimator and system variables also informed our advisory committee's proposal that we adopt a new rule of evidence aimed specifically at eyewitness identification testimony. New rule 617 first identifies nine nonexclusive factors (based on "estimator variables" identified in the literature) for a judge to use in assessing the reliability and thus admissibility of eyewitness identification evidence: (1) "opportunity to observe the suspect committing the crime"; (2) impaired attention caused by a weapon or other distraction; (3) physical and mental capacity to make an observation; (4) awareness that a crime was taking place and resulting effect on a witness's ability to "perceive, remember, and relate it correctly"; (5) cross-racial identification; (6) length of time between the original observation and identification; (7) instances of identification or failure to identify a suspect, and later consistency; (8) exposure to opinions, photographs, or other information influencing the independence of the identification; and (9) "any other aspect of the identification" affecting reliability. UTAH R. EVID. 617(b). Under rule 617(b), the court is to exclude eyewitness identification evidence if the party challenging the evidence "shows that a factfinder . . . could not reasonably rely on the eyewitness identification." Id.
¶43 Rule 617 also incorporates "system variables." These variables are incorporated by the rule in its identification of factors relevant to determining whether a photo array, lineup, or showup was "unnecessarily suggestive." Id. 617(c). If an identification procedure is "contested," the court is to "determine whether the identification procedure was unnecessarily suggestive or conducive to mistaken identification" based on the system and estimator variables set forth in the rule. Id. "If so, the eyewitness identification must be excluded unless the court, considering" both estimator and system variables, "finds that there is not a substantial likelihood of misidentification." Id.
¶44 By taking account of both the "estimator variables" and "system variables" that have been identified in scholarly literature, rule 617 thus allows a trial judge to more effectively assess whether such variables have undermined the reliability of a given eyewitness account. In light of the evolving nature of academic discourse on this subject, however, the rule does not paint courts into a corner. Rule 617 makes clear that the factors it identifies to account for estimator and system variables are nonexhaustive: "As scientific research advances, other factors in addition to those outlined . . . may be considered" by judges in performing their gatekeeping function. Id. 617 advisory committee notes.
¶45 A threshold focus on the rules of evidence furthers the principle of constitutional avoidance—by frontloading the question of admissibility under our rules of evidence in advance of any constitutional inquiry. And it yields the kind of flexibility called for by the court of appeals in its decision below and by the parties in their briefing. See State v. Lujan, 2015 UT App 199, ¶ 10 n.1, 357 P.3d 20 (referring to developments in "scientific and legal research regarding the reliability of eyewitness identification" testimony over the years since our decision in Ramirez and suggesting that the Ramirez framework "must be revisited" by this court); see also Clopten, 2015 UT 82, ¶ 80, 362 P.3d 1216 (Lee, A.C.J., concurring) (the evolution of eyewitness identification science "can and should be dealt with by an amendment to our rules of evidence").
¶46 While the threshold standard of admissibility of eyewitness testimony is provided by the Utah Rules of Evidence, the governing case law also preserves a role for due process. When eyewitness identification evidence is secured by "unnecessarily suggestive" police action, the federal Due Process Clause adds a constitutional backstop to our rules of evidence. Perry v. New Hampshire, 565 U.S. 228, 235 (2012). And our case law has established a similar backstop under the due process clause of the Utah Constitution. See, e.g., State v. Ramirez, 817 P.2d 774, 779, 784 (Utah 1991); State v. Hubbard, 2002 UT 45, ¶¶ 25-26, 48 P.3d 953.
¶47 The backstop test under federal law bars eyewitness evidence if the taint of suggestive police procedures created a "substantial likelihood of misidentification." Neil v. Biggers, 409 U.S. 188, 201 (1972). In assessing that likelihood, the court is to weigh the "indicia of reliability" against the "corrupting effect of the police-arranged suggestive circumstances." Perry, 565 U.S. at 232.
¶48 The Supreme Court has identified factors to consider in that assessment. These factors include: "the opportunity of the witness to view the criminal at the time of the crime, the witness' degree of attention, the accuracy of his prior description of the criminal, the level of certainty demonstrated at the confrontation, and the time between the crime and the confrontation." Id. at 239 n.5 (citation omitted). These factors are to be applied in assessing "on a case-by-case basis whether improper police conduct created a 'substantial likelihood of misidentification.'" Id. at 239 (quoting Biggers, 409 U.S. at 201). Where the court finds such a likelihood in light of the "totality of the circumstances," due process requires the exclusion of the evidence. Id.
¶49 While this court did expand on the Supreme Court's "factors" in Ramirez—with factors that we deemed to "more precisely define the focus of the relevant [due process] inquiry," 817 P.2d at 781—these factors are not constitutionally required because they are not rooted in the text or original understanding of the Utah Constitution. See supra ¶¶ 26-28. We reinforce that holding here. But we nonetheless preserve the standards set forth in Ramirez as a matter of stare decisis as clarified above. Thus, we hold that the Ramirez factors are entitled to stare decisis respect, and will not be overridden, insofar as they provide "guidance" of relevance to the purpose for which they have been applied in our case law—as possible considerations in assessing whether evidence produced as a result of suggestive police activity should be excluded on the ground that it leads to a substantial likelihood of misidentification. But unless and until the framework for a backstop due process test is rooted in the original meaning of the Utah due process clause, these factors remain just that—guidance, rather than a constitutionally required test.
In so holding we do not rule out the possibility of a determination, in a future case in which the question is squarely presented, that state constitutional standards under the Utah due process clause differ from federal due process standards under Neil v. Biggers, 409 U.S. 188, 201 (1972). See Alpine Homes, Inc. v. City of West Jordan, 2017 UT 45, ¶ 16, 424 P.3d 95 (explaining that even where "the Utah clause is similar to the federal clause, we do not presume that federal court interpretations of federal Constitutional provisions control the meaning of identical provisions in the Utah Constitution" (citation and internal quotation marks omitted)); Jensen ex rel. Jensen v. Cunningham, 2011 UT 17, ¶ 46, 250 P.3d 465 (emphasizing that even though "some of the language of our state and federal constitutions is substantially the same, similarity of language does not indicate that this court moves in 'lockstep' with the United States Supreme Court's [constitutional] analysis" (alteration in original) (citation and internal quotation marks omitted)). We are of course not bound to follow precedent on federal due process in our formulation of state due process standards. And we may thus depart from the federal formulation if and when we are presented with state constitutional analysis rooted in the original meaning of the Utah due process clause. See South Salt Lake City v. Maese, 2019 UT 58, ¶ 18, 450 P.3d 1092 (clarifying that "[w]hen we interpret constitutional language, we start with the meaning of the text as understood when it was adopted"); Zimmerman v. Univ. of Utah, 2018 UT 1, ¶ 25, 417 P.3d 78 (noting that we interpret the Utah Constitution by examining its "text . . . as understood when it was adopted in the late nineteenth century"); Neese v. Utah Bd. of Pardons & Parole, 2017 UT 89, ¶ 96, 416 P.3d 663 (explaining that we "interpret the Constitution according to how the words of the document would have been understood by a competent and reasonable speaker of the language at the time of the document's enactment" (alteration in original) (citation omitted)).
¶50 In so stating we reinforce points we have made throughout this opinion. We endorse the need for ongoing adjustment and refinement of the standards of reliability and admissibility of eyewitness identification testimony in response to scholarly developments in this field. Yet we emphasize that the principal, threshold basis for such adjustment and refinement is through amendments to our rules of evidence.
¶51 We also emphasize the predominant role for our rules of evidence in the run of cases. The standard for admissibility under our rules of evidence is, if anything, more protective than the constitutional due process standard. Cf. Ramirez, 817 P.2d at 784 (declining to engage in "separate Biggers federal analysis" in light of the court's conclusion that the state due process clause "is certainly as stringent as, if not more stringent than, the federal analysis"). Specifically, the estimator and system variables that have been incorporated into rule 617, and could be considered under rule 403, encompass all of the factors relevant to the due process inquiry (and more). It seems likely that whenever there is a "substantial likelihood of misidentification" under the due process framework there will also be a basis for exclusion under our rules of evidence. And this will allow our courts to avoid the constitutional inquiry in the run of cases, and to keep the focus on the standards of admissibility under our rules of evidence.
¶52 We have clarified and reformulated the framework for the analysis of the admissibility of eyewitness identification testimony in Utah. Ordinarily that might call for a remand to allow a lower court to apply our revised legal standard to the facts of this case in the first instance. We see no need for a remand in a case like this one, however, because we endorse an alternative basis for affirming the trial verdict even assuming (for the sake of argument) error in the admission of the eyewitness testimony in this case. We conclude, specifically, that any arguable error was harmless beyond a reasonable doubt in light of other substantial evidence connecting Lujan to the crime in question.
This is the federal standard for harmlessness of constitutional errors under Chapman v. California, 386 U.S. 18, 24 (1967). We apply this standard here based on the arguendo assumption that there may have been a constitutional violation in admitting eyewitness identification testimony that could be thought to have run afoul of the above-stated due process standard. --------
¶53 We can evaluate the harmlessness of the eyewitness identification testimony by considering "the importance of the witness' testimony in the prosecution's case, whether the testimony was cumulative, the presence or absence of evidence collaborating or contradicting the testimony of the witness on material points, the extent of cross-examination otherwise permitted, and, of course, the overall strength of the prosecution's case." State v. Villarreal, 889 P.2d 419, 425-26 (Utah 1995) (citation omitted). And here we harbor no reasonable doubt that the jury would still have convicted Lujan even without the testimony from the eyewitness.
¶54 Even without Lujan's eyewitness identification, the jury would still have had the description of the robber and corroboration from the police that Lujan largely matched that description. The jury also would have heard that Lujan was found when police followed a trail of liquid on the ground from where the stolen car had been parked and leading north out of the driveway. Significantly, the jury would have heard that the trail of liquid led them to find the stolen car parked a few blocks away near an elementary school—and to find Lujan "curled into a ball" and hiding next to a fenced-off air conditioning unit in the adjacent school yard. Lujan was the only person found in the area, and he largely fit the description of the robber. And when the police found Lujan he expressed the same concern the robber had raised earlier that night—stating "somebody is following me."
¶55 It is also significant that Lujan challenged only the admission of the in-court and show-up identifications given by the eyewitness. With this in mind, we may presume for purposes of our harmlessness analysis that the jury would have heard a separate identification given by the eyewitness—the identification given in the lineup. Thus, the jury would also have been told that four months after the robbery, two suspects were identified from a lineup, and that only one of them was the only person located within minutes of the early-morning robbery a short distance from the stolen car, and who largely matched the description given to the police minutes before.
¶56 In light of all the evidence we conclude that any error in admitting the eyewitness identification testimony was harmless beyond a reasonable doubt. The in-court and showup identifications were undoubtedly important to the prosecution's case. But we think that evidence was cumulative and largely unnecessary. And in light of the other evidence implicating Lujan we hold that any potential error in the district court's admission of his eyewitness identification was harmless beyond a reasonable doubt. We affirm the trial verdict against Lujan on this basis.
¶57 Eyewitness identification testimony presents difficult problems for our courts. Our past decisions could be read to suggest that the admissibility of such evidence should be judged in the first instance under a due process standard framed by factors identified in our opinion in State v. Ramirez, 817 P.2d 774 (Utah 1991). We now repudiate that notion. We hold that the admissibility of this kind of evidence is to be measured in the first instance by our rules of evidence. And we clarify that in cases in which eyewitness testimony is procured by unnecessarily suggestive police misconduct, the role of the due process inquiry is a limited one—a secondary backstop to the threshold question of admissibility under our rules of evidence.