Laurie N. Feldman, special deputy assistant state's attorney, with whom, on the brief, were Gail P. Hardy, state's attorney, and Anthony Bochiccio, senior assistant state's attorney, for the appellant (state). Lisa J. Steele, assigned counsel, for the appellee (defendant). Nicole E. Feit and David W. Ogden, pro hac vice, filed a brief for the American Psychological Association as amicus curiae. Todd D. Fernow, Timothy H. Everett and John T. Walkley, Milford, filed a brief for the Connecticut Criminal Defense Lawyers Association as amicus curiae. Charles D. Ray, Hartford, filed a brief for the Innocence Project as amicus curiae.
Laurie N. Feldman, special deputy assistant state's attorney, with whom, on the brief, were Gail P. Hardy, state's attorney, and Anthony Bochiccio, senior assistant state's attorney, for the appellant (state).
Lisa J. Steele, assigned counsel, for the appellee (defendant).
Nicole E. Feit and David W. Ogden, pro hac vice, filed a brief for the American Psychological Association as amicus curiae.
Todd D. Fernow, Timothy H. Everett and John T. Walkley, Milford, filed a brief for the Connecticut Criminal Defense Lawyers Association as amicus curiae.
Charles D. Ray, Hartford, filed a brief for the Innocence Project as amicus curiae.
ROGERS, C.J., and PALMER, ZARELLA, EVELEIGH, McDONALD, ESPINOSA and VERTEFEUILLE, Js.
PALMER, J.Following an incident in which the defendant, Troy Artis, allegedly aided two other individuals in assaulting and seriously wounding the victim, Alexis Otero, a jury found the defendant guilty of accessory to assault in the first degree by means of a dangerous instrument in violation of General Statutes §§ 53a–59 (a)(1) and 53a–8 (a). The trial court rendered judgment in accordance with the jury verdict, and the defendant appealed to the Appellate Court, claiming, inter alia, that the trial court improperly had denied his motion to suppress Otero's out-of-court and in-court identifications of the defendant as one of his assailants. State v. Artis, 136 Conn.App. 568, 585, 47 A.3d 419 (2012). The Appellate Court, with one judge dissenting in part, agreed with the defendant, concluding that the trial court's denial of the defendant's motion to suppress Otero's identifications violated the defendant's right to due process because the use of an unnecessarily suggestive procedure by the police rendered the identifications unreliable. See id. at 608, 47 A.3d 419. Relying on State v. Gordon, 185 Conn. 402, 441 A.2d 119 (1981), cert. denied, 455 U.S. 989, 102 S.Ct. 1612, 71 L.Ed.2d 848 (1982), in which this court concluded that “sound judicial policy requires reversal whenever the erroneous admission of an unnecessarily suggestive and unreliable identification has violated a defendant's constitutional rights”; id. at 420, 441 A.2d 119 ; the Appellate Court reversed the trial court's judgment. See State v. Artis, supra, at 608–609, 613, 617, 47 A.3d 419. The Appellate Court further determined that, even if the rule of per se reversibility that this court adopted in Gordon was not applicable, the defendant nevertheless was entitled to a new trial because the state had failed to meet its burden of establishing that its use of Otero's identification testimony was harmless beyond a reasonable doubt. Id. at 617, 441 A.2d 119. We granted the state's petition for certification to appeal, limited to the following three issues: First, “[d]id the Appellate Court majority properly determine that admission of [Otero's] in-court and out-of-court identifications following a suggestive police display of the defendant's photograph was a ... due process violation under Manson v. Brathwaite, 432 U.S. 98, 97 S.Ct. 2243, 53 L.Ed.2d 140 (1977) ?” State v. Artis, 307 Conn. 909, 910, 53 A.3d 999 (2012). Second, “[i]f the answer to the first question is affirmative, should this court expressly overrule the holding [in Gordon] that harmless error review is unavailable for identification evidence ...?” Id. Third, “if so, did the Appellate Court majority properly determine that the identification evidence was not harmless?” Id. We need not address the first issue because, even if we assume, without deciding, that the trial court improperly denied the defendant's motion to suppress Otero's identification testimony, we conclude, contrary to our holding in Gordon, that the improper admission of such evidence is subject to harmless error analysis. We also conclude that the state's use of Otero's identification testimony was harmless. Accordingly, we reverse the judgment of the Appellate Court.
The trial court imposed a total effective sentence of nine years imprisonment, to run concurrently with the term remaining on a preexisting sentence, followed by eight years of special parole. State v. Artis, 136 Conn.App. 568, 575, 47 A.3d 419 (2012).
In Manson v. Brathwaite, supra, 432 U.S. at 98, 97 S.Ct. 2243, the United States Supreme Court held that the admission of an eyewitness' out-of-court identification, obtained as a result of an unnecessarily suggestive identification procedure, that is, a procedure that is both suggestive and not justified by exigent circumstances, violates a defendant's right to due process under the federal constitution when “the ‘indicators of [a witness'] ability to make an accurate identification’ are ‘outweighed by the corrupting effect’ of law enforcement suggestion....” Perry v. New Hampshire, ––– U.S. ––––, 132 S.Ct. 716, 725, 181 L.Ed.2d 694 (2012), quoting Manson v. Brathwaite, supra, at 116, 97 S.Ct. 2243. Furthermore, an in-court identification by an eyewitness whose out-of-court identification was unreliable as a result of an unnecessarily suggestive identification procedure also is inadmissible unless it “is purged of the taint of the defective pretrial procedure by establishment of the fact that it is based [on] disassociated and independent observation.”State v. Smith, 165 Conn. 680, 685, 345 A.2d 41 (1974).
The opinion of the Appellate Court sets forth the following procedural history and facts that the jury reasonably could have found. “At approximately 11 p.m. on February 14, 2008, [Otero] drove some of his friends to Club Blu on Ann Street in [the city of] Hartford where he was sometimes employed as a bouncer. Approximately [one-half] hour later, Otero walked two blocks from Club Blu to Club NV near the corner of Allyn and High Streets.
“[That] same evening, Christina Miano also went to Club NV together with her ... boyfriend, Robert Acevedo, and his sister, Anna Acevedo, and Anna's boyfriend, the defendant. Robert Acevedo drove the four of them to Hartford in his silver Infiniti.... At approximately 11 p.m., as the four walked toward Club NV, the defendant, who carried a knife on his belt, asked Robert Acevedo for the keys to the Infiniti so he could put his knife [away] before undergoing a security check at Club NV. Miano and Anna Acevedo went on ahead to Club NV, where they socialized apart from the defendant and Robert Acevedo.
“Near closing time, Otero, who knew Miano, visited with her for several minutes. At the time, he [was unaware] that Miano had arrived with Robert Acevedo, whom he did not know. Otero routinely photographed people at nightclubs for a radio station website, and that night he took a photograph of Miano and Anna Acevedo.
“At closing time, Otero left Club NV to return to Club Blu. Miano left Club NV at approximately the same time and got into the Infiniti with Robert Acevedo and Anna Acevedo. She sat in the front seat next to Robert Acevedo as the three, uncertain of the defendant's whereabouts, waited for him to join them. Miano saw Otero and beckoned for him to come speak with her, which he did. Once the conversation was over, as Otero crossed High Street on his way to Club Blu, Robert Acevedo drove the Infiniti straight toward him. Otero jumped onto the sidewalk and yelled an obscenity at Robert Acevedo, who apologized. According to Miano, Robert Acevedo thought that Otero had ‘disrespect[ed]’ him by talking with her.
“Shortly thereafter, the defendant opened the rear passenger door of the Infiniti from the outside, entered the automobile briefly and quickly exited to confront Otero. The defendant and Otero exchanged profanities, and the defendant then punched Otero in the face and shoulder as the two men engaged in a face-to-face fist-fight that lasted somewhere between two and ten seconds. Seconds after the fistfight commenced, Robert Acevedo and Anna Acevedo got out of the Infiniti and approached the defendant and Otero. Otero was then struck from behind, causing him to fall to the sidewalk. While he was on the sidewalk, Otero was on all fours covering his head. For approximately ten to twenty seconds, he felt three or four people assault him from different directions. Miano got out of the automobile when she saw Robert Acevedo, Anna Acevedo and the defendant all crowded around Otero while he was on the ground.... Miano then ‘grabbed Anna off’ Otero, and the two women began to argue. During the assault on Otero, Miano did not see a weapon or a knife. At some point, she dropped her cell phone.
“As Hector Robles, a Hartford police officer, walked toward the group of people surrounding Otero, others on the street called out, ‘cops....' The fight broke up, and the crowd dispersed. The defendant, Miano, Robert Acevedo and Anna Acevedo got into the Infiniti and drove away. Anna Acevedo continued to express displeasure with Miano and wanted to fight her. Robert Acevedo instructed the two women to stop, and he took Miano to her mother's home in order to separate the two women. According to Miano, while they were driving from the scene, Anna Acevedo asked the defendant, ‘where'd that blood come from?’ Miano stated that she never looked at the defendant again after he got back into the automobile. Later, another police officer found blood splattered on the sidewalk.
“When Otero got up from the sidewalk, the Infiniti was gone. He saw a cell phone and put it in his pocket. He also saw that his hand was bleeding heavily. He walked to Club Blu where he knew he would find a police officer and familiar people to help him. Jessie Rego, who was working at the door of Club Blu, observed that the cut in Otero's thumb was so deep that he could see the bone. Rego also saw blood coming from Otero's stomach. A police officer found a trail of blood between Club NV and Club Blu.
“Otero was transported by ambulance to Hartford Hospital (hospital). Although Otero has no recollection of being transported to or arriving at the hospital, he remembers going into an operating room, where he was treated for seven puncture wounds to his torso, arms and hand. Twenty sutures and fifty staples were required to close Otero's wounds. As a result of his injuries, Otero is no longer able to exercise as he once did, and walking is difficult due to an injury he sustained to his knee.
“Although Otero has no recollection of talking to a police officer at the hospital, he was, in fact, interviewed by Sergeant Jeff Rousseau soon after arriving at the hospital. Rousseau testified that Otero told him that his assailants, two men and a woman, were in a newer, gray Infiniti. He described both men as light-skinned black males, approximately twenty-seven to twenty-eight years old. The stockier of the two men had freckles on his face. According to Rousseau, Otero was not sure that he could identify the perpetrators, but he did not rule out identification. Otero gave Rousseau Miano's cell phone.
“The defendant was arrested ... and subsequently was charged with assault in the first degree while aided by two or more persons in violation of ... § 53a–59 (a)(4), conspiracy to commit assault in the first degree while aided by two or more persons in violation of General Statutes §§ 53a–48 and 53a–59 (a)(4), accessory to assault in the first degree in violation of §§ 53a–8 and 53a–59 (a)(1), and conspiracy to commit assault in the first degree with a dangerous instrument in violation of §§ 53a–48 and 53a–59 (a)(1).” (Footnotes altered.) State v. Artis, supra, 136 Conn.App. at 570–74, 47 A.3d 419.
Robert Acevedo and Anna Acevedo also were charged in separate cases. Robert Acevedo pleaded guilty to conspiracy to commit assault in the first degree and was sentenced to five years imprisonment. Anna Acevedo pleaded guilty to assault in the third degree, reckless endangerment in the first degree, unlawful restraint in the second degree, threatening in the second degree and breach of the peace in the second degree, and received a suspended sentence.
During trial, the defendant filed a motion to suppress Otero's out-of-court identification and to preclude any in-court identification, claiming that the out-of-court identification was the product of an unnecessarily suggestive procedure and that any subsequent in-court identification would be tainted by the improper out-of-Court identification. At the hearing on the defendant's motion, Otero testified “that in the weeks following the incident he received secondhand and thirdhand reports ... [with the] names of people who might have been involved [in the assault], including the name ‘Hershey,’ which he was told was the defendant's street name. Otero stated that in May, 2008, while at the Hartford police station, he was shown a photographic array of eight individuals from which he was not able to make a positive identification. The defendant's photograph was not part of this array. Otero testified that after he had been told by others, secondhand and thirdhand, that the defendant had been involved in the attack ... he looked the defendant up on the [D]epartment of [C]orrection website and discovered that the defendant was incarcerated. He stated that on the day he gave Hartford Detective [Jerry] Bilbo the defendant's name, Bilbo then brought up the defendant's photograph on his computer. Otero testified that while he expected to see several photographs on the computer screen, there was only one photograph, that of the defendant, and that upon seeing the defendant's photograph, he immediately identified him as the initial assailant.
“Later in the trial, Miano testified that Hershey was the street name for Robert Acevedo, who she said had been her boyfriend at the time of the incident and the operator of the car that nearly struck Otero.” State v. Artis, supra, 136 Conn.App. at 590 n. 17, 136 Conn.App. 568.
“[Bilbo also testified at the suppression hearing. He explained] that in May, 2008, he showed Otero a photographic array with eight photographs that did not include the defendant. From this array, Otero was not able to make any positive identifications, although he did indicate that two of the photographs were similar to the person who attacked him. One of the photographs [that Otero] tentatively selected ... was that of Robert Acevedo. Bilbo indicated, as well, that in June, 2008, he prepared a photographic array for Otero and that this array included the defendant's photograph. However, he did not show this array to Otero because he did not believe that Otero could identify any of his attackers. Nevertheless, Bilbo testified that six or seven months later he did show Otero a booking photograph of the defendant, with the name ‘Artis' printed across the front of the shirt of the person depicted, while telling Otero that it was a photograph of the defendant, that the defendant was a suspect in the case, and that he was seeking an arrest warrant for him. Contrary to Otero's testimony, Bilbo claimed that Otero was not able to identify the defendant from this one photograph even though he told Otero that it was a photograph of the defendant. Bilbo stated, as well, that when shown the photograph of the defendant, Otero responded that he did not know who his attackers were and acknowledged that he could not identify them. Bilbo did not, at any time before or after he showed Otero the single booking photograph, display the photographic array to him that included the defendant's photograph.
“Confronted with the inconsistency between Otero's testimony that he positively identified the defendant from the one photograph shown to him and Bilbo's testimony that Otero did not make an identification, the court [found] ... that Otero's testimony was more credible in this regard than Bilbo's. The court, therefore, concluded that Otero had made an out-of-court pretrial identification of the defendant as the initial assailant.” (Footnotes altered.) Id. at 590–92, 47 A.3d 419. Following the suppression hearing testimony, the trial court denied the defendant's motion to suppress Otero's out-of-court identification and to preclude any subsequent in-court identification of the defendant. Although the trial court found that Otero's initial identification of the defendant was unnecessarily suggestive, the court further determined that his identification was reliable and, therefore, admissible, and that any subsequent in-court identification of the defendant by Otero also would be admissible. Following the trial court's ruling, Otero, while testifying before the jury, identified the defendant as the man with whom he had engaged in a fistfight before being repeatedly stabbed. Otero further described his initial, out-of-court identification of the defendant. Miano, who testified after Otero, also identified the defendant, who was well-known to her, as the man who initially had confronted Otero. In addition, she testified that, after observing that Otero was on the ground, she saw Robert Acevedo, Anna Acevedo, and the defendant fighting with him. Neither Robert Acevedo nor Anna Acevedo testified at the defendant's trial. At the conclusion of the trial, the court granted the defendant's motion for a judgment of acquittal with respect to the charge of assault in the first degree while aided by two or more persons. The jury found the defendant guilty of accessory to assault in the first degree by means of a dangerous instrument and not guilty of the two remaining conspiracy charges.
In Manson v. Brathwaite, supra, 432 U.S. at 98, 97 S.Ct. 2243, the court identified the following nonexclusive list of factors that are relevant to a determination of whether, under the totality of the circumstances, an out-of-court identification resulting from an unnecessarily suggestive police procedure is nevertheless reliable: “[T]he 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 114, 97 S.Ct. 2243. The trial court expressly considered these factors in reaching its conclusion that Otero's identification of the defendant was reliable even though the police had employed an unnecessarily suggestive identification procedure.
On appeal to the Appellate Court, the defendant challenged the trial court's denial of his motion to suppress. In a divided opinion, the Appellate Court majority agreed with the defendant, concluding that the unnecessarily suggestive identification procedure used by Bilbo had created a substantial likelihood of misidentification. See id. at 593, 595–617, 47 A.3d 419. In reaching its determination, the Appellate Court majority stated that “[i]t would be difficult to conceive of a less neutral or more preemptive identification process than ... that [employed by Bilbo]”; id. at 593, 47 A.3d 419 ; and reasoned that the trial court had “failed to give proper weight to the corrupting effect of this most inappropriate identification confirmation procedure.” Id. at 607–608, 47 A.3d 419. “[A]pplication of the factors [enumerated] in Manson v. Brathwaite, supra, 432 U.S. at 114–16, 97 S.Ct. 2243, considered against the backdrop of the extraordinarily overbearing manner of the identification procedure, leads [the court] to the conclusion that the pretrial identification of the defendant by Otero was not reliable and that Otero's subsequent in-court identification was not sufficiently removed from the taint of the earlier out-of-court identification to be independently reliable.” (Footnote added.) State v. Artis, supra, 136 Conn.App. at 608, 47 A.3d 419.
The defendant also raised a claim of evidentiary insufficiency with respect to the charge of accessory to assault in the first degree. See State v. Artis, supra, 136 Conn.App. at 575, 47 A.3d 419. The Appellate Court rejected this claim; id. at 585, 47 A.3d 419 ; which is not at issue in the present appeal.
See footnote 5 of this opinion.
Adhering to this court's holding in State v. Gordon, supra, 185 Conn. at 419–20, 441 A.2d 119, the Appellate Court majority also concluded that the defendant was entitled to a new trial regardless of whether the improper admission of Otero's identification testimony was harmless. See State v. Artis, supra, 136 Conn.App. at 608–609, 613, 47 A.3d 419. The court further explained that, even if harmless error review were applicable to the admission of unnecessarily suggestive and unreliable eyewitness identification testimony, the state could not meet its burden of demonstrating that the admission of Otero's identification testimony was harmless beyond a reasonable doubt. See id. at 617, 47 A.3d 419. Specifically, the Appellate Court concluded that the following factors demonstrated that the improper admission of Otero's identification testimony influenced the result of the trial: (1) Otero's “identifications of the defendant were key elements, if not the centerpiece, of the state's case”; id.; (2) eyewitness identification evidence generally has a powerful influence on a jury's deliberations; id. at 616, 47 A.3d 419 ; (3) Miano did not provide strong independent evidence of the defendant's guilt because “her testimony regarding the circumstances of the altercation was confused and imprecise, her ability to observe the events accurately was unclear, and her familiarity with Otero together with her relationship with Robert Acevedo reasonably could have put her objectivity in doubt for the fact finders”; id. at 614, 47 A.3d 419 ; and (4) the state had relied on Otero's testimony in closing argument to the jury. Id. at 616–17, 47 A.3d 419.
Judge Lavine dissented in part from the majority opinion of the Appellate Court. He concluded that the trial court properly had denied the defendant's motion to suppress because Otero's out-of-court identification of the defendant, although the product of a suggestive police procedure, was admissible because it bore sufficient indicia of reliability. See id. at 618, 629, 47 A.3d 419 (Lavine, J., concurring in part and dissenting in part). Judge Lavine further concluded that, even if Otero's identification testimony had been improperly admitted, harmless error review of the state's use of that testimony was not barred under Gordon because, inter alia, this court had overruled Gordon 's rule of automatic reversal sub silentio in State v. Milner, 206 Conn. 512, 536 n. 11, 539 A.2d 80 (1988). See State v. Artis, supra, 136 Conn.App. at 632–33, 47 A.3d 419 (Lavine, J., concurring in part and dissenting in part). Judge Lavine finally concluded that the admission of Otero's testimony was harmless, stating that, “[b]ecause the defendant was convicted for his participation in the three-on-one altercation, and [Otero] could not identify any of the individuals involved in that altercation, the jury must have believed Miano's testimony. Simply put, in light of the other evidence, [Otero's] identification of the defendant as the assailant in the one-on-one altercation provided only incidental support for the defendant's conviction.” (Emphasis in original.) Id. at 642, 47 A.3d 419 (Lavine, J., concurring in part and dissenting in part).
On appeal to this court following our grant of certification, the state first challenges the Appellate Court's determination that the trial court improperly had denied the defendant's motion to suppress Otero's identification testimony. The state further contends that, even if that identification testimony was improperly admitted into evidence, we should expressly overrule Gordon and reverse the judgment of the Appellate Court because the admission of Otero's testimony was harmless error. Although it is a close question as to whether Otero's out-of-court identification of the defendant was reliable and, therefore, admissible, despite the procedure by which the police obtained it—a procedure that the state concedes was unnecessarily suggestive—we need not decide that issue because, even if we assume that Otero's identification testimony should have been suppressed, the state's use of that testimony is subject to harmless error review, and the state has proven beyond a reasonable doubt that the admission of the testimony was harmless. I
In support of its claim that the Appellate Court incorrectly concluded that Otero's identification of the defendant was unreliable, the state argues that the Appellate Court improperly substituted its own judgment concerning the weight to be afforded the various factors relevant to the reliability of Otero's identification. The state also provides an extensive analysis in explaining why, in its view, this court should take a cautious approach with respect to the utility of social science research on the subject of eyewitness identifications. See, e.g., United States v. Johnson, 745 F.3d 227, 228 (7th Cir.2014) (although “th[e] court has suggested that police show photographs sequentially rather than as part of an array ... [s]ome recent research has called into question the view that sequential presentation of photographs is superior to photo[graphic] spreads” [citations omitted] ). We do not address this issue, however, in view of our conclusion that, even if Otero's identification testimony should have been excluded as unreliable, its admission was harmless.
We first address the state's claim that we should overrule State v. Gordon, supra, 185 Conn. at 419–20, 441 A.2d 119, insofar as it rejected harmless error analysis in favor of a rule of per se reversibility in cases involving the state's use of unreliable eyewitness identification evidence that is the product of an unnecessarily suggestive police procedure. According to the state, “[r]eversing convictions based on harmless evidence creates, rather than corrects, erroneous outcomes.... The harmless error doctrine avoids wrongful convictions without giving guilty perpetrators the windfall of appellate reversal.” (Citations omitted; emphasis in original.) The defendant urges us to decline the state's request to reconsider and reject our holding in Gordon because unreliable eyewitness identification testimony is “extremely powerful evidence” that invariably compromises the jury's fact-finding function. We agree with the state that the introduction of such testimony in violation of a defendant's right to due process is not one of the rare instances of structural error in which the automatic reversal of a conviction is warranted.
Revisiting prior precedent implicates the doctrine of stare decisis, which “counsels that a court should not overrule its earlier decisions unless the most cogent reasons and inescapable logic require it.... Stare decisis is justified because it allows for predictability in the ordering of conduct, it promotes the necessary perception that the law is relatively unchanging, it saves resources and it promotes judicial efficiency.... It is the most important application of a theory of decisionmaking consistency in our legal culture and ... is an obvious manifestation of the notion that decisionmaking consistency itself has normative value.” (Internal quotation marks omitted.) State v. Salamon, 287 Conn. 509, 519, 949 A.2d 1092 (2008).
Nevertheless, “[w]hen a previous decision clearly creates injustice, the court should seriously consider whether the goals of stare decisis are outweighed, rather than dictated, by the prudential and pragmatic considerations that inform the doctrine to enforce a clearly erroneous decision.... The court must weigh [the] benefits of [stare decisis] against its burdens in deciding whether to overturn a precedent it thinks is unjust.... It is more important that the court should be right upon later and more elaborate consideration of the cases than consistent with previous declarations.... Moreover, [e]xperience can and often does demonstrate that a rule, once believed sound, needs modification to serve justice better.... Indeed, [i]f law is to have current relevance, courts must have and exert the capacity to change a rule of law when reason so requires.... In accordance with these principles, we have not hesitated to revisit and overrule our prior holdings, including prior holdings applicable to criminal matters ... once we are convinced that they were incorrect and unjust.” (Citations omitted; internal quotation marks omitted.) Id. at 520–21, 949 A.2d 1092 ; see also Jaiguay v. Vasquez, 287 Conn. 323, 351 n. 22, 948 A.2d 955 (2008) ( “[s]tare decisis ... is neither an end in itself nor an inexorable command”). In the present case, we must decide whether these principles dictate either that we reaffirm our holding in Gordon or that we overrule it.
In Gordon, a jury found the defendant, Mitchell Gordon, guilty of sexual assault in the first degree and robbery in the third degree predicated on, inter alia, the victim's out-of-court identification of Gordon as her assailant shortly after the incident, as well as her in-court identification of him. See State v. Gordon, supra, 185 Conn. at 403–404, 407, 417–19, 441 A.2d 119. On appeal to this court, we concluded that the initial identification procedure, pursuant to which officers informed the victim that Gordon had been located near the scene of the crime and displayed him alone in a room to the victim from behind a one-way mirror at the police station, was unnecessarily suggestive and that the identification was unreliable. See id. at 414–18, 441 A.2d 119. We further determined that the victim's in-court identification of Gordon was not based on observation independent of the highly suggestive out-of-court identification procedure. See id. at 418–19, 441 A.2d 119. Finally, we rejected the state's contention that the admission of the victim's identifications was harmless error, stating: “Ordinarily the burden of establishing that harm resulted from a trial court error rests on the appellant.... However, there are some constitutional rights so basic to a fair trial that their infraction can never be treated as harmless error.... If error touches a less basic constitutional right, we sometimes apply the harmless error exception, but only sparingly, in a few, discrete circumstances....
“Thus we refuse to expand our harmless constitutional error doctrine to the discrete circumstances of unnecessarily suggestive and unreliable identifications, the admission of which significantly impairs the truth finding function of the jury.... Were we to do so we would fail to correct negligent infractions of constitutional rights and tempt some public officials to overstep the law in their zeal to convict the guilty. Some would yield to such temptation. The devastating nature of both negligently and deliberately obtained, unreliable eyewitness identifications would inevitably lead to the conviction of innocent persons. Hence sound judicial policy requires reversal whenever the erroneous admission of an unnecessarily suggestive and unreliable identification has violated a defendant's constitutional rights.” (Citations omitted; internal quotation marks omitted.) Id. at 419–20, 441 A.2d 119.
Justice Anthony J. Armentano, joined by Justice Arthur H. Healey, dissented in part in Gordon on the ground that the improper admission of eyewitness identification testimony should be subject to harmless error analysis and that the improperly admitted identification testimony in that case was harmless beyond a reasonable doubt. See State v. Gordon, supra, 185 Conn. at 426, 441 A.2d 119 (Armentano, J., concurring in part and dissenting in part).
Upon reexamination of Gordon, we conclude that its reasoning is fundamentally flawed for two reasons. First, our holding was based on the faulty premise that harmless error review only applies “in a few, discrete circumstances.” (Internal quotation marks omitted.) Id. at 419, 441 A.2d 119. As we explain hereinafter, automatic reversal, not harmless error review, is the exceptional remedy for instances of constitutional impropriety. Second, the state's use of unreliable eyewitness identifications resulting from unduly suggestive police procedures is not one of the rare circumstances necessitating a new trial, as such identification evidence does not inevitably undermine the integrity or legitimacy of the adjudicatory process. Rather, reviewing courts are fully capable of ascertaining whether the admission of such tainted evidence, like the improper admission of any other evidence, affected the fact finder's ultimate determination. Logic and reason compel the conclusion, therefore, that appellate courts should not reverse an otherwise trustworthy conviction merely because it was based in part on improperly obtained and unreliable eyewitness testimony, if the state can establish that the erroneous admission of that evidence was harmless.
Although the court in Gordon concluded that harmless error review was applicable only when a defendant's “less basic constitutional right [s]” are infringed on; (internal quotation marks omitted) id. at 419, 441 A.2d 119 ; we agree with Judge Lavine that “the legal landscape supporting the premise of the ... holding [in Gordon ]—that, when a defendant's constitutional rights have been violated, harmless error analysis is the exception to the general rule, only to be used sparingly, in a few, discrete circumstances; [id.]—has changed substantially since Gordon was decided. As [this court] explained in State v. Jenkins, 271 Conn. 165, [186–87] 856 A.2d 383 (2004) : It is well settled that most improprieties, even those of constitutional magnitude, can be harmless and, therefore, do not require the reversal of a defendant's conviction.... [T]he appellate harmless error doctrine is rooted in that fundamental purpose of our criminal justice system—to convict the guilty and acquit the innocent. The harmless error doctrine recognizes the principle that the central purpose of a criminal trial is to decide the factual question of the defendant's guilt or innocence ... and promotes public respect for the criminal process by focusing on the underlying fairness of the trial rather than on the virtually inevitable presence of immaterial error.... Accordingly, we forgo harmless error analysis only in rare instances involving a structural defect of constitutional magnitude.... Structural defect cases defy analysis by harmless error standards because the entire conduct of the trial, from beginning to end, is obviously affected.... Arizona v. Fulminante, [499 U.S. 279, 309–10, 111 S.Ct. 1246, 113 L.Ed.2d 302 (1991) ]....
* * *
“Therefore, the statement in State v. Gordon, supra, 185 Conn. at 419, 441 A.2d 119, that Connecticut appellate courts sometimes apply the harmless error exception, but only sparingly, in a few, discrete circumstances, lacks viability.” (Citations omitted; emphasis in original; internal quotation marks omitted.) State v. Artis, supra, 136 Conn.App. at 635–37, 47 A.3d 419 (Lavine, J., concurring in part and dissenting in part); see also Rose v. Clark, 478 U.S. 570, 579, 106 S.Ct. 3101, 92 L.Ed.2d 460 (1986) (“if the defendant had counsel and was tried by an impartial adjudicator, there is a strong presumption that any other errors that may have occurred are subject to harmless-error analysis”).
As Judge Lavine explained, the court's assertion in Gordon that appellate courts infrequently apply the harmless error doctrine to constitutional violations is an inaccurate statement of the law, at least as that law has developed since Gordon was decided. In view of the fact that our holding in Gordon is founded on a flawed premise, we must determine whether that holding nevertheless is sustainable on the ground that the admission of unnecessarily suggestive and unreliable eyewitness identifications is one of the rare forms of structural error to which harmless error review does not apply. As we now explain, the improper admission of such evidence does not constitute structural error.
We previously have stated that structural defect cases “contain a defect affecting the framework within which the trial proceeds, rather than simply an error in the trial process itself.... Such errors infect the entire trial process ... and necessarily render a trial fundamentally unfair.... Put another way, these errors deprive defendants of basic protections without which a criminal trial cannot reliably serve its function as a vehicle for [the] determination of guilt or innocence ... and no criminal punishment may be regarded as fundamentally fair.” (Citations omitted; internal quotation marks omitted.) State v. Jenkins, supra, 271 Conn. at 187, 856 A.2d 383. Thus, “when the consequences of the deprivation of the defendant's constitutional right are necessarily unquantifiable and indeterminate, [the deprivation of that right] unquestionably qualifies as structural error.” (Internal quotation marks omitted.) State v. Lopez, 271 Conn. 724, 738, 859 A.2d 898 (2004). “A structural error creates a defect in the trial mechanism such that, while it is virtually impossible to pinpoint the exact harm, it remains abundantly clear that the trial process was flawed significantly. For this reason, [e]rrors of this magnitude are per se prejudicial and require that the underlying conviction be vacated.” (Emphasis added; internal quotation marks omitted.) Id. at 739, 859 A.2d 898. Examples of such structural errors include, among others, racial discrimination in the selection of a grand jury or petit jury and the denial of a defendant's right to counsel, right to a public trial, or right to self-representation. See, e.g., Batson v. Kentucky, 476 U.S. 79, 100, 106 S.Ct. 1712, 90 L.Ed.2d 69 (1986) (racial discrimination in selection of petit jury); Vasquez v. Hillery, 474 U.S. 254, 263–64, 106 S.Ct. 617, 88 L.Ed.2d 598 (1986) (racial discrimination in selection of grand jury); Waller v. Georgia, 467 U.S. 39, 49 n. 9, 104 S.Ct. 2210, 81 L.Ed.2d 31 (1984) (denial of right to public trial); Faretta v. California, 422 U.S. 806, 836, 95 S.Ct. 2525, 45 L.Ed.2d 562 (1975) (denial of right to self-representation); Gideon v. Wainwright, 372 U.S. 335, 344–45, 83 S.Ct. 792, 9 L.Ed.2d 799 (1963) (denial of right to counsel); see also Tumey v. Ohio, 273 U.S. 510, 535, 47 S.Ct. 437, 71 L.Ed. 749 (1927) (denial of right to impartial judge).
Another example of structural error, namely, instructional error concerning the reasonable doubt standard, is illustrative. As the United States Supreme Court explained in Sullivan v. Louisiana, 508 U.S. 275, 113 S.Ct. 2078, 124 L.Ed.2d 182 (1993), the “right to a jury verdict of guilt beyond a reasonable doubt is ... a basic protection [the] precise effects [of which] are unmeasurable, but without which a criminal trial cannot reliably serve its function....” (Citation omitted; internal quotation marks omitted.) Id. at 281, 113 S.Ct. 2078. The court further stated that an incorrect definition of reasonable doubt in a jury instruction “vitiates all the jury's findings. A reviewing court can only engage in pure speculation—its view of what a reasonable jury would have done. And when it does that, the wrong entity judge[s] the defendant guilty.” (Emphasis omitted; internal quotation marks omitted.) Id. Accordingly, the court concluded that “[t]he deprivation of [the] right [to a jury verdict of guilty beyond a reasonable doubt], with consequences that are necessarily unquantifiable and indeterminate, unquestionably qualifies as structural error.” (Internal quotation marks omitted.) Id. at 281–82, 113 S.Ct. 2078.
In reaching its conclusion, the court in Sullivan distinguished improper reasonable doubt instructions from “errors [that] occur during the presentation of the case to the jury, and [that] may therefore be quantitatively assessed in the context of other evidence presented....” (Citation omitted; internal quotation marks omitted.) Id. at 281, 113 S.Ct. 2078. When this type of impropriety occurs, the defendant's conviction will not be reversed if the state demonstrates, in light of independent evidence establishing the defendant's guilt, that the violation of the defendant's constitutional right was harmless beyond a reasonable doubt. See, e.g., State v. Mitchell, 296 Conn. 449, 459–60, 996 A.2d 251 (2010). As we discussed previously, most constitutional violations are subject to such harmless error review. See, e.g., Moore v. Illinois, 434 U.S. 220, 232, 98 S.Ct. 458, 54 L.Ed.2d 424 (1977) (harmless error review applies to admission of pretrial identification in violation of right to counsel); Milton v. Wainwright, 407 U.S. 371, 377–78, 92 S.Ct. 2174, 33 L.Ed.2d 1(1972) (admission of postindictment confession in violation of right to counsel subject to harmless error review); Chambers v. Maroney, 399 U.S. 42, 53, 90 S.Ct. 1975, 26 L.Ed.2d 419 (1970) (admission of evidence in violation of fourth amendment does not warrant reversal if it was harmless); Harrington v. California, 395 U.S. 250, 252–54, 89 S.Ct. 1726, 23 L.Ed.2d 284 (1969) (no reversal required when admission of evidence in violation of right to confront witnesses constituted harmless error); State v. Mitchell, supra, at 459, 996 A.2d 251 (harmless error doctrine applies to admission of statements taken in violation of Miranda v. Arizona, 384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694  ).
The state's use of an unreliable eyewitness identification that was the product of an unnecessarily suggestive procedure does not infect or permeate the entire trial process in the same manner as structural errors do; rather, the use of such evidence simply results in the introduction of additional, albeit improper, evidence into the proceedings. A reviewing court can examine the relative weight and import of such evidence to determine whether it may have influenced the jury's verdict in light of the other evidence adduced by the state. Furthermore, we reject the defendant's contention that the highly persuasive nature of eyewitness identification evidence mandates the conclusion that its improper admission is tantamount to structural error in that harmless error analysis would be an inadequate remedy for the due process violation. On the contrary, even the unconstitutional admission of a defendant's involuntary confession, which “is probably the most probative and damaging evidence that can be admitted against him”; Arizona v. Fulminante, supra, 499 U.S. at 296, 111 S.Ct. 1246 ; is subject to harmless error review. Id. at 295–96, 111 S.Ct. 1246. In sum, we see no reason to treat the improper admission of eyewitness identification evidence as structural error because it simply is not error of a structural nature. We also see no reason to treat the admission of such evidence as qualitatively different, for purposes of harmless error review, from any other improperly admitted evidence. Moreover, it is significant that, because of the constitutional magnitude of the error, the burden falls on the state to prove that the admission of the tainted identification was harmless beyond a reasonable doubt. See, e.g., State v. Rose, 305 Conn. 594, 625–26, 46 A.3d 146 (2012). Therefore, if there is any reasonable doubt that the admission of the identification was harmful, the defendant will be entitled to a new trial.
This conclusion accords with the view of federal and sister state courts, which uniformly apply harmless error analysis to claims that the admission of an unnecessarily suggestive and unreliable eyewitness identification deprived a defendant of a fair trial. See, e.g., Manson v. Brathwaite, supra, 432 U.S. at 118 n. *, 97 S.Ct. 2243 (Stevens, J., concurring); Commonwealth
v. Jones, 423 Mass. 99, 105–106, 666 N.E.2d 994 (1996) ; State v. Van Egmond, 206 Neb. 356, 359, 293 N.W.2d 74 (1980) ; People v. Johnson, 80 N.Y.2d 798, 799, 599 N.E.2d 682, 587 N.Y.S.2d 278 (1992) ; State v. Souza, 110 R.I. 261, 268–69, 292 A.2d 214 (1972). Indeed, the defendant has not identified a single case other than Gordon, and our research has not disclosed one, in which a court has held that the improper admission of such identification evidence is a type of structural error requiring per se reversal. Indeed, this court never has applied Gordon to grant an automatic reversal of a conviction in a case involving eyewitness identification testimony, the use of which violated principles of due process. On the contrary, since our decision in Gordon, we have indicated that harmless error review would apply in such circumstances. State v. Milner, supra, 206 Conn. at 536 n. 11, 539 A.2d 80.
It appears that our holding in Gordon may have been predicated on policy considerations above and beyond any constitutional requirements. See State v. Gordon, supra, 185 Conn. at 420, 441 A.2d 119 (“sound judicial policy requires reversal [of a conviction] whenever the erroneous admission of an unnecessarily suggestive and unreliable identification has violated a defendant's constitutional rights”). To the extent that Gordon rests on such independent policy concerns, we are not persuaded that they outweigh the policy favoring harmless error review. As we have recognized, “[a] criminal trial is a search for truth”; (internal quotation marks omitted) State v. Morales, 232 Conn. 707, 723, 657 A.2d 585 (1995) ; and, when the state's improper use of inadmissible evidence, including unreliable eyewitness identification testimony, does not impair that truth-seeking function—that is, when there is no reasonable possibility that the result of the trial would have been different without that evidence—preservation of the defendant's conviction serves the strong public interest in convicting the guilty without unfairly prejudicing the defendant.
We acknowledge the powerful effect that eyewitness identification testimony has on juries and recognize that the improper admission of that evidence will constitute harmful error in many instances, particularly when there is no other such eyewitness identification testimony. See, e.g., Watkins v. Sowders, 449 U.S. 341, 352, 101 S.Ct. 654, 66 L.Ed.2d 549 (1981) (Brennan, J., dissenting) (“ ‘[E]yewitness testimony is likely to be believed by jurors, especially when it is offered with a high level of confidence, even though the accuracy of an eyewitness and the confidence of that witness may not be related to one another at all. All the evidence points rather strikingly to the conclusion 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!” ’ ” [Emphasis omitted.] ). Nevertheless, the improper use of eyewitness identification testimony does not implicate the structural integrity of a trial, and we cannot say that such testimony is always so powerful and prejudicial, considering that testimony in light of the other evidence, that harmless error review is necessarily inadequate to safeguard the defendant's right to a fair trial. For these reasons, we improperly concluded otherwise in Gordon, which we hereby overrule.
Judge Lavine provided the following persuasive illustration of this point: “For example, imagine a bank robbery case in which the out-of-court identification made by a teller is unnecessarily suggestive and unreliable but is admitted at trial. The bank robber is clearly depicted on [a video recording]; he is found outside the bank in possession of marked bills; his mother testifies that he admitted to her that he robbed the bank; five independent witnesses who were inside the bank identify him and he confesses on television. Under Gordon, harmless error analysis would not be available following a conviction of the bank robber, and [a reviewing] court would be required to reverse the conviction.” State v. Artis, supra, 136 Conn.App. at 634 n. 15, 47 A.3d 419 (Lavine, J., concurring in part and dissenting in part).
The defendant also asserts that we should not overrule Gordon because “[t]here is no deterrent effect when an appellate court, having concluded that [an identification] procedure was unnecessarily suggestive, and the identification unreliable, nevertheless upholds the conviction under the harmless error doctrine.” We disagree. The suppression of unnecessarily suggestive and unreliable identifications is sufficient to deter police overreaching in the same manner and to the same extent that the suppression of improperly obtained confessions, the admission of which also is reviewed for harmless error, adequately deters police coercion. See Arizona v. Fulminante, supra, 499 U.S. at 295–96, 111 S.Ct. 1246. Moreover, harmless error analysis creates no incentive for law enforcement officers to overstep the law, as such officials cannot know whether their actions subsequently will be deemed harmless.
Having concluded that harmless error review applies to the admission of unreliable eyewitness identifications that are the product of unnecessarily suggestive procedures, we now must determine whether the purportedly improper admission of Otero's identification testimony was harmless beyond a reasonable doubt. The state maintains that the testimony was indeed harmless because Miano's testimony, including her identification of the defendant, with whom she was acquainted, provided strong, independent evidence of the defendant's guilt. The state further argues that Otero's identification of the defendant as the man with whom he had engaged in a fistfight before being stabbed did not influence the jury's finding of guilt because Otero specifically “disclaimed any knowledge of whether the defendant participated in the assault for which he was convicted,” and, therefore, the jury, in finding the defendant guilty of accessory to assault in the first degree by means of a dangerous instrument, must have relied solely on Miano's identification of the defendant as one of the three individuals who accosted Otero while he was on the ground. The defendant maintains that the state's reliance on Miano's testimony as proof of harmlessness is misplaced because Miano was biased against the defendant, her testimony was “imprecise and confused,” her ability to observe accurately was questionable, and she acknowledged that she was under the influence of alcohol on the evening of the incident. We agree with the state.
The following additional facts are relevant to our resolution of this issue. At trial, Otero testified that, after being knocked to the ground, he was assaulted by three or four individuals and that, although he was unable to see any of his assailants, he heard Miano telling them to stop. He further stated that he had had no contact with Miano since the night of the assault. Miano also testified as to what she had observed that evening, stating that Robert Acevedo, her boyfriend at the time, drove her, Anna Acevedo and the defendant, who was dating Anna Acevedo, to Club NV at around 11 p.m. At that time, she had known the defendant for approximately one to two months, through her friendship with Anna Acevedo. After leaving Club NV, Miano briefly spoke to Otero while she, Anna Acevedo, and Robert Acevedo were in Robert Acevedo's car waiting for the defendant. Miano testified that, although she had been “tipsy” earlier from consuming two or three alcoholic beverages, her vision and hearing were not impaired after she left Club NV.
As Otero was walking away from Robert Acevedo's vehicle, Robert Acevedo, who Miano believed was upset with Otero for having spoken to her, attempted to strike Otero with his vehicle. The defendant then arrived and, after briefly entering the car, left to confront Otero. Miano, who was distracted by her cell phone, was not aware of the resulting scuffle until she saw the defendant pushing Otero while they were standing in front of the vehicle. At this point, Robert Acevedo and Anna Acevedo got out of the vehicle. As soon as Miano noticed that Otero had fallen to the ground, she also exited the vehicle. Miano then observed Robert Acevedo, Anna Acevedo, and the defendant fighting with Otero, although she acknowledged that she never saw any of them actually strike Otero. Miano also testified that, as she was attempting to pull Anna Acevedo off of Otero and help him stand up, police officers approached the scene. Miano, Robert Acevedo, Anna Acevedo, and the defendant then quickly returned to the vehicle, and Robert Acevedo drove to Miano's house. Soon thereafter, Miano ceased all contact with the Acevedos and the defendant. Finally, Miano testified that she initially had avoided the police when they sought her assistance in connection with their investigation of the incident because she feared that Robert Acevedo, the defendant, or one of his friends might harm her or her children if she did so.
The principles governing our review of the state's claim are well established. “[W]hether an error is harmful depends on its impact on the trier of fact and the result of the case.... This court has held in a number of cases that when there is independent overwhelming evidence of guilt, a constitutional error would be rendered harmless beyond a reasonable doubt.... When an [evidentiary] impropriety is of constitutional proportions, the state bears the burden of proving that the error was harmless beyond a reasonable doubt.... [W]e must examine the impact of the evidence on the trier of fact and the result of the trial.... If the evidence may have had a tendency to influence the judgment of the jury, it cannot be considered harmless.... That determination must be made in light of the entire record [including the strength of the state's case without the evidence admitted in error].” (Internal quotation marks omitted.) State v. Mangual, 311 Conn. 182, 214–15, 85 A.3d 627 (2014).
As we noted previously, the defendant was convicted of being an accessory to assault in the first degree by means of a dangerous instrument. A person is an accessory to assault in the first degree by means of a dangerous instrument when he has the specific intent to cause serious physical injury to an individual and solicits, requests, commands, importunes or intentionally aids another person, who, using a dangerous instrument, causes serious injury to that individual while also possessing the specific intent to do so. See General Statutes §§ 53a–59 (a)(1) and 53a–8 (a) ; see also State v. Williams, 237 Conn. 748, 754, 679 A.2d 920 (1996) (discussing intent element of § 53a–59 [a] ).
Applying these principles to the present case, we conclude that, even if the admission of Otero's identification testimony was improper, such testimony was harmless in light of Miano's independent identification of the defendant, whom she knew, as the individual who initially had confronted Otero and, more importantly, as one of the three individuals who subsequently assaulted Otero while he was on the ground. In contrast, Otero merely identified the defendant, a stranger to him, as the man with whom he initially had engaged in a fist-fight; Otero specifically disclaimed any knowledge of the identity of the individuals who assaulted him after he was knocked to the ground. Consequently, in finding the defendant guilty of accessory to assault in the first degree by means of a dangerous instrument, the jury must have credited the testimony of Miano, who drove with the defendant to and from Club NV and had known him socially for one to two months. Accordingly, it is highly unlikely that Otero's identification of the defendant as the man with whom he had a brief fistfight had any effect on the jury's finding of guilt with respect to the charge of assault in the first degree as an accessory. Cf. State v. Mitchell, supra, 296 Conn. at 462–63, 996 A.2d 251 (purportedly improper admission of defendant's statements into evidence was harmless error because statements were of limited probative value and victim identified defendant); State v. Dupigney, 78 Conn.App. 111, 120–22, 826 A.2d 241 (admission of evidence identifying defendant as shooter, even if improper, was nevertheless harmless beyond reasonable doubt because, inter alia, three independent witnesses also identified defendant as shooter), cert. denied, 266 Conn. 919, 837 A.2d 801 (2003).
The defendant's contention that “the jury could have reasonably doubted [Miano's] objectivity” because she was “friends with Otero and intervened in the scuffle to pull Anna [Acevedo] off of him” is belied by the record. At trial, Miano stated that she briefly had encountered Otero, who was a friend of the father of her children, only twice in her life prior to seeing him at Club NV on the night of the assault, and Otero testified that he and Miano had not been in contact since then. Indeed, Miano even forgot Otero's last name during her testimony. Suffice it to say that the record is devoid of any evidence that Miano was biased against the defendant as a result of her extremely attenuated relationship with Otero.
Although the state, during closing argument, mentioned Otero's identification of the defendant, the state also argued to the jury that Miano's testimony alone was more than sufficient to establish that the defendant was one of the individuals who had surrounded and then assaulted Otero. Moreover, the state relied heavily on Miano's identification of the defendant in arguing that the state had proven the defendant's guilt beyond a reasonable doubt. Finally, defense counsel cross-examined Otero extensively and impeached him through the testimony of Bilbo, the police officer who directly contradicted Otero's claim that he made a definitive out-of-court identification of the defendant.
We conclude, therefore, that, even if the defendant's due process rights were violated by the state's use of Otero's identification testimony, the admission of that evidence was harmless beyond a reasonable doubt. Accordingly, we agree with the state that the defendant's conviction should be affirmed.
The judgment of the Appellate Court is reversed and the case is remanded to that court with direction to affirm the judgment of the trial court.
In this opinion the other justices concurred.