State v. Marquez

967 A.2d 56 (2009) 291 Conn. 122

STATE of Connecticut
v.
Julian MARQUEZ.

No. 17663.

Supreme Court of Connecticut.

Argued September 3, 2008. Decided April 14, 2009.

Conrad Ost Seifert, special public defender, for the appellant (defendant).

Michele C. Lukban, senior assistant state's attorney, with whom, on the brief, were Gail P. Hardy, state's attorney, and Edward R. Narus, senior assistant state's attorney, for the appellee (state).

KATZ, PALMER, VERTEFEUILLE, ZARELLA and SCHALLER, Js.

ZARELLA, J.

The defendant, Julian Marquez, appeals from the judgment of conviction, rendered after a jury trial, of one count of felony murder in violation of General Statutes § 53a-54c, two counts of robbery in the first degree in violation of General Statutes § 53a-134 (a)(2), and one count of attempt to commit robbery in the first degree in violation of § 53a-134 (a)(2) and General Statutes § 53a-49. On appeal, the defendant claims that the trial court's denial of his motion to suppress two eyewitness identifications violated his right to due process of law under both the fourteenth amendment to the United States constitution^[1] and article first, § 8, of the Connecticut constitution.^[2] Specifically, the defendant challenges the trial court's ruling that those identifications were reliable under the totality of the circumstances test set forth by the United States Supreme Court in Neil v. Biggers, 409 U.S. 188, 93 S. Ct. 375, 34 L. Ed. 2d 401 (1972), and Manson v. Brathwaite, 432 U.S. 98, 97 S. Ct. 2243, 53 L. Ed. 2d 140 (1977),^[3] even though the court found that the police had used an unnecessarily suggestive identification procedure. The defendant also claims that, even if this court concludes that the trial court ruled correctly under existing law, we nevertheless should exercise our supervisory authority to mandate the adoption and use of new and purportedly more accurate identification procedures.^[4]

The state urges this court to uphold the trial court's denial of the defendant's motion to suppress the identifications. Specifically, the state argues that the trial court correctly concluded that the identifications were reliable under the totality of the circumstances and asserts that there is no basis for this court to exercise its supervisory authority to mandate new identification procedures.^[5] As an alternative ground for affirmance, the state also argues that the trial court improperly concluded that the identification procedures used by the police were unnecessarily suggestive.^[6] We agree with the state and affirm the judgment of the trial court on this alternative ground.

In its memorandum of decision on the defendant's motion to suppress, the trial court made the following findings of fact. On the evening of Friday, December 19, 2003, Mark Clement and his friend, Christopher Valle, were visiting at the apartment of a mutual friend, Miguel Delgado, Jr., at 134 Babcock Street in Hartford. The three men, along with various others, socialized regularly on weekends at Delgado's apartment. The apartment itself was on the third floor of the building and could be accessed by visitors through a front door that opened into a lighted common hallway at the top of the interior staircase.

Delgado's apartment was relatively small, consisting of a short interior hallway leading from the front door directly into the living room, a small game room adjacent to the living room, a kitchen to the rear of the living room and a bedroom off of the kitchen. On the evening of December 19, the living room was illuminated only indirectly by light emanating from the kitchen and game room. There was a couch on the back wall of the living room that faced the front door.

After arriving at the apartment, Delgado, Clement, Valle and another man named Amauri Escobar primarily remained in the game room playing video games and drinking alcoholic beverages.^[7] While several other people visited the apartment intermittently throughout the night, only these four men were present just prior to the robbery.^[8]

At around midnight on December 20, as Valle was preparing to leave the apartment, he exited the game room to say goodnight to Delgado, who was standing just inside the front door attempting to get rid of two men who stood facing him in the common hallway. As Valle approached the front door, one of the strangers, a Hispanic male in his early twenties wearing braids and black clothing, pointed a handgun directly at him and entered the apartment, forcing Valle and Delgado backward into the living room and ordering them to sit on the couch. Shortly thereafter, the intruders entered the game room, announced their presence and ordered Clement and Escobar to join Valle and Delgado. Once everyone returned to the living room, the intruders were only one or two feet away from their victims for a period of several minutes. Importantly, both Valle and Clement had multiple opportunities to see the faces of the perpetrators, particularly the gunman. Valle initially saw the gunman in the brightly lit common hallway while Delgado was trying to get rid of him. Clement observed the gunman's face when the gunman entered the lighted game room and announced that "it was a stickup." Further, while Valle and Clement were seated on the couch, they both viewed the intruders' faces with the light from the kitchen and the game room providing illumination.

The intruders then ordered all of them to surrender their money and jewelry, which Clement, Valle and Escobar did promptly. Delgado, however, offered only a small amount of marijuana, insisting that that was all he had. The intruders, who were dissatisfied with this offer, apparently were convinced that Delgado had money and drugs stashed elsewhere in the apartment, and demanded to be taken to inspect the back bedroom. Delgado, who was now standing to the side of the couch and blocking the entrance to the kitchen, refused. Delgado suddenly rushed at the gunman and grabbed his arm. A struggle ensued, during which Delgado forced the gunman back toward the front of the apartment, and a shot rang out. Valle jumped to his feet from his position on the couch and began to head for the perceived safety of the game room. He then heard a second shot and saw Delgado fall to his knees on the floor. Valle made it to the game room with Clement behind him, while Escobar apparently escaped out of the rear exit of the apartment.

Valle heard a third shot fired, and, once it was apparent that the intruders had fled, he eventually emerged from the game room. Feeling around on the floor in the relative darkness of the hall near the front door, Valle discovered Delgado lying in a pool of blood. When he turned the living room light on, Valle realized that Delgado was critically wounded and called the police. Immediately after the incident, both Valle and Clement expressed to police investigators their confidence that they could identify the gunman.

On December 23, 2003, while making his regular visit to his parole officer, Valle was startled to observe the defendant at the office of the parole officer, immediately recognizing him as the gunman who had robbed him four days earlier. Valle reported his observation to personnel on duty in the office. This information was relayed to Detective Patricia Beaudin of the Hartford police department, who was leading the investigation into the incident. On the basis of this information, a photographic array was produced consisting of photographs of eight men fitting the description that Valle had provided, including that of the defendant.

Beaudin and her partner, Detective Ezequiel Laureano, contacted Valle and asked him to come to the police station to view the photographs in the array. Prior to having him look at the photographic array, Beaudin instructed Valle simply to look at the photographs and tell her if he recognized anyone, and that it was fine if he did not. Neither Beaudin nor Laureano told Valle that he had to select a photograph, and they did not indicate in any way which photograph he should pick. In addition, consistent with the detectives' instructions, there was a notice prominently printed at the bottom of the photographic array that provided: "You have been asked to look at this group of photographs. The fact that they are shown to you should not influence your judgment. You should not conclude or guess that the photographs contain the person who committed the offense under investigation. You are not obligated to identify anyone. It is just as important to free innocent persons from suspicion as to identify guilty parties. Please do not discuss this case with other witnesses nor indicate in any way that you have, or have not identified someone."

Valle immediately selected the photograph depicting the defendant, indicating that he was sure that the individual pictured was the gunman. Although acknowledging some superficial differences between the defendant's photograph and the others in the array, Valle denied that such discrepancies played any roll in his choice.

Four days later, Beaudin contacted Clement and requested that he view the photographic array that Valle had previously viewed.^[9] The detectives followed an almost identical procedure with Clement, and either read aloud or pointed out the same prominent notice at the bottom of the board on which the photographic array was mounted. Clement believed that the array contained photographs of known robbers but did not know if photographs of either of the persons who had robbed him would be included in the display. He further indicated that, although neither of the detectives pressured him to select any photograph or indicated in any way that he should choose a particular photograph, he did feel that a suspect was probably in the array and that he should "pick somebody...."

Clement found himself immediately drawn to one photograph but was concerned about speaking up too quickly and identifying the wrong person. He described how he eliminated all but two of the photographs as possibilities and how he kept returning to the photograph that originally had garnered his attention. Clement noted that he based his identification almost exclusively on the defendant's eyes, which he recognized as the eyes of the gunman who had robbed him and his companions. He indicated that the choice he had made was based on his own "gut feeling," derived from his personal observations of the gunman during the robbery. When he finally did inform Beaudin of his choice, about which he felt fairly confident, she informed him that he had done well because he had chosen the same photograph as Valle.

On the basis of these identifications, an arrest warrant was issued for the defendant, which was executed on December 30, 2003. The state subsequently filed a five count information charging the defendant with one count of felony murder, three counts of robbery in the first degree, and one count of attempt to commit robbery in the first degree.

Prior to trial, the defendant filed a motion to suppress as evidence any pretrial or in-court identifications offered by the state. The defendant argued that the procedures used in connection with the photographic array identifications were unnecessarily suggestive and unreliable and that any in-court identification by Valle and Clement would be irreparably tainted by such prior, purportedly improper identifications. The state responded that the identification procedures were not unnecessarily suggestive and that, even if they were, the identifications themselves were sufficiently reliable under the totality of the circumstances.

The trial court held a hearing on the motion on December 7, 2005, and heard testimony from Valle, Clement and Beaudin. In addition, the trial court considered several scientific articles and reports regarding the suggestiveness and reliability of various identification methods. On December 20, 2005, the court heard arguments on the motion, and, on January 4, 2006, the court made an oral ruling denying the defendant's motion to suppress. The trial court issued a "[c]orrected" memorandum of decision on the defendant's motion to suppress on March 20, 2006.

In its memorandum of decision, the trial court determined that the identification procedures used were unnecessarily suggestive but nonetheless sufficiently reliable under the totality of the circumstances to be admissible at trial. The court based its determination primarily on what it termed the "universal judgment of the relevant scientific community" that sequential identification procedures, pursuant to which a witness views photographs or live suspects one at a time, are superior to the traditional lineup or photographic array in which a witness views all of the individuals simultaneously. The court was persuaded by the suggestion in the literature that sequential procedures may be superior because they limit the operation of "``relative judgment,'" whereby a witness may be tempted to choose the photograph or individual who looks the most like their memory of the perpetrator relative to the others viewed through a process of elimination, rather than by actually comparing each individual to their memory of the perpetrator. Sequential procedures reduce or eliminate this tendency, it is argued, by depriving a witness of the opportunity to compare subjects to each other and by forcing the witness to rely strictly on his memory rather than on the relative judgment process.^[10]

The trial court also was "troubled" by the fact that the identification procedures in this case were not "double blind." To qualify as double-blind, a photographic array must be administered by an uninterested party without knowledge of which photograph represents the suspect. Again relying on the scientific literature, the court found that "[t]he risk of producing a misidentification in such circumstances, due to conscious or unconscious bias by a highly interested person administering the procedure, is so well established in the relevant scientific literature that experts have strongly recommended that all pretrial identification procedures be conducted only by persons who do not know which member of the lineup or photospread is the suspect." The court was concerned that Beaudin's statement to Clement congratulating him on choosing the same photograph as Valle was outward evidence of a bias in the process that subtly and perhaps subconsciously infected the procedure before Clement had made his choice.^[11]

For the foregoing reasons, and on the basis of the scientific research,^[12] the trial court essentially concluded that the traditional identification procedure that the police used in this case was per se unnecessarily suggestive. The court further directed that "[p]olice ... conducting photo[graphic] identifications should henceforth strive to eliminate the danger of misidentification arising from the simultaneous showing of multiple [photographs] by making all such showings sequentially." Despite finding these flaws in the identification process in the present case, the court nonetheless examined the totality of the factual circumstances under the Manson test, determined that the identifications by Valle and Clement were reliable, and denied the defendant's motion to suppress.

Thereafter, the jury found the defendant guilty of the felony murder charge, two of the robbery charges and the attempted robbery charge. The trial court rendered judgment in accordance with the jury verdict and sentenced the defendant to a total effective term of fifty years imprisonment, execution suspended after thirty-five years, and five years probation. This appeal followed.

We begin by noting that we need not reach the defendant's claim that the trial court improperly concluded that the challenged identifications were reliable under the Manson test because we decide this case on the state's alternative ground. In this regard, the state claims that the trial court improperly determined that the identification procedures were unnecessarily suggestive, whereas the defendant agrees with the court's conclusion. We elect to address this aspect of the court's ruling, rather than simply uphold the court's analysis of reliability, in order to confront what we view as the trial court's establishment of a per se rule with respect to the question of the suggestiveness of the identification procedures at issue. Although we appreciate the trial court's laudable desire to improve the dependability of the eyewitness identification process, we cannot agree that the procedures used in this case were unnecessarily suggestive.

Turning to the applicable legal principles, we first observe that the defendant invoked his due process rights under both the fourteenth amendment to the United States constitution and article first, § 8, of the Connecticut constitution in support of his motion to suppress. The trial court's memorandum of decision does not differentiate its analysis between these sources of law. This is the correct approach as this court explicitly has held that article first, § 8, provides no greater protection than the federal constitution in the realm of identification procedures. State v. Ledbetter, 275 Conn. 534, 568, 881 A.2d 290 (2005), cert. denied, 547 U.S. 1082, 126 S. Ct. 1798, 164 L. Ed. 2d 537 (2006). We reaffirm the congruence between the protections afforded by our state constitution and the federal constitution in the area of pretrial identification and therefore proceed to analyze the question in the same fashion under both provisions.

Although we have never directly addressed the standard for reviewing whether a photographic array is unnecessarily suggestive,^[13] we are influenced by federal precedent and the approach taken by our sister states. The United States Supreme Court has held that "the ultimate question as to the constitutionality of ... pretrial identification procedures ... is a mixed question of law and fact...." Sumner v. Mata, 455 U.S. 591, 597, 102 S. Ct. 1303, 71 L. Ed. 2d 480 (1982). More specifically, the Fifth Circuit Court of Appeals recently has affirmed that "a court must determine whether the pretrial identification was impermissibly suggestive.... Such an analysis is a mixed question of law and fact." (Citation omitted.) Coleman v. Quarterman, 456 F.3d 537, 544 (5th Cir. 2006), cert. denied, 549 U.S. 1343, 127 S. Ct. 2030, 167 L. Ed. 2d 772 (2007); accord Lee v. Keane, Docket No. 01-2136, 2002 WL 31520115, *1-2, 2002 U.S.App. LEXIS 23588, *4 (2d Cir. November 13, 2002), cert. denied sub nom. Lee v. Fischer, 540 U.S. 869, 124 S. Ct. 191, 157 L. Ed. 2d 125 (2003); Armstrong v. Young, 34 F.3d 421, 427 (7th Cir.1994), cert. denied, 514 U.S. 1021, 115 S. Ct. 1369, 131 L. Ed. 2d 224 (1995). "We review the [trial] [c]ourt's decision for abuse of discretion, applying clear error review to its underlying factual findings and plenary review to its conclusions drawn from such facts." United States v. Mathis, 264 F.3d 321, 331(3d Cir.2001), cert. denied, 535 U.S. 908, 122 S. Ct. 1211, 152 L. Ed. 2d 148 (2002). Our sister states that have addressed this question are generally in accord. See, e.g., People v. Hogan, 114 P.3d 42, 49 (Colo. App.2005) ("The ultimate question as to the constitutionality of pretrial identification procedures is a mixed question of law and fact. Thus, we give deference to the trial court's finding of historical fact ... but may give different weight to those facts and may reach a different conclusion."), review denied, Docket No. 05SC46, 2005 Colo. LEXIS 597 (Colo. June 27, 2005); State v. Mack, 255 Kan. 21, 26, 871 P.2d 1265 (1994) ("[a]n eyewitness identification due process determination is a mixed question of law and fact that should be reviewed de novo"); McClenton v. State, 167 S.W.3d 86, 96 (Tex.App.2005) ("[t]he admissibility of an identification is a mixed question of law and fact that we review de novo"); see also State v. Moore, 334 S.C. 411, 418, 513 S.E.2d 626 (App. 1999) (Howell, C.J., dissenting) ("[w]hether an eyewitness identification is sufficiently reliable is a mixed question of law and fact subject to de novo review on appeal"), rev'd in part on other grounds, 343 S.C. 282, 540 S.E.2d 445 (2000).

Given the weight and uniformity of the preceding authority, as well as its commonsense appeal, we conclude that a claim of an unnecessarily suggestive pretrial identification procedure is a mixed question of law and fact. With respect to our review of the facts, we further note that, because the issue of the suggestiveness of a photographic array implicates the defendant's constitutional right to due process, we undertake a "scrupulous examination of the record to ascertain whether the findings are supported by substantial evidence." State v. Mullins, 288 Conn. 345, 364, 952 A.2d 784 (2008); cf. State v. Ledbetter, supra, 275 Conn. at 547, 881 A.2d 290 ("[b]ecause the issue of the reliability of an identification involves the constitutional rights of an accused ... we are obliged to examine the record scrupulously to determine whether the facts found are adequately supported by the evidence and whether the [trial] court's ultimate inference of reliability was reasonable" [internal quotation marks omitted]).

Our analysis proceeds in three parts. In part I of this opinion, we examine the approach taken by the trial court and, in particular, the scientific foundation underlying its conclusions. In part II of this opinion, we more fully describe the appropriate legal framework for analyzing suggestiveness and proceed to apply those principles to the facts of this case. Finally, in part III of this opinion, we address the defendant's request that this court exercise its supervisory authority to mandate the implementation of specific eyewitness identification procedures.

I

We begin our analysis with the state's claim that the trial court improperly concluded that a photographic array conducted pursuant to the traditional simultaneous identification procedure by an interested administrator is inherently unnecessarily suggestive.^[14] We agree. To the extent that the trial court's decision implies that the simultaneous display of photographs in an array by a police officer with specific knowledge of the case is per se unnecessarily suggestive, it is incorrect.^[15]

This court has, for some time, maintained a stern test for suggestiveness: "An identification procedure is unnecessarily suggestive only if it gives rise to a very substantial likelihood of irreparable misidentification." (Emphasis added; internal quotation marks omitted.) State v. Cook, 262 Conn. 825, 832, 817 A.2d 670 (2003), quoting State v. Reid, 254 Conn. 540, 555, 757 A.2d 482 (2000). We take this opportunity to revisit our definition of suggestiveness. The phrase we have been using to define "unnecessarily suggestive" seems to have its origins in Simmons v. United States, 390 U.S. 377, 88 S. Ct. 967, 19 L. Ed. 2d 1247 (1968), in which the United States Supreme Court declared that "convictions based on eyewitness identification at trial following a pretrial identification by photograph will be set aside on that ground only if the photographic identification procedure was so impermissibly suggestive as to give rise to a very substantial likelihood of irreparable misidentification."^[16] Id., at 384, 88 S. Ct. 967. The first use of this language by this court appears to have occurred in State v. Theriault, 182 Conn. 366, 438 A.2d 432 (1980), in which we resolved the claim of the defendant, Norman A. Theriault, that the trial court improperly declined to suppress a pretrial identification in violation of "his due process rights because it was impermissibly suggestive and gave rise to a substantial likelihood of irreparable misidentification." (Emphasis added.) Id., at 371, 438 A.2d 432. We concluded that the one-on-one "show-up" procedure used in Theriault was inherently suggestive as well as unnecessary under the circumstances. Id., at 372-73, 438 A.2d 432. We upheld the trial court's ruling, however, concluding that the identification, although suggestive, was nonetheless sufficiently reliable. Id., at 375, 438 A.2d 432.

In Theriault, we examined the suggestiveness of the identification procedures, in addition to the overall reliability of the identification itself, under the totality of the circumstances; see id., at 371-72, 438 A.2d 432; in order to determine whether the challenged identification violated Theriault's due process rights on the ground that it "gave rise to a substantial likelihood of irreparable misidentification." Id., at 371, 438 A.2d 432. There is no indication in Theriault that we used this phrase to define suggestiveness.

The first instance in which this court conflated these two concepts apparently occurred in State v. Williams, 203 Conn. 159, 523 A.2d 1284 (1987). In Williams, we stated that "[a]n identification procedure is unnecessarily suggestive when it ``give[s] rise to a very substantial likelihood of irreparable misidentification.'" Id., at 174, 523 A.2d 1284. Part of this statement was derived from our opinion in State v. Fullwood, 193 Conn. 238, 243-44, 476 A.2d 550 (1984). This language also can be traced from Ledbetter back to State v. Outlaw, 216 Conn. 492, 501, 582 A.2d 751 (1990), which contains the exact same language as Williams and which also cited to Fullwood. In our view, defining "unnecessarily suggestive" in this manner has been the result of a misstep. It is illogical and inconsistent with the authority on which it purports to be based.

We next set forth what we believe is the correct approach to this issue. First, there appears to be a consensus with regard to the overall analytical framework to be used in considering a claim of this sort: "In determining whether identification procedures violate a defendant's due process rights, the required inquiry is made on an ad hoc basis and is two-pronged: first, it must be determined whether the identification procedure was unnecessarily suggestive; and second, if it is found to have been so, it must be determined whether the identification was nevertheless reliable based on examination of the totality of the circumstances." (Internal quotation marks omitted.) State v. Theriault, supra, 182 Conn. at 371-72, 438 A.2d 432; see also Manson v. Brathwaite, supra, 432 U.S. at 107, 97 S. Ct. 2243 ("[T]he first inquiry [is] whether the police used an impermissibly suggestive [identification] procedure.... If so, the second inquiry is whether, under all the circumstances, that suggestive procedure gave rise to a substantial likelihood of irreparable misidentification."); United States v. DeCologero, 530 F.3d 36, 62 (1st Cir.) ("we first determine whether the identification procedure was impermissibly suggestive, and if it was, we then look to the totality of the circumstances to decide whether the identification was reliable"), cert. denied, ___ U.S. ___, ___, 129 S. Ct. 513, 615, 172 L. Ed. 2d 376, 469 (2008). We continue to endorse and adhere to this widely utilized analytical approach.

In the seminal case of Neil v. Biggers, supra, 409 U.S. at 188, 93 S. Ct. 375, the Supreme Court explained the overarching concern that courts face when assessing a challenged identification procedure: "It is ... apparent that the primary evil to be avoided is ``a very substantial likelihood of irreparable misidentification.'... It is the likelihood of misidentification which violates a defendant's rights to due process...." Id., at 198, 93 S. Ct. 375, quoting Simmons v. United States, supra, 390 U.S. at 384, 88 S. Ct. 967. As courts apply the two-pronged test to determine if a particular identification procedure is so suggestive and unreliable as to require suppression, they always should weigh the relevant factors against this standard. In other words, an out-of-court eyewitness identification should be excluded on the basis of the procedure used to elicit that identification only if the court is convinced that the procedure was so suggestive and otherwise unreliable as to give rise to a very substantial likelihood of irreparable misidentification. See Simmons v. United States, supra, at 384, 88 S. Ct. 967.

The critical question for our present purposes is what makes a particular identification procedure "suggestive" enough to require the court to proceed to the second prong and to consider the overall reliability of the identification. This is a straightforward question that does not appear to have received a very direct answer. There are, however, two factors that courts have considered in analyzing photographic identification procedures for improper suggestiveness. The first factor concerns the composition of the photographic array itself. In this regard, courts have analyzed whether the photographs used were selected or displayed in such a manner as to emphasize or highlight the individual whom the police believe is the suspect. See, e.g., State v. Williams, supra, 203 Conn. at 176, 523 A.2d 1284 (multiple photographs of same individual in same or subsequent photographic arrays possibly suggestive "when, in the context of the entire array, the recurrence unnecessarily emphasizes the defendant's photograph"); State v. Fullwood, supra, 193 Conn. at 247, 476 A.2d 550 (to be unnecessarily suggestive, variations in array photographs must highlight defendant to point that it affects witness' selection); State v. Gold, 180 Conn. 619, 656, 431 A.2d 501 ("[when] a feature is placed on the defendant's photograph in order to make the picture conform to the witness' description of the criminal he or she had seen, the identification proceeding has been held to have been rendered highly suggestive"), cert. denied, 449 U.S. 920, 101 S. Ct. 320, 66 L. Ed. 2d 148 (1980); see also United States v. DeCologero, supra, 530 F.3d at 62 (at first step in two-pronged test, court "consider[s] whether the photo[graphic] array included, as far as was practicable, a reasonable number of persons similar in appearance to the suspect"); United States v. Rattler, 475 F.3d 408, 413 (D.C.Cir.2007) (court must "examine the suggestivity of irregularities between the subjects in the array").

The second factor, which is related to the first but conceptually broader, requires the court to examine the actions of law enforcement personnel to determine whether the witness' "attention was directed to a suspect because of police conduct.... In considering this [factor, the court should] look to the effects of the circumstances of the pretrial identification, not whether law enforcement officers intended to prejudice the defendant." (Citation omitted; internal quotation marks omitted.) Howard v. Bouchard, 405 F.3d 459, 470 (6th Cir.2005), cert. denied, 546 U.S. 1100, 126 S. Ct. 1032, 163 L. Ed. 2d 871 (2006). It stands to reason that police officers administering a photographic identification procedure have the potential to taint the process by drawing the witness' attention to a particular suspect. This could occur either through the construction of the array itself or through physical or verbal cues provided by an officer. See, e.g., State v. Fullwood, supra, 193 Conn. at 248, 476 A.2d 550 (irregularity in defendant's photograph not suggestive because it did not "signal to the witnesses that the defendant was the person whom the police believed to be the perpetrator of the robbery"); see also Simmons v. United States, supra, 390 U.S. at 385, 88 S. Ct. 967 ("[t]here is no evidence to indicate that the witnesses were told anything about the progress of the investigation, or that the [law enforcement] agents in any other way suggested which persons in the pictures were under suspicion"); State v. Ledbetter, 185 Conn. 607, 612, 441 A.2d 595 (1981)^[17] (no basis for claiming that "display itself was suggestive or that [the administering officer] was suggestive in any respect in the selection process"); State v. Gold, supra, 180 Conn. at 656, 431 A.2d 501 ("[a] procedure is unfair which suggests in advance of identification by the witness the identity of the person suspected by the police" [internal quotation marks omitted]).

We hereby clarify that a determination as to whether a particular identification procedure is "unnecessarily suggestive" must focus on the foregoing factors. The phrase "very substantial risk of irreparable misidentification" must be understood as the overall standard for suppressing an out-of-court identification. By improperly making the test of suggestiveness so rigorous, we essentially have made the reliability prong of the analysis vestigial. Obviously, any identification resulting from a procedure that fails our current test for suggestiveness could never be considered sufficiently reliable under the totality of the circumstances to warrant admissibility. The suggestiveness prong should be less stringent and more focused on the mechanics of the photographic array itself as well as the behavior of the administering officers. This approach is more logical, gives some meaning to the term "suggestive" that accords with common experience, and can be readily applied by trial courts.

We stress that this is not a "best practices" test. In other words, the test does not require a court to engage in a relative value judgment of various possible identification techniques and settle on the one that it believes bears the least risk of mistake, a decision that would be prone to being revised or second-guessed as the scientific debate evolves and new studies become available. See, e.g., State v. Nunez, 93 Conn.App. 818, 832, 890 A.2d 636 (2006) ("[t]he question ... is not whether a double-blind, sequential identification procedure is less suggestive than the traditional procedures ... but ... whether the traditional procedures are unnecessarily suggestive under [the Connecticut] constitution"), cert. denied, 278 Conn. 914, 899 A.2d 621, cert. denied, 549 U.S. 906, 127 S. Ct. 236, 166 L. Ed. 2d 186 (2006); see also State v. Fullwood, supra, 193 Conn. at 244, 476 A.2d 550 ("[i]t has been generally recognized that the presentation of several photographs to witnesses, including that of the suspect ... is by itself a nonsuggestive and constitutionally acceptable practice, in the absence of any unfairness or other impropriety in the conduct of the exhibit" [internal quotation marks omitted]). Nor does this test require law enforcement personnel to alter their procedures every time a fresh scientific study suggests that a new identification procedure might lead to more reliable results. Moreover, although our analysis focuses principally on two key functional aspects of the eyewitness identification process,^[18] we stress that it is the entire procedure, viewed in light of the factual circumstances of the individual case, that must be examined to determine if a particular identification is tainted by unnecessary suggestiveness. The individual components of a procedure cannot be examined piecemeal but must be placed in their broader context to ascertain whether the procedure is so suggestive that it requires the court to consider the reliability of the identification itself in order to determine whether it ultimately should be suppressed.

In the present case, although the facts adduced at the hearing on the motion to suppress are essentially undisputed, the true controversy involves the potential suggestiveness of the chosen procedures in light of the scientific data presented to the trial court, which the court clearly found to be overwhelmingly persuasive in fashioning its categorical rule. It is this evidence that we now review to determine whether the court's conclusion that the photographic arrays and the detectives' method of presenting them were indeed unnecessarily suggestive.

The defendant presented four scientific documents to the trial court in support of his contention that the simultaneous, single-blind procedures used in this case were inherently suggestive. The state responds by presenting this court with two additional documents indicating that the science in this area is less than settled.^[19] We will briefly discuss each article or report, organizing our discussion chronologically on the basis of the date that the document was published and beginning with the scientific documents presented by the defendant.

The authors of the first document; see G. Wells et al., "Eyewitness Identification Procedures: Recommendations for Lineups and Photospreads," 22 Law & Hum. Behav. 603 (1998); discuss potential problem areas with traditional procedures and eventually make four recommendations that they believe would improve the accuracy and reliability of eyewitness identifications. The authors admit to relying "heavily on relative-judgment theory, which describes a process by which eyewitnesses make lineup identifications." Id., at 613. They note that "[t]here is good empirical evidence" in support of the relative judgment theory; id.; and their recommendations are aimed primarily at countering the negative effect that the relative judgment process can have on the accuracy of eyewitness identifications.

The authors' first recommendation is that "[t]he person who conducts the lineup or photospread should not be aware of which member of the lineup or photospread is the suspect." Id., at 627.^[20] The authors admit, however, that they are "aware of no studies indicating that lineup and photospread administrators are affecting the identification behaviors of eyewitnesses in actual cases"; id., at 628; and caution that this recommendation should be taken "somewhat on face value...." Id. The authors' second recommendation is that a warning be given to eyewitnesses indicating that the suspect may or may not be present in the lineup or photospread, and that the witness therefore should not feel compelled to make an identification. Id., at 629. The third recommendation concerns the structure of the lineup or photospread itself and advises that "[t]he suspect should not stand out in the lineup or photospread as being different from the distracters based on the [eyewitness'] previous description of the culprit or based on other factors that would draw extra attention to the suspect." Id., at 630. Finally, the authors recommend that the identification administrator collect a statement from the witness after an identification is made regarding the witness' level of confidence in his selection. Id., at 635. The authors of this article expressly decline to recommend the use of sequential identification procedures, noting that "[t]he superiority of the sequential over the simultaneous procedure is evident primarily under conditions [when recommendation two] (warning the eyewitness that the culprit might not be present) and [recommendation three] (distractors fitting the description) are violated...." (Citation omitted.) Id., at 640. Furthermore, the authors note that the adoption of a sequential procedure without the adoption of double-blind testing actually could lead to a greater error rate than that realized with a simultaneous procedure without double-blind testing. Id.

The second document offered by the defendant; United States Dept. of Justice, Eyewitness Evidence: A Guide for Law Enforcement (1999);^[21] contains a fairly brief section covering suspect identification procedures. Id., at p. 29. Interestingly, although the report provides different instruction sets for conducting sequential and simultaneous lineups and photographic arrays, it does not make a recommendation favoring one procedure over the other. Moreover, the report makes no mention of the use of double-blind testing procedures.

The third document on which the defendant relies is from the Canadian Journal of Police and Security Services. J. Turtle, R. Lindsay & G. Wells, "Best Practice Recommendations for Eyewitness Evidence Procedures: New Ideas for the Oldest Way to Solve a Case," 1 Canadian J. Police & Security Serv. 5 (2003).^[22] In addition to advocating that warnings be given to the witness to reduce the effect of the relative judgment process and providing practical guidelines for composing lineups, the article also recommends the use of blind, sequential identification procedures. Blind procedures are recommended because of the potential for an officer with knowledge of the investigation to transmit information or expectations inadvertently to the witness, thereby leading the witness to make a particular selection.^[23]

The article cites studies indicating that, although simultaneous identification procedures are three times more likely to yield misidentifications than sequential procedures, sequential procedures also yield lower correct identification rates than simultaneous procedures. Moreover, the article highlights a number of circumstances in which the use of a sequential procedure "may be no better or even worse than the traditional simultaneous line-up." Such circumstances include (1) identifications by child witnesses, who can become confused by a sequential procedure, (2) scenarios in which a witness is asked to identify multiple perpetrators, and (3) situations involving "cross-race identifications," in which a witness is asked to identify a person of a different race. Although the authors of this article clearly advocate the use of sequential procedures generally, the authors nevertheless conclude that "the sequential line-up has not been demonstrated to show its superiority under these conditions and, in fact, some data exist suggesting that there may be some disadvantage to using the procedure under these conditions. Until more and better data are available, we do not recommend using sequential line-ups in these particular situations."

The defendant's final scientific document; see G. Wells & E. Olson, "Eyewitness Testimony," 54 Ann. Rev. Psychol. 277 (2003); is a relatively brief review of much of the material discussed in connection with the previous documents. The authors strongly support the use of "might or might not be present" instructions, which, they note, have been shown to reduce mistaken identification rates significantly in lineups in which the perpetrator is absent. (Internal quotation marks omitted.) Id., at 286. The authors also show a preference for sequential line-up procedures, noting the tendency of such procedures to reduce "the chances of mistaken identifications in culprit-absent lineups by nearly one half" while also reducing "accurate identification rates in culprit-present lineups." Id., at 288. The authors discuss and support the use of double-blind testing procedures, although they stress that such procedures are especially necessary when a sequential identification procedure is utilized. Id., at 289. The authors conclude by lamenting the paucity of "real-world data" in this field of research. Id., at 290.

The state offers two documents presumably intended to highlight the lack of scientific consensus in the eyewitness identification field. The first document, which was a report to the Illinois legislature; see S. Mecklenburg, Report to the Legislature of the State of Illinois: The Illinois Pilot Program on Sequential Double-Blind Identification Procedures (2006) (Mecklenburg Report); was the product of a year long pilot program conducted at three police departments in the Chicago area.^[24] The results of this field study, which the author characterized as surprising, were that "sequential, double-blind procedures resulted in an overall higher rate of known false identifications than did the simultaneous lineups."^[25] (Emphasis in original.) Id., p. iv. The author concluded that "the sequential, double-blind method [could not] be regarded as superior to the simultaneous method"; id., at p. 64; and emphasized the need for further study. Id., at p. 65.^[26]

The second document that the state submits is a 2006 article from Psychology, Public Policy, and Law. See D. McQuiston-Surrett, R. Malpass & C. Tredoux, "Sequential vs. Simultaneous Lineups: A Review of Methods, Data, and Theory," 12 Psychol., Pub. Policy & L. 137 (2006). In this article, written after the release of the Mecklenburg Report, the authors maintain some reservations about the methodologies and significance of that report but nonetheless conclude that "the literature concerning [simultaneous lineups] versus [sequential lineups] may be underdeveloped in some important ways...." Id., at 141. In addition, "the research base for [sequential lineups] may not be sufficiently developed from a methodological or theoretical point of view to ... advocate for its implementation to the exclusion of other procedures." Id., at 162.^[27]

Presented with the foregoing research, the trial court considered several factors in determining that the procedures used in this case were unnecessarily suggestive. First, the court concluded, in the abstract, that "the simultaneous showing of all photo[graphs] to each witness on a single ... board created an unnecessary risk of producing irreparable misidentifications by enabling the witnesses to make side-by-side comparisons of the photo[graphs], and thus to select one of them simply by eliminating all the others in an unreliable exercise of relative judgment." The court based this judgment on the "unchallenged findings of the scientific research studies...."^[28]

The trial court also was "troubled" by the fact that Beaudin, the lead detective investigating the robbery, administered both identification procedures. The court expressed "concern ... based [on] the undisputed judgment and recommendation of well respected scientific researchers" that an officer with knowledge of the investigation, particularly, the identity of the suspect, runs the risk of intentionally or inadvertently "injecting bias into the identification process ... and thus of producing irreparable misidentifications."

Contrary to the trial court, we conclude that the scientific evidence regarding the value of sequential procedures is more nuanced and uncertain than portrayed by the defendant, and, therefore, it cannot definitively answer the question of whether the procedures used in this case were unnecessarily suggestive.^[29] For instance, the research indicates that, in multiple perpetrator scenarios, the use of sequential identification procedures may not be advisable, or even practical: "[I]f multiple perpetrators were involved in the crime and more than one suspect is to be shown to the witness, it is not clear how a sequential procedure should be used, and traditional methods have not been shown to be inferior in such cases." J. Turtle, R. Lindsay & G. Wells, supra, at 1 Canadian J. Police & Security Serv. 5. In this case, for example, the detectives knew from eyewitness statements that the robbery had been committed by two individuals. As a result, the value of using a sequential procedure is at least questionable. Moreover, although the scientific community recommends the use of a double-blind identification procedure, and such a procedure has intuitive appeal, we never have held that the failure to use such a procedure carries such a substantial risk of misidentification that its use must be required to avoid unnecessary suggestiveness.^[30]

Upon consideration of the scientific literature, we conclude that one thing is clear, namely, that the judgment of the relevant scientific community with respect to eyewitness identification procedures is far from universal or even well established, and that the research is in great flux.^[31] Indeed, when the reported research was seemingly more uniform, we still found that "[t]he scientific studies are not definitive."^[32]State v. Ledbetter, supra, 275 Conn. at 568, 881 A.2d 290. The more recent research offered by the state muddies the water further and only confirms this view. Thus, this continues to be an issue particularly ill suited to generic, bright line rules. Indeed, we repeatedly have insisted that this inquiry be made on an ad hoc basis, and we affirm that the courts of this state should continue to evaluate "whether individual identification procedures are unnecessarily suggestive on the basis of the totality of the circumstances surrounding the procedure, rather than replacing that inquiry with a per se rule." Id., at 574, 881 A.2d 290. We agree with the Appellate Court that, until the scientific research produces more definitive answers with respect to the effects of various procedures, "[d]ue process does not require the suppression of a photographic identification that is not the product of a double-blind, sequential procedure."^[33]State v. Smith, 107 Conn.App. 666, 674, 946 A.2d 319, cert. denied, 288 Conn. 902, 952 A.2d 811 (2008).

II

Having concluded that the trial court overemphasized the significance of the scientific research and improperly applied a per se analysis to the challenged identifications, we now turn to an examination of the procedures at issue.

The defendant made several claims of suggestiveness at the suppression hearing that were based on the factual context of this case. The defendant asserted that the photographic array itself was in fact unnecessarily suggestive insofar as the composition of the array unfairly highlighted him and was designed to promote his identification by the witnesses. Specifically, the defendant claimed that the array was unnecessarily suggestive because his photograph was distinctive. He specifically asserted that his photograph was somewhat brighter in appearance than the other photographs and had a white height scale in the background, whereas six of the other seven photographs contained visible height scales that were black in color. The defendant further argued that Beaudin's administration of the identification procedure was flawed because she knew which photograph represented the suspect. The defendant points to the confirmatory comment that Beaudin made to Clement after he selected the defendant's photograph as evidence of a bias on her part that must have tainted the entire process.

In its memorandum of decision, the trial court addressed the defendant's claims and determined that the photographic arrays used in this case were fair and did not draw attention to the defendant's photograph. The court concluded that "the array used in this case did not unfairly highlight the defendant or promote his identification by the witnesses in any way." The trial court, however—perhaps as a corollary to its finding with respect to the importance of utilizing double-blind procedures—was impressed by the testimony regarding Beaudin's reaction to Clement's selection. The court characterized Beaudin's comment to Clement that he "did good because that was the same guy [Valle] picked," as an indication of a real "risk of unfairness due to administrator bias...." The court determined that this was a revelation of the detective's bias, which "risked ... unfairly bolstering the witness' confidence in the strength of his ... identification ... thus making it harder for the defense to test the true certainty with which he made that identification on cross-examination...."^[34] In the same passage of its memorandum of decision, however, the trial court noted that Beaudin's comment was made after the identification was complete and Clement had expressed his confidence in the selection and, thus, that "the utterance was not shown to have tainted the witness' identification when it was initially made."^[35]

In assessing the constitutionality of challenged eyewitness identifications, we engage in a case-by-case review. "[T]he required inquiry is made on an ad hoc basis and is two-pronged: first, it must be determined whether the identification procedure was unnecessarily suggestive; and second, if it is found to have been so, it must be determined whether the identification was nevertheless reliable based on an examination of the totality of the circumstances." (Internal quotation marks omitted.) State v. Ledbetter, supra, 275 Conn. at 547-48, 881 A.2d 290, quoting State v. Cook, supra, 262 Conn. at 832, 817 A.2d 670.^[36] As we noted in part I of this opinion, an out-of-court identification will be excluded only if the procedures utilized to obtain that identification "[give] rise to a very substantial likelihood of irreparable misidentification." (Internal quotation marks omitted.) State v. Ledbetter, supra, at 548, 881 A.2d 290, quoting State v. Cook, supra, at 832, 817 A.2d 670. We conclude that any analysis of unnecessary suggestiveness must be conducted in light of the totality of the circumstances and must focus specifically on the presentation of the photographic array itself as well as the behavior of law enforcement personnel to determine if the procedure was designed or administered in such a way as to suggest to the witness that a particular photograph represents the individual under suspicion.

In evaluating the suggestiveness of a photographic array, "a court should look to both the photographs themselves and the manner in which they were presented to the identifying witness." Hodges v. Commonwealth, 45 Va.App. 735, 774, 613 S.E.2d 834 (2005), rev'd on other grounds, 272 Va. 418, 634 S.E.2d 680 (2006). We consider the following nonexhaustive factors in analyzing a photographic array for unnecessary suggestiveness: "(1) the degree of likeness shared by the individuals pictured ... (2) the number of photographs included in the array ... (3) whether the suspect's photograph prominently was displayed or otherwise was highlighted in an impermissible manner... (4) whether the eyewitness had been told that the array includes a photograph of a known suspect ... (5) whether the eyewitness had been presented with multiple arrays in which the photograph of one suspect recurred repeatedly ... and (6) whether a second eyewitness was present during the presentation of the array." (Citations omitted.) State v. Randolph, 284 Conn. 328, 385-86, 933 A.2d 1158 (2007). It is important to note, however, that "[p]hotographs will often have distinguishing features. The question ... is not whether the defendant's photograph could be distinguished from the other photographs... but whether the distinction made it unnecessarily suggestive." State v. Nunez, supra, 93 Conn.App. at 828, 890 A.2d 636; see also Hodges v. Commonwealth, supra, at 774, 613 S.E.2d 834 ("[a] valid lineup does not require that all the suspects or participants be alike in appearance and have the same description as long as nothing singles the accused out from the rest").

We find it significant that the trial court's analysis essentially ignores the fact that both photographic arrays contained a conspicuous "might or might not be present" warning, indicating to each witness that the perpetrator was not necessarily among those pictured and that the witnesses should not feel obligated to choose someone. The presence of such a warning is a consistent recommendation of the scientific literature that we have reviewed and is deemed to counteract effectively the tendency of witnesses to use relative judgment. See, e.g., G. Wells et al., supra, 22 Law & Hum. Behav. at 629-30 (recommending warnings and explaining how they obviate need for sequential procedure); see also United States Dept. of Justice, supra, at p. 31(indicating importance and purpose of instructions). Moreover, this court expressly has endorsed, without mandating, the use of such warnings and has recognized their potential prophylactic effect against the dangers of the relative judgment process. "[W]e recognize that [certain] studies ... strongly militate in favor of an affirmative warning to witnesses that the perpetrator may or may not be among the choices in the identification procedure.... [T]rial court[s], as part of [their] analysis, should consider whether the identification procedure administrator instructed the witness that the perpetrator may or may not be present in the procedure...." (Citations omitted.) State v. Ledbetter, supra, 275 Conn. at 574-75, 881 A.2d 290. Significantly, we have concluded that, even when police not only fail to give such a warning but affirmatively inform the witness that the suspected culprit is in the lineup, there is no presumption of suggestiveness. See State v. Reid, supra, 254 Conn. at 556, 757 A.2d 482. "[E]ven if a court finds that the police expressly informed witnesses that the defendant would be in the array, our courts have found the identification procedure unnecessarily suggestive only when other factors exist that otherwise emphasize the defendant's photograph." State v. Owens, 38 Conn.App. 801, 811, 663 A.2d 1094, cert. denied, 235 Conn. 912, 665 A.2d 609 (1995).

In the present case, Clement testified at the suppression hearing that he remembered that Detective Beaudin had read him the warning before he viewed the photographic array. He further testified that the detectives did nothing to influence his decision or to direct his attention to any particular photograph, and that he selected the defendant's photograph from the array solely on the basis of his "gut feeling." Moreover, the trial court credited Clement's testimony that "he was not at all sure that the true perpetrator would be in the array" and observed that, "although [Clement] believed that the police would not have invited him to view photo[graphs] if they did not at least have a suspect in mind, he was not at all sure that the true perpetrator would be in the array, and so he commendably took his time in order not to implicate an innocent man."

Valle testified that he did not remember reading the warning or having it read to him prior to identifying the defendant from the photographic array. His identification was unique, however, insofar as he originally had reported spotting the defendant at the parole office, and that it was this information that led to the defendant's inclusion in the photographic array in the first place. In light of this independent source for his identification, it is not surprising that Valle did not remember any warnings being given, or any other specifics about the identification procedure.^[37]

Although there is no evidence that Valle heeded the warning written on the photographic array, there is ample evidence that he needed no such warning. Under these somewhat unusual circumstances, including Valle's chance encounter with the defendant at the parole office and Valle's immediate selection of the defendant from the array, it is clear that Valle's selection of the defendant's photograph was not influenced by relative judgment because he simply did not have the time or the need to engage in a process of elimination.

Furthermore, as we previously stated, the failure to use a double-blind procedure does not automatically render an identification suspect, particularly when, as in the present case, there is no evidence that the detectives conducting the procedure influenced the witnesses in any discernible way prior to their making the identification.^[38] Moreover, we agree with the Appellate Court, which has held that "[t]he police officer's telling the victim that she had identified the suspect after she positively identified the defendant as her assailant does not render the identification procedure unnecessarily suggestive." (Emphasis in original.) State v. Smith, supra, 107 Conn.App. at 675, 946 A.2d 319. Thus, although Beaudin's comment to Clement may affect the weight or even the admissibility of a subsequent in-court identification, it is irrelevant to our analysis regarding the suggestiveness of the procedure itself.^[39]

In view of the totality of the circumstances, we are convinced that the trial court improperly concluded that the identification procedures used in this case were so flawed as to present "a very substantial likelihood of irreparable misidentification." (Internal quotation marks omitted.) State v. Randolph, supra, 284 Conn. at 385, 933 A.2d 1158. The detectives employed traditional procedures and proceeded in a neutral fashion. There is no evidence that they attempted to influence, consciously or subconsciously, the outcome of the identification process, and the witnesses' testimony bears this out. The photographic arrays themselves were not designed or presented in an unfair or suggestive manner. Furthermore, we emphasize the importance of the warnings provided on the photographic arrays themselves or read aloud by the detectives, which served to counter any tendency of the witnesses to engage in the process of relative judgment. We conclude that the procedures employed in this case, although not ideal, were within the acceptable parameters of effective and fair police work, and satisfy the requirements of due process.

III

Finally, we turn to the defendant's contention that this court should exercise its supervisory authority to mandate new identification procedures in the interests of justice. The defendant urges this court to implement three specific procedural changes: (1) the double-blind identification procedure; (2) the sequential display of live suspects or photographs; and (3) a prohibition on police informing witnesses, after they identify a suspect, that the individual that they chose is the person whom police believe is the culprit. We decline the defendant's request to exercise our supervisory authority in this manner.

We first note the reluctance with which we have occasionally exercised our supervisory authority. "Our supervisory powers are not a last bastion of hope for every untenable appeal. They are an extraordinary remedy to be invoked only when circumstances are such that the issue at hand, while not rising to the level of a constitutional violation, is nonetheless of utmost seriousness, not only for the integrity of a particular trial but also for the perceived fairness of the judicial system as a whole." (Emphasis in original; internal quotation marks omitted.) State v. Hines, 243 Conn. 796, 815, 709 A.2d 522 (1998), quoting State v. Holloway, 209 Conn. 636, 645, 553 A.2d 166, cert. denied, 490 U.S. 1071, 109 S. Ct. 2078, 104 L. Ed. 2d 643 (1989). "We ordinarily invoke our supervisory powers to enunciate a rule that is not constitutionally required but that we think is preferable as a matter of policy." State v. Ledbetter, supra, 275 Conn. at 578, 881 A.2d 290.

We are not persuaded that this case presents an appropriate forum for the exercise of our supervisory authority. We are not convinced that allowing law enforcement officers to engage in identification procedures such as those used in the present case presents a threat to the "perceived fairness of the judicial system as a whole." (Internal quotation marks omitted.) State v. Hines, supra, 243 Conn. at 815, 709 A.2d 522. Our thorough review of the scientific research offered by the parties reconfirms the opinion we held in Ledbetter, namely, that "[t]he circumstances surrounding the various identification procedures present too many variables for us to conclude that a per se rule is appropriate." State v. Ledbetter, supra, 275 Conn. at 574, 881 A.2d 290. We believe that the development and implementation of identification procedures "should continue to be the province of the law enforcement agencies of this state." Id. We also reiterate that "the trial courts should continue to determine whether individual identification procedures are unnecessarily suggestive on the basis of the totality of the circumstances surrounding the procedure, rather than replacing that inquiry with a per se rule." Id. Any residual element of suggestiveness or untrustworthiness that does not bear a "very substantial likelihood of irreparable misidentification"; (internal quotation marks omitted) id., at 548, 881 A.2d 290; is "customary grist for the jury mill. Juries are not so susceptible that they cannot measure intelligently the weight of identification testimony that has some questionable feature." (Internal quotation marks omitted.) State v. Garner, 270 Conn. 458, 469, 853 A.2d 478 (2004).

This opinion is not a blanket endorsement of any particular identification procedure. We emphasize that we have not created a per se rule approving of the procedures used in this case but, rather, have evaluated those specific procedures within the totality of the circumstances and have not found them to be unnecessarily suggestive. We would, of course, encourage the state's law enforcement agencies to maintain currency in the latest research in this field and to adapt their policies to implement the most accurate, reliable and practical identification procedures available. In fact, we believe that the scientific research and common sense suggest that the employment of double-blind procedures, whenever reasonably practicable, is preferable to the use of an interested administrator because such procedures avoid the possibility of influencing the witness, whether intentionally or unintentionally, and thereby tainting the accuracy of any resulting identification. At this time, however, we continue to review suggestiveness on a case-by-case basis using the established standards.

The judgment is affirmed.

In this opinion VERTEFEUILLE and SCHALLER, Js., concurred.

KATZ, J., concurring.

Although I agree with the majority's conclusion that the judgment of conviction of the defendant, Julian Marquez, must be affirmed because the admission of the witness identifications in question did not violate due process, I disagree with its decision to affirm the conviction on the basis of the state's alternative ground for affirmance rather than on the issue presented in the defendant's appeal. Like the majority, I recognize that the trial court's discussion of the simultaneous photographic array^[1] suggests that, despite the accompanying warnings, the array was unnecessarily suggestive. If, indeed, the trial court had reached that conclusion without the support of other factual findings, it would have been inconsistent with our holding in State v. Ledbetter, 275 Conn. 534, 580, 881 A.2d 290 (2005), cert. denied, 547 U.S. 1082, 126 S. Ct. 1798, 164 L. Ed. 2d 537 (2006), in which we held that, standing alone, a nonsequential procedure accompanied by appropriate warnings is not unnecessarily suggestive. I also agree with the majority that the trial court had no authority to issue what could be perceived as a directive to local police departments. Because, however, the trial court also reasonably relied on the fact that the detective in charge of the investigation not only administered the procedure but also conveyed approval of the identification made by one of the witnesses as a basis for its determination that the identification procedure was unnecessarily suggestive, I would not conclude, as the majority does, that the trial court abused its discretion when it determined that the identification procedure as a whole was unnecessarily suggestive. More importantly, this court need not decide this case by rejecting the trial court's determination of unnecessary suggestiveness. Indeed, in my view, this approach is both unnecessary and unwise, as it signals our approval of the use of an interested administrator^[2] and closes off debate on witness identification procedures at a time when a clearer consensus about those procedures is just beginning to emerge from scientific research. Therefore, I would limit consideration to the trial court's determination that the identification was reliable based upon the totality of the circumstances, as presented by the posture of the defendant's appeal, and would affirm the judgment on that basis.

I

More than three years ago, in State v. Ledbetter, supra, 275 Conn. at 546-47, 881 A.2d 290, this court was asked to conclude that the failure to warn a witness that the perpetrator might not be present at an identification procedure renders that procedure per se unnecessarily suggestive, thereby overruling our decision in State v. Reid, 254 Conn. 540, 556, 757 A.2d 482 (2000) (witness identification procedure utilizing photographic array not unnecessarily suggestive when witness had been informed that suspect was present in array). In support of this proposed conclusion, various amici curiae presented this court with academic research that underscored the dangers of misidentification inherent in the use of identification procedures, including lineups, show-ups and photographic arrays that contained no warning to the witness that the perpetrator may not be present. Id., at 569-70, 757 A.2d 482. We declined to adopt such a per se rule, reasoning that "[t]he circumstances surrounding the various identification procedures present too many variables for us to conclude that a per se rule is appropriate." Id., at 574, 757 A.2d 482. Instead, we held that "trial courts should continue to determine whether individual identification procedures are unnecessarily suggestive on the basis of the totality of the circumstances surrounding the procedure...." Id. In Ledbetter, we further directed trial courts, however, to consider the results of research concerning whether the identification administrator informs the witness that the perpetrator may or may not be present. Id., at 574-75, 757 A.2d 482. Because of the import of the studies, we also invoked our supervisory authority to mandate that, if the administrator has not provided such a warning, trial courts must issue a jury instruction explaining that there may be an increased likelihood of misidentification under such circumstances. Id., at 579, 757 A.2d 482. A necessary implication of our holding in Ledbetter, however, is that if a witness identification procedure employs a warning instruction, then, standing alone, it would not be considered unnecessarily suggestive, and that trial courts must continue to examine the facts and circumstances of each case to determine if other factors are present that would render the procedure improper. See id., at 574-75, 757 A.2d 482.

Four months after we decided Ledbetter, the trial court in the present case was presented with the following identification procedure in the defendant's trial for, inter alia, felony murder and robbery in the first degree. Two witnesses separately were shown a simultaneous photographic array that contained a printed warning in conformity with our holding in Ledbetter from which they each identified the defendant as the perpetrator who possessed a firearm. The identification procedures were conducted by Detective Patricia Beaudin of the Hartford police department, the detective in charge of the criminal investigation, who informed the second of the two witnesses, Mark Clement, at the conclusion of the procedure that he "``did good'" because he had selected the same individual whom the other witness, Christopher Valle, had chosen.

In support of his motion to suppress the identifications on the ground, inter alia, that these procedures were unnecessarily suggestive, the defendant introduced into evidence four scientific research papers that called into question both the use of an interested administrator and the use of a simultaneous photographic array because of the risks of misidentification they present. The state did not object to the introduction of this evidence and did not dispute the results of the studies. After considering the identification procedure, and in light of the scientific evidence that had been introduced, the trial court denied the defendant's motion to suppress the identifications, finding that, although the procedure was unnecessarily suggestive, it nevertheless was reliable under the totality of the circumstances.

In its analysis of the procedure, the trial court relied primarily on two factors.^[3] First, the court reviewed the scientific research presented and concluded that using an interested lineup administrator^[4] creates an "unnecessary risk of injecting bias into the identification process, and thus of producing irreparable misidentifications."^[5] The court then found that, in the present case, those risks specifically had been realized when the administrator, Beaudin, who also had served as the lead investigator in the case, contaminated the process by announcing to Clement during the identification procedure, in essence, that he correctly had identified the defendant, and that he had identified the same person that Valle had identified, thus reinforcing Clement's choice. The trial court was particularly troubled by this fact because Clement noticeably had hesitated before selecting the defendant's photograph. The court noted that, by commenting on Clement's choice, Beaudin unwittingly had bolstered Clement's confidence and rendered it impossible for the defense to test the accuracy of his memory and made it more difficult for the jury to assess the reliability of the identifications.

Second, the trial court found, on the basis of the unchallenged studies,^[6] that the use of the simultaneous photographic array itself was unnecessarily suggestive, despite the inclusion of the printed warning, because it allowed Clement to engage in a "relative judgment" process, creating an unnecessary risk of a misidentification.^[7] The court then set forth the following cautionary note to guide police in improving identification procedures: "Police personnel conducting photo[graphic] identifications should henceforth strive to eliminate the danger of misidentification arising from the simultaneous showing of multiple photo[graphs] by making all such showings sequentially." On the basis of these factors, the trial court concluded that the identification procedure as a whole was unnecessarily suggestive.

I recognize that, although the trial court properly considered the studies as evidence and found potential risks inherent in the use of the simultaneous (nonsequential) array, it failed to make specific findings that such risks were present in the present case. Therefore, to the extent that the trial court may have predicated its finding of unnecessary suggestiveness solely on the use of the simultaneous photographic array, such a determination would be contrary to our holding in State v. Ledbetter, supra, 275 Conn. at 575, 881 A.2d 290. Moreover, the trial court's admonition to police department personnel may be read as an attempt to establish a per se rule requiring sequential identification procedures as an exercise of supervisory authority. Although the research suggests that sequential identification procedures are superior to nonsequential ones, except in certain circumstances; see footnotes 11 and 12 of this concurring opinion; it is well established that trial courts are not empowered to bind police departments, much less their sister tribunals, by virtue of their rulings. See J.M. Lynne Co. v. Geraghty, 204 Conn. 361, 369, 528 A.2d 786 (1987); McDonald v. Rowe, 43 Conn.App. 39, 43, 682 A.2d 542 (1996); see also State v. Anderson, 255 Conn. 425, 438, 773 A.2d 287 (2001) ("[a]ppellate courts possess an inherent supervisory authority over the administration of justice" [emphasis added]). Accordingly, I would reject any attempt made by the trial court to establish a per se rule.

In my view, however, the trial court made its determination of unnecessary suggestiveness by examining the procedure as a whole, such that the use of the simultaneous array was evaluated in concert with the other factors present. With respect to the use of the interested administrator, the research presented clearly indicated substantial risks of misidentification, which the state chose not to address, and those risks specifically were realized when Beaudin praised Clement as essentially having made the correct choice in identifying the defendant. Moreover, Beaudin's comment occurred immediately after the identification had been made, thus having an indelible effect on Clement's memory. For the reasons set forth by Justice Palmer in his concurring opinion, it is clear that the trial court viewed Beaudin's comment as having contaminated the identification procedure by "unfairly bolstering the witness' confidence in the strength of his photo[graphic] identification and the solidity of his basis in memory for it...." Because I cannot conclude that the trial court abused its discretion in finding that it rendered the procedure as a whole unnecessarily suggestive; State v. Ledbetter, supra, 275 Conn. at 548, 881 A.2d 290 (noting abuse of discretion standard required to review decisions on admitting witness identifications); I would decide the case on the basis of the issue presented in the defendant's appeal, namely, whether the identifications nevertheless were reliable under the totality of the circumstances.

Additionally, I would confine the resolution of this case to the defendant's appeal for another reason—my concern that, by approving the procedures at issue in the present case, we may discourage, if not halt, development of jurisprudence surrounding witness identification procedures. I am mindful that our holding in Ledbetter was predicated on the state of the research as it then existed. Since that time, however, a great deal of research has been done on the effects of various factors on the accuracy of witness identifications. From the research presented to this court and my own independent review of available scientific literature, it is clear that science has uncovered a number of flaws in current identification procedures.

For example, research studies clearly support the hypothesis that using an individual who is aware of the identity of the suspect to conduct the identification process contaminates the process, even when the individual makes no conscious effort to influence the witness' identification.^[8] Researchers have documented a number of mechanisms by which a lineup administrator unintentionally may influence the outcome of the results.^[9] Moreover, experimental evidence has demonstrated that, when a lineup administrator erroneously believes that a particular member of the lineup is the perpetrator, witnesses tend to select that member rather than the true perpetrator. G. Wells & D. Quinlivan, "Suggestive Eyewitness Identification Procedures and the Supreme Court's Reliability Test in Light of Eyewitness Science: 30 Years Later," 33 Law & Hum. Behav. 1, 8 (2009). Although further analysis of realworld data should be completed, and the use of blind administrators; see footnote 2 of this concurring opinion; may be impractical to implement in certain situations,^[10] it is clear that, when reasonably possible, it is preferable to employ a blind administrator to avoid inadvertent contamination of the identification.

Additionally, research indicates that sequential identification procedures generally appear to be superior to simultaneous identification procedures, and some studies go so far as to recommend that they be adopted by law enforcement agencies.^[11] Nevertheless, the research has identified specific situations in which sequential identification procedures may produce fewer accurate results than simultaneous identification procedures,^[12] and therefore these situations should be studied to determine the extent to which they render sequential procedures inadvisable.

Although the state chose not to dispute the research in the proceedings in the trial court, it presented to this court a report that cast doubt on the value of sequential, blind identification procedures and concluded that field experiments did not replicate the favorable results obtained when utilizing such procedures in a controlled laboratory environment. S. Mecklenburg, "Report to the Legislature of the State of Illinois: The Illinois Pilot Program on Sequential Double-Blind Identification Procedures" (March 17, 2006) pp. i-ii (hereafter Illinois Report). Ultimately, however, the conclusions in that report have been discredited as the product of an unsound, unscientific methodology that does not support the conclusions reached therein.^[13] The report's author later conceded that "the purpose of the Illinois field study was not to isolate the effect of one factor upon lineup results," and reiterated the need for further field studies to evaluate those factors in greater detail. S. Mecklenburg, "Addendum to the [Illinois Report]" (June 19, 2006), pp. 3-4.

It is therefore clear that witness identification research, although evolving, is converging toward a consensus. It is equally clear that, in light of the evolving research, this court should avoid closing off debate on witness identification procedures by signaling its approval of procedures on which research currently casts doubt, especially when there is no need to do so. Ultimately, as science progresses and is able to offer more concrete recommendations, witness identification procedures may need to be revised to ensure that they produce accurate results in accordance with due process guarantees.

In light of these concerns, I would reaffirm the case-by-case approach that we endorsed in State v. Ledbetter, supra, 275 Conn. at 575, 881 A.2d 290, allowing trial courts to examine the studies presented to them and to consider those studies under the particular facts as they arise in any given case. I would limit the court's consideration in this appeal to the trial court's determination that the identification was reliable based upon the totality of the circumstances, as presented by the posture of the defendant's appeal. Accordingly, I turn to that claim.

II

The defendant contends that the trial court improperly determined that the identifications made by the two witnesses, Valle and Clement, were reliable under the totality of circumstances. Specifically, the defendant notes that, with respect to their ability to see the perpetrator at the time of the crime, the lighting was "quite poor" because the light in the dining room of the victim's apartment was off. The defendant also points out that, although the descriptions provided by the witnesses generally matched his appearance, they lacked specific descriptions of facial features, including hair and eye color, and he contends that his photograph was highlighted with respect to others in the photographic array. In light of this court's acknowledgment in State v. Ledbetter, supra, 275 Conn. at 576, 881 A.2d 290, that "the correlation between witness confidence and accuracy tends to be weak, and witness confidence can be manipulated," the defendant suggests that the confidence level of Valle and Clement should be given little weight. The defendant also asserts that the studies he submitted to the trial court indicated that the identification procedures themselves rendered the identification unreliable. Finally, he maintains that Clement's identification was biased directly by the postidentification statement by Beaudin confirming that Clement had identified the same individual that Valle had identified. We conclude that the trial court properly determined that the identifications were reliable.

The standard of review and legal principles for decisions to admit witness identifications is well settled: "[W]e will reverse the trial court's [evidentiary] ruling ... only where there is an abuse of discretion or where an injustice has occurred ... and we will indulge in every reasonable presumption in favor of the trial court's ruling.... Because the inquiry into whether evidence of pretrial identification should be suppressed contemplates a series of factbound determinations, which a trial court is far better equipped than this court to make, we will not disturb the findings of the trial court as to subordinate facts unless the record reveals clear and manifest error." (Internal quotation marks omitted.) Id., at 548, 881 A.2d 290. The admission of a witness' identification into evidence, however, implicates a defendant's constitutional rights, and thus "we are obliged to examine the record scrupulously to determine whether the facts found are adequately supported by the evidence and whether the court's ultimate inference of reliability was reasonable." (Internal quotation marks omitted.) Id., at 547, 881 A.2d 290.

If the trial court has determined that the identification procedure was unnecessarily suggestive, as in the present case, it must then determine "whether the identification was nevertheless reliable based on an examination of the totality of the circumstances.... The defendant bears the burden of proving both that the identification procedures were unnecessarily suggestive and that the resulting identification was unreliable." (Internal quotation marks omitted.) State v. Ortiz, 252 Conn. 533, 553, 747 A.2d 487 (2000).

"[R]eliability is the linchpin in determining the admissibility of identification testimony.... Manson v. Brathwaite, 432 U.S. 98, 114, 97 S. Ct. 2243, 53 L. Ed. 2d 140 (1977). To determine whether an identification that resulted from an unnecessarily suggestive procedure is reliable, the corruptive effect of the suggestive procedure is weighed against certain factors, such as the opportunity of the [witness] to view the criminal at the time of the crime, the [witness'] degree of attention, the accuracy of [the witness'] prior description of the criminal, the level of certainty demonstrated at the [identification] and the time between the crime and the [identification]." (Internal quotation marks omitted.) State v. Cook, 262 Conn. 825, 836-37, 817 A.2d 670 (2003).

Turning to the first two factors of reliability in the present case, the trial court made several factual findings that support its determination that, at the time of the crime, the witnesses had ample opportunity to view the perpetrator with a sufficient degree of attention. Valle first observed the perpetrator with the gun just before the robbery as the perpetrator stood in a lighted common hallway of the apartment building talking to the victim. He next observed the perpetrator during the course of the robbery over a period of several minutes in the living room from a distance of one to two feet as they sat facing each other. Clement first saw the perpetrator as he entered the victim's game room, which was well lit, and next saw him as Clement sat with Valle on the living room couch for several minutes. The living room was lit adequately by light coming from adjacent rooms, once the witnesses' eyes had adjusted to the admittedly low light level,^[14] and the trial court found that "the light from the kitchen must have illuminated the faces of the robbers, helping to fix them in [the witnesses'] memory." Both witnesses had stated to the police that based on their observations, they would be able to identify the perpetrator if they saw him again. Thus, the record supports the trial court's finding that they had adequate opportunities to view the defendant with a sufficient degree of attention. This conclusion is consistent with this court's previous holdings that a witness may make a reliable identification based on a brief interval when the witness was able to view the criminal clearly; State v. Morgan, 274 Conn. 790, 804-805, 877 A.2d 739 (2005) (identification reliable when witness "got a good look" at perpetrator as he ran past her and pulled down his mask); and when the witness had observed the criminal over a period of several minutes from a sufficiently close distance. State v. Crosswell, 223 Conn. 243, 268, 612 A.2d 1174 (1992) (identification reliable when witnesses observed masked perpetrator for several minutes with their attention directed at perpetrator).

Turning to the third factor of reliability, both witnesses gave consistent, detailed descriptions of the perpetrator, which matched that of the defendant. Valle described the perpetrator with the gun as a Hispanic male in his early twenties, taller and slimmer than his companion, with braids in his hair and dressed entirely in black. Clement described the gunman as tall, Hispanic, in his early twenties, wearing all black clothing, with "corn rows" and a medium build. Although neither witness included certain facial features, such as eye or hair color, in their descriptions, the descriptions were consistent, both with each other and with the appearance of the defendant.

With respect to the fourth factor of reliability, the trial court found that both witnesses' level of certainty was high. Valle had informed the police investigators immediately after the robbery that he would be able to identify the perpetrator again, and he later recognized the defendant at a chance encounter at the office of Valle's parole officer. At the subsequent photographic identification procedure, he made the identification "immediately and with great confidence." Clement testified that, when he was shown the photographic array at the identification procedure, his eyes immediately were drawn to the defendant's photograph. Although he did not identify the defendant as the perpetrator at that moment, Clement testified that he had hesitated only because he wanted to be sure that he identified the correct individual. The trial court found that Clement's hesitation reasonably stemmed from his "genuine concern ... that he not identify an innocent man," and that he was nevertheless confident in his identification.

Finally, Valle's identification occurred only four days after the crime, while Clement's occurred eight days after the crime. In light of both witness' opportunity to view the perpetrator and their confidence levels, it was unlikely that the witnesses' memories would have faded in that short period of time. Compare State v. Ortiz, supra, 252 Conn. at 555, 747 A.2d 487 (three month period reliable in light of witness' opportunity to view defendant and witness' level of certainty); State v. Howard, 221 Conn. 447, 454, 604 A.2d 1294 (1992) (two and one-half month period reliable when witnesses had ample opportunity to view assailants and their prior physical descriptions matched).

To the extent that the defendant in the present case points to some facts that could have cast some doubt on the reliability of the identifications, the trial court gave adequate consideration to them in making its determination. With respect to the effect of the statements by Beaudin to Clement confirming that he had identified the same individual that Valle had selected, the trial court reasonably found that "Clement is a particularly credible, forthcoming witness ... [and the] court has every confidence that his ability to recall and relate such details honestly will remain unaffected by ... Beaudin's unfortunate mistake...." In light of its findings, the trial court reasonably concluded that the identifications were reliable.

Accordingly, I concur in the judgment concluding that the admission of the identifications did not violate due process.

PALMER, J., concurring.

I agree with the majority that the due process rights of the defendant, Julian Marquez, were not violated by the state's use of the photographic identifications at issue. Consequently, I also agree that the judgment of conviction rendered by the trial court must be affirmed. Because I reach my conclusion concerning the defendant's due process claim by a somewhat different route than the majority, however, and because I do not agree with the majority's reading of the trial court's decision in one important respect,^[1] I am unable to join the majority opinion. I therefore concur in the result.

The trial court concluded that the identification procedures employed by the police in the present case were unnecessarily suggestive for two reasons.^[2] The first reason on which the trial court relied was the fact that the witness, Mark Clement,^[3] was shown a simultaneous rather than a sequential photographic array; that is, the photographs were displayed to him all at the same time rather than one at a time.^[4] The second reason on which the trial court based its finding of unnecessary suggestiveness was comprised of two components: First, the police did not employ a double-blind identification procedure,^[5] and, second, the detective who conducted the procedure commended Clement on his selection of the defendant's photograph because a second witness, Christopher Valle, previously had selected that same photograph.

With respect to the trial court's first finding, I agree with the majority and the state that, under our holding in State v. Ledbetter, 275 Conn. 534, 881 A.2d 290 (2005), cert. denied, 547 U.S. 1082, 126 S. Ct. 1798, 164 L. Ed. 2d 537 (2006), the trial court improperly concluded that the nonsequential identification procedure used by the police in this case was unnecessarily suggestive.^[6] Indeed, the procedure that the police used in the present case was less suggestive than the procedure that we approved of in Ledbetter because Clement, in contrast to the witness in Ledbetter, was expressly advised that he "should not conclude or guess that the photographs contain the person who committed the offense under investigation. You are not obligated to identify anyone. It is just as important to free innocent persons from suspicion as to identify guilty parties."^[7] Because there is nothing in the record of this case relating to the use of the nonsequential identification procedure that, following our opinion in Ledbetter, may be deemed to be unduly suggestive, the trial court's contrary finding is unsupportable.^[8]

The trial court's second reason for concluding that the procedure was unnecessarily suggestive is based on the fact that the procedure was conducted by the lead investigator in the case, namely, Detective Patricia Beaudin. As the trial court explained, it is "the undisputed judgment and recommendation of well respected scientific researchers that the practice of having persons with special knowledge of a criminal investigation administer identification procedures in the course of it must not be permitted in order to ensure against biasing those procedures and tainting any identifications to which they lead." Indeed, in the present case, Beaudin did, in fact, improperly advise Clement that he had "[done] good" by identifying the same person Valle identified.

I would not decide whether the failure of the police to use a double-blind identification procedure, at least without good cause for not doing so, is sufficient, standing alone, to warrant a finding of unnecessary suggestiveness. Because of the real risk that an investigator who, like Beaudin, knows the identity of the suspect may consciously or unconsciously "transmit information or expectations inadvertently to the witness, thereby leading the witness to make a particular selection," it is clear—in fact, it appears to be undisputed—that use of a double-blind procedure is preferable to the approach that the police followed in the present case.^[9] For purposes of this case, however, we need not decide whether the use of the double-blind approach, when practicable, may be required to avoid a finding of unnecessary suggestiveness, because, in the present case, the failure of the police to employ that approach resulted in the very harm that its use is designed to prevent: Beaudin, knowing the identity of the actual suspect, effectively relayed that information to Clement. Thus, as the trial court aptly explained, "[h]ere ... the risk of unfairness due to administrator bias was not just hypothetical, but real, for on at least one proven occasion ... Beaudin acted [on] her bias in a clear and inexcusably prejudicial manner—specifically, when she reacted to... Clement's identification of the defendant by commending him on choosing the defendant's photo[graph] and informing him that ... Valle had selected the same person." (Emphasis added.) In fact, she did so despite the following admonition on the photographic array itself: "Please do not discuss this case with other witnesses nor indicate in any way that you have, or have not identified someone."

It is true, of course, that Beaudin improperly conveyed this information to Clement after Clement had selected the defendant's photograph from the array. Thus, as the trial court stated, that information "was not shown to have tainted the witness' identification when it was initially made." As the trial court further observed, however, "[w]hat [Beaudin] risked by her conduct, however, was unfairly bolstering the witness' confidence in the strength of his photo[graphic] identification and the solidity of his basis in memory for it, thus making it harder for the defense to test the true certainty with which he made that identification on cross-examination, and correspondingly more difficult for the jury to assess the true strength and reliability of that identification in the totality of the circumstances." In other words, Beaudin's remark gave rise to an undue risk that the extent or degree to which Clement was confident about the accuracy of his identification would be skewed in favor of the state. The record, therefore, fully supports the finding of the trial court that Beaudin's statement was "clear[ly] and inexcusably prejudicial...." This is particularly significant because, even without such improper reinforcement, witnesses often are more confident of their identification after the fact. E.g., United States v. Williams, 522 F.3d 809, 812 (7th Cir.2008). Moreover, "[p]eople confuse certitude with accuracy and so are led astray." Id., at 811. Because Clement's level of confidence in his identification of the defendant cannot be separated from the identification itself—an identification is only as valuable as the witness' confidence in it—Beaudin's comment affected Clement's identification in a very real sense. That fact is not altered merely because the extent of Beaudin's influence on the identification may not become manifest until Clement's testimony at trial.^[10] Under the circumstances, therefore, I agree with the trial court's conclusion that "Beaudin's violation of the warning printed on her own photo[graphic] [array] cannot simply be treated as a harmless irrelevancy." Consequently, I believe that the trial court justifiably predicated its finding of unnecessary suggestiveness on this impropriety.

The majority maintains that Beaudin's statement did not render the identification procedure unnecessarily suggestive because the statement was made after Clement's initial identification of the defendant. According to the majority, once that initial identification had been made, the identification procedure was over, and nothing that occurred thereafter can be deemed to bear on the issue of whether the procedure was unnecessarily suggestive. For several reasons, I disagree with this narrow view of when a police administered eyewitness identification procedure ends. Before setting forth my reasons for reaching this conclusion, however, I first address a second, threshold point of disagreement with the majority, namely, the majority's assertion that the trial court itself did not treat Beaudin's comment to Clement as part of the pretrial identification procedure. As the following explication of the trial court's decision reveals, that decision leaves no room for doubt that the trial court viewed Beaudin's comment as occurring during the identification procedure.

At the outset of its decision, the trial court set forth with specificity the claims that the defendant raised in connection with his motion to suppress: "[T]he defendant moved this court, under [Practice Book §] 41-13 ... the fourteenth amendment to the United States constitution and article I, [§] 8, of the constitution of Connecticut, to suppress as evidence all pretrial and in-court identifications of him as a perpetrator of the crimes charged against him in this case." Thus, the two identifications involving Clement that the defendant moved to suppress were (1) the pretrial identification of the defendant that Clement made in connection with the out-of-court procedure administered by Beaudin, and (2) any in-court identification of the defendant that Clement might make during the course of the defendant's trial.

The trial court then explained the legal bases of the defendant's motion to suppress: "As grounds for [his] motion, the defendant has alleged, more particularly, that: (1) the procedure by which each challenged pretrial identification was obtained was unnecessarily suggestive; (2)... any in-court identification by a witness who previously identified him in an unnecessarily suggestive pretrial identification procedure would be irreparably tainted by the prior illegal identification and, thus, would have no independent basis; and (3)... each challenged identification ... must be suppressed for failure to meet minimum constitutional standards of reliability for procedurally tainted identification evidence, as announced by the United States Supreme Court in Neil v. Biggers, 409 U.S. 188, 93 S. Ct. 375, 34 L. Ed. 2d 401 (1972), and enforced under the due process clauses of our state and federal constitutions." Thus, with respect to the pretrial identification procedure that Beaudin administered, the defendant claimed that that procedure was unnecessarily suggestive. With respect to any subsequent in-court identification, the defendant claimed that it would be irreparably tainted due to the fact that the earlier, pretrial identification was unnecessarily suggestive. Finally, with respect to both identifications, the defendant maintained that they were unreliable and, therefore, must be suppressed. It is readily apparent, therefore, that all of the defendant's claims relate directly to the allegedly improper pretrial identification procedure that Beaudin administered; the defendant's motion to suppress contains no other claims, and the trial court did not purport to address any other claims.

After making extensive factual findings, the trial court set forth the applicable law in a section entitled "The Controlling Legal Standard." This section is relatively brief, and provides in relevant part: "To prevail on a motion to suppress identifications under the due process clause of the fourteenth amendment to the United States constitution or its Connecticut constitutional counterpart, in article I, [§] 8, of the constitution of Connecticut, the defendant must prove by a fair preponderance of the evidence both that a government administered pretrial identification procedure was unnecessarily suggestive and, if it was, that any subsequent identification to which it led or may lead in the future was or will be unreliable in the totality of the circumstances.... The reliability of an identification procedure is considered under various factors, such as 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.... Reliability is the linchpin in determining the admissibility of the identification testimony." (Citations omitted; internal quotation marks omitted.)

This is the only legal standard that the trial court mentions in its decision; at no time does the decision refer to a different legal test. Thus, consistent with the claims that the defendant raises in his motion to suppress, the law that the trial court applied pertains only to the allegedly unlawful pretrial identification procedure conducted by Beaudin and its effect on any subsequent in-court identification of the defendant.

The trial court then set forth in detail the parties' claims and responses. Thereafter, the court analyzed the defendant's claims concerning the pretrial identification of the defendant that Clement had made during the procedure administered by Beaudin. As I noted previously, the trial court concluded that the procedure was unnecessarily suggestive because of Beaudin's use of a nonsequential photographic array and because it was administered by a police officer who (1) had knowledge of the investigation and the suspect, and (2) commended Clement on his identification of the defendant immediately after Clement had made it.

The state claimed that Beaudin's comment to Clement did not render the identification procedure unnecessarily suggestive because (1) it occurred after Clement's identification of the defendant, and (2) in any event, the comment did no more than apprise Clement of what he ultimately would have learned prior to trial anyway, that is, that a second witness also had identified the defendant. In response to the state's first contention, the trial court acknowledged that "[t]he state correctly asserts that ... Beaudin's utterance came after the witness had selected the defendant's photo[graph]. Thus, the utterance was not shown to have tainted the witness' identification when it was initially made." The court nevertheless was unpersuaded that the comment had not adversely affected the identification procedure, explaining: "What [Beaudin] risked by her conduct, however, was unfairly bolstering [Clement's] confidence in the strength of his photo[graphic] identification and the solidity of his basis in memory for it, thus making it harder for the defense to test the true certainty with which he made that identification on cross-examination, and correspondingly more difficult for the jury to assess the true strength and reliability of that identification in the totality of the circumstances." In the trial court's view, therefore, Beaudin's comment had infected the identification procedure with unfairness due to the fact that the comment created an undue risk that Clement's trial testimony about his photographic identification of the defendant would be skewed in favor of the state because the comment improperly had bolstered his confidence in that identification. The trial court next addressed, and also rejected, the state's argument that Clement inevitably would have found out about the second witness' identification of the defendant in any event.

Having concluded that various "aspects of the procedures by which the challenged photo[graphic] identifications were obtained were unnecessarily suggestive," the trial court then applied the second prong of the applicable two part test for determining the constitutionality of identifications, namely, the reliability prong. The court determined that neither Clement's pretrial identification nor any subsequent in-court identification "was thereby rendered unreliable in the totality of the circumstances in which it was made." Turning to Clement's pretrial identification of the defendant, the trial court stated in relevant part: "In reference to ... Clement, although the unnecessary suggestiveness of ... Beaudin's postidentification commendation of him for selecting the [correct photograph] is well established on this record, it is equally well established that he [like the second witness, Valle] had a reliable independent basis for making his challenged photo[graphic] identification of the defendant based solely [on] his memory of the gunman's face."^[11] (Emphasis added.) The italicized language demonstrates beyond any doubt that the trial court found that Beaudin's comment had rendered the pretrial identification procedure unnecessarily suggestive. Moreover, unless the trial court had made such a finding, it would have had no reason to explain why Clement's identification of the defendant was reliable notwithstanding Beaudin's comment. Indeed, as I noted previously, all of the defendant's claims, including his claim concerning Beaudin's comment, emanated directly from the allegedly improper pretrial identification procedure that Beaudin administered, and there is nothing in the trial court's decision to suggest that the court treated the comment as representing an issue separate and apart from that procedure. In fact, the contrary is true: The record definitively establishes that the court considered the comment as part of the procedure itself.

The fact that the court treated the comment as having been made during that pretrial procedure is again confirmed by its analysis and resolution of the defendant's claim that any in-court identification by Clement should be suppressed on the ground that it merely would be the product of the unnecessarily suggestive pretrial identification procedure. With respect to this issue, the trial court stated: "As for future in-court identifications of the defendant by ... Clement, whose confidence in his own identification may well have been bolstered by ... Beaudin's postidentification comment to him, the court agrees with the state that ... Clement is a particularly credible, forthcoming witness ... who still appears to be quite capable of explaining the mental processes by which he made his challenged identification. The court has every confidence that his ability to recall and relate such details honestly will remain unaffected by... Beaudin's unfortunate mistake ... and thus will be sufficiently reliable in the totality of the circumstances not to require their suppression as evidence." The court's express references to Beaudin's comment in its analysis of the reliability prong reflects the court's earlier finding that the comment had rendered the pretrial identification procedure unnecessarily suggestive. Finally, once again, if the trial court had not made such a finding, there would have been no reason for the court to have turned to the reliability of the in-court identification.

It is crystal clear, therefore, that the trial court concluded that Beaudin's comment had rendered Clement's pretrial identification of the defendant unnecessarily suggestive, and that the comment had affected both that identification and any subsequent in-court identification of the defendant by Clement. Of course, as a practical matter, the unnecessarily suggestive nature of the procedure and its effect on Clement was likely to manifest itself in Clement's trial testimony about his pretrial identification of the defendant and, in particular, in his testimony about his level of confidence in that identification. Insofar as the administration of the identification procedure unfairly bolstered Clement's confidence in his pretrial identification of the defendant, that unfairness also was likely to manifest itself in Clement's testimony at trial about his level of confidence in any in-court identification of the defendant.^[12]

The trial court's determination that Beaudin's comment was part of the pretrial identification procedure and, therefore, must be analyzed under the two part test applicable to claims concerning the constitutionality of such identification procedures, is amply supported by the record. Beaudin, the procedure administrator, made her improper comment immediately after Clement had selected the defendant's photograph from the array and while he was continuing to interact with Beaudin about the subject matter of the procedure, that is, his identification of a suspect. The comment, therefore, occurred toward the end of the identification procedure but not before it finally had concluded. As long as the witness remains with the police and is discussing his identification of the suspect with the officer in charge of administering the procedure, there is no legitimate reason to presume that the identification procedure has concluded. Indeed, such interaction between the witness and the police may serve to taint the witness' identification, as it did in the present case, by bolstering the witness' confidence that he had, in fact, identified the actual perpetrator. As I stated previously, the confidence that a witness has in his identification is a critical aspect of the identification itself; indeed, the two are inextricably interrelated. Thus, Beaudin's improper comment to Clement concerning the identification that Clement had just made rendered the identification procedure unnecessarily suggestive because of the high likelihood that Beaudin's remark would serve to bolster Clement's confidence in his identification— the very same identification that had been the subject of the procedure at issue. Under the circumstances presented, the majority's insistence on treating Beaudin's comment as entirely separate and distinct from the identification procedure makes little sense.

Furthermore, to the extent that this court must decide precisely when an identification procedure is complete, there are strong policy reasons to conclude that it has not ended when, as in the present case, the police and the witness are discussing the witness' identification immediately after the witness identifies the suspect. In Manson v. Brathwaite, 432 U.S. 98, 111-13, 97 S. Ct. 2243, 53 L. Ed. 2d 140 (1977), the United States Supreme Court identified several considerations that it deemed relevant to the determination of how best to deal with pretrial identification evidence when the police have obtained that evidence by use of a procedure that is unnecessarily suggestive.^[13] The court characterized the first such factor as its "concern [generally] with the problems of eyewitness identification. Usually, the witness must testify about an encounter with a total stranger under circumstances of emergency or emotional stress. The witness' recollection of the stranger can be distorted easily by the circumstances or by later actions of the police." Id., at 112, 97 S. Ct. 2243. The court identified the second factor as police deterrence, that is, discouraging the police from engaging in conduct that is likely to lead to an unreliable identification. Id. The third and final factor is the effect that a particular approach will have on the administration of justice, including the extent to which the fact finder may be denied access to reliable evidence. Id., at 112-13, 97 S. Ct. 2243.

I believe that all of these considerations are advanced by a determination that the identification procedure in the present case had not concluded before Beaudin commended Clement on his identification of the defendant. With respect to the first such consideration, the present case exemplifies how the "later actions of the police"; id., at 112, 97 S. Ct. 2243; can have an unfair distorting effect on the original identification by altering the witness' confidence in that identification in a manner favorable to the state. Viewing the identification procedure more broadly to include Beaudin's improper comment to Clement also would promote the strong public interest in deterring misconduct by the police in their interactions with eyewitnesses at the very time that the police are engaging those eyewitnesses in identification procedures. Finally, a determination that the identification procedure had not concluded before Beaudin made her comment would result in no adverse effect on the administration of justice because the state still would have the opportunity to establish that the comment did not so taint the procedure, under the totality of the circumstances, as to require suppression of Clement's identification of the defendant.^[14]

The majority contends that "[t]he problems inherent" in the view that I advocate "are manifest, as there are many events that may, and often do, occur prior to or during trial that may reinforce or otherwise affect the witness' level of confidence in his recollection." Footnote 11 of the majority opinion. The majority further observes that "[n]one of these situations, however, presents a basis for excluding the identification"; id.; rather, they provide "customary grist for the jury mill." (Internal quotation marks omitted.) Id. I fully agree with the majority that examples abound of postidentification events that may influence the witness' confidence in his identification; indeed, it is equally possible to conceive of circumstances in which a witness' recollection of a particular suspect is influenced by events that occur prior to his participation in the procedure administered by the police. I also agree with the majority that such influences generally are addressed through cross-examination only. These examples, however, in no way undermine the fact that special steps must be taken to ensure that police sponsored identification procedures are undertaken with care, and that they are administered in an unbiased, evenhanded manner. This is true for obvious reasons, including the fact that witnesses are likely to place great weight on what the police have to say about who committed the crime under investigation. Furthermore, under the controlling two-pronged test pursuant to which reliability is the overriding consideration, it is highly unlikely that a comment of the kind made by Beaudin, standing alone, ever will result in suppression of the identification.^[15] Both the state and the defendant, however, have a strong interest in seeing to it that police administered identification procedures are fair, neutral and do not skew either the witness' initial identification of a suspect or the witness' confidence in that identification. Unfortunately, the approach that the majority adopts does not advance that objective. Nevertheless, for the reasons set forth by Justice Katz in her concurring opinion, I agree that the trial court properly determined that the photographic identifications at issue in the present case were reliable under the totality of the circumstances and, therefore, admissible at trial.^[16] Accordingly, I concur in the judgment.

I note, finally, that, despite the significant weight that juries tend to give eyewitness identification testimony, all too often, those identifications are inaccurate. As the United States Court of Appeals for the Third Circuit recently has stated, "[i]t is widely accepted by courts, psychologists and commentators that ``[t]he identification of strangers is proverbially untrustworthy.' [F. Frankfurter, The Case of Sacco and Vanzetti: A Critical Analysis for Lawyers and Laymen (Grosset and Dunlap 1962 Ed.) p. 30] (``What is the worth of identification testimony even when uncontradicted?... The hazards of such testimony are established by a formidable number of instances in the records of English and American trials. These instances are recent-not due to the brutalities of ancient criminal procedure.'); see also United States v. Wade, 388 U.S. 218, 228, 87 S. Ct. 1926, 18 L. Ed. 2d 1149 (1967) (stating that ``[t]he vagaries of eyewitness identification are well-known; the annals of criminal law are rife with instances of mistaken identification'); [C. Huff et al., ``Guilty Until Proven Innocent: Wrongful Conviction and Public Policy,' 32 Crime & Delinq. 518, 524 (1986)] (``the single most important factor leading to wrongful conviction in the United States ... is eyewitness misidentification'). The recent availability of post-conviction DNA tests demonstrate[s] that there have been an overwhelming number of false convictions stemming from uninformed reliance on eyewitness misidentifications. In 209 out of 328 cases [64 percent] of wrongful convictions identified by a recent exoneration study, at least one eyewitness misidentified the defendant. [S. Gross et al., ``Exonerations in the United States: 1989-2003,' 95 J.Crim. L. & Criminology 523, 542 (2004)]. In fact, ``mistaken eyewitness identifications are responsible for more wrongful convictions than all other causes combined.' [A. Yarmey, ``Expert Testimony: Does Eyewitness Memory Research Have Probative Value for the Courts?,' 42 Canadian Psychol. 92, 93 (2001)]. ``[E]yewitness evidence presented from well-meaning and confident citizens is highly persuasive but, at the same time, is among the least reliable forms of evidence.' Id....

"Even more problematic, ``jurors seldom enter a courtroom with the knowledge that eyewitness identifications are unreliable.' [Note, ``Process v. Outcome: The Proper Role of Corroborative Evidence in Due Process Analysis of Eyewitness Identification Testimony,' 88 Cornell L.Rev. 1097, 1099 n. 7 (2003)]. Thus, while science has firmly established the ``inherent unreliability of human perception and memory,' [id., at 1102] ... this reality is outside ``the jury's common knowledge,' and often contradicts jurors' ``commonsense' understandings.... [Id., at 1105 n. 48].... To a jury, ``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!"' Watkins v. Sowders, 449 U.S. 341, 352, 101 S. Ct. 654, 66 L. Ed. 2d 549 (1981) (Brennan, J., dissenting)...." United States v. Brownlee, 454 F.3d 131, 141-42 (3d Cir.2006).

Moreover, as the United States Supreme Court recognized more than forty years ago, "[a] major factor contributing to the high incidence of miscarriage of justice from mistaken identification has been the degree of suggestion inherent in the manner in which the prosecution presents the suspect to witnesses for pretrial identification. A commentator has observed that ``[t]he influence of improper suggestion upon identifying witnesses probably accounts for more miscarriages of justice than any other single factor—perhaps it is responsible for more such errors than all other factors combined.' ... Suggestion can be created intentionally or unintentionally in many subtle ways. And the dangers for the suspect are particularly grave when the witness' opportunity for observation was insubstantial, and thus his susceptibility to suggestion the greatest." (Citation omitted.) United States v. Wade, supra, 388 U.S. at 228-29, 87 S. Ct. 1926.

Because of the potentially grave consequences of witness misidentification, it is imperative that all reasonable efforts be made to ensure that the identification procedures used by the police are fair and neutral and that, to the fullest extent possible, those procedures promote reliable identifications. Of course, this endeavor is the shared responsibility of the police and the courts, both of which must accord paramount importance to emerging scientific research and literature in this critical area, particularly with respect to the use of identification procedures that are not sequential and not double-blind.^[17] For purposes of this case, however, the defendant has failed to establish that the identification procedures employed by the police rendered the challenged identifications so unreliable as to constitute a violation of due process.

Accordingly, I concur.

SCHALLER, J., concurring.

I concur and join in the majority's decision to affirm the judgment of the trial court. I write separately, however, to highlight several points, in the hope that doing so will provide some benefit to the trial bench and bar. Because eyewitness identification issues involving pretrial procedures arise frequently, our trial courts deserve guidance from this court in such matters.^[1] In order to maximize the usefulness of the combined opinions, I will attempt to synthesize the basic consensus.

First, I agree with the majority's decision to decide this appeal by addressing the state's alternative ground for affirmance, namely, that the trial court improperly determined that the identification procedures used by the police were unnecessarily suggestive, rather than to decide this appeal by addressing the reliability issue raised by the defendant, Julian Marquez. It would be feasible—and, in some respects, simpler—to affirm the trial court's decision by determining that the court properly concluded that the identifications were reliable under the totality of the circumstances test set forth by the United States Supreme Court in Neil v. Biggers, 409 U.S. 188, 93 S. Ct. 375, 34 L. Ed. 2d 401 (1972), and Manson v. Brathwaite, 432 U.S. 98, 97 S. Ct. 2243, 53 L. Ed. 2d 140 (1977). The rationale for a determination of reliability is concisely set forth in part II of Justice Katz' concurring opinion. Although I do not disagree with that rationale, affirming on that basis is not prudent, in my mind, because it would fail to address the trial court's disregard of the test set forth in State v. Ledbetter, 275 Conn. 534, 574, 881 A.2d 290 (2005), cert. denied, 547 U.S. 1082, 126 S. Ct. 1798, 164 L. Ed. 2d 537 (2006). Although the trial court conceivably may have been able to ground its conclusion on the totality of circumstances test, it did not do so. Instead, it chose to base its decision solely on the research material presented to it. If there is a single important message to be gleaned from the present case, it is that trial courts are bound to apply the totality of circumstances test in these situations.

Second, I emphasize my agreement with the majority opinion that the trial court abused its discretion in concluding that the identification procedures employed in the present case were unnecessarily suggestive. I do not find it useful, however, to establish whether the trial court purported to establish a per se rule or whether the trial court's admonition to the police was, in fact, a directive. The trial court, in my view, was persuaded by the scientific literature presented by the defendant, which supported his claim that simultaneous lineups administered by an official who was directly involved in the investigation are unnecessarily suggestive. Although I appreciate that much of the recent research raises very serious questions about the use of simultaneous photographic arrays and the use of an interested administrator, the trial court failed to provide a totality of circumstances analysis establishing that either of those methods rendered the identification procedures in the present case unnecessarily suggestive.

Finally, and again in light of the different viewpoints expressed in the majority and concurring opinions, I emphasize that this controversy does not alter the essential holding of Ledbetter, that is, that the inquiry for determining whether an identification procedure was unnecessarily suggestive is still a case-by-case, factually intensive, totality of the circumstances standard. See id. In other words, none of the opinions in this case should be read to establish any per se rules, either universally authorizing or universally disapproving of any identification procedure. My goal, in this concurring opinion, is to highlight precisely what I see as the trial court's misstep in this case, and to offer general guidance to trial courts to help them avoid making similar missteps.

Preliminarily, I wish to emphasize what I consider to be the strongest argument in favor of deciding this case on the presented ground for appeal, rather than on the state's alternative ground for affirmance. The question raised by the alternative ground, whether the trial court improperly concluded that the identification procedures were unnecessarily suggestive, requires that we address two features of identification procedures, the validity and reliability of which are currently called into question by scientific studies. See, e.g., G. Wells & D. Quinlivan, "Suggestive Eyewitness Identification Procedures and the Supreme Court's Reliability Test in Light of Eyewitness Science: 30 Years Later," 33 Law & Hum. Behav. 1, 8 (2009) (describing ways interested photographic lineup administrator can unintentionally or intentionally influence eyewitness identification); Technical Working Group for Eyewitness Evidence, United States Dept. of Justice, "Eyewitness Evidence: A Guide for Law Enforcement" (October, 1999) p. 9 (interested administrator's "unintentional cues [e.g., body language, tone of voice] may negatively impact the reliability of eyewitness evidence"); see also sources cited in footnote 11 of Justice Katz' concurring opinion (standing for proposition that sequential identification procedures are significantly more accurate than simultaneous identification procedures). Both the majority opinion and Justice Katz' concurring opinion acknowledge that the evidence in this area of study is in a state of flux. In light of the evolving nature of the scientific research on this issue, it can be argued that it is unhelpful to reach the issue of the propriety of these procedures in the present case, when doing so may run the risk of suggesting that the identification procedures at issue in this case provide a reference point or model for conducting future identification procedures. Put simply, the jury is still out on the continued validity of those procedures, especially the interested administrator feature.^[2] Moreover, because the factual scenario involving the identification procedures in this case does not provide any helpful guidance to trial courts in judging the propriety of the procedures generally, I believe that this case does not provide an appropriate reference point or model for future investigations. The majority opinion, however, calls attention to this point and carefully avoids indicating either approval or disapproval of these two identification procedures. The majority analysis focuses on the specific factual circumstances of the present case, and highlights the fact that the trial court failed to do so.^[3]

I offer a few observations regarding the trial court's ruling that the identification procedures employed in the present case were unnecessarily suggestive, first explaining why, in my view, the trial court's decision constituted an abuse of discretion. In reviewing a trial court's decision regarding the admissibility of identification evidence under the abuse of discretion standard, "we will indulge in every reasonable presumption in favor of the trial court's ruling." (Internal quotation marks omitted.) State v. Ledbetter, supra, 275 Conn. at 548, 881 A.2d 290. We clearly have stated that "trial courts [must] determine whether individual identification procedures are unnecessarily suggestive on the basis of the totality of the circumstances surrounding the procedure, rather than replacing that inquiry with a per se rule." Id., at 574, 881 A.2d 290. Such a determination, therefore, necessarily must be grounded on a factually intensive inquiry, not on generalizations, and not solely on scientific literature. This is the principle that the trial court overlooked in arriving at its conclusion that the identification procedures in the present case were unnecessarily suggestive.

The trial court did not support its determination that the identification procedures at issue in the present case were unnecessarily suggestive by reference to the totality of the circumstances surrounding the procedures, but instead relied solely on the facts that: (1) both procedures involved simultaneous photographic lineups rather than sequential lineups; and (2) both lineups were conducted by an interested administrator, rather than in a double-blind manner, by an uninterested person who did not know who the suspect was and whether the suspect was present in the lineup. In concluding that these features alone rendered the procedures unnecessarily suggestive, the court relied exclusively on scientific literature that criticizes both types of procedures. See, e.g., G. Wells, M. Small & S. Penrod et al., "Eyewitness Identification Procedures: Recommendations for Lineups and Photospreads," 22 Law & Hum. Behav. 603, 614 (1998) (explaining correlation between simultaneous photographic lineups and relative judgment); J. Turtle, R. Lindsay & G. Wells, "Best Practice Recommendations for Eyewitness Evidence Procedures: New Ideas for the Oldest Way to Solve a Case," Can. J. of Police and Security Services (March, 2003) pp. 5-19 (same); G. Wells & E. Olson, "Eyewitness Testimony," 54 Ann. Rev. Psychol. 277, 289 (February, 2003) (risk of misidentification due to unconscious bias of interested administrator of identification procedure). Other than noting that the procedures employed involved a simultaneous photographic array and an interested administrator, the trial court pointed to no particular facts that would support a determination that the procedures "[gave] rise to a very substantial likelihood of irreparable misidentification." (Internal quotation marks omitted.) State v. Cook, 262 Conn. 825, 832, 817 A.2d 670 (2003). In fact, the factual circumstances highlighted by the trial court tend to lead to the opposite conclusion. Specifically, the court found that the array did not unfairly highlight the defendant, and that both photographic arrays contained the instruction recommended by Ledbetter, that the suspect may or may not be included in the lineup. State v. Ledbetter, supra, 275 Conn. at 575, 881 A.2d 290.

I emphasize that I am not convinced that the trial court created a per se rule by concluding that the procedures at issue in this case were unnecessarily suggestive. Rather, it was the trial court's failure to ground its determination on the totality of the circumstances that, in my view, constituted an abuse of discretion. The one distinguishing factual circumstance noted by the trial court, that the detective who administered the photographic arrays remarked to one of the witnesses, Mark Clement, after he had identified the defendant from the photographic array, that he had "[done] good" because he identified the same person picked out by the other witness, is irrelevant because the detective's validation of Clement's identification occurred after the fact.^[4] In the absence of some reference to particular aspects of the identification procedures that rendered them unnecessarily suggestive, I believe that the trial court abused its discretion in relying solely on scientific studies, rather than grounding its decision on the totality of the circumstances surrounding the identification.^[5]

Finally, in light of the different perspectives offered by the majority opinion and the concurring opinions, I offer the following guidance to trial courts as they undertake to discern a route through the virtual forest of opinions in the present case. The problem is twofold: what to do when confronted with an identification procedure that employs one of these two methods; and how to deal with scientific literature discussing identification procedures. As to the first question, the answer remains the same as it was following Ledbetter. In determining whether an identification procedure is unnecessarily suggestive, the court must make its inquiry "on the basis of the totality of the circumstances surrounding the procedure, rather than replacing that inquiry with a per se rule." State v. Ledbetter, supra, 275 Conn. at 574, 881 A.2d 290. In other words, the inquiry remains what it always has been, a factually intensive, case-by-case determination made on the basis of the totality of the circumstances.

As to the second issue, Ledbetter specifically authorized the trial court to consider scientific studies concerning an instruction to a witness that the perpetrator may or may not be present. Id. Beyond that, although Ledbetter did not preclude trial courts from reviewing scientific studies that are offered by the parties regarding the admissibility of identification testimony, it is clear that trial courts are not authorized to rely on scientific studies in order to create new rules. See, e.g., id., at 568, 881 A.2d 290 (defendant not obligated to present to trial court scientific studies in support of argument that Connecticut should abandon Biggers factors on state constitutional grounds because "the trial court was bound to apply the Biggers factors in its analysis").

Finally, the mere fact that persuasive scientific literature may be presented to the trial court does not relieve the court of its obligation to ground its decision as to whether the procedure was unnecessarily suggestive on the factual circumstances surrounding the identification procedure. In Ledbetter, the court left no doubt that "[t]he circumstances surrounding the various identification procedures present too many variables for us to conclude that a per se rule is appropriate." Id., at 574, 881 A.2d 290. The court, however, did identify one circumstance as giving rise to the need for an instruction by the trial court warning the jury of the risk of misidentification, namely, when an administrator indicates that a suspect is present in the identification procedure. Id., at 579, 881 A.2d 290. It seems likely that, as scientific knowledge increases in the field of eyewitness identification, other elements also will be identified as requiring remedial steps. It is noteworthy that the remedial instruction would be unnecessary if disinterested, double-blind administrators were required in such procedures. Although the court in Ledbetter was reluctant to restrict the province of the law enforcement agencies of the state in their application of the processes of eyewitness identifications, consensus in research studies may lead to such restrictions in the future. Although the court in Ledbetter did not establish any safeguards against the "``relative judgment process,'" it was aware of the dangers of such a process. Id., at 572, 881 A.2d 290 (research indicating that eyewitnesses tend to identify person from lineup who looks most like culprit relative to other members of lineup).

"The vagaries of eyewitness identification are well-known; the annals of criminal law are rife with instances of mistaken identification. Mr. Justice Frankfurter once said: What is the worth of identification testimony even when uncontradicted? The identification of strangers is proverbially untrustworthy. The hazards of such testimony are established by a formidable number of instances in the records of English and American trials. These instances are recent—not due to the brutalities of ancient criminal procedure." (Internal quotation marks omitted.) Id., at 577, 881 A.2d 290. This reinforces the idea that the appellate courts of this state must remain vigilant to developments in this field of inquiry by engaging in careful study of the scientific literature.^[6] Further developments may require the exercise of supervisory authority. Until that time, each challenge to an identification procedure must be addressed in accordance with our well established rule requiring that a trial court conduct a factually intensive inquiry, focusing on the totality of the circumstances, to determine whether the procedure was unnecessarily suggestive, and, if so, whether the identification was nevertheless reliable.

NOTES

[1] The fourteenth amendment to the United States constitution provides in relevant part: "No State shall ... deprive any person of life, liberty or property, without due process of law...."

[2] Article first, § 8, of the Connecticut constitution provides in relevant part: "No person shall ... be deprived of life, liberty or property without due process of law...."

[3] In Biggers, the United States Supreme Court held that when there is a finding of unnecessary suggestiveness, the court considers "whether under the ``totality of the circumstances' the identification was reliable.... [T]he factors to be considered in evaluating the likelihood of misidentification include the opportunity of the witness to view the criminal at the time of the crime, the witness' degree of attention, the accuracy of the witness' prior description of the criminal, the level of certainty demonstrated by the witness at the confrontation, and the length of time between the crime and the confrontation." Neil v. Biggers, supra, 409 U.S. at 199-200, 93 S. Ct. 375. In Manson, the court reaffirmed the holding of Biggers, stating that "reliability is the linchpin in determining the admissibility of identification testimony for both pre- and post-Stovall confrontations. [See Stovall v. Denno, 388 U.S. 293, 87 S. Ct. 1967, 18 L. Ed. 2d 1199 (1967).] The factors to be considered are set out in Biggers." Manson v. Brathwaite, supra, 432 U.S. at 114, 97 S. Ct. 2243. For convenience, we hereinafter refer to this test as the Manson test.

[4] The defendant also asserts that Manson should be modified or overruled to the extent that it would allow the admission of eyewitness identifications procured through procedures similar to those used by the police in this case. As the defendant concedes, however, it is beyond the authority of this court to overrule a decision of the United States Supreme Court with respect to an issue of federal law. See, e.g., State v. Ledbetter, 275 Conn. 534, 559, 881 A.2d 290 (2005), cert. denied, 547 U.S. 1082, 126 S. Ct. 1798, 164 L. Ed. 2d 537 (2006). Therefore, we decline to entertain this argument.

[5] The state also responds to the defendant's argument that Manson should be modified or overruled. For the reason set forth in footnote 4 of this opinion, we need not address the state's argument on this point.

[6] Although the state notified this court of its intention to offer alternative grounds for affirmance by filing a statement of such grounds pursuant to Practice Book §§ 63-4(a)(1) and 84-11, this particular claim was not specified in the state's filing. Nevertheless, we will consider this additional alternative ground for affirmance inasmuch as the defendant has not argued that he was prejudiced by the state's failure to include this claim in its statement. This determination is bolstered by the fact that the claim was fully briefed and argued before this court, and the defendant had the opportunity to, and did, respond to this issue. See, e.g., Dickinson v. Mullaney, 284 Conn. 673, 682 n. 4, 937 A.2d 667 (2007); State v. DeLoreto, 265 Conn. 145, 152 n. 11, 827 A.2d 671 (2003).

[7] Clement recalled drinking only one twenty-two ounce beer during the course of the evening, whereas Valle testified that he and the other three men shared the contents of a one-half pint bottle of Hennessy cognac and a small bottle of hypnotic liquor.

[8] Delgado's girlfriend also was present, but she remained in the bedroom in the back of the apartment during the robbery.

[9] The arrays that Clement and Valle viewed differed only in that the position of the defendant's photograph in each array had changed.

[10] In the literature, this process often is referred to as "absolute judgment."

[11] In my view, Justice Palmer's concurrence is based on a misunderstanding of the facts of this case and a misapplication of the standard set forth in Ledbetter. The crux of Justice Palmer's disagreement with the majority is his view that Beaudin's comment to Clement, although made after Clement had already confirmed his selection of the defendant's photograph from the array, "gave rise to an undue risk that the extent or degree to which Clement was confident about the accuracy of his identification would be skewed in favor of the state." Suggestiveness, however, refers to the witness' initial selection of an individual from a photographic array or lineup. In other words, the term "suggestive" signifies that the police have done something during the course of the identification procedure to suggest that the witness should choose a specific individual. Once Clement chose the defendant's photograph, however, the procedure had ended, and it no longer was possible for Beaudin to suggest anything. The trial court recognized this when it concluded that Beaudin's "utterance was not shown to have tainted the witness' identification when it was initially made." Justice Palmer strains to justify his position by improperly shifting the focus from the impact of police conduct on the out-of-court identification to the impact of the out-of-court identification on the subsequent in-court identification, thus unduly expanding the temporal and conceptual scope of what constitutes an identification procedure administered by the police.

The problems inherent in Justice Palmer's view are manifest, as there are many events that may, and often do, occur prior to or during trial that may reinforce or otherwise affect the witness' level of confidence in his recollection. For instance, the witness may see a news report of the suspect's arrest, or he may disregard the warnings that police typically provide and compare notes with other witnesses about his identification of a particular suspect. Moreover, when a witness takes the stand, his recollection may be affected by his observation of the individual he selected during the identification procedure sitting at the defense table. All of these scenarios, and countless others, carry the potential for affecting the "witness' confidence in the strength of his photo[graphic] identification and the solidity of his basis in memory for it, thus making it harder for the defense to test the true certainty with which he made the identification on cross-examination...." None of these situations, however, presents a basis for excluding the identification. To the contrary, the defendant may elicit from the witness, on cross-examination, any factors that might have influenced the witness' confidence in the accuracy of his initial pretrial identification. The jury is then in the best position to weigh the probative value of the identification in light of these potential influences. This is truly "customary grist for the jury mill. Juries are not so susceptible that they cannot measure intelligently the weight of identification testimony that has some questionable feature." (Internal quotation marks omitted.) State v. Garner, 270 Conn. 458, 469, 853 A.2d 478 (2004).

[12] Although Justice Katz concedes in her concurrence that the trial court "failed to make specific findings" that the risks purportedly inherent in simultaneous presentations "were present in the present case," she nonetheless concludes that "the trial court made its determination of unnecessary suggestiveness by examining the procedure as a whole, such that the use of the simultaneous array was evaluated in concert with the other factors present." The only other "factor" the concurrence refers to, however, is the group of studies indicating that, generally, the use of an uninterested administrator may lead to more accurate results. Despite her best efforts, however, Justice Katz cannot highlight a single fact that the trial court relied on in this case to determine that the identification procedures were unnecessarily suggestive. Although the trial court expressed concern about Beaudin's confirmatory comment to Clement, it specifically found that this comment did not taint the accuracy of the initial identification, with which our suggestiveness analysis is concerned.

[13] We have, however, set forth the standard of review for the reliability prong of the Manson test. In State v. Wooten, 227 Conn. 677, 631 A.2d 271 (1993), we stated that "we examine the legal question of reliability with exceptionally close scrutiny and defer less than we normally do to the related fact finding of the trial court." (Internal quotation marks omitted.) Id., at 688, 631 A.2d 271, quoting State v. Gordon, 185 Conn. 402, 416, 441 A.2d 119 (1981), cert. denied, 455 U.S. 989, 102 S. Ct. 1612, 71 L. Ed. 2d 848 (1982).

[14] In his reply brief, the defendant claims that the state "never argued that a sequential showing would not have been superior to a simultaneous showing," and quotes the trial court's memorandum of decision for the proposition that "the state does not dispute the findings of scientific researchers that sequential identification procedures are more reliable than simultaneous identification [procedures]." (Internal quotation marks omitted.) The defendant claims that the state has thus waived its ability to contest this point on appeal, citing State v. Cruz, 269 Conn. 97, 106, 848 A.2d 445 (2004), for the principle that a party cannot challenge on appeal an alleged error that it has induced. Although we agree with this principle, we find it has no application in the present case. First, the state did generally contest the defendant's argument that the identification procedures used in this case were unnecessarily suggestive. Second, there is a distinct difference between inducing an error and failing to address a discrete argument that the court eventually relies on in making its decision. Finally, we note that the appropriate legal standard for determining unnecessary suggestiveness does not require that the court decide whether a particular procedure is more reliable or superior but, rather, whether the procedure employed is designed or administered in such a way as to suggest to the witness that a particular individual is the correct choice. See, e.g., State v. Gold, 180 Conn. 619, 656, 431 A.2d 501, cert. denied, 449 U.S. 920, 101 S. Ct. 320, 66 L. Ed. 2d 148 (1980). Thus, the significance that the defendant attaches to the state's failure to address this argument is misplaced.

[15] In her concurrence, Justice Katz appears to agree generally with our characterization of the trial court's analysis. The sole factual ground on which Justice Katz bases her disagreement with us that the trial court implemented a per se rule is her belief that "the trial court also reasonably relied on the fact that the detective in charge of the investigation not only administered the procedure but also conveyed approval of the identification made by one of the witnesses as a basis for its determination that the identification procedure was unnecessarily suggestive...." The concurrence, however, overlooks the fact that the court determined, unequivocally, that Beaudin's "utterance was not shown to have tainted the witness' identification when it was initially made." This is the court's key finding relating to the suggestiveness of the procedure itself with respect to the use of an interested administrator. Justice Katz appears to misread the court's memorandum of decision when she declares that "[t]he trial court was particularly troubled by [Beaudin's utterance] because Clement noticeably had hesitated before selecting the defendant's photograph." This statement is flatly contradicted by the court's statement that "the only reason [Clement] felt such hesitation was his genuine concern, which was reasonable under the circumstances, that he not identify an innocent man.... [H]e was not at all sure that the true perpetrator would be in the array, and so he commendably took his time in order not to implicate an innocent man." The trial court concluded with a resounding endorsement of the quality of Clement's identification: "The court thus agrees with the state that his selection of the defendant was not made by the exercise of relative judgment ... but by the process of positive selection based [on] his personal memory of the gunman's face." Moreover, when the court did examine the photographic arrays actually used in the present case, it determined that they "did not unfairly highlight the defendant or promote his identification by the witnesses in any way." It is, therefore, unsurprising that Justice Katz fails to point to any basis in fact for the trial court's determination that the procedures used in this case were so flawed as to present a "very substantial risk of irreparable misidentification." State v. Reid, 254 Conn. 540, 555, 757 A.2d 482 (2000).

[16] Although similar language was employed by the court in Stovall v. Denno, 388 U.S. 293, 301-302, 87 S. Ct. 1967, 18 L. Ed. 2d 1199 (1967), this is the first occasion that the court used this exact phrase in this precise context.

[17] The similarity in names between this case and the 2005 Ledbetter case cited previously in this opinion is purely coincidental. Hereinafter, all references to Ledbetter are to the 2005 case.

[18] We focus primarily on the presentation of photographs in either a simultaneous or sequential fashion, and whether the person administering the procedure is "blind" to the identity of the suspect.

[19] Although the studies that the state presents were not published when the trial court made its ruling, both parties had the opportunity to supply this court with the most current research and to respond to the research presented by the opposing side in their briefs and at oral argument. Furthermore, "it is appropriate for this court to survey relevant scientific data as that data [have] been reported in the decisions of other courts and in the scientific literature. Such a survey does not amount to fact-finding by this court." State v. Ledbetter, supra, 275 Conn. at 568, 881 A.2d 290; see also State v. Porter, 241 Conn. 57, 94-95, 698 A.2d 739 (1997) (even when no evidence is presented in trial court, appellate court can take notice of relevant scientific literature), cert. denied, 523 U.S. 1058, 118 S. Ct. 1384, 140 L. Ed. 2d 645 (1998).

[20] The authors reach this recommendation by analogizing lineups with scientific experiments, in which double-blind procedures are widely accepted when necessary to ensure that the tester does not consciously or unconsciously act so as to bias the results. G. Wells et al., supra, 22 Law & Hum. Behav. at 627.

[21] Intended as a practical field guide for law enforcement personnel, this document provides instructions and advice regarding (1) the composition of lineups (photographical and live), (2) model instructions that should be administered to the eyewitness prior to attempting an identification, (3) the method of conducting the identification procedure, and (4) the proper method of recording the identification results. United States Dept. of Justice, supra, pp. at 29-38.

[22] As its title suggests, this article is intended to marshal the relevant scientific theory together with existing recommendations guiding the training of police services in North America and elsewhere in the world and to transform them into practical recommendations that can be used by law enforcement personnel with the goal of increasing accurate eyewitness identifications.

As we were unable to obtain the exact pagination for this article, we do not provide pinpoint citations. The text of this article may be found at page 105 of volume one of the appendix to the defendant's brief to this court.

[23] The authors note the beneficial use of double-blind procedures in scientific experiments and clinical drug trials, analogizing the eyewitness identification process to such endeavors.

[24] The pilot program was designed to compare the effectiveness of the sequential, double-blind method with the traditional, nonblind (or, more accurately, single-blind) simultaneous lineup procedure. See S. Mecklenburg, supra, p. ii. In her concurrence, Justice Katz claims that "the conclusions [of the Mecklenburg Report] have been discredited as the product of an unsound, unscientific methodology that does not support the conclusions reached therein." We believe that this is an overblown and inaccurate assessment of the criticism of the report. Although some commentators have sharply criticized the methodologies employed in the report, at least one prominent researcher in the field has recognized that "partisans on both sides of the debate over procedures have unfairly dismissed some criticism and praise of the ... [r]eport as reflecting nothing more than the scientific commentators' stubborn loyalty to their own preexisting beliefs." D. Schacter et al., "Policy Forum: Studying Eyewitness Investigations in the Field," 32 Law & Hum. Behav. 3, 4 (2007). Other prominent academics in the field, commenting on the debate surrounding the supposed inadequacies of the report, have declared: "Given that there is so much left unresolved we believe it premature to advocate policy change, especially since the policy communities are so dispersed and since psychological science will both take a black eye and have difficulty implementing alternative policies if current advocacy is found to be incorrect, oversold or both." S. Ross & R. Malpass, "Moving Forward: Response to ``Studying Eyewitness Investigations in the Field,'" 32 Law & Hum. Behav. 16, 17 (2007). The same authors opined that the purported methodological flaw was not particularly important, in light of the purposes of the Mecklenburg Report: "After carefully examining the arguments and the available research, we find little evidence that the blind confound is important even for an academic interpretation of the Illinois study." Id. Thus, we think it is hyperbole to state that the Mecklenburg Report has been "discredited" or that its conclusions are unsupported. We believe the very controversy and debate engendered by this report is but further evidence that all of the research in this area must be taken with a substantial dose of salt.

[25] The author of the report also collected and analyzed surveys from officers in the field, highlighting practical challenges in implementing the procedures.

[26] In the appendix to his reply brief, the defendant included an article that is highly critical of the methodologies employed in the Mecklenburg Report. That article calls for further, better designed field studies and recognizes that "[a] standoff has arisen" in the field. See D. Schacter et al., "Policy Forum: Studying Eyewitness Investigations in the Field," 32 Law & Hum. Behav. 3, 4 (2007).

[27] This article also highlights the inherent uncertainty in laboratory studies in which many aspects of the study methodology that may significantly impact results are unknown or underreported. See McQuiston-Surrett, R. Malpass & C. Tredoux, supra, at 12 Psychol., Pub. Policy & L. 160-61.

[28] The trial court was even more direct in its oral ruling on the motion to suppress: "The bottom line on that, and I'll just state for the record for you now, is that I am concerned about two aspects of the procedures [the ``nonblind process' and simultaneous presentation of photographs] which we[re] utilized here because of their generic potential to cause unfair prejudice in [the] identification process." (Emphasis added.)

[29] The defendant's contrary contention notwithstanding, it is appropriate for this court to engage in close scrutiny of the scientific evidence presented to the trial court; see State v. Ledbetter, supra, 275 Conn. at 568, 881 A.2d 290; and to review the legal conclusions drawn from such evidence de novo. See State v. Porter, 241 Conn. 57, 94-95, 698 A.2d 739 (1997), cert. denied, 523 U.S. 1058, 118 S. Ct. 1384, 140 L. Ed. 2d 645 (1998). The defendant's citation to our opinion in Schoonmaker v. Cummings & Lockwood of Connecticut, P.C., 252 Conn. 416, 747 A.2d 1017 (2000), simply does not support his proposition that our review of this evidence would constitute "an unwarranted interpretation of the evidence before the trial court."

[30] For instance, in a case very similar to the present case, both factually and in terms of the claims raised, the Appellate Court concluded, rather persuasively, that, "[g]iven the limited number of studies on the subject [at that time], [the court is] not convinced ... that [the] state constitution requires ... [the] adopt[ion] [of] double-blind, sequential identification procedures because the traditional procedures are unnecessarily suggestive." State v. Nunez, supra, 93 Conn.App. at 832, 890 A.2d 636; see also State v. Nieves, 106 Conn.App. 40, 50, 941 A.2d 358 ("[d]ue process does not require the suppression of a photographic identification that is not the product of a double-blind, sequential procedure"), cert. denied, 286 Conn. 922, 949 A.2d 482 (2008). As we have noted in this opinion, we are convinced that the research is even more unsettled today in light of the introduction of the Mecklenburg Report and the other field reports cited therein. See S. Mecklenburg, supra, pp. at 42-43 (referring to field studies conducted in Hennepin County, Minnesota, and Queens, New York).

[31] In her concurrence, Justice Katz expresses a concern that our decision will "discourage, if not halt, development of jurisprudence surrounding witness identification procedures," and "[close] off debate on witness identification procedures at a time when a clearer consensus about those procedures is just beginning to emerge from scientific research." We do not believe that this decision risks any such result. In fact, we believe that our approach clearly evinces an open-mindedness with respect to the import of the continuing scientific debate, and that our adherence to Ledbetter's ad hoc approach is the most conducive method for ensuring that the jurisprudence in this area does not become ossified. We believe that this opinion encourages trial courts to continue to consider the evolving scientific evidence, in light of the facts and circumstances of each individual case, to determine if a particular identification procedure is unnecessarily suggestive.

In her concurrence, Justice Katz highlights several articles, the product of her own independent research, that she claims either supports the defendant's assertions or calls into question the methodology of the Mecklenburg Report. See footnotes 8 through 13 and accompanying text of Justice Katz' concurrence. Notwithstanding our belief that Justice Katz has cherry picked these articles and that their results and recommendations are much more ambivalent than Justice Katz portrays them to be, the very existence of these articles serves as further support for our conclusion that the science of eyewitness identification procedures is far too unsettled to support a per se approach. The additional research that Justice Katz presents simply adds more silt to the already muddy water. To the extent that Justice Katz cites further studies in support of the supremacy of sequential, double-blind procedures, we remain convinced that the very availability of such research emphasizes the uncertain state of the science. If the issue were definitively decided, the scientific debate would have ceased. Justice Katz has demonstrated that it has not. Thus, we remain tethered to the ad hoc approach mandated by Ledbetter until the scientific evidence reaches a point of uniformity with respect to the unnecessarily suggestive nature of simultaneous, single-blind procedures such that due process requires us to jettison such procedures in favor of a demonstrably superior alternative. We are not yet at that point.

[32] In Ledbetter, we engaged in a very thorough review of several aspects of the relevant scientific research, particularly, the validity of the fourth Biggers factor, namely, witness certainty, which we found to be a relatively unreliable measure of reliability, as well as the effect that a warning can have on procedures that are not double-blind and the process of relative judgment. See generally State v. Ledbetter, supra, 275 Conn. at 566-75, 881 A.2d 290. We are confident that our current understanding of the uncertain state of the science is consistent with the letter and tone of Ledbetter.

[33] In her concurrence, Justice Katz expresses concern that, by deciding this case on the state's alternative ground for affirmance, our opinion "signals our approval of the use of an interested administrator and closes off debate on witness identification procedures...." The plain language of the Appellate Court's conclusion in State v. Smith, 107 Conn.App. 666, 674, 946 A.2d 319, cert. denied, 288 Conn. 902, 952 A.2d 811 (2008), is clearly to the contrary. Holding that due process does not require a finding that procedures that are not sequential or double-blind are unnecessarily suggestive is quite different from declaring such procedures forever immune from challenge. We expressly leave open the possibility that, in a future case, under different circumstances, the procedures utilized by the police in the present case could violate due process. We are especially cognizant of the theoretical and commonsense value of utilizing a blind administrator whenever practical. The unequivocal intent of the court's conclusion in Smith, as well as the thrust of this entire opinion, however, is to reinforce our holding in Ledbetter that each eyewitness identification scenario must be judged on its own facts.

[34] In his concurrence, Justice Palmer reads this statement and similar statements to signify that the trial court believed that Beaudin's comment was "part of the pretrial identification procedure" and thus susceptible to review for suggestiveness. In fact, the entire basis for Justice Palmer's concurrence rests on the most slender of reeds. What Justice Palmer fails to recognize is that the trial court was responding to a motion to suppress two distinct pieces of evidence. The defendant sought exclusion of both the out-of-court identification, which was derived from the photographic identification procedure at issue, and Clement's subsequent in-court identification. The court confirmed this fact in the first paragraph of its memorandum of decision, referring to the defendant's "[motion] to suppress as evidence all pretrial and in-court identifications" on the grounds "that: (1) the procedure by which each pretrial identification was obtained was unnecessarily suggestive; [and] (2) that any in-court identification by a witness who previously identified him in an unnecessarily suggestive pretrial identification procedure would be irreparably tainted by the prior illegal identification...." (Emphasis added.) In his concurrence, Justice Palmer asserts that "[i]t is crystal clear ... that the trial court concluded that Beaudin's comment had rendered Clement's pretrial identification of the defendant unnecessarily suggestive...." In fact, the record demonstrates precisely the opposite. After determining that the photographic identification procedure used by the police in this case "created an unnecessary risk of producing irreparable identifications" because of the use of a simultaneous photographic array the administration of which was not double-blind, the trial court explicitly found that "Beaudin's utterance came after the witness had selected the defendant's photograph ... [and] [t]hus ... was not shown to have tainted the witness' identification when it was initially made." (Emphasis added.) Although Justice Palmer places great emphasis on the word "initially," it is clear from the very next sentence that the court was using that word to refer specifically to the fact that Beaudin's remark did not affect the out-of-court identification. The court stated in that next sentence: "What [Beaudin] risked by her conduct ... was unfairly bolstering [Clement's] confidence in the strength of his photo[graphic] identification and the solidity of his basis in memory for it, thus making it harder for the defense to test the true certainty with which he made that identification ... and correspondingly more difficult for the jury to assess the true strength and reliability of that identification...." This potential harm clearly refers to the prejudicial effect that Beaudin's comment might have on Clement's subsequent in-court identification of the defendant. As the trial court noted, however, the physical evidence derived from the photographic identification, i.e., the photographic array itself with Clement's initials next to a circled photograph of the defendant, was not in any way tainted by Beaudin's remark because it was produced before she made the remark. Clement's later memory of the procedure is simply irrelevant to a determination of whether the procedure itself was suggestive, and any suggestion to the contrary in the trial court's memorandum of decision is merely the product of a confusion of the issues.

In making a determination of suggestiveness, we look to whether the allegedly suggestive aspect of the identification procedure unnecessarily emphasized the defendant's photograph or otherwise indicated which individual the police considered a suspect. See, e.g., State v. Gold, supra, 180 Conn. at 656, 431 A.2d 501. This is not an inquiry into the potential in-court prejudice that a defendant may suffer because of police conduct or other events occurring subsequent to the witness' selection but, rather, a determination of whether the selection itself is suspect because it may be the product of the suggestive conduct or display. Moreover, although comments such as Beaudin's may not be a "harmless irrelevancy," we must be careful to pinpoint the potential harm and identify why that harm might be relevant. Only then can a court determine the appropriate remedy. To be clear, the purportedly prejudicial effect created by Beaudin's comment that the trial court identified was the potentially reduced efficacy of defense counsel's cross-examination of Clement because of his allegedly heightened sense of certitude and not the inherent suggestiveness of the procedure leading to Clement's selection of the defendant's photograph from the array in the first place. Thus, we conclude that the trial court correctly determined that Beaudin's remark had no bearing on the prior photographic identification. To the extent that Justice Palmer feels bound by some of the trial court's more equivocal statements that seemingly conflate these issues, however, we simply note that this court is not obliged to perpetuate a perceived error: "Our oath is to do justice, not to perpetuate error." (Internal quotation marks omitted.) Mendillo v. Board of Education, 246 Conn. 456, 507, 717 A.2d 1177 (1998).

[35] Justice Palmer not only fails to recognize the significance of this explicit finding but appears to ignore it entirely when he declares in his concurrence that "the trial court justifiably predicated its finding of unnecessary suggestiveness on [Beaudin's comment]." It is patently contradictory to claim that the trial court "predicated its finding" of suggestiveness on an impropriety that it found did not taint the witness' identification at the time it was made. The trial court actually predicated its finding of unnecessary suggestiveness on "the unchallenged findings of the scientific research studies ... that the simultaneous showing of all photo[graphs] to each witness on a single photo[graphic] [array] created an unnecessary risk of producing irreparable misidentifications...." Moreover, although the court expressed a concern on the basis of "the undisputed judgment and recommendation of well respected scientific researchers that the practice of having persons with special knowledge of a criminal investigation administer identification procedures" was undesirable because the procedure might be biased and the subsequent identification tainted, the court found that, in this case, the identification was not tainted. Furthermore, the trial court also examined the photographic array itself and determined that it "did not unfairly highlight the defendant or promote his identification by the witnesses in any way." Although Justice Palmer decries "the majority's insistence on treating Beaudin's comment as entirely separate and distinct from the identification procedure," he fails to recognize that the basis for our treatment of Beaudin's statement is the trial court's determination that the comment did not infect Clement's initial selection of the defendant from the photographic array. It is this initial identification, and the procedure used to elicit that identification, with which our suggestiveness analysis is concerned. Police conduct occurring after the witness' definitive selection of a defendant's photograph from an array is more appropriately dealt with in cross-examination, where the credibility of the initial identification and the reliability of the witness' subsequent recollection of his level of confidence in that identification can be tested in the crucible of jury scrutiny.

[36] The first case in which we used this language to describe the test for the constitutionality of pretrial identification procedures appears to be State v. Theriault, supra, 182 Conn. at 371-72, 438 A.2d 432, which we decided in 1980.

[37] Valle testified as follows:

"Q. Okay. And on that occasion, what, if anything, did [Detective Beaudin] describe for you to do in terms of looking at photographs?

"A. I don't really remember. The first set [of photographs] they laid down, I already knew who it was."

[38] In Justice Katz' concurrence, Justice Katz lists "many ways that a lineup administrator can influence an eyewitness' identification decision"; (internal quotation marks omitted) footnote 9 of Justice Katz' concurrence; and declares that "the research submitted by the defendant universally indicates that the use of an interested administrator ... contaminates the process, even when the individual makes no conscious effort to influence the identification...." (Citation omitted.) Footnote 5 of Justice Katz' concurrence. Making the same mistake as the trial court, Justice Katz fails to refer us to any evidence that such "contamina[tion]" actually occurred in this case. It is just this type of generalizing that Ledbetter cautions against and that leads to unnecessary per se rules.

[39] We strongly discourage such police commentary, especially if it made prior to the witness' identification, because it could, under some circumstances, taint the identification process to such an extent that any subsequent identification would be subject to suppression as the product of an unnecessarily suggestive procedure. That is not the case here, however.

[1] A simultaneous photographic array displays multiple photographs together on a single page to be shown to a witness, in contrast to a sequential identification procedure in which individual photographs are displayed to a witness one at a time.

[2] I use the term interested administrator to indicate an individual who knows the identity of the suspect in a lineup or photographic array and the term blind or disinterested administrator to indicate an individual who does not possess that knowledge. See footnote 4 of this concurring opinion.

[3] The court also discussed, but ultimately rejected, the defendant's claim that his photograph had been highlighted unfairly in the photographic array. It examined the features of the photographs, with particular focus on the details of the background of the photographs, finding that all of the photographs therein were consistent with each other and with the descriptions previously provided by the witnesses.

[4] The research generally refers to the individual responsible for conducting an identification procedure as a "lineup administrator" or an "administrator," irrespective of whether the procedure involves a photographic array, a live group of persons presented together, or individuals presented one at a time.

[5] As I explain further in footnote 8 of this concurring opinion, the research submitted by the defendant universally indicates that the use of an interested administrator; see footnote 2 of this concurring opinion; contaminates the process, even when the individual makes no conscious effort to influence the identification, substantially increasing the likelihood of misidentification.

[6] As I explain in more detail in footnote 11 of this concurring opinion, the scientific studies indicate that, as a general matter, sequential identification procedures are significantly more accurate than simultaneous identification procedures.

[7] As we noted in State v. Ledbetter, supra, 275 Conn. at 572, 881 A.2d 290, to which the trial court in the present case referred in its analysis, the relative judgment process allows a witness to compare each lineup participant to his memory of the perpetrator and determine which lineup participant most closely resembles the perpetrator. We stated then that "the relative judgment process exerts a significant influence in eyewitness identifications.... The problem with the relative judgment process ... is that it includes no mechanism for deciding that the culprit is none of the people in the lineup." (Citation omitted; internal quotation marks omitted.) Id. Consequently, in lineups that do not contain the actual perpetrator, witnesses invariably make a mistaken identification, choosing some participant in the lineup rather than realizing that the perpetrator is not present. See id., at 572-73, 881 A.2d 290 (noting that in experiments, when witnesses were shown lineup containing actual perpetrator, 92 percent of witnesses correctly identified perpetrator, but when lineup did not contain actual perpetrator, 100 percent of witnesses identified someone other than actual perpetrator).

[8] See, e.g., G. Wells & E. Olson, "Eyewitness Testimony," 54 Ann. Rev. Psychol. 277, 289 (2003) (summarizing research studies that concluded interested lineup administrators communicated their knowledge about suspects to witnesses, influencing outcome of identification); G. Wells, M. Small & S. Penrod et al., "Eyewitness Identification Procedures: Recommendations for Lineup and Photospreads," 22 Law & Hum. Behav. 603, 628 (1998) ("we know from experiments that a photospread administrator's behaviors ... can lead eyewitnesses to falsely identify that person as the culprit"); Technical Working Group for Eyewitness Evidence, United States Dept. of Justice, "Eyewitness Evidence: A Guide for Law Enforcement" (October, 1999) p. 9. ("investigator's unintentional cues [e.g., body language, tone of voice] may negatively impact the reliability of eyewitness evidence").

[9] "There are many ways that a lineup administrator can influence an eyewitness' identification decision. For instance, the eyewitness might call out the number of a filler photo[graph], and the lineup administrator, knowing that the photo[graph] is a mere filler, might urge the witness to make sure she has looked at all the photo[graphs] before making a decision.... In contrast, the mere utterance of the number of the suspect's photo[graph] could yield a very different reaction from the lineup administrator, such as ``Good, tell me what you remember about that guy.'... Even without speaking, a lineup administrator can influence an eyewitness through facial expressions and body movements such as head nodding or head shaking.... These discretionary behaviors by the lineup administrator are not necessarily intentional and the lineup administrator might not even be aware that she or he is doing it. Instead, these are natural behaviors that testers display when they think that they know the correct answer or have expectations about how the tested person will or should behave." G. Wells & D. Quinlivan, "Suggestive Eyewitness Identification Procedures and the Supreme Court's Reliability Test in Light of Eyewitness Science: 30 Years Later," 33 Law & Hum. Behav. 1, 8 (2009).

[10] For example, in a two person police department, although one officer might serve as a disinterested administrator in the first lineup, once that officer observes which individual a witness chooses, that officer is no longer blind for the purpose of subsequent lineups. See, e.g., Technical Working Group for Eyewitness Evidence, United States Dept. of Justice, "Eyewitness Evidence: A Guide for Law Enforcement" (October, 1999) p. 9 ("[b]lind procedures ... may be impractical for some jurisdictions to implement"); S. Mecklenburg, "Report to the Legislature of the State of Illinois: The Illinois Pilot Program on Sequential Double-Blind Identification Procedures" (March 17, 2006) p. 63 ("[t]he advantages of blind administrators as a recommended practice should be further explored and compared to optimal instructions and technological procedures to prevent feedback"). In such circumstances, however, it may be possible to employ technology-based tools, such as computer software, to conduct the identification procedures, at least in the case of photographic arrays. See, e.g., O. MacLin, L. Zimmerman & R. Malpass, "PC—Eyewitness and the Sequential Superiority Effect: Computer-Based Lineup Administration," 29 Law & Hum. Behav. 303 (2005).

[11] See, e.g., G. Wells, M. Small & S. Penrod et al., "Eyewitness Identification Procedures: Recommendations for Lineups and Photospreads," 22 Law & Hum. Behav. 603, 639 (1998) ("[w]hen compared to the usual simultaneous procedure, it is clear that the sequential procedure produces a lower rate of mistaken identifications [in perpetrator-absent lineups] with little loss in the rate of accurate identifications [in perpetrator-present lineups]"); J. Turtle, R. Lindsay & G. Wells, "Best Practice Recommendations for Eyewitness Evidence Procedures: New Ideas for the Oldest Way to Solve a Case," Can. J. of Police and Security Services (March, 2003), pp. 22-23 ("A rapidly growing body of research indicates that sequential line-up presentation makes it extremely difficult to use a relative judgment strategy. As a result, false identifications by eyewitnesses occur at a dramatically reduced rate while, fortunately, the rate of accurate identifications is not reduced significantly."); Technical Working Group for Eyewitness Evidence, United States Dept. of Justice, "Eyewitness Evidence: A Guide for Law Enforcement" (October, 1999) p. 9 ("[s]cientific research indicates that identification procedures such as lineups and photo[graphic] arrays produce more reliable evidence when the individual lineup members or photographs are shown to the witness sequentially—one at a time—rather than simultaneously").

[12] See, e.g., G. Wells & E. Olson, "Eyewitness Testimony," 54 Ann. Rev. Psychol. 277, 288-89 (February, 2003) ("A recent meta-analysis of [twenty-five] studies comparing simultaneous and sequential lineups showed that the sequential lineup reduced the chances of mistaken identifications in culprit-absent lineups by nearly one half.... Unfortunately, the sequential technique was also associated with a reduction in accurate identification rates in culprit-present lineups." [Citations omitted.]); S. Mecklenburg, "Report to the Legislature of the State of Illinois: The Illinois Pilot Program on Sequential Double-Blind Identification Procedures" (March 17, 2006) pp. i-ii (Evaluating a sequential and double-blind identification procedure against a simultaneous, nonblind identification procedure and concluding: "[T]he sequential, double-blind method results in a loss of accurate identifications when compared to the simultaneous method.... [T]here are five categories in which the sequential, double-blind method may not be superior to the simultaneous procedure."); Technical Working Group for Eyewitness Evidence, United States Dept. of Justice, "Eyewitness Evidence: A Guide for Law Enforcement" (October, 1999) p. 9 ("there is not a consensus on any particular method or methods of sequential presentation that can be recommended as a preferred procedure" [emphasis in original]).

[13] See, e.g., Office of the Attorney General, State of Wisconsin, "Response to Chicago Report on Eyewitness Identification Procedures" (July 21, 2006) p. 4 (noting that, because "the [study in the Illinois Report] did not separately compare double-blind simultaneous to double-blind sequential [it instead attempts to compare non-blind simultaneous to double-blind sequential], there is no way to ascertain whether the different rates of selection of suspects and fillers in the two tested conditions were the result of the difference between double-blind and non-blind procedures, or between sequential and simultaneous procedures"); D. Schacter, R. Dawes & L. Jacoby et al., "Policy Forum: Studying Eyewitness Investigations in the Field," 32 Law & Hum. Behav. 3, 5 (2008) ("[i]f the [study in the Illinois Report] was not designed to address the question of what happens in a blind/simultaneous line-up, given its centrality to the issue, then our assessment is that the Illinois study addressed a question ... that is not worth addressing, because the results do not inform everyday practice in a useful manner").

In February, 2008, the academic journal Law and Human Behavior, recognizing the impact that the Illinois Report was having on eyewitness identification research, published commentaries on that study. An introduction article to those commentaries summarized the findings, noting that each of the commentaries concluded that additional field research on eyewitness identification procedures is necessary, and several of the papers made specific recommendations to improve the methodology used in future research. B. Cutler & M. Kovera, "Introduction to Commentaries on the Illinois Pilot Study of Lineup Reforms," 32 Law. & Hum. Behav. 1, 1-2 (2008). The most common criticism of the study was that the methodology of the study was inadequate to support the conclusions reached because it was impossible to determine which variable caused the results. See S. Ross & R. Malpass, "Moving Forward: Response to ``Studying Eyewitness Investigations in the Field,'" 32 Law & Hum. Behav. 16, 19 (2008) ("[t]reating the [study in the Illinois Report] as an attempt to answer the scientific questions in the [simultaneous] versus [sequential] controversy is erroneous, inappropriate and fruitless"); G. Wells, "Field Experiments on Eyewitness Identification: Towards a Better Understanding of Pitfalls and Prospects," 32 Law & Hum. Behav. 6, 7 (2008) ("we cannot be certain whether the results ... are attributable to the sequential versus simultaneous difference or to the double-blind versus non-blind difference"); D. Schacter, R. Dawes & L. Jacoby et al., supra, 32 Law & Hum. Behav. at 4 ("[T]he [methodology] has devastating consequences for assessing the real-world implications of this particular study [in the Illinois Report].... [I]t is critical to determine whether the seemingly better result from the simultaneous procedure is attributable to properties of the simultaneous procedure itself, or to the influence of the non-blind administrator."). One commentary further discussed the effects of interested administrators on identification procedures, noting that "testers influence the person they test in ways that are consistent with the testers' expectations, assumptions, hopes, and so on"; G. Wells, supra, at 8; and provided additional evidence that feedback from an administrator at the time of the identification procedure alters the witness' perceived confidence levels as well as his actual memory of events. Id., at 10.

[14] The defendant does not challenge explicitly the factual findings of the trial court in his brief, asserting only that "[u]pon plenary review, the overall poor lighting and not seeing the perpetrators' faces very well should cause these two [reliability] factors to be given light weight." To the extent that he contends that the findings are clearly erroneous, they are supported adequately by the record. Valle testified that the hallway lighting was "bright as hell," and that, although the living room light was not turned on, light from adjacent rooms illuminated the living room so that "you could see pretty good."

[1] As I explain more fully hereinafter, I disagree with the majority's conclusion that the trial court found that Beaudin's comment to Clement occurred after the identification procedure had concluded and, therefore, that the comment was not a part of that procedure.

[2] The following two part test governs our review of the defendant's due process claim. "In determining whether identification procedures violate a defendant's due process rights, [t]he required inquiry is made on an ad hoc basis and is two-pronged: first, it must be determined whether the identification procedure was unnecessarily suggestive; and second, if it is found to have been so, it must be determined whether the identification was nevertheless reliable based on an examination of the totality of the circumstances.... The defendant bears the burden of proving both that the identification procedures were unnecessarily suggestive and that the resulting identification was unreliable.... Generally, [t]he exclusion of evidence from the jury is ... a drastic sanction, one that is limited to identification testimony which is manifestly suspect." (Internal quotation marks omitted.) State v. Ortiz, 252 Conn. 533, 553, 747 A.2d 487 (2000).

[3] The police used the same essential identification procedures for each of the two eyewitnesses, Clement and Christopher Valle. For ease of reference, I discuss those procedures insofar as the police employed them in obtaining Clement's identification of the defendant as the perpetrator. My ultimate conclusion that the defendant's due process rights were not violated, however, applies equally to Valle's identification of the defendant.

[4] As Justice Katz explains in her concurrence, "[a] simultaneous photographic array displays multiple photographs together on a single page to be shown to a witness, in contrast to a sequential identification procedure in which individual photographs are displayed to a witness one at a time." Footnote 1 of Justice Katz' concurrence.

[5] The majority explains the double-blind procedure as follows: "To qualify as double-blind, a photographic array must be administered by an uninterested party without knowledge of which photograph represents the suspect."

[6] In Ledbetter, this court approved of the use of a nonsequential identification procedure even though it was not accompanied by a warning to the witness that the array may or may not contain the suspected perpetrator. See State v. Ledbetter, supra, 275 Conn. at 574-75, 881 A.2d 290. Concluding, however, that "some action is necessary to mitigate the risks of such procedures"; id., at 575, 881 A.2d 290; we deemed it appropriate "to exercise our supervisory authority to require an instruction to the jury in those cases [in which] the identification procedure administrator fails to provide such a warning, unless no significant risk of misidentification exists." Id. That instruction includes language alerting the jury to the fact that the failure of the administrator to advise the witness that the suspected perpetrator may or may not be in the array may increase the likelihood that the witness will select one of the individuals in the array even when the suspect is not in the array. Id., at 579, 881 A.2d 290.

[7] This warning appeared on the photographic array itself. As I explained in footnote 6 of this opinion, the nonsequential identification procedure at issue in Ledbetter included no such warning.

[8] Our opinion in Ledbetter was decided only a few months prior to the trial court's ruling on the defendant's motion to suppress the photographic identifications in the present case. As I noted previously, the procedure that we approved of in Ledbetter was manifestly more suggestive than the procedure that the police used in the present case because, in Ledbetter, the administrator failed to warn the witness that the suspect may or may not be included in the procedure. See State v. Ledbetter, supra, 275 Conn. at 570 n. 23, 881 A.2d 290. Although I agree with Justice Katz that, with certain exceptions, sequential identification procedures are superior to nonsequential identification procedures, I do not believe that the scientific research in this area is sufficiently clear or compelling that we now are required to revoke our approval of the kind of nonsequential procedure employed in Ledbetter and in the present case, especially in view of the fact that, in the present case, the witnesses were fully advised that they should not assume that the suspect's photograph was included in the array, that they should feel no obligation to identify anyone and that it is no less important to clear innocent persons of suspicion than it is to identify guilty persons.

[9] Of course, there may be times when the use of such a procedure is impossible or impracticable in light of the circumstances.

[10] Although the majority finds that Clement "felt fairly confident" about his choice, the trial court noted that Clement had testified that "the sole basis for his identification of the defendant as the gunman was his own ``gut feeling,' based [on] his personal observations of the gunman during the robbery." Clement also testified that the police never had told him that he had to identify someone in the photographic array and never had indicated to him that he should give special consideration to any particular photograph. Nevertheless, Clement did acknowledge that he believed that the police would not have shown him the photographic array unless it contained a suspect's photograph and, further, that he felt he "had to pick somebody"....

[11] The trial court then proceeded to elaborate as to why Clement's photographic identification of the defendant was reliable under the totality of the circumstances.

[12] Despite the absolute clarity of the trial court's determination that Beaudin's comment was apart of the pretrial identification procedure, the majority asserts that the record demonstrates precisely the opposite. Footnote 34 of the majority opinion. The majority's explanation for its assertion reveals that there simply is nothing in the record to substantiate it. Indeed, in reaching its unfounded conclusion, the majority ignores the relevant portions of the record that I have identified and hypothesizes that "any suggestion" in the decision of the trial court that the court was treating Beaudin's comment as part of the pretrial identification procedure was "merely the product of a confusion of the issues [by the trial court]." Id. This latter assertion, along with the majority's reference to "the trial court's more equivocal statements that seemingly conflate [the relevant] issues"; id.; is wholly unwarranted. The trial court's thirty-four page memorandum of decision on the defendant's motion to suppress is characteristically scholarly, thoughtful and thorough, and most definitely not the product of any confusion. The fact is that, in seeking to justify its own faulty analysis, the majority, which first describes the court's decision as "``clearly'" reaching one conclusion and then, when it suits its purpose to do so, reverses course and characterizes the trial court's statements as "equivocal" and "the product of ... confusion"; id.; unfairly maligns the perfectly logical analytical approach taken by the trial court.

The majority's fundamental misunderstanding of the trial court's decision is exemplified by the majority's mistaken reading of that decision as containing a finding by the trial court that the prejudice attributable to Beaudin's comment did not affect Clement's pretrial identification of the defendant. In support of this conclusion, the majority specifically relies on the following language from the trial court's decision: "What [Beaudin] risked by her conduct ... was unfairly bolstering [Clement's] confidence in the strength of his photo[graphic] identification and the solidity of his basis in memory for it, thus making it harder for the defense to test the true certainty with which he made that identification on cross-examination, and correspondingly more difficult for the jury to assess the true strength and reliability of that identification...." On the basis of this language, the majority asserts: "This potential harm clearly refers to the prejudicial effect that Beaudin's comment might have on Clement's subsequent in-court identification of the defendant," and not on "Clement's later memory of the [pretrial identification] procedure...." Footnote 34 of the majority opinion. The majority's conclusion is manifestly and demonstrably wrong, as it is readily apparent from the very language of the trial court's decision—on which the majority itself relies—that the trial court was concerned about the defendant's memory of and confidence in his pretrial identification of the defendant. In fact, the language of the trial court's decision could not be clearer in this regard. The trial court determined that Beaudin's comment created an undue risk of "unfairly bolstering [Clement's] confidence in the strength of his photo[graphic] identification and the solidity of his basis in memory for [that identification]," thereby making it more difficult "for the defense to test the true certainty with which he made that identification on cross-examination...." (Emphasis added.) In other words, Beaudin's comment gave rise to an unfair risk of affecting Clement's trial testimony about his pretrial identification of the defendant because that comment likely would have a distorting effect on his memory of that out-of-court identification of the defendant. There is absolutely nothing in the language on which the majority relies to suggest that the trial court was referring to the prejudicial effect that Beaudin's comment was likely to have on any in-court identification of the defendant; rather, it is perfectly clear from the language of the trial court that the court was referring to the effect that the comment would have on Clement's in-court trial testimony about his pretrial, out-of-court identification of the defendant.

[13] The court in Manson considered these factors in deciding whether to adopt a per se test pursuant to which a court would be required to exclude identification evidence that had been obtained by use of unnecessarily suggestive procedures or, alternatively, a totality of the circumstances test pursuant to which the use of an unnecessarily suggestive identification procedure by the police would not require suppression of the identification unless that identification was unreliable upon consideration of all of the relevant circumstances. Manson v. Brathwaite, supra, 432 U.S. at 109-13, 97 S. Ct. 2243. The court ultimately concluded that, because "reliability is the linchpin in determining the admissibility of identification testimony"; id., at 114, 97 S. Ct. 2243; the proper test requires an evaluation of the totality of the circumstances. Id., at 113-14, 97 S. Ct. 2243.

[14] I also note that one of the factors that a court must consider in determining whether the identification is sufficiently reliable to be admissible under the totality of the circumstances—the second prong of the two part test that must be applied following a determination that the identification procedure was unnecessarily suggestive—is the "level of certainty [that the witness] demonstrated" in connection with his identification of the suspect. Manson v. Brathwaite, supra, 432 U.S. at 114, 97 S. Ct. 2243; see also id., at 114-16, 97 S. Ct. 2243 (identifying five factors to be considered in evaluating reliability of identification under totality of circumstances, including witness' degree of certainty in identification). Under the test adopted by the majority, however, the fact that the witness' confidence in his identification is likely to have been affected significantly by virtue of the kind of comment that Beaudin had made in the present case is insufficient even to move the court's inquiry forward to a consideration of the totality of the circumstances. I do not believe that this approach is consistent with the test for determining the reliability of the identification.

[15] Indeed, suppression will be required only if such a comment, when considered with any other police impropriety that occurs during the identification procedure, renders the identification manifestly suspect under the totality of the circumstances. Although that eventuality is not likely, if it occurs, I see no reason why suppression would be inappropriate or unwarranted. I therefore disagree with the majority's assertion that a comment such as the one that Beaudin made can never be more than grist for cross-examination. In the overwhelming number of cases—the present one included—that is exactly what it ultimately will be. But when such a comment, in combination with some other unnecessarily suggestive police conduct that occurs at the identification procedure itself, results in an identification that is unreliable under the totality of the circumstances, the state should be precluded from using that identification at trial.

[16] Because I conclude that Valle's pretrial identification of the defendant also was reliable, I do not address the issue of whether the failure of the police to use a double-blind identification procedure with respect to Valle rendered that procedure unnecessarily suggestive.

[17] In contrast to Justice Katz, I do not read the majority opinion as signaling a contrary view.

[1] When the court speaks in multiple voices rather than a single voice, the usefulness of the decision, apart from deciding the particular case, is limited. Even though all of the opinions in this case are in accord as to affirming the trial court's decision, each opinion reaches the result by a different route.

[2] Although the studies are mixed as to the benefits of simultaneous and sequential procedures under varying circumstances, there is no research literature that supports the use of an interested administrator. No matter how expedient it may be for law enforcement agencies to use interested investigators in this role, that practice imperils the integrity of the identifications.

[3] In any event, I disagree with Justice Katz' conclusion that the trial court did not abuse its discretion.

[4] I disagree with Justice Palmer's conclusion in his concurring opinion that the detective's remark rendered the procedure unnecessarily suggestive. When the detective made the remark, Clement already had identified the defendant, without any information as to the result of the other witness' identification. Although it is possible that the detective's remark may have bolstered Clement's confidence in the accuracy of his identification of the defendant during testimony at trial, that possibility properly would be raised during cross-examination of Clement, and used to impeach his testimony, rather than to render the identification inadmissible. As the majority opinion notes, the rationale offered to support that conclusion has the effect of improperly shifting the focus of attention from the impact of police conduct on the pretrial identification to the impact of the pretrial identification on the in-court identification. Moreover, it unnecessarily causes potential confusion about the boundary of the police identification procedure, a factor that is not an issue in this case.

[5] The trial court also stated in its memorandum of decision that "[p]olice personnel conducting photographic identifications should henceforth strive to eliminate the danger of misidentification arising from the simultaneous showing of multiple photographs by making all such showing sequentially." The majority opinion characterizes this statement as a directive. My reading of the statement, however, is that, while it could be read to issue a directive to police personnel, it need not be accorded that meaning. It is possible to interpret this statement as the court merely indicating its opinion as to what procedures would be most prudent for police personnel to employ. To the extent, however, that the statement could be read to issue a directive to police personnel, it would, of course, exceed the authority of the trial court.

[6] The Ledbetter opinion has been criticized for failing to give appropriate effect to the evidence of scientific studies that question the reliability of eyewitness identification. C. TerBeek, "A Call for Precedential Heads: Why the Supreme Court's Eyewitness Identification Jurisprudence is Anachronistic and Out-Of-Step with the Empirical Reality," 31 Law & Psychol. Rev. 21, 49-50 (2007).