Decision & Order The Court conducted pretrial hearings in this matter on April 16, 19, and August 21, 2024, pursuant to Dunaway v. New York, 441 U.S. 200 (1979), United States v. Wade, 388 U.S. 218 1967), and People v. Huntley, 62 N.Y.2d 134 (1984). For the reasons stated herein, Brown’s motion to suppress his identification in two photo array procedures is granted because the photo array was unduly suggestive. Brown’s pretrial motion to suppress pursuant to Huntley is denied. The Dunaway prong of the motion is held in abeyance. An independent source hearing with complainant Argylis Oriach is ordered before trial. FINDINGS OF FACT The Court heard the testimony from three witnesses: Detective Marc Devito, Detective William Sciara, and complainant Fabrizio Ellis. The hearing testimony is summarized, in pertinent part, as follows: Det. Devito has been employed by the NYPD for 12 years, including six years at the Brooklyn Robbery Squad investigating pattern robberies. He has participated in over 300 robbery investigations, and he was previously with the Queens North Anti-Crime Unit as a plainclothes officer. He was the lead investigator on the two robberies involved in the present case. The first robbery involved complainant Ellis, on January 28, 2022, within the confines of Brooklyn’s 76th Precinct. The second robbery involved compliant Oriach, on March 2, 2022, within confines of Brooklyn’s 79th Precinct. Det. Devito was assigned the January 28th robbery on March 2, 2022. See Tr. Apr. 16, 2024, at 25:4-6. On March 13, 2022, Det. Devito interviewed Ellis via cell phone, after reviewing Ellis’ complaint report, his 911 call, snapshots of Venmo transactions, and other materials. Id. at 25:7-26:1. According to Det. Devito, Ellis told him that he was walking on the street when he was approached by a “young looking, light skinned male, wearing a black jacket, light blue G-Star jeans, [and] white sneakers.” Id. at 26:2-8 (emphasis added). The individual grabbed Ellis by the shoulder, told him that he had a gun, and demanded Ellis’s wallet and that Ellis unlock his cell phone. Id. Ellis, who is white, described to Det. Devito that his assailant’s race was black, and that Ellis would be unable to make any identification because the person had a black mask on. See id. 56:18-57:15 (enumerating a description of black, young, thin build, black jacket, light blue jeans, white sneakers, and black mask); 68:1-4, 19-24; and 72:5-8. Ellis also provided Det. Devito with screenshots of unauthorized Venmo transactions on Ellis’ bank account showing cash transfers and recipient information, including the name “Kamal,” and an entity called Queens Nails with an address of 732 Bristol Street in Brooklyn. See Tr. Apr. 16, 2024, at 26:9-16; 27:17-27:1. According to Det. Devito, NYPD records linked Brown to that address, making him a person of interest in the investigation. Id. at 28:7-20; 28:21-29:6. On March 4, 2022, Det. Devito was notified by the 76th Precinct that Brown had been arrested on unrelated matter. See Tr. Apr. 16, 2024 at 32:14-33:1. Det. Devito, along with a Det. Napolitano, interviewed Brown that afternoon at the 76th Precinct Detective Squad. See id. at 33:2-34:3. As borne out by a video recording of the interview, Brown was read his Miranda rights, agreed to answer questions, and gave pedigree information, but denied involvement in any robbery. See id. at 34:4-37:21. On March 17, 2022, Det. Devito was assigned the March 2nd robbery of complainant Oriach. See Tr. Apr. 16, 2024, at 29:21-23. On March 21, 2022, he interviewed Oriach, who again described being “approached by a light-skinned male,” approximately 20-25 years old, thin build, with a black jacket with hood up, black pants, and a black mask. Id. at 29:24-30:15; 67:13-25. 69:12-14, 20-22 (emphasis added). The assailant said to Oriach, in sum and substance, “I have a gun. Give me your phone and unlock it.” Id. at 30:10-15. Oriach also said that he felt a hard object in the man’s pocket, which Oriach thought was a gun, and that there had been unauthorized transfers on Oriach’s bank account, including to a payee named “Kamal Brown.” Id. at 30:14-21 and 31:7-23. Based on this information, where both complainants gave similar descriptions of a black, thin, young, “light-skinned male,” Det. Devito created a photo array. Tr. Apr. 16, 2024 at 38:11-18; compare id. at 26:2-8 and 30:10-15. Using the NYPD Photo Management System, Det. Devito started with a mugshot of Brown. Id. As for the array fillers, Det. Devito testified that the photo management system has a “similar image search” inquiry to “generate similar-looking mugshots” for fillers based on the selected suspect photograph. Id. at 9:24-11:7; see id. at 81:1-82:8. However, Det. Devito did not use the “similar image search” function. He also did not use the photo management system’s skin complexion filter. Instead, Det. Devito searched for fillers by selecting filters for “male, black, approximately 19 to 23 years old, dark brown hair,” id. at 38:16-23, “[b]ecause [he] felt those looked most close to what Kamal Brown looks like.” Id. at 39:1-8. As for the skin complexion filter, Det. Devito initially stated that he did not know whether the NYPD photo management system had such a feature. “I am not sure. There are a number of filters. It is very possible that there is one.” Tr. Apr. 16 at 82:9-11. Later, the People conceded that the system does have a skin complexion filter, but that Det. Devito did not use it. Tr. Aug. 19 at 15:21-24; People Ex. 17 (photo manager skin filters include, inter alia, Dark Brown, Light Brown, Fair, and Pale); see also Tr. Aug. 19 at 11:21-12:1 (Det. Devito testimony about the skin complexion filter). Having conceded the existence of a skin complexion filter, Det. Devito stated that he did not use it because “[complexion] can be filtered out as I am looking through the [approximately 9,000] photos that populated” from his search query for fillers. Tr. Apr. 16 at 70:16-22. Also, as opposed to being “not sure” whether system had such of filter, Det. Devito now stated that the filter is “not 100 percent reliable.” Tr. Aug. 19 at 14:6-11. “You can add as many filters as you want; it doesn’t necessarily mean that the photos that are then generated are automatically going to be suitable to the subject, the subject’s photo.” Tr. Aug. 19 at 14:6-11. Rather than using either the similar image search function, or the skin complexion filter, Det. Devito “picked which ones I felt looked closest to Kamal Brown.” Tr. Apr. 16 at 78:8-12; see also id. at 82:12-17 (“[T]hey matched what Kamal Brown’s picture looked like most. I wanted to generate the most possibilities possible so that I had enough options to pick from to try and make — to try to use the images that looked most like Kamal Brown.”) Based on Det. Devito’s search, the photo management system generated approximately 9,000 possible fillers. Id. at 38:24-25. According to Det. Devito, he did not use any additional filters because it would narrow the pool of potential fillers. Tr. Aug. 19 at 12:19-13:7; 13:15-20. Det. Devito then hand selected five fillers from the approximately 9,000 photographs, although could not recall how many photographs he reviewed, or how long it took to assemble the array. Id. at 39:9-10; Tr. Aug. 19 at 20:12-21:14. The photo management system randomly assigned Brown to position 3 in the array, on the top right, as well as the location of the fillers. See Id. at 39:11-13; 14:23-15:4. The following is Det. Devito’s photo array: People Ex. 2. On April 6, 2022, Det. Devito interviewed Oriach again to ask how much of the mask covered the perpetrator’s face. See Tr. Apr. 16, 2024 at 80:3-81:5. During that interview, Oriach indicated that he “possibly could” make an identification. Id. at 79:22-80:2; 81:13-21. That same day, in a double-blind identification procedure conducted by Det. Sciara, Oriach picked out Brown in the array. See Tr. Apr. 16, 2024 at 39:16-41:21. After learning the results, Det. Devito prepared a probable cause I-Card for Brown’ arrest. See id. at 43:14-19. On April 11, 2022, Devito also interviewed Ellis again about the placement of the perpetrator’s mask. Tr. Apr. 16 at 81:22-24. According to Ellis, based on that conversation, “[i]t made me feel that I can do an ID.” Id. at 18:11-17. Later that day, Ellis identified Brown in the same photo array, in a double-blind procedure conducted by Police Officer Rumph of the Brooklyn Special Victims Unit. Tr. Apr. 16, 2024 at 43:20-45:6. Ellis said, “[a]t first, I was between a couple of suspects that had a similar feature and, eventually, I wanted to mimic the mask so I put my hand over the mouths to basically simulate that.” Tr. Apr. 19 at 22:19-23. Ellis said that “it was between two people…one and three…and eventually I was leaning in the direction of number three.” Id. at 23:15-20. In his confidence statement, Ellis indicated, “I am in between two suspects, about halfway,” that he was “halfway confident that [he] selected correctly.” Id. at 41:8-22. On April 12, 2022, police arrested Brown in Queens on an unrelated matter. On April 14, 2022, after being notified about Brown’s Queens arrest, Det. Devito arrested Brown on the current case. Id. at 49:9-50:8. Conclusions of Law In a motion to suppress evidence, “the People have the burden of going forward to show the legality of the police conduct in the first instance.” People v. Dunbar, 188 A.D.3d 1247 (App. Div. 2d Dept. 2020) (quoting People v. Whitehurst, 25 N.Y.2d 389 [1969]); see People v. Berrios, 28 N.Y.2d 361 (1971). Once that burden has been met, the burden of persuasion is on the defendant to prove, by a preponderance of credible evidence, that the police conduct was unlawful. People v. Thomas, 291 A.D.2d 462, 463 (App. Div. 2d Dept. 2002); see People v. Berrios, 28 N.Y.2d 361 (1971); People v. Baldwin, 25 N.Y.2d 66 (1969). The primary issue raised during the pre-trial hearings is whether the photo array utilized to identify Brown was unduly suggestive. Unduly suggestive pretrial identifications “violate due process and therefore are not admissible to determine the guilt or innocence of an accused.” People v. Chipp, 75 N.Y.2d 327, 335 (1990); see also Wade, 388 U.S. 218 at 235. To determine whether a pretrial identification procedure is unduly suggestive, courts must engage in a long-standing burden-shifting analysis. “While the People have the initial burden of going forward to establish…the lack of any undue suggestiveness in a pretrial identification procedure, it is the defendant who bears the ultimate burden of proving that the procedure was unduly suggestive.” Chipp, 75 N.Y.2d at 335; see also People v. Marryshow, 162 A.D.3d 1313, 1314 (App. Div. 3d Dept. 2018). Brown contends that the photo array used in this matter is unduly suggestive because he is the only individual pictured, out of all six images, who appears young, thin, and with a light complexion. A review of the photo array shows that Brown, in slot number 3, is significantly lighter in complexion than any of the other five filler photos. Fillers 1 and 2 appear young and thin, but they clearly have dark complexions. Fillers 4 and 5 have a camera flash in the center of their faces, but they have dark complexions as well, particularly on their cheeks and necks, and appear notably older and heavier than Brown. Filler 6 is young and thin, but he also has a dark complexion. All men appear to be clothed similarly. None are wearing masks. None have any hoods on or any head coverings. They also have different hair lengths. The Court of Appeals has said that “whether a lineup is unduly suggestive is a mixed question of law and fact.” People v. Perkins, 28 N.Y.3d 432, 436 (2016). The law does not require that the fillers’ photos be identical to that of the accused. See e.g. People v. Starks, 91 A.D.3d 975 (App. Div. 2d Dept. 2012); People v. Howard, 50 A.D.3d 823 (App. Div. 2d Dept. 2008). However, “a photographic display is suggestive where some characteristic of one picture draws the viewer’s attention to it, indicating that the police have made a particular selection.” People v. Robert, 184 A.D.2d 597, 598 (App. Div. 2d Dept. 1992). “[A] witness’s prior description is but one factor a court should consider” in its determination. Id. at 437 (holding that the defendant’s dreadlocks were so distinctive compared to the rest of the “fillers” that they rendered a lineup unduly suggestive, especially because two complainants had described dreadlocks in their initial descriptions of the perpetrator to police); but see People v. Mata, 174 A.D.3d 647, 647 (App. Div. 2d Dept. 2019) (finding that the defendant’s ponytail did not render a lineup unduly suggestive, as “he was positioned in a way that obscured it”). Other factors to be considered are skin tone and apparent age. See People v. Miller, 199 A.D.2d 422, 423 (App. Div. 2d Dept. 1993) (finding that skin tone is a factor); People v. Jones, 24 Misc.3d 1229(A) at *5 (Sup. Ct. Kings Cty. 2009) (Del Giudice, J.) (suppressing a lineup identification due to the fillers appearing significantly older than the defendant); People v. Newman, 48 Misc.3d 1212(A) at *6 (Sup. Ct. Kings Cty. 2015) (Del Giudice, J.) (holding the same). The Court finds that Brown has met his burden of demonstrating that the photo array procedure was impermissibly suggestive and that Brown would be singled out for identification. See People v. Holley, 26 N.Y.3d 514, 521-25 (2015). Det. Devito testified that he based his fillers on Brown’s appearance and that he looked for individuals who matched Brown’s age, build, hair, race, and complexion — even though the complainants never described the perpetrator’s hair because his head was covered with a hood, and even though five out of the six individuals in the photo array have dark complexions with varying apparent ages and builds. Also, Det. Devito expressed that while there exists a filter for skin tone, he opted not to use that filter to retain as many individuals as possible in the search pool. He also did not use the similar image feature to generate fillers. Rather, he then manually reviewed an unknown number of photographs to selected fillers he purportedly believed to be close in appearance to Brown. Further, even upon selecting a photograph from the array, complainant Fabrizio Ellis was only “about halfway confident” that he “selected correctly” according to the NYPD Photo Array Viewing Report in evidence. People’s Ex. 3. This Court is convinced that the photo array was unduly suggestive because Brown was the only light-skinned, thin, young black man included in the photo array. The description of the perpetrator as light-skinned, thin, and young were the few factors that Det. Devito had to incorporate into the photo array. However, the Court cannot help but observe that none of the individuals in the photo array, save for Brown, match that description. Moreover, while Det. Devito manually selected fillers who he perceived as “light-skinned,” he did not use the NYPD Photo Management System skin complexion filter or the similar image function, and, instead, chose fillers who were significantly darker than Brown from a pool of approximately 9,000 photographs. Det. Devito’s failure to utilize the skin complexion filter, or select similarly toned fillers, runs afoul of the Division of Criminal Justice Services Identification Procedures for photo arrays. See Identification Procedures: Photo Arrays and Line-ups, Municipal Police Training Council Model Policy and Identification Procedures Protocols and Forms at 2 (NYS DCJS, June 2017) (“[F]illers should be similar in appearance to the suspect in the array.”)1 The concern with Ellis’s identification is particularly troublesome. Courts at both the trial and appellate levels have long grappled with how to approach the phenomenon known as cross-racial identification bias (also known as the cross-race effect), whereby the likelihood of misidentification between individuals of different races is significantly higher than identifications between individuals of the same race. See Johnson, “Cross-Racial Identification Errors in Criminal Cases,” 69 Cornell L. Rev. 934 (1984). Cross-racial identification bias is generally accepted by the scientific community, and the results of scientific studies demonstrating cross-racial identification bias are even frequently admissible in courts. See People v. Boone, 30 N.Y.3d 521, 528-9 (2017) (citing to John C. Brigham et al., The Influence of Race on Eyewitness Memory in Rod C. L. Lindsey et al., Handbook of Eyewitness Psychology, Vol. 2 at 258 [2014]) (“[I]n a survey of psychologists with expertise in eyewitness identification, 90 percent of the experts believed that empirical evidence of the cross-race effect was sufficiently reliable to be presented in court”); see also People v. Williams, 14 Misc.3d 571 (Sup. Ct. Kings Cty. 2006) (permitting expert testimony about cross-racial identification bias after conducting a Frye hearing); People v. Radcliffe, 196 Misc.2d 381 (Sup. Ct. Bronx Cty. 2003) (granting a motion in limine for an expert witness to testify about issues with cross-racial identifications). It is within the context of this this heightened likelihood for error, taken together with notably different fillers in the instant photo array, that this Court makes its decision. Since Brown was singled out by the inclusion of dissimilar fillers, the Court finds that his identification in both procedures was the result of the unduly suggestive photo array. Accordingly, any testimony of Brown being identified in the photo array by either Ellis or Oriach is suppressed. Moreover, Ellis’ in court identification is precluded since he first told Det. Devito that he could not make an identification, and second was only able to identify in the array with only fifty percent confidence. As a result, the Court cannot permit an in-court identification since it is tainted by the unduly suggestive photo array. An independent source hearing is permitted as to complainant Oriach, since there was no lack of confidence expressed in his selection of Brown, notwithstanding problems with the photo array. The suppression of all noticed statements to police, pursuant to Huntley, is also denied. The People have the burden of proving the voluntariness of an individual’s statements beyond a reasonable doubt for their use in evidence at trial. See People v. Guilfors, 21 NY3d 205, 208 (2013); People v. Anderson, 42 NY2d 35 (1977); People v. Valerius, 31 NY2d 51 (1972); Huntley, 15 NY2d at 72; CPL §60.45(1). Det. Devito testified, and body-worn camera shows, that Miranda warnings were properly administered to Brown, and he expressed clear understanding of them. Furthermore, at no time was Brown threatened or coerced during the interview. Accordingly, Brown’s motion to suppress noticed statements to police are denied. Any unnoticed statements are precluded. The Court’s decision as to the Dunaway prong of Brown’s motion is held in abeyance pending the outcome of the independent source hearing with Oriach. The Court finds a lack of probable cause with respect to the charges involving Ellis, although the People have leave to seek charges against Brown for criminal possession of stolen property related to the Venmo transactions involving Ellis. This constitutes the decision and order of the Court. Dated: November 4, 2024
