United States v. Nolan, No. 16-3423 (2d Cir. 2020)
Annotate this Case
Justia Opinion Summary
Want to stay in the know about new opinions from the Second Circuit US Court of Appeals?
Sign up for free summaries delivered directly to your inbox. Learn More ›
You already receive new opinion summaries from Second Circuit US Court of Appeals. Did you know we offer summary newsletters for even more practice areas and jurisdictions? Explore them here.
The Second Circuit reversed the district court's denial of defendant's 28 U.S.C. 2255 motion to vacate his conviction on the ground of ineffective assistance of counsel. The court held that defendant received ineffective assistance because his lawyers did almost nothing to challenge the eyewitness identification testimony that formed the core of the Government's case, even though the identifications bore glaring indicia of unreliability. Furthermore, defendant received ineffective assistance because his counsel did not seek to exclude or object to the admission of a highly prejudicial and dubiously relevant photo of defendant posing with what appears to be a handgun. Because defendant was prejudiced by counsel's deficient performance, the court vacated his conviction and remanded for further proceedings.
Download PDF
16-3423-pr(L) United States of America v. Ralph Nolan 1 UNITED STATES COURT OF APPEALS 2 FOR THE SECOND CIRCUIT 3 _____________________ 4 5 AUGUST TERM 2019 6 7 (ARGUED: NOVEMBER 4, 2019 DECIDED: APRIL 15, 2020) 8 9 Nos. 16-3423-pr(L), 18-1113-pr(CON) 10 11 _____________________ 12 13 UNITED STATES OF AMERICA 14 Appellee 15 16 17 -v.- 18 19 RALPH NOLAN 20 Defendant-Appellant. 21 22 1 1 Before: SACK and HALL, Circuit Judges, and RAKOFF, District Judge. 1 2 _______________________ 3 4 5 6 7 8 9 10 11 12 13 14 Defendant-Appellant Ralph Nolan challenges the denial by the district court (Daniels, J.) of his motion under 28 U.S.C. § 2255 to vacate his conviction on the ground of ineffective assistance of counsel. Nolan argues that he received ineffective assistance because his lawyers did almost nothing to challenge the eyewitness identification testimony that formed the core of the Government’s case, even though the identifications bore glaring indicia of unreliability. Nolan also argues that he received ineffective assistance because his counsel did not seek to exclude or object to the admission of a highly prejudicial and dubiously relevant photo of the defendant posing with what appears to be a handgun. We agree with Nolan as to both arguments. Accordingly, the judgment of the district court is REVERSED, Nolan’s conviction is VACATED, and the case is REMANDED for further proceedings consistent with this opinion. _______________________ 15 16 17 18 19 APPEARING FOR APPELLANT: SUSAN J. WALSH (Yannick Allan Grant, on the brief), Vladeck, Raskin & Clark, P.C., New York, NY 20 21 22 23 24 25 APPEARING FOR APPELLEE: RICHARD COOPER, Assistant United States Attorney (Karl Metzner, Assistant United States Attorney, on the brief), for Geoffrey S. Berman, United States Attorney for the Southern District of New York, New York, NY _______________________ 26 27 28 29 30 Judge Jed S. Rakoff, of the United States District Court for the Southern District of New York, sitting by designation. 1 2 1 2 3 RAKOFF, District Judge: 4 Supreme Court recognized over a half-century ago, “[t]he vagaries of eyewitness 5 identification are well-known; the annals of criminal law are rife with instances 6 of mistaken identification.” United States v. Wade, 388 U.S. 218, 228 (1967); see 7 also Simmons v. United States, 390 U.S. 377, 383-84 (1968); Manson v. Brathwaite, 8 432 U.S. 98 (1977). The results can be devastating. For example, as Appellant here 9 points out, according to the Innocence Project, eyewitness misidentification was Eyewitness identification testimony is notoriously prone to error. As the 10 present in an astonishing 71 percent of the cases in which subsequent DNA 11 testing established the factual innocence of wrongfully convicted defendants. See 12 Innocence Project, Eyewitness Identification Reform, www.innocenceproject.org/ 13 eyewitness-identification-reform (Last Visited Mar. 16, 2020). 14 In the instant case, even though many of the typical causes of mistaken 15 eyewitness identifications were apparent, defendant’s trial counsel did almost 16 nothing to challenge the introduction of such identifications or combat these 17 problems. We conclude that given the obvious materiality of the eyewitness 18 testimony in the case at bar, this failure amounted to ineffective assistance of 19 counsel, requiring reversal of the district court’s judgment, vacatur of Nolan’s 3 1 conviction, and remand to the district court for further proceedings consistent 2 with this opinion. 3 In brief summary, on April 10, 2015, a jury sitting in the Southern District 4 of New York, relying almost entirely on eyewitness identifications, convicted 5 Defendant-Appellant Ralph Nolan of conspiracy to commit Hobbs Act robbery 6 in violation of 18 U.S.C. § 1951 (Count One), attempted Hobbs Act robbery in 7 violation of 18 U.S.C. § 1951 (Count Two), and brandishing a firearm during and 8 in relation to a crime of violence in violation of 18 U.S.C. § 924(c)(1)(A)(ii) (Count 9 Three). Specifically, Nolan was found to have joined in an armed robbery of an 10 apartment occupied by a family, some of whose members had been dealing 11 drugs. Four of the five adults present in the apartment at the time of the 12 robbery—all testifying at Nolan’s trial pursuant to grants of immunity— 13 identified Nolan as one of the robbers. On September 28, 2016, the district judge 14 sentenced Nolan to 120 months’ imprisonment followed by three years of 15 supervised release. Nolan is currently serving his sentence. 16 All four identifications bore significant indicia of unreliability. The robbers 17 were partially disguised. They carried guns, on which the eyes of the victims 18 were likely focused. Initially, the victims were unable to give investigators a 4 1 detailed description of the robbers beyond noting that they were light-skinned or 2 Hispanic. The four victims did not identify Nolan (who is white) as one of the 3 intruders until they saw his photo in a photo array presented to them more than 4 a month after the crime. Even then, at least one of the victims did not firmly 5 identify Nolan until law enforcement allowed that victim to view photos of 6 Nolan on Facebook. That victim discussed with two other victims her 7 identification of Nolan and showed the other two victims his Facebook photo 8 before these victims were asked to identify Nolan from a photo array. 9 Nolan’s defense counsel nonetheless did virtually nothing to contest the 10 admissibility of these identifications. In particular, defense counsel abandoned a 11 pre-trial motion to preclude the eyewitness identifications for reasons that 12 counsel has failed to explain. And both then and after the testimony had been 13 introduced at trial, defense counsel failed to call or even consult an expert 14 witness who could have informed the judge and jury about the multiple, well- 15 established ways in which these identifications were unreliable. 16 Arguing that his lawyers’ errors prejudiced the outcome of his trial, Nolan 17 petitioned the district court under 28 U.S.C. § 2255 for a writ of habeas corpus on 18 the ground of ineffective assistance of counsel. See Strickland v. Washington, 466 5 1 U.S. 668 (1984). On February 20, 2018, the district court (Daniels, J.) denied the 2 petition without a hearing. No. 14-cr-555 (GBD), 2018 WL 1166726 at *1 (S.D.N.Y. 3 Feb. 20, 2018). Nolan appeals from this denial. 2 We have jurisdiction under 28 4 U.S.C. §§ 2253 and 2255(d), and we review Nolan’s ineffective assistance claims 5 de novo. Pham v. United States, 317 F.3d 178, 182 (2d Cir. 2003). For the 6 following reasons, we reverse the district court and vacate Nolan’s conviction. Background 7 By finding Nolan guilty, Nolan’s jury effectively found that on December 8 9 16, 2013, Nolan and an accomplice committed an armed home invasion of an 10 apartment in the Bronx, known by the robbers to be home to a family, some of 11 whose members had been dealing drugs, from whom the robbers were seeking 12 to steal marijuana and other valuables. (Evidence at trial also established the Additionally, on direct appeal, Nolan challenges some of the district court’s evidentiary rulings, and he also challenges his conviction on Count Three on the ground that attempted Hobbs Act robbery is not a “crime of violence” under the definition in 18 U.S.C. § 924(c)(3)(A). See United States v. Davis, 139 S. Ct. 2319 (2019). Prior to oral argument, this Court consolidated Nolan’s direct appeal with his § 2255 petition. Because we grant Nolan’s ineffective assistance claims and vacate his conviction on all three counts, we need not and do not consider his alternative grounds for relief. Instead, we remand the matter so that, if the Government chooses to retry Nolan, the district court can consider these alternative grounds with the benefit of additional briefing on recent developments in the case law interpreting § 924(c). 2 6 1 involvement of a third co-conspirator, Shaun Odiot, who had canvassed the 2 apartment earlier that day but did not enter it during the robbery.) According to 3 trial testimony, upon entering the apartment, Nolan and the other robber 4 threatened the victims in the apartment with guns and demanded marijuana and 5 money. In the course of the robbery, Nolan is alleged to have pistol-whipped one 6 victim, leaving her unconscious. He also purportedly ripped cable wires out of 7 the wall and used them to bind the hands of the victims. According to the 8 victims’ testimony, Nolan stole a small amount of marijuana as well as some 9 electronic devices from the apartment. 10 There were five adult victims in the apartment at the time of the robbery: 11 Lorraine Scroggins, the primary resident; Desiree Scroggins, Lorraine’s daughter; 12 Christopher Martinez, Desiree’s boyfriend; Sandra Martinez, Christopher’s 13 mother; and Deyanira Beriguete, a home health aide who took care of Lorraine, 14 who is quadriplegic. Christopher’s and Desiree’s infant daughter was also in the 15 apartment during the robbery. 16 In the immediate aftermath of the crime, several of the victims indicated 17 that the robbers may have been Hispanic, and none identified Nolan (who is 18 white and who was previously known to two of the victims) as one of the 7 1 culprits. 3 Law enforcement initially showed Desiree, Lorraine, and Christopher 2 photo arrays that did not include Nolan’s picture, and all three of these victims 3 said that they did not recognize anyone in the photos as one of the robbers. Law enforcement did not add Nolan’s picture to the photo array until 4 5 several weeks later. Specifically, after the NYPD detective investigating the 6 robbery began to suspect that Odiot was the third co-conspirator, the detective 7 “searched [Odiot’s] address for anybody that had been arrested using that 8 address in the past,” Det. E. Deloren, trial tr. at 449, and created a new photo 9 array including people who turned up in that search and who matched the 10 general description of the other robber. This new photo array included a picture 11 of Nolan. The detective first showed this new photo array to Desiree. She testified 12 13 that she identified Nolan from his photo as one of the robbers but was at first 14 only about 75% sure in her identification. The detective then went into another 15 room and showed the photos to Lorraine, who identified Nolan as the robber. At After the robbery, the NYPD initially circulated an “information needed” poster identifying the robbers as two Hispanic males. At trial, however, Sandra testified that one of the robbers was white. Desiree testified that she had told law enforcement shortly after the robbery that the perpetrators were two Hispanic men, but that one had light skin. 3 8 1 that point, Desiree entered the room where Lorraine was. Lorraine, referring to 2 Nolan’s photo, asked Desiree, “that’s him, right?” L. Scroggins, trial tr. at 152. 3 Desiree and Lorraine, in the detective’s presence, then looked at photos of Nolan 4 on his Facebook page. 4 And after viewing Nolan’s Facebook photos, Desiree 5 testified, she became certain of her identification. On the same day, according to Desiree’s testimony, she discussed her 6 7 identification of Nolan with Sandra and Christopher Martinez, to whom law 8 enforcement had not yet showed a photo array that included Nolan’s picture. 9 She also showed them Nolan’s Facebook photos. The following day, NYPD showed a “six-pack” photo array containing 10 11 Nolan’s photograph to Sandra, who identified Nolan and testified that she was 12 “pretty confident” in her identification. S. Martinez, trial tr. at 626. Sandra also 13 testified that, at that point, she recognized Nolan as the person she knew of as 14 “White Boy” from her old neighborhood. About two weeks later, law Lorraine and Desiree were able to locate Nolan’s Facebook page, according to Lorraine’s testimony, because the detective had told her Nolan’s name. There is inconsistent testimony about the detective’s exact role in the Facebook search, but at the very least he was aware of what was happening and did not immediately stop it, and he may have initiated and/or encouraged Desiree and Lorraine to look at Nolan’s photos on Facebook. 4 9 1 enforcement showed the photo array to Christopher, who also identified Nolan 2 as one of the robbers and stated that he recognized him as “White Boy.” 3 Christopher testified that he was confident in his identification. He also testified 4 that his mother, Sandra, had shown him a picture of Nolan before law 5 enforcement showed him the photo array. 5 The Government heavily relied in its summation on the victims’ 6 7 identifications of Nolan, arguing to the jury that: There’s absolutely no dispute in this trial that this heinous crime happened. . . . The only issue that’s really in dispute is whether Ralph Nolan was one of the robbers, and not one, not two, not three, but four of the victims took that witness stand and told you he was. Back closer to the robbery, when they talked to the police, they looked at his picture and said that’s the guy who robbed us, and right here in this courtroom you saw them, one after the other, they took the witness stand, they looked around the courtroom, and they said he did it, he did it. 8 9 10 11 12 13 14 15 16 17 Trial tr. at 765. Deyanira Beriguete, Lorraine’s home health aide, and the only one of the five victims not granted immunity for her testimony, was also shown the photo array but did not make an identification. When asked at trial whether she would be able to recognize either of the two individuals who committed the robbery, she replied, “I don’t know.” D. Beriguete, trial tr. at 193. 5 10 The evidence against Nolan beyond these eyewitness identifications was 1 2 limited. But among such evidence, the Government introduced a photo from 3 Nolan’s Facebook page of him holding what appears to be a black handgun, but 4 which the jury was told was in fact a BB gun. The Government also discussed 5 this photo in its summation, arguing that it evidenced “the defendant’s . . . access 6 to firearms and knowledge of them.” Trial tr. at 840. 7 Defense counsel argued a theory of mistaken identity at trial. But as 8 relevant to this appeal, defense counsel made little effort to exclude the evidence 9 discussed here. Counsel initially made a pre-trial motion to preclude in-court 10 identification testimony by Lorraine and Desiree Scroggins and Christopher and 11 Sandra Martinez. But when instructed by the trial court that, were it to preclude 12 the identifications, it would also preclude any questioning about the 13 identification techniques used by law enforcement, such as permitting the 14 victims to view Facebook photos of Nolan, defense counsel withdrew the motion 15 in favor of allowing “it . . . all [to] come in.” Trial tr. at 50. Defense counsel then 16 attempted to impeach the credibility of the identifications on cross-examination 6 17 and argued on summation that the identifications were not reliable. For example, counsel cross-examined Desiree about her initial statement to the police that the robbers were Hispanic, her statement that she was only 75% sure 6 11 1 Defense counsel also did not seek to exclude the Facebook BB gun photo, 2 nor did counsel object to its introduction at trial. Before trial, the Government 3 moved in limine to permit introduction of the photo as direct evidence of the 4 crime, or alternatively as evidence under Fed. R. Evid. 404(b) of the defendant’s 5 knowledge of and access to firearms. The Government also moved in limine to 6 introduce a small portion of the video recording of Nolan’s post-arrest interview 7 with law enforcement in which Nolan made the incriminating statement that he 8 knew suspected co-conspirator Odiot. In response, defense counsel, rather than 9 seeking to exclude the photo of Nolan holding a gun, moved under Fed. R. Evid. 10 106, the so-called “rule of completeness,” to introduce the entire post-arrest of her identification of Nolan based on the photo array, and the fact that she and her mother looked at Facebook photos of Nolan. Counsel cross-examined Christopher about his initial identification of the robbers as Hispanic and the fact that he did not recognize “White Boy” at the time of the robbery. Counsel also cross-examined Lorraine about looking at Nolan’s Facebook photos. And finally, counsel cross-examined Sandra about the identification of both robbers as Hispanic, the fact that she and Christopher did not recognize “White Boy” during the crime, and the fact that she viewed and discussed Nolan’s Facebook photos with Desiree and Christopher before identifying Nolan from a photo array. The defense also called to the stand Lorraine’s son, Robert, who was in the room with her and the detective when she made her photo array identification of Nolan. The defense apparently expected him to testify that Lorraine was uncertain in her identification until she viewed the Facebook photos, but he did not actually say this. 12 1 video, including portions where Nolan denied involvement in the charged crime. 2 Significantly, another portion of the video included an exchange in which law 3 enforcement showed Nolan the BB gun photo and discussed it with him. Ruling 4 on these motions, the district court essentially gave both sides what they sought, 5 admitting the entire post-arrest video under Rule 106 and also admitting the BB 6 gun photo, reasoning that introducing the segments of the post-arrest video 7 where Nolan and the law enforcement officer discuss the BB gun photo would 8 “minimize[] any potential prejudice” from showing the jury the photo itself. Tr. 9 of April 1, 2015 hearing, at 9. 10 Discussion 11 The Eyewitness Identifications 12 To succeed on an ineffective assistance claim, a defendant must 13 demonstrate, first, that “in light of all the circumstances,” the acts or omissions of 14 trial counsel “were outside the wide range of professionally competent 15 assistance,” and, second, that “there is a reasonable probability that, but for 16 counsel’s unprofessional errors, the result of the proceeding would have been 17 different.” Strickland, 466 U.S. at 690, 694. This test presents a high bar for a 18 defendant, as Strickland instructs courts to apply a “presumption of effective 13 1 performance.” Greiner v. Wells, 417 F.3d 305, 326 (2d Cir. 2005). “The standard 2 of Strickland ‘is rigorous, and the great majority of habeas petitions that allege 3 constitutionally ineffective counsel founder on [it].’” Bell v. Miller, 500 F.3d 149, 4 155 (2d Cir. 2007) (alteration in original) (quoting Lindstadt v. Keane, 239 F.3d 5 191, 199 (2d Cir. 2001)). 6 The first ground on which Nolan challenges his trial counsel’s 7 effectiveness is that counsel failed to adequately challenge the eyewitness 8 identifications that formed the core of the Government’s case. Specifically, Nolan 9 argues that he received ineffective assistance because his lawyers (1) failed to 10 pursue a pre-trial motion to preclude the in-court identification testimony by the 11 four victims, and (2) failed to hire an expert on eyewitness testimony or seek to 12 introduce expert testimony about the unreliability of eyewitness identifications 13 in the circumstances presented by this case. Despite the very high bar for an 14 ineffective assistance claim, we agree with both arguments. 15 As to the first argument, trial counsel’s strategy of abandoning defendant’s 16 pretrial motion to exclude the eyewitness testimony of the four victims in favor 17 of trying to impeach such testimony at trial was professionally unreasonable, 18 thereby satisfying Strickland’s first prong. Indeed, it was contrary to the 14 1 protections that the Supreme Court has erected to deal with the widely 2 recognized problems with eyewitness testimony. Following the Court’s decision 3 in United States v. Wade, 388 U.S. 218 (1967), which held that the Sixth 4 Amendment guarantees a defendant the right to have counsel present during a 5 post-indictment “lineup” identification procedure, id. at 228-39, a series of 6 subsequent cases recognized that due process requires trial courts to scrutinize 7 whether the pretrial identification procedures employed in a particular case were 8 “so impermissibly suggestive as to give rise to a very substantial likelihood of 9 irreparable misidentification.” Simmons, 390 U.S. 377, 384 (1968); see also Stovall 10 v. Denno, 388 U.S. 293 (1967), overruled on other grounds by Griffith v. 11 Kentucky, 479 U.S. 314 (1987). 12 A growing body of scientific research, moreover, has clarified and 13 expanded what factors a court should examine in determining whether to 14 exclude eyewitness identification testimony. As this Court has observed, this 15 “literature indicates that certain circumstances surrounding a crime — including 16 the perpetrator’s wearing a disguise, the presence of a weapon, the stress of the 17 situation, the cross-racial nature of the crime, the passage of time between 18 observation and identification, and the witness’s exposure to [the] defendant 15 1 through multiple identification procedures — may impair the ability of a witness 2 . . . to accurately process what she observed.” Young v. Conway, 698 F.3d 69, 78- 3 79 (2d Cir. 2012); see generally National Research Council, Identifying the 4 Culprit: Assessing Eyewitness Identification (2014) (hereinafter “Identifying the 5 Culprit”). 7 All of the impairing factors referred to in Young were present here. The 6 7 perpetrators were wearing disguises that partially obscured their facial features; 8 one robber was wearing a “half ski mask” and the other was wearing a “skully.” 9 They were armed, thus suggesting the widely-recognized phenomenon known 10 as “weapon focus,” by which the presence of a weapon at the crime scene tends 11 to draw the eyewitnesses’ attention, distracting them from the face of the 12 perpetrator. See Identifying the Culprit at 93-94; State v. Henderson, 27 A.3d 872, 13 904-05 (N.J. 2011). The aggressive behavior of the robbers toward their victims 14 plainly placed the latter under stress, and, as the Supreme Court of New Jersey 15 pointed out in revising its eyewitness procedures, research has shown that “an Such research has already led the Supreme Courts of two states to substantially change their approach to evaluating eyewitness identifications. State v. Henderson, 27 A.3d 872 (N.J. 2011) (replacing the existing test for admissibility of eyewitness identifications with one that incorporates the findings of scientific research on eyewitness reliability); State v. Lawson, 291 P.3d 673 (Or. 2012) (same). 7 16 1 eyewitness under high stress is less likely to make a reliable identification of the 2 perpetrator.” Id. at 904; see also, e.g., State v. Lawson, 291 P.3d 673, 744 (Or. 3 2012); Charles A. Morgan III et al., Accuracy of Eyewitness Memory for Persons 4 Encountered During Exposure to Highly Intense Stress, 27 Int’l J. L. & Psychiatry 5 265 (2004). 6 Also, while the victims were black (the Scrogginses) and Hispanic (the 7 Martinezes), the individual identified as a robber (Nolan) was white. It is well 8 established that eyewitnesses are materially less accurate when identifying 9 individuals of a different race, see Lawson, 291 P.3d at 688; Henderson, 27 A.3d 10 at 907 (citing academic sources), or a different ethnicity, see Stephanie J. Platz & 11 Harmon M. Hosch, Cross-Racial/Ethnic Eyewitness Identification: A Field 12 Study,18 J. Applied Soc. Psychol. 972 (1988); Identifying the Culprit at 96. But cf. 13 Henderson, 27 A.3d at 917 (distinguishing what it called “the dearth of social 14 scientific research in the field of cross-ethnic bias from the convincing social 15 science data demonstrating the potential unreliability of cross-racial 16 identifications.” (internal quotation marks omitted)). Further still, many weeks 17 elapsed between the time of the robbery and the time the victims were shown the 18 photo array containing Nolan’s photo, and it is also well established that 17 1 eyewitness identifications are materially less accurate if made more than a short 2 time after the crime. Lawson, 291 P.3d at 688; see also, e.g., Kenneth A. 3 Deffenbacher et al., Forgetting the Once-Seen Face: Estimating the Strength of an 4 Eyewitness’s Memory Representation, 14 J. Experimental Psychol.: Applied 139 5 (2008). 6 Finally, and perhaps most egregiously, the police employed highly 7 irregular procedures in pursuing the witnesses’ identification of Nolan, 8 potentially biasing the victims’ identifications by, for example, allowing them to 9 talk among themselves about Nolan’s identification and allowing them to view 10 his photos on Facebook. Studies have demonstrated that the memories of 11 eyewitnesses are extremely susceptible to contamination by external information, 12 a common source of which is “cowitness interaction.” See Lawson, 291 P.3d at 13 710 (citing Elin M. Skagerberg, Co-Witness Feedback in Line-Ups, 21 Applied 14 Cognitive Psychol. 489 (2007)). 15 “Even the best criminal defense attorneys would not defend a particular 16 client in the same way.” Strickland, 466 U.S. at 689. But in light of the foregoing 17 characteristics of the witnesses’ identification testimony here, we conclude that 18 1 an effective defense counsel would have vigorously contested the admissibility 2 of the testimony. 3 Here, however, defense counsel, after initially bringing such a motion, 4 abandoned it on the theory that the defense would be better off impeaching the 5 eyewitnesses at trial. We do not understand how he could have reached that 6 conclusion. If defense counsel had succeeded in the exclusion motion, it would 7 appear that the case would have been effectively over in light of the 8 Government’s heavy reliance on the eyewitness identifications. And even if 9 defense counsel had lost the motion, they would have educated the judge as to 10 the frailty of the identifications and would, of course, still have been free to fully 11 impeach the eyewitnesses at trial. Even making “every effort . . . to eliminate the 12 distorting effects of hindsight,” Bell, 500 F.3d at 156 (quoting Strickland, 466 U.S. 13 at 689), we conclude that this strategy fell outside “the wide range of reasonable 14 professional assistance.” Strickland, 466 U.S. at 689. 15 Counsel also failed to render reasonable professional assistance by 16 neglecting to call or even consult an expert to testify about the unreliability of the 17 eyewitness identifications under the egregious circumstances presented in this 19 1 case. 8 This Court came to a similar conclusion in a comparable case. In Bell, the 2 eyewitness was the victim of an attack, and endured significant blood loss and a 3 lengthy period of sedation before identifying his assailant. See generally 500 F.3d 4 149. Although the victim’s identification was virtually the only evidence against 5 defendant Bell at his trial, Bell’s counsel did not consult an expert who could 6 have opined on whether the victim’s identification was reliable despite the 7 medical trauma he had experienced. Id. at 155-56. This Court held that counsel’s 8 performance was ineffective, as an expert could have explained that the 9 eyewitness’s identification was highly unreliable for reasons “beyond the ken of 10 the typical juror.” Id. at 156 (quoting People v. Taylor, 552 N.E.2d 131, 135 (N.Y. 11 1990)). Here, similarly, the eyewitness testimony was sufficiently unreliable in 12 13 ways not readily apparent to a lay jury. For example, as noted above, persons of 14 a given race or color are not nearly as good at perceiving and remembering the 15 fine facial features of someone of a different race or color as they are at In support of Nolan’s § 2255 petition to the court below, one of Nolan’s trial counsel wrote a letter to the trial court, conceding that he had rendered ineffective assistance on this ground. United States v. Nolan, No. 14-cr-555 (GBD), Dkt. 115, Ex. 1 (Oct. 9, 2015). 8 20 1 perceiving such features of someone of their own race or color. But it appears to 2 us unlikely that this is common knowledge among typical jurors. Moreover, if 3 defense counsel had consulted an expert on eyewitness identifications, in light of 4 the state of the relevant science referred to above, they would surely have 5 become more aware of the factors causing the eyewitness identifications here to 6 be problematic, and would not have so readily abandoned their exclusion 7 motion. Indeed, the expert might well have provided an affidavit greatly 8 strengthening that motion. 9 To be sure, Strickland ordinarily does not require defense counsel to call 10 any particular witness. See United States v. Best, 219 F.3d 192, 201 (2d Cir. 2000) 11 (“[W]e [do not] see anything unreasonable in counsel’s decision not to call the 12 potential witnesses.”). But under the unusual circumstances presented here, as in 13 in Bell, counsel could not render effective assistance without input from an 14 expert. Counsel therefore had a duty at least to consult an expert and consider 15 whether to call her to the stand. 16 As to Strickland’s second prong, we find it abundantly clear that counsel’s 17 actions so prejudiced the outcome of the trial that we must grant Nolan’s 18 ineffective assistance claim. See 466 U.S. at 694. Without the identifications, the 21 1 probative value of the Government’s remaining evidence against Nolan would 2 have been limited at best. 9 Therefore, had defense counsel successfully excluded 3 the eyewitness identifications and Government nonetheless proceeded with the 4 prosecution, a reasonable jury would likely have rendered an acquittal, assuming 5 that the judge would have even let the case go to the jury. The only possible argument that Nolan might not have been prejudiced by 6 7 the performance of defense counsel — and the argument that persuaded the 8 district court — that we can conjure is that the jury instructions told the jury all 9 they needed to know about eyewitness identifications. Among other factors, the 10 district court instructed the jury to “ask whether the witness was able to observe 11 and had an adequate opportunity to observe the person who committed the 12 crimes charged[,] . . . whether the witness is positive in the identification, . . . 13 whether the witness gave a description of the person who committed the crime, 14 and . . . [whether] the identification was made under circumstances that may In addition to the prejudicial Facebook photo discussed below, the Government introduced evidence that Nolan knew co-conspirator Odiot, that Nolan — without knowledge that the police considered Odiot a suspect — told the police that Odiot had set him up, and that Nolan admitted to being known as “White Boy.” The Government also drew a connection between the fact that the victims had been tied up with cable wire and the fact that Nolan worked as a cable installer. None of this comes close to establishing guilt by any standard, let alone beyond a reasonable doubt. 9 22 1 have influenced the witness.” 2018 WL 1166726 at *5. But this was not sufficient 2 to cure the aforementioned problems with the identifications that would not 3 have been apparent to a lay jury. As this Court has noted, “[m]any of these 4 factors are counterintuitive and, therefore, not coterminous with ‘common 5 sense.’” Young, 698 F.3d at 79. A lay juror would not know, for example, about 6 the likely impact on perception of extreme stress and weapon focus, nor would 7 the juror necessarily understand that the detective’s identification practices were 8 highly suggestive. Accordingly, we find more than “a reasonable probability that 9 . . . the result of the proceeding would have been different” had defense counsel 10 effectively contested the eyewitness identification testimony. 466 U.S. at 694. 11 The Facebook “Gun” Photo 12 Although the foregoing is sufficient to warrant reversal, there is still 13 another, independent ground for such relief. Nolan argues that the performance 14 of defense counsel at trial also failed to pass constitutional muster because 15 counsel made no effort to preclude the Government from introducing a highly 16 prejudicial Facebook photo of Nolan posing with what appears to be a handgun, 17 but which was actually a BB gun. 10 Nolan also argues that counsel was ineffective for failing to move to exclude two other pieces of “bad act” evidence: first, a reference he made in his post- 10 23 1 We agree. Although Strickland instructs courts to begin with a 2 “presumption that, under the circumstances, the challenged action ‘might be 3 considered sound trial strategy,’” 466 U.S. at 689 (quoting Michel v. Louisiana, 4 350 U.S. 91, 101 (1955)), we again see no strategic rationale for defense counsel’s 5 failure to move to exclude the photo. To be clear, we do not fault defense counsel 6 for seeking to admit the favorable portions of the post-arrest video, and had the 7 district court already ruled that the Facebook BB gun photo was admissible, then 8 it would seem to have become a reasonable fallback strategy for defense counsel 9 to move to admit the entire video under Rule 106. But as in the analysis above, 10 we fail to understand why defense counsel did not first seek to exclude the gun 11 photo. arrest interview to a juvenile conviction, which he went on to explain was the result of a fistfight among teenagers, and second, a fleeting reference to a prior arrest of Nolan in the testimony of the NYPD detective who conducted the photo array sessions. While we are troubled by the admission of the prior conviction without a limiting instruction, see United States v. McCallum, 584 F.3d 471, 478 (2d Cir. 2009), we find it unlikely that this evidence was so prejudicial in the eyes of the jury that counsel’s decision not to move to preclude it constituted ineffective assistance. In any event, we need not address the question in detail since we grant Nolan’s claim as to the Facebook BB gun photo. We also need not reach Nolan’s third independent ground for his ineffective assistance claim — his counsel’s failure to introduce purported alibi testimony from the defendant’s father. 24 1 Indeed, the district court may well have granted the motion. The photo is 2 not direct evidence of the charged crime, inasmuch as the photo does nothing to 3 place Nolan at the time and place of the robbery. And the Government’s 4 proffered rationale — that the photo is supposedly probative of Nolan’s access 5 to, knowledge of, and comfort with firearms — is weak on its face (especially in 6 light of the fact that it is a BB gun) and must be weighed against the obvious 7 prejudice to the defendant that would result from showing the jury a photo of a 8 defendant charged with armed robbery posing with a gun of any kind. See Fed. 9 R. Evid. 403. In contrast to the district court, then, we see no valid ground for 10 defense counsel not to have at least attempted to have the court exclude the 11 photo. See 2018 WL 1166726 at *13. 12 We think defense counsel’s failure to seek a limiting instruction for the 13 photo also constituted ineffective assistance of counsel. We are mindful that in 14 some cases, effective counsel might elect not to seek a limiting instruction for fear 15 that “the instruction may call the jury’s attention to the harmful evidence.” PRL 16 USA Holdings, Inc. v. U.S. Polo Ass’n, Inc., 520 F.3d 109, 117 (2d Cir. 2008). But 17 such reasoning does not apply here. As noted above, the Government moved in 18 limine for the court to admit the photo. This should have put defense counsel on 25 1 notice that the Government planned to focus on this evidence at trial. And 2 indeed, the Government did highlight the photo in its summation, arguing to the 3 jury that the photo provided evidence of the defendant’s “access to firearms and 4 knowledge of them.” Trial tr. at 840. This portion of the summation in effect 5 implied that the jury could consider the photo as evidence of Nolan’s propensity 6 to commit gun crimes. Given all of this, counsel’s decision never to seek a 7 limiting instruction for the photo fell below the level of reasonable professional 8 effectiveness required by Strickland. We also find that defense counsel’s handling of the photo, both in not 9 10 seeking to exclude it and in not seeking a limiting instruction to accompany its 11 admission, was sufficiently prejudicial to Nolan that “there is a reasonable 12 probability that . . . the result of the proceeding would have been different” had 13 the photo not been admitted. 11 Strickland, 466 U.S. at 694. Concededly, the photo 14 was of much less significance to the Government’s case than the eyewitness 15 identifications. But it nonetheless likely served to elicit an emotional reaction Nolan also argues that counsel erred by inadequately investigating the circumstances of the photo, as well as Nolan’s claim that the gun in the photo was a BB gun. But since the Government in its summation told the jury that the weapon was a BB gun, any such error had limited prejudicial effect. 11 26 1 among the jurors in light of the fact that the defendant was being tried on a 2 charge of armed robbery. 3 4 Deciding Ineffectiveness of Counsel We are mindful, of course, of the Supreme Court's instruction, to which 5 this Court often refers, that “in most cases a motion brought under § 2255 is 6 preferable to direct appeal for deciding claims of ineffective assistance.” Massaro 7 v. United States, 538 U.S. 500, 504 (2003). As noted in footnote 2, above, though, 8 the defendant made a § 2255 motion in the district court, which the court denied. 9 And on May 4, 2018, the defendant obtained an order from this Court 10 consolidating his direct and § 2255 appeals. The § 2255 ineffectiveness of counsel 11 claim is thus properly before us for resolution. 12 Conclusion 13 For the foregoing reasons, we hold that Ralph Nolan did not receive the 14 effective assistance of counsel guaranteed by the Sixth Amendment. We reverse 15 the district court’s denial of Nolan’s § 2255 motion, vacate his conviction on all 16 three counts, and remand the case to the district court for further proceedings 17 consistent with this opinion. 27
Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
