COMMONWEALTH, Plaintiff-Appellant, KYLE JOHNSON, Defendant-Appellee.

Transcription

1 No. SJC THE SUPREME JUDICIAL COURT OF MASSACHUSETTS COMMONWEALTH, Plaintiff-Appellant, v. KYLE JOHNSON, Defendant-Appellee. ON DIRECT APPELLATE REVIEW BRIEF AND APPENDIX OF THE INNOCENCE PROJECT AND THE INNOCENCE NETWORK AS AMICI CURIAE SUPPORTING DEFENDANT-APPELLEE Karen A. Newirth THE INNOCENCE PROJECT 40 Worth Street Suite 701 New York, New York, Telephone (212) Facsimile (212) R.J. Cinquegrana BBO # Choate, Hall & Stewart LLP Two International Place Boston, MA Telephone (617) Facsimile (617) rjcinquegrana@choate.com James L. Brochin Jennifer H. Wu PAUL, WEISS, RIFKIND, WHARTON & GARRISON LLP 1285 Avenue of the Americas New York, New York, Telephone (212) Facsimile (212) jbrochin@paulweiss.com jwu@paulweiss.com Counsel for Amici Curiae the Innocence Project and the Innocence Network

2 TABLE OF CONTENTS QUE STIONS PRESENTED STATEMENT OF INTEREST STATEMENT OF THE CASE..., 4 SUMMARY OF ARGUMENT ARGUMENT I. THIS COURT SHOULD REVISIT, REAFFIRM, AND EXPAND UPON COMMONWEALTH V. JONES IN LIGHT OF THE 2013 SUPREME JUDICIAL COURT STUDY GROUP'S REPORT AND RECOMMENDATIONS A. The Jones Court Correctly Focused on the Reliability and Fundamental Fairness of Eyewitness Identifications, Regardless of State Action B. The SJC Study Group's Recommendations Provide a Framework for Courts to Scrutinize Unreliable Eyewitness Identifications The Study Group Recommended that Courts Suppress Unreliable Identifications, Regardless of Suggestive Circumstances The Study Group Also Recommended that Courts Hold Pretrial Hearings to Vet Eyewitness Identifications, Even Where State Actors Did Not Cause Suggestive Circumstances This Case Presents an Ideal Opportunity for Addressing the Study Group's Recommendations on the Admissibility of Eyewitness Testimony C. The SJC Study Group's Recommendations Should Be Adopted with Slight Modifications to Most Appropriately Reflect Principles of Social Science

3 II. THIS COURT SHOULD REVISIT AND ELIMINATE THE INDEPENDENT SOURCE DOCTRINE IN LIGHT OF THIS COURT'S DECISIONS IN COMMONWEALTH V. CRAYTON AND COMMONWEALTH V. COLLINS AND THE SJC STUDY GROUP'S REPORT AND RECOMMENDATIONS CONCLUSION CERTIFICATE OF COMPLIANCE APPENDIX A ii

4 TABLE OF AUTHORITIES Page(s) Cases Commonwealth v. Botelho, Mas s ( ) State v. Chen, 27 A.3d 930 (N.J. 2011)... 24, 31 Commonwealth v. Collins, Mass ( ) , 3 3, 3 7 Commonweal th v. Cowans, 52 Mass. App. Ct. 811 (2001) Commonwealth v. Crayton, Mass ( 2014 ) l, 3 3 Commonweal th v. Gomes, Mass ( ) passim State v. Henderson, 27 A.3d 872 (N.J. 2011) , 37 Holmes v. South Carolina, 547 U.S. 319 (2006) Commonwealth v. Johnson, Mas s ( ) Commonwealth v. Jones, Mass. 9 9 ( ) passim Commonwealth v. Jules, Mass ( ) State v. Lawson, 291 P.3d 673 (Or. 2012)... 10, 31 Maher v. Town of Ayer, 463 F. Supp. 2d 117 (D. Mass. 2006) Manson v. Brathwaite, 432 U.S. 98 (1977) iii

5 Miller v. City of Boston, 297 F. Supp. 2d 361 (D. Mass. 2003) N.E. Physical Therapy Plus, Inc. v. Liberty Mut. Ins. Co., Mass ( ) , 31 Commonwealth v. Payne, Mas s ( ) Commonwealth v. Polk, Mass. 2 3 ( ) People v. Santiago, 958 N.E.2d 874 (N.Y. 2011) United States v. Wade, 388 U.S. 218 (1967) Commonwealth v. Walker, 460 Mass. 590 (2011)... 15, 16, 21 Other Authorities Brandon Garrett, Eyewitnesses and Exclusion, 65 Vand. L. Rev. 451 (2012)... 7, 38 Fred Dickey, Worst-Case Scenario, Los Angeles Times, gazine/tm (June 25, 2000) Gary L. Wells & Amy L. Bradfield, "Good, You Identified the Suspect": Feedback to Eyewitnesses Distorts their Reports of the Witnessing Experience, 83 J. Applied Psychol. 360 (1998) Gary L. Wells & Deah S. Quinlivan, Suggestive Eyewitness Identification Procedures and the Supreme Court's Reliability Test in Light of Eyewitness Science: 30 Years Later, 33 Law & Hum. Behav. 9 (2008) , 29 Innocence Project, Dennis Maher, 25 iv

6 Innocence Project, Wilton Dedge, 23 Jennifer N. Sigler & James V. Couch, Eyewitness Testimony and the Jury Verdict, 4 N. Am. J. Psychol. 143 (2002) Massachusetts Court System, Supreme Judicial Court Announces New Standing Committee on Eyewitness Identification (Jan. 12, 2015), html Melissa Boyce et al., Belief of Eyewitness Identification Evidence, in 2 Handbook of Eyewitness Psychology: Memory for People 501 (R.C.L. Lindsay et al. eds., 2007) Michael R. Leippe et al., Cueing Confidence in Eyewitness Identifications: Influence of Biased Lineup Instructions and Pre Identifica tion Memory Feedback Under Varying Lineup Conditions, 33 Law & Hum. Behav. 194 (2009) National Academy of Sciences, Identifying the Culprit: Assessing Eyewitness Identification, available at # passim Saul M. Kassin, Daniel Bogart, and Jacqueline Kerner, Confessions that Corrupt: Evidence from the DNA Exoneration Case Files, 23 Psychol. Sci. 41 (2012) , 27 Steve D. Charman et al., The Dud Effect: Adding Highly Dissimilar Fillers Increases Confidence in Lineup Identifications, 35 Law & Hum. Behav. 479 (2011) v

7 Steven E. Clark et al., Lineup Administrator Influences on Eyewitness Identification Decisions, 15 J. Experimental Psychol.: Applied 63 (2009) Supreme Judicial Court Study Group on Eyewitness Evidence, Report and Recommendations to the Justices (July 25, 2013), available at yewitness-evidence-report-2013.pdf.... passim vi

8 QUESTIONS PRESENTED In a case involving an eyewitness identification made without the participation of the Commonwealth, whether the principles of Commonwealth v. Jones, 423 Mass. 99 (1996), should be revisited in light of the 2013 report and recommendations of the Supreme Judicial Court's Study Group on Eyewitness Evidence; for purposes of an in-court identification, whether the court should revisit the independent source doctrine in light of Commonwealth v. Crayton, 470 Mass. 228 (2014), and Commonwealth v. Collins, 470 Mass. 255 (2014). STATEMENT OF INTEREST The Innocence Network (the "Network") is an association of organizations dedicated to providing pro bono legal and/or investigative services to prisoners for whom evidence discovered post-conviction can provide conclusive proof of innocence. The 70 current members of the Network represent hundreds of prisoners with innocence claims in all 50 states, the District of Columbia, and Puerto Rico, as well as

9 Australia, Canada, France, Ireland, Italy, the Netherlands, New Zealand and Taiwan. 1 1 The member organizations that are signatories to this brief include the Actual Innocence Clinic, Alaska Innocence Project, Arizona Innocence Project, Arizona Justice Project, The Association in Defence of the Wrongly Convicted (Canada), California Innocence Project, Center on Wrongful Convictions, Committee for Public Counsel Services Innocence Program, Connecticut Innocence Project/Post-conviction Unit, The Duke Center for Criminal Justice and Professional Responsibility, The Exoneration Initiative, Georgia Innocence Project, Hawai'i Innocence Project, Idaho Innocence Project, Illinois Innocence Project, Innocence Project, Innocence Project France, Innocence Project at UVA School of Law, Innocence Project New Orleans, Innocence Project New Zealand, Innocence Project Northwest Clinic, Innocence Project of Florida, Innocence Project of Iowa, Innocence Project of Texas, Irish Innocence Project at Griffith College, Italy Innocence Project, Justicia Reinvindicada, Kentucky Innocence Project, Knoops' Innocence Project (the Netherlands), Life After Innocence, Loyola Law School Project for the Innocent, Michigan Innocence Clinic, Michigan State Appellate Defender Office, Wrongful Conviction Units, Mid-Atlantic Innocence Project, Midwest Innocence Project, Minnesota Innocence Project, Mississippi Innocence Project, Montana Innocence Project, Nebraska Innocence Project, New England Innocence Project, New Mexico Innocence and Justice Project at the University of New Mexico School of Law, North Carolina Center on Actual Innocence, Northern California Innocence Project, Office of the Public Defender, State of Delaware, Office of the Ohio Public Defender, Wrongful Conviction Project, Ohio Innocence Project, Oklahoma Innocence Project, Osgoode Hall Innocence Project (Canada), Pennsylvania Innocence Project, Reinvestigation Project, Resurrection After Exoneration, Rocky Mountain Innocence Center, Sellenger Centre Criminal Justice Review Project (Australia), Taiwan Association for Innocence, Thurgood Marshall School of Law Innocence Project, University of Baltimore Innocence Project Clinic, 2

10 Its affiliate, the Innocence Project, is an organization dedicated primarily to providing pro bono legal and related investigative services to indigent prisoners whose actual innocence may be established through post-conviction evidence. It has a specific focus on exonerating long-incarcerated individuals through use of DNA evidence, including newly-developed DNA testing methods. Amici are dedicated to improving the accuracy and reliability of the criminal justice system in future cases. To that end, Amici have researched the causes of wrongful convictions and, based on this research, pursue legislative, judicial, and administrative reform initiatives designed to enhance the truthseeking functions of the criminal justice system. Eyewitness misidentification has been the leading contributing cause of wrongful convictions, occurring in 236 of the 330 wrongful convictions identified through post-conviction DNA testing. Thus, the University of British Columbia Law Innocence Project (Canada), University of Miami Law Innocence Clinic, Wake Forest University Law School Innocence and Justice Clinic, West Virginia Innocence Project, Western Michigan University Cooley Law School Innocence Project, Wisconsin Innocence Project, Witness to Innocence, and Wrongful Conviction Clinic at Indiana University School of Law. 3

11 Network and the Innocence Project have a compelling interest in ensuring that courts apply a legal framework that adequately protects criminal defendants from the substantial risk of wrongful conviction based on misidentification. In this case, Amici seek to present a broad perspective on the issues presented in the hope that the Court will adopt a legal framework that will minimize the risk of future wrongful convictions. STATEMENT OF THE CASE The Innocence Network adopts the facts as recited in the brief of Defendant-Appellee Kyle Johnson. SUMMARY OF ARGUMENT Amici the Innocence Network and the Innocence Project submit this amici curiae brief in response to the Court's questions presented. In Part I and subpart A, Amici discuss this Court's holding in Jones, 423 Mass. 99, where this Court correctly concluded that the admissibility of eyewitness identification testimony that is tainted by suggestive circumstances "should not turn on whether government agents had a hand in causing" the suggestive circumstances. Id. at 109. Jones was prescient in that it based its holding on reliability and 4

12 fundamental fairness, rather than the cause of suggestive circumstances. Social science research has confirmed that eyewitness identifications that are tainted by suggestive circumstances are unreliable, regardless of the cause of the suggestion. This section concludes that this Court should reaffirm and expand upon Jones in this case. (pp ) Subpart B of Part I discusses the recommendations of the Supreme Judicial Court's Study Group on Eyewitness Evidence, and concludes that certain of the Study Group's recommendations should be adopted in this case, with slight modifications. For example, the Study Group recommended that courts suppress unreliable identifications, regardless of whether there are suggestive circumstances, and regardless of the cause of suggestive circumstances, if any. Social science research confirms that factors other than suggestion can render an eyewitness identification highly unreliable, and that suggestion caused by private actors can be just as problematic as suggestion caused by state actors. Actual cases of wrongful conviction also confirm that these recommendations are necessary. Amici therefore respectfully submit that this Court should expand upon 5

13 Jones as recommended by the Study Group. Subpart B also argues that this case presents an ideal opportunity to revisit Jones and to consider the SJC Study Group's recommendations, because this case squarely presents these issues, and because these issues were properly preserved here. (pp ) Subpart C presents proposed modifications to the Study Group's recommendations. Specifically, Amici respectfully request that this Court abandon the "confrontation" and "uncorroborated" requirements for pretrial hearings. Social science research and actual cases of wrongful conviction confirm that these modifications are appropriate and necessary. Amici also urge this Court to adopt a larger framework for approaching eyewitness testimony that would guide the lower courts in utilizing a burden-shifting regime, considering the totality of the circumstances, and in implementing intermediate remedies. (pp ) Finally, Part II argues that the "independent source doctrine" is undermined by social science and by this Court's recent decisions. The doctrine is premised on flawed assumptions about the way memory is stored and the way witnesses report on the circumstances of the crime and prior identifications. 6

14 As a result, Amici urge this court to adopt an exclusionary rule for all in-court identifications that arise out of impermissible out-of-court identifications. (pp ) ARGUMENT This Court has "long recognized" that mistaken eyewitness identifications are the "'primary cause of erroneous convictions, outstripping all other causes combined."' Commonwealth v. Gomes, 470 Mass. 352, 359, (2015) (quoting Commonwealth v. Martin, 447 Mass. 274, 293 (2006) (Cordy, J., dissenting)); Commonwealth v. Johnson, 420 Mass. 458, 465 (1995). Indeed, a study of the first 250 DNA exonerations concluded that over 75 percent of those wrongful convictions involved mistaken eyewitness identification. Brandon Garrett, Eyewitness and Exclusion, 65 Vand. L. Rev. 451, 472 (2012). The Court's recognition of the "dangers" of eyewitness testimony, Johnson, 420 Mass. at 465, is based on forty years of scientific research indisputably establishing that eyewitness identifications can be undermined by a number of factors, many of which are beyond the control of state actors. For example, this Court has recognized a "near consensus" among social 7

15 scientists that witnesses' memories are "'highly malleable and can readily be altered by information received by witnesses both before and after an identification procedure.'" Gomes, 470 Mass. at 373 (quoting Report of the Special Master at 30-31, State v. Henderson, 27 A.3d 872 (N.J. 2011) (No. A-8-08)). Other factors beyond the control of law enforcement that social science research has shown to impact reliability, known as "estimator" variables, include high stress, the presence of a weapon, whether the perpetrator wore a disguise, whether the witness and perpetrator are of different races or ethnicities, and the passage of time. See, e.g., Henderson, 27 A.3d at ; SJC Report at These principles have been borne out in actual cases of wrongful conviction. In the case of Herman Atkins, for example, a rape victim erroneously identified Mr. Atkins as her attacker after seeing a wanted poster with his photo in the police department. 2 "SJC Report" refers to the Report and Recommendations to the Justices of the Supreme Judicial Court Study Group on Eyewitness Evidence (the "Study Group"). Supreme Judicial Court Study Group on Eyewitness Evidence, Report and Recommendations to the Justices (July 25, 2013), available at 8

16 Though not caused by law enforcement, these suggestive circumstances led the woman to honestly, but mistakenly, identify Mr. Atkins at trial, with confidence. Based on this eyewitness testimony, which was corroborated by two other mistaken eyewitnesses, Mr. Atkins was sentenced to 47 years and 8 months in prison for a rape he did not commit; he remained in prison for 13 years before DNA evidence exonerated him. Fred Dickey, Worst-Case Scenario, Los Angeles Times (June 25, 2000), Suggestive circumstances, even when beyond the control of law enforcement, tend to artificially inflate a witness's testimony regarding the witness's opportunity to view the suspect and confidence in the identification, as scientists have shown. (SJC Report at ) As a result, suggestion can have an especially pernicious effect on the reliability of such identifications. Moreover, although suggestion that occurs during law enforcement-orchestrated identification procedures is often recorded or witnessed by defense counsel or the accused himself, 9

17 there is often no record of suggestion that occurs outside of the control of state actors. Suggestion can come from a variety of private actors, including the news media, co-witnesses, friends and family, and social media. The Jones Court ref erred to a chance viewing of the defendant in the news media as "neutral." Jones, 423 Mass. at In fact, media reports are often anything but neutral. See, e.g., State v. Lawson, 291 P.3d 673, 698 (Or. 2012) ("It was not until after [the witness] had seen a newspaper article with a picture of defendant, and was later brought by police to.. view defendant in person, that she was able to identify him. Those instances introduce further uncertainty as to whether [the identification] was based on her brief initial viewing of the perpetrator, or on the numerous subsequent viewings of defendant under circumstances that were highly suggestive of his guilt."); People v. Santiago, 958 N.E.2d 874, 878, (N.Y. 2011) (holding that, where newspaper article made it clear that defendant had been identified by the victim and had been arrested, and where newspaper photo showed defendant in handcuffs, accompanied by police officers, it was possible that witness's later 10

18 identification was tainted by the newspaper). The danger of suggestive circumstances lies even in the most seemingly benign situations, which is precisely why the Court should address the issue here. I. THIS COURT SHOULD REVISIT, REAFFIRM, AND EXPAND UPON COMMONWEALTH V. JONES IN LIGHT OF THE 2013 SUPREME JUDICIAL COURT STUDY GROUP'S REPORT AND RECOMMENDATIONS Nearly twenty years ago, in Commonwealth v. Jones, this Court recognized that the reliability of eyewitness identifications can be undermined by suggestive circumstances, regardless of whether they are caused by law enforcement. 423 Mass. at There, the Court found that an accidental viewing of the defendant by the witness, where the defendant was shackled to a confessed participant, and the two matched the descriptions provided by the witness, constituted a "highly suggestive" confrontation. Id. at 105. Given the absence of an independent source for the witness's in-court identification of the defendant, this Court found its admission to be an error that undermined the fundamental fairness of Jones' trial. Id. at The Court correctly concluded, If a witness is involved in a highly suggestive confrontation with a defendant 11

19 and that witness's in-court identification of the defendant is not shown to have a basis independent of that confrontation, the admissibility of the witness's proposed testimony identifying the defendant should not turn on whether government agents had a hand in causing the confrontation. The evidence would be equally unreliable in each instance. Id. at 109. Notwithstanding this conclusion, the Court did not set forth procedures for reliability hearings when no state action was involved in suspect identifications. This Court should take this opportunity to do so now. A. The Jones Court Correctly Focused on the Reliability and Fundamental Fairness of Eyewitness Identifications, Regardless of State Action The Jones Court began its analysis from a position of measured skepticism arising out of the well-settled proposition that eyewitness identification testimony is particularly problematic. Indeed, the Court quoted the Supreme Court's famous observation in United States v. Wade, 388 U.S. 218, 228 (1967), that "the vagaries of eyewitness identification are well-known; the annals of criminal law are rife with instances of mistaken identification." 423 Mass. at 109 (citing Johnson, 420 Mass. at 466). 12

20 Precisely because of this risk of misidentification, the Court in Jones appropriately looked to reliability as the touchstone of an admissible identification. The key consideration is fairness. From that perspective, an unreliable identification is equally problematic whether it is a function of abusive or otherwise suggestive law enforcement tactics or, as was the case in Jones, mere happenstance. See id. at 109. This Court in Jones also correctly acknowledged that the tools of the trial process for combatting weak evidence cannot be relied upon to protect against the dangers of eyewitness misidentification. The Court noted that "[t]his is not a case in which crossexamination and a judge's jury instruction concerning eyewitness identification testimony can fairly protect the defendant from the unreliability of [the witness's] identification.n Id. at 110. The Court thus implicitly recognized the tendency of juries to systematically apply more weight to eyewitness identification evidence than is warranted, even when presented with information calling the identification into serious doubt: 13

21 [An instruction from the Judge] would not.. have mitigated the effect of the admission of an eyewitness identification that the judge had concluded was based on highly suggestive confrontations and was not shown to have a reliable basis independent of these confrontations. [The witness's] identification of the defendant had the potential of leading the jury to a different conclusion than they would have reached without it. No instruction from a judge could have confidently guarded against that risk. Id. at 111. Social science research confirms that jurors tend to "overbelieve" eyewitness testimony. Jennifer N. Sigler & James V. Couch, Eyewitness Testimony and the Jury Verdict, 4 N. Am. J. Psychol. 143, 146 (2002). Indeed, identification evidence "has been shown to be comparable to or more impactive than physical evidence. and even sometimes confession evidence." Melissa Boyce et al., Belief of Eyewitness Identification Evidence, in 2 Handbook of Eyewitness Psychology: Memory for People 501, 505 (R.C.L. Lindsay et al. eds., 2007). Notwithstanding the acknowledgement that jury instructions may not adequately educate jurors about the dangers of eyewitness testimony, the Jones Court did not establish any alternative procedures or mechanisms to guide lower courts. As a result, courts 14

22 have subsequently struggled to find consistency in applying the holding of Jones. Without a clear legal framework for analyzing the admissibility of eyewitness evidence, defendants are unable to challenge unreliable identifications that are the product of suggestive circumstances caused by private actors, and the lower courts' treatment of these identifications is unpredictable. A framework for consistently analyzing these identifications would reduce the likelihood of wrongful conviction and would give defendants equal opportunity to challenge these identifications. B. The SJC Study Group's Recommendations Provide a Framework for Courts to Scrutinize Unreliable Eyewitness Identifications As recently as 2011, this Court cited Jones with approval, recognizing its holding that "common-law principles of fairness dictate that an in-court identification arising from 'especially suggestive circumstances' should not be admitted in evidence even where the police were not responsible for the suggestive confrontation." Commonweal th v. Walker, Mas s , ( ). The Walker Court also announced its "willingness to revisit our jurisprudence," and convened a study group to 15

23 recorrunend procedures for examining unreliable eyewitness identifications. Id. at 604 n.16, 606. The Study Group issued its Report and Recorrunendations to the Justices of the Court on July 25, The Study Group found it problematic that the current case law ''[l]eft unclear" whether a pretrial screening is permissible "where the identification may well be unreliable.. because of the circumstances of the viewing, e.g., estimator variables." (SJC Report at 40.) The Study Group explained: [T]he holding in Jones - that certain eyewitness evidence is too unreliable to enter into evidence - has never been sufficiently developed. Given what scientific research now shows about the impact that certain estimator variables can have on reliability, and in light of persistent myths about how memory works, the holding of Jones can no longer remain just a tantalizing possibility; it must be elaborated. (Id. at 41 (emphasis added).) 1. The Study Group Recommended that Courts Suppress Unreliable Identifications, Regardless of Suggestive Circumstances In response to this uncertainty, the Study Group recorrunended a more concrete procedure for directly addressing these types of identifications. Specifically, it recommended that trial courts should 16

24 suppress both out-of-court and in-court identifications where the defendant proves by a "preponderance of the evidence that the pretrial eyewitness identification is unreliable, taking into account the totality of the circumstances in the case at bar, including system and estimator variables." (Id. at 111.) Under this approach, courts can take estimator variables into account at pretrial hearings and have the flexibility to exclude identifications that are "too unreliable" to admit into evidence, regardless of the presence of, or impact of, factors within the control of law enforcement, known as "system variables." 3 Gary L. Wells & Deah S. Quinlivan, Suggestive Eyewitness Identification Procedures and the Supreme Court's Reliability Test in Light of Eyewitness Science: 30 Years Later, 33 Law & Hum. Behav. 9, 8-9 (2008). 3 To the extent this policy contradicts this Court's holding in Commonwealth v. Payne, 426 Mass. 692, & n.3 (1998), we respectfully submit that Payne should be overruled on the ground that the SJC Report demonstrates that unreliable identifications can have an unwarranted impact on jurors, even if there were no suggestive circumstances involved in the identification. (See, e.g., SJC Report at 18, ) 17

25 2. The Study Group Also Reconunended that Courts Hold Pretrial Hearings to Vet Eyewitness Identifications, Even Where State Actors Did Not Cause Suggestive Circumstances The Study Group also called for an expanded pretrial judicial inquiry into the reliability of eyewitness identifications, with "the focus. on ensuring that reliable evidence is presented to the fact finder whether or not misconduct is implicated." (SJC Report at 3, 81 ("[W]itness memory is equally susceptible to contamination by nonstate actors."); see also id. at (recognizing that postidentification feedback ''such as news accounts identifying the suspect as the perpetrator, conversations with other witnesses, or pretrial witness preparation sessions" can "falsely inflate witness confidence in the reports they tender regarding many of the factors conunonly used by courts and jurors to gauge eyewitness reliability").) The Study Group expanded upon Jones by reconunending that a pretrial reliability hearing be required in certain cases where suggestive circumstances were caused by someone other than a member of law enforcement. All together, the Study Group reconunended expanding the circumstances in which a pretrial screening is 18

26 warranted into four specific categories, two of which do not require any showing of police misconduct, consistent with the holding and rationale of Jones. 4 These two categories are: (1) when the defendant demonstrates that a witness was "involved in a highly suggestive confrontation with the defendant" regardless of state action; and (2) when the out-ofcourt identification is "uncorroborated" and the defendant "makes a showing of.. estimator variables[] casting doubt on the reliability of the identification." (Id. at 110 (emphasis added).) 5 Using pretrial hearings to test eyewitness identifications is consistent with this Court's stated 4 To the extent these recommendations contradict this Court's holding in Commonwealth v. Jules, 464 Mass. 478, 490 (2013), we respectfully submit that Jules should be overruled, because the Court's rationale was based on the erroneous assumptions that cross-examination and jury instructions can cure the defects of an unreliable identification, and that jurors have the knowledge and tools necessary to accurately discount unreliable identifications. (See, e.g., SJC Report at 3-4 & n.5, 40 (citing Jones, 423 Mass. at 110).) We also note that the question of admissibility of the identification in Jules was not preserved and arose though the lens of an ineffective assistance of counsel claim. 5 The Study Group also recommended two other circumstances where courts should require pretrial hearings. (SJC Report at 110.) We support those recommendations, subject to the proposed modifications set forth in our comments to the report, which we attach as Appendix A. (See App'x at 8-10.) 19

27 goal of ensuring that only reliable evidence is admitted, reflects the consensus of social science research, and is in keeping with the critical gatekeeping function of trial courts. Seer e.g., N.E. Physical Therapy Plusr Inc. v. Liberty Mut. Ins. Co., 466 Mass. 358, 359 (2013) (holding that, based on "the gatekeeper role of a trial judge,. it is within a judge's discretion to consider the reliability of evidence"); Commonwealth v. Polk, 462 Mass. 23, 31 (2012). 3. This Case Presents an Ideal Opportunity for Addressing the Study Group's Recommendations on the Admissibility of Eyewitness Testimony Earlier this year, this Court stated, "With the Study Group Report completed and the comments to that report received, it is now time to do what we declared we were willing to do with respect to. jury instructions." Gomes, 470 Mass. at So too, 6 This Court has also signaled its approval of the Study Group's report when, at the recommendation of the Study Group, it appointed twelve individuals to a newly constituted Standing Committee on Eyewitness Identification, which will "periodically meet to assess the evolving science and law of eyewitness identification and make appropriate recommendations to the Justices." Massachusetts Court System, Supreme Judicial Court Announces New Standing Committee on Eyewitness Identification, (Jan. 12, 2015), 20

28 here, with respect to the admissibility of unreliable identifications. This Court has limited its "willingness to revisit [its] jurisprudence," however, to cases where the issue in question has been preserved and can be examined directly, instead of solely through the lens of an ineffective assistance of counsel claim. Walker, 460 Mass. at 606. The case at bar is such a case. Here, the trial judge excluded both the out-ofcourt and in-court identifications, and the Commonwealth appealed to the Appeals Court. Because this case comes before this Court on interlocutory appeal, the Court will not be burdened by a complicated trial record. In addition, the case cleanly presents the question of the judicial treatment of identifications tainted by suggestive circumstances that were not caused by law enforcement, giving this Court the opportunity to revisit, reaffirm, and expand upon Jones. announces-new-standing-committee-on-eyewitnessidentification.html. 21

29 C. The SJC Study Group's Recommendations Should Be Adopted with Slight Modifications to Most Appropriately Reflect Principles of Social Science The Study Group unquestionably had the right goals in mind, and correctly identified the issues that need to be addressed. We respectfully submit that these recommendations should now be adopted with slight modifications. The Study Group recommended that defendants should be entitled to a pretrial screening in at least four sets of circumstances. In one of these four categories, the Study Group recommended a pretrial hearing "when the defendant makes a showing that a witness was involved in a highly suggestive confrontation with the defendant independent of any police involvement." Jones, 423 Mass. 99).) (SJC Report at 110 (citing Although the Study Group was correct in relying on the Court's holding in Jones that the admissibility of a witness's identification of the defendant "should not turn on whether government agents had a hand" in the identification, Jones, 423 Mass. at 109, its proposal is incomplete. First, suggestion that emanates from a source other than a state actor should not be tested with a 22

30 higher standard - "highly suggestive" - than the "suggestive" standard reserved for confrontations with state actors. As described above, and borne out by the social science research and DNA exonerations, human memory is affected by suggestion no matter the source. For this reason, the standard necessary to receive a pre-trial hearing should be no different based on the source of suggestion. Second, suggestion may occur without a "confrontation" if the witness has been exposed to, for example, rumors that the defendant is the culprit, or has made his or her own investigation in person, using social media or websites such as those that collect mug shots. 7 Accordingly, we respectfully 7 The case of Wilton Dedge is illustrative of suggestive circumstances that occured without a "confrontation" with the defendant. There, a rape victim saw a man in a convenience store four days after the crime. She told her sister that the man resembled her attacker, but that her attacker was shorter. The victim's sister told her that the man in the store was named Walter Hedge. Police arrested a man named Walter Dedge, but when the victim's sister saw Walter Dedge, she said that the man in the store was actually Walter's brother Wilton Dedge. The victim then identified Wilton Dedge in a photographic array. The victim's identification of Mr. Dedge was "corroborated" by microscopic hair comparison, snitch testimony, and dog sniffing evidence. Wilton Dedge served 22 years before DNA evidence exonerated him. Innocence Project, Wilton Dedge, 23

31 suggest that the Court expand upon the Study Group's recommendation to include not only a "suggestive confrontation with the defendant" but also any suggestive information obtained by the witness from other sources. Compare, e.g., State v. Chen, 27 A.3d 930, 943 (N.J. 2011) (adopting rule that, "to obtain a pretrial hearing, a defendant must present evidence that the identification was made under highly suggestive circumstances that could lead to a mistaken identification"). 8 Additionally, the Study Group recommended a pretrial screening where "the pretrial eyewitness identification is uncorroborated" and the defendant makes a showing that estimator variables undermine the identification's reliability. (SJC Report at 110.) But the Study Group's insistence that this protection apply only in the case where the identification is not imprisonment/wilton-dedge (last visited Sept. 22, 2015). 8 In Chen, the court held that the identification circumstances were highly suggestive and warranted a pretrial hearing. 27 A.3d at 944. There, the witness's husband suspected that his ex-girlfriend was the assailant and showed the witness multiple pictures of his ex-girlfriend from her website and suggested that she might be the perpetrator on multiple occasions. Id. 24

32 corroborated by other evidence actually compounds the problem the recommendation was meant to prevent. The reality is that convictions in many DNA exoneration cases do not rest solely on a single witness's identification. Rather, there often are multiple contributing causes of wrongful conviction, including multiple mistaken eyewitnesses, flawed forensics, mistaken informant testimony, and/or false confessions. 9 Of the 236 DNA exonerations stemming from convictions involving mistaken identifications, 77 (33 percent) involved multiple witnesses misidentifying the same innocent person and 124 (53 percent) involved at least one other evidentiary error (for example, forensics, false confessions, 9 For example, in the case of Dennis Maher, a rape victim in Lowell, Massachusetts told police that the perpetrator attacked her with a knife, and that he was wearing a red, hooded sweatshirt and a khaki militarystyle jacket. The police stopped Dennis Maher on the night of the attack. He was wearing a red, hooded sweatshirt, and an army field jacket and military issue knife were found in his car. The victim then identified Mr. Maher in a photographic array. The sweatshirt, jacket, and knife "corroborated" the identification, and Mr. Maher was convicted. After 19 years in prison, Mr. Maher was exonerated on the basis of DNA evidence. The Innocence Project, Dennis Maher, (last visited Sept. 22, 2015); see also supra n.7 (identification of Wilton Dedge was "corroborated" by microscopic hair comparison, snitch testimony, and dog sniffing evidence). 25

33 informants) that "corroborated" the misidentification. Indeed, recent research suggests that multiple errors are common in wrongful conviction cases, and that such errors are not independent of one another; one error often corrupts other, seemingly unrelated pieces of evidence. See Saul M. Kassin, Daniel Bogart, and Jacqueline Kerner, Confessions that Corrupt: Evidence from the DNA Exoneration Case Files, 23 Psychol. Sci. 41, (2012). 10 Kassin et al. established the corrupting effect of erroneous evidence first by examining the types of cases most likely to have multiple types of erroneous evidence, and then by examining the order in which that evidence emerged. They found that eyewitness 1 0 Numerous exonerated defendants, whose wrongful convictions were based on eyewitness identifications and "corroborating" evidence, were convicted in Massachusetts. See, e.g., Commonwealth v. Cowans, 52 Mass. App. Ct. 811, 813 (2001) (exonerated defendant's conviction rested on a flawed fingerprint comparison and two mistaken eyewitnesses); Maher v. Town of Ayer, 463 F. Supp. 2d 117 (D. Mass. 2006) (exonerated defendant was identified by three separate victims); Miller v. City of Boston, 297 F. Supp. 2d 361, 365 (D. Mass. 2003) (exonerated defendant's conviction rested on victim's identification, which was corroborated by a flawed serology test); see also Kassin et al., 23 Psychol. Sci. at 44 (suggesting that "the assumption that the other evidence on record is independent of the confession.. is incorrect[;] that the other evidence may be tainted[;] and that the appearances of corroboration and sufficiency may be more illusory than real"). 26

34 misidentification cases were second-most likely to involve multiple types of erroneous evidence (following false confession cases). 11 In addition, they found that, in 46 cases involving multiple types of faulty evidence, confession evidence appeared first in 65 percent of the cases, and eyewitness evidence appeared first in 26 percent of the cases. Id. at 43 & Table 2. These findings suggest that eyewitness identification evidence (and false confession evidence) corrupt other kinds of evidence. Id. As Professors Kassin, Bogart, and Kerner wrote, "[E)yewitnesses tainted by extrinsic information cannot accurately estimate the extent of that influence, which suggests that self-report cannot be used to diagnose corruption once it occurs." Id. at In addition, a rule that predicates the admission of a particular piece of evidence - or whether the defendant is even entitled to a hearing challenging the admission of such evidence - on the strength or 11 Kassin found that 45 percent of the 163 eyewitness identification cases he examined involved forensic-science error, 6 percent involved informant error and 52 percent involved no other errors. This analysis excluded cases that involved both eyewitness misidentification and false confession evidence. Id. at & Table 1. 27

35 weakness of the Corrunonwealth's other evidence, before that evidence has even been proffered or subject to the crucible of adversarial testing, raises significant due process concerns. See, e.g., Holmes v. South Carolina, 547 U.S. 319, 331 (2006). Accordingly, a defendant should be entitled to a pretrial hearing if he makes a showing that estimator variables cast doubt on the reliability of an identification, regardless of the perceived strength of the other evidence against him. In addition, the Court should provide guidance to the lower courts for conducting pretrial hearings in a manner that is consistent with prevailing social science research. First, as the proponent of the identification evidence, the Commonwealth should carry the initial burden of establishing baseline reliability. Doing so would require eyewitness identifications to be treated the same as other trace evidence. (See SJC Report at 45 (noting that some members of the Study Group "felt that the trace evidence analogy most accurately captures the nature of eyewitness evidence and forms the appropriate basis for analyzing reliability").) The Study Group declined to reconunend this approach due to concerns 28

36 that it would place a limitation on defendants' Massachusetts due process rights, as articulated in Walker. (Id. at 46.) That concern does not apply to pretrial hearings that are warranted for reasons unrelated to state action, however. For these pretrial hearings, where due process does not apply under current law, 12 fundamental fairness requires that the proponent of the evidence, who has control of most of the facts relating to reliability, carries the initial burden. Second, this Court should encourage lower courts to consider the totality of the circumstances, including all relevant system and estimator variables, in light of evolving social science research, in assessing the reliability of eyewitness identifications. The lower courts should not consider factors that have no basis in scientific research, such as self-reports about the witness's opportunity to view the perpetrator or about his/her confidence in the identification. Wells & Quinlivan, 33 Law & Hum. Behav. at 9 ("Psychological scientists are highly 12 Although the issue is beyond the scope of the Court's questions presented, the Court could, in this case, conclude that due process protects defendants from the admission of eyewitness identifications that are tainted by suggestion caused by private actors. 29

37 skeptical of retrospective self-reports because of well-known tendencies for such reports being at odds with objective facts."). This approach should be flexible in light of the fact that scientific research is dynamic. The lower courts should also be free to consider new variables on which there is a "general acceptance" in the scientific community. 13 Finally, as the Study Group recommended, this Court should encourage the lower courts to implement intermediate remedies for reducing the likelihood that an unreliable identification will lead to a wrongful conviction, rather than restrict themselves to outright suppression as the sole remedy. (SJC Report at 48, 112 ("[T]he traditional 'in/out' option on the admission of eyewitness evidence is an improper restraint on the exercise of judicial discretion.").) When courts do not find that identification evidence is so unreliable as to require suppression, but have some concerns about reliability or the prejudicial 13 This standard should be similar to the Frye standard for admission of expert testimony, and should not be as high as the "near consensus" standard the Court applied in Gomes, where the Court was considering how to instruct a jury, because the Gomes "near consensus" standard is reserved for scientific principles that are so generally accepted that juries are required to accept them. See Gomes, 470 Mass. at

38 effect of the evidence, they should consider other remedies, such as the admission of expert testimony, Lawson, 291 P.3d at 761; partial suppression, see id. at 759; preclusion of in-court identifications, see infra at pp ; and science-based jury instructions given at appropriate times (including before eyewitness testimony), see generally Gomes, 470 Mass In adopting a new legal framework for the regulation of eyewitness identification evidence, this Court should not be confined to the recommendations in the SJC Report. Rather, this Court should also reaffirm the requirement that trial courts exercise their gatekeeping functions to ensure that only reliable evidence is admitted. See, e.g., N ~ E. Physical Therapy Plus, 466 Mass. at Accordingly, we respectfully propose, in addition to the modified recommendations of the SJC Report discussed above, that this Court adopt the National 14 See also Chen, 27 A.3d at 932 (considering "[r]ecent social science research reveal[ing] that suggestive conduct by private actors.. can undermine the reliability of eyewitness identifications and inflate witness confidence in light of the court's traditional gatekeeping role to ensure that unreliable, misleading evidence is not presented to jurors"). 31

39 Academy of Sciences' recommendation that trial courts "make basic inquiries when eyewitness identification evidence is offered." (NAS Report at ) 15 As the NAS Report recommended, such an inquiry, although necessarily case-specific, would include, "at a minimum, [consideration of] prior lineups, what information had been given to the eyewitness before the lineup, what instructions had been given to the eyewitness in connection with administering the lineup, and whether the lineup had been administered 'blindly.'" (Id. at 110.) If these basic inquiries raise issues, (Id.) [A] judge could conduct a pre-trial hearing to review the reliability and admissibility of eyewitness identification evidence and to assess how it should be treated at trial if found admissible. If indicia of unreliable eyewitness identifications are present, the judge should apply applicable law in deciding whether to exclude the identifications or whether some lesser sanction is appropriate. 15 "NAS Report" refers to the 2014 report of the National Academy of Sciences on eyewitness identifications. See National Academy of Sciences, Identifying the Culprit: Assessing Eyewitness Identification, available at (last visited Sept. 22, 2015). 32

40 Therefore, we respectfully submit that this Court should reaffirm and expand upon Jones, adopt the Study Group's recommendations for pretrial hearings with the above modifications, and provide the lower courts with guidance for conducting pretrial hearings and for implementing intermediate remedies. II. THIS COURT SHOULD REVISIT AND ELIMINATE THE INDEPENDENT SOURCE DOCTRINE IN LIGHT OF THIS COURT'S DECISIONS IN COMMONWEALTH V. CRAYTON AND COMMONWEALTH V. COLLINS AND THE SJC STUDY GROUP'S REPORT AND RECOMMENDATIONS Last year, this Court in Crayton recognized the inherently suggestive nature of in-court identifications. Commonwealth v. Crayton, 470 Mass. 228, 235 (2014). Indeed, this Court acknowledged that in-court identifications are as suggestive as showups, noting the requirement of a showing of "good reason" before allowing show-up identifications. Id. at Accordingly, this Court in Crayton and Collins adopted a new framework that bars in-court identifications where no out-of-court identification occurred and where the eyewitness "made something less than an unequivocal positive identification of the defendant during a nonsuggestive identification procedure." Commonwealth v. Collins, 470 Mass. 255, 265 (2013); Crayton, 470 Mass. at These 33

41 rulings reflected the SJC Study Group's conclusions regarding the inherently suggestive nature of in-court identifications and went a step further than its recorrunendation that "in-court identification[s should] not be permitted except, in the judge's discretion, on redirect examination, in rebuttal, or in other circumstances where the defendant challenges the witness's ability to make such [an] identification.u (SJC Report at 48, 113.) The National Academy of Sciences likewise criticized the use of in-court identifications, emphasizing that they "do not reliably test an eyewitness' memoryu due to the fact that the witness can ''easily see where the defendant is sitting.u (NAS Report at 36 n.28.) Based on these findings, the NAS recorrunended that identifications should not occur for the first time in the courtroom, and that if an incourt identification occurs at trial, the jury should be informed fully about the facts of any prior identifications. (Id. at ) Notwithstanding the above, under this Court's decision in Commonwealth v. Botelho, 369 Mass. 860 (1976), a witness whose out-of-court identification was excluded as impermissibly suggestive may 34

42 nonetheless be permitted to make an in-court identification if the Commonwealth can prove that such in-court identification is the product of a "source independent" of the tainted procedure and is reliable. Id. at 868. However, like the two-step Manson test, Manson v. Brathwaite, 432 U.S. 98 (1977), developed just one year later, the "independent source" doctrine is outdated, fundamentally flawed, and undermined by prevailing social science research. The dangers of the "independent source" doctrine are rooted in the impact of suggestive circumstances. These dangers are present regardless of the source of suggestion or if those circumstances rise to the level required for suppression. Scientific research has conclusively demonstrated that one cannot separate the effects of suggestion from a person's memory of an event. Indeed, in one of the most well-known of all of the studies on eyewitness identification, "Good, You Identified the Suspect, " Gary Wells and colleagues demonstrated that suggestive circumstances alter participants' memory for the event and memory of the identification procedure, in addition to inflating confidence in that memory. See Gary L. Wells & Amy L. Bradfield, "Good, You Identified the Suspect": 35

43 Feedback to Eyewitnesses Distorts their Reports of the Witnessing Experience, 83 J. Applied Psychol. 360 ( 1998). Subsequent research has demonstrated that other types of suggestion or post-event information have similar effects on memory. 16 This research makes clear that when a witness is subject to suggestion or postevent information, it is impossible to determine what part of the memory is "independent" of that suggestion or information and what is a product of it. Perhaps most dangerously, this research shows that people are unaware that their memories have been changed. As the New Jersey Supreme Court concluded in Henderson, "The irony of the current test is that the more suggestive the procedure, the greater the chance eyewitnesses 16 Steve D. Charman et al., The Dud Effect: Adding Highly Dissimilar Fillers Increases Confidence in Lineup Identifications, 35 Law & Hum. Behav. 479 ( 2 011) (concluding that the presence of "highly dissimilar fillers inflates witnesses' confidence in a mistaken identification"); Steven E. Clark et al., Lineup Administrator Influences on Eyewitness Identification Decisions, 15 J. Experimental Psychol.: Applied 63 (2009) (concluding that even statements that are ostensibly cautionary can increase the likelihood of an inaccurate identification) ; Michael R. Leippe et al., Cueing Confidence in Eyewitness Identifications: Influence of Biased Lineup Instructions and Pre-Identification Memory Feedback Under Varying Lineup Conditions, 33 Law & Hum. Behav. 194 (2009). 36

44 will seem confident and report better viewing conditions. Courts in turn are encouraged to admit identifications based on criteria that have been tainted by the very suggestive practices the test aims to deter." Henderson, 27 A.3d at 918. In addition, the "danger of confusion of source memory" whenever an eyewitness participates in multiple identification procedures makes the notion of an "independent source" for a memory a fiction. See Collins, 470 Mass. at 262 n.9. In other words, the "independent source" doctrine is based on a flawed premise that memory is like a videotape and can be spliced into recordings of different moments in time. (See SJC Report at 61 ("It is a common misconception that a person's memory operates like a videotape, recording an exact copy of everything a person sees").) See also Gomes, 470 Mass. at 369 ("The central principle that has emerged from over 2,000 published studies over the past thirty years is that 'memory does not function like a videotape, accurately and thoroughly capturing and reproducing a person, scene or event Memory is, rather[,] a constructive, dynamic and selective process.'"(quoting SJC Report at 15)). Prevailing social science 37

45 research teaches us that the impermissible out-ofcourt identification fatally taints all future identifications. (See NAS Report at 65 (finding that a "witness' inevitable interactions with law enforcement and legal counsel, not to mention communications from journalists, family, and friends, have the potential to significantly modify the witness' memory of faces encountered and of other event details at the scene of the crime").) See also Garrett, 65 Vand. L. Rev. at (deconstructing problems with independent source doctrine and concluding, "The very idea that a courtroom identification could be seen as 'independent' is anomalous.'"). Thus, because memories cannot be separated into independent parts, there can be no truly "independent" source for any future identification. This Court and the Study Group have recognized that numerous social science studies have highlighted flaws in the legal framework for approaching eyewitness testimony. The Study Group's recommendations and this Court's decisions implementing those recommendations have started the process of aligning the law with modern science and 38

46 correcting these flaws. Eliminating the "independent source" rule would be a further step in that direction. Because the "independent source" doctrine is inconsistent with the modern understanding of human memory, this Court should eliminate it and adopt an exclusionary rule that precludes all in-court identifications that arise out of impermissible outof-court identifications. CONCLUSION For all of the reasons listed herein, the Innocence Project and the Innocence Network respectfully urge this Court to: (1) Affirm the trial court's decision to exclude the out-of-court and in-court identifications of the Defendant; (2) Reaffirm Jones, holding that identifications arising out of "especially suggestive" circumstances should not be admitted even where law enforcement was not responsible for the suggestive circumstances; (3) Adopt the SJC Study Group's recommendations for the circumstances under which a pretrial hearing is appropriate, with slight 39

47 modifications to parts B and D that reflect the current state of social science research; (4) employ a burden-shifting regime that places the burden of establishing baseline reliability on the proponent of the evidence, adopt a totality-of-the-circumstances approach for scrutinizing eyewitness testimony, and encourage courts to utilize intermediate remedies; and (5) Eliminate the independent source doctrine and adopt an exclusionary rule that precludes all in-court identifications that arise out of impermissible out-of-court identifications. Dated: September 23, 2015 /s/ James L. Brochin James L. Brochin Jennifer H. Wu PAUL, WEISS, RIFKIND, WHARTON & GARRISON LLP 1285 Avenue of the Americas New York, New York, Telephone (212) Facsimile (212) jbrochin@paulweiss.com jwu@paulweiss.com Choate, Hall & Stewart LLP Two International Place Boston, MA Telephone (617) Facsimile (617) rjcinquegrana@choate.com 40

48 Karen A. Newirth THE INNOCENCE PROJECT 40 Worth Street Suite 701 New York, New York, Telephone (212) Facsimile (212) rg Attorneys for Amici Curiae the Innocence Project and the Innocence Network 41

49 CERTIFICATE OF COMPLIANCE Under Massachusetts Rule of Appellate Procedure 16(k), I, R.J. Cinquegrana, hereby certify that the preceding brief complies with the rules of court that pertain to the filing 42

50 CERTIFICATE OF SERVICE I, R.J. Cinquegrana, hereby certify that on September 23, 2015, I caused a true copy of the foregoing to be served by first-class mail on the following attorneys: Timothy J. Cruz District Attorney, Plymouth District Carolyn A. Burbine Assistant District Attorney BBO # Belmont Street Brockton, MA Edward Crane Attorney for the Defendant BBO # Law Off ice of Kevin Crane 104 Mount Auburn Street P.O. Box Cambridge, MA Thomas R. Glover Attorney for the Defendant 273 Hanover Street Suite 15 Hanover, MA I also certify that on September 23, 2015, I caused a true copy of the foregoing to be served by hand delivery to the Clerk for the Supreme Judicial Court. Signed under the perjury this 23rd day of 43

51 APPENDIX A

52 Tel Fax January 24, 2014 Christine P. Burak, Esq. Legal Counsel to the Chief Justice Supreme Judicial Court John Adams Courthouse One Pemberton Square Boston, MA Re: Supreme Judicial Court Study Group on Eyewitness Evidence Report and Recommendations to the Justices Dear Ms. Burak: The Innocence Project respectfully submits these comments to the Supreme Judicial Court Study Group on Eyewitness Evidence's Report and Recommendations to the Justices (hereinafter, the "Report") for the Court's consideration. The Innocence Project is a national litigation and public policy organization dedicated to exonerating wrongfully convicted individuals through DNA testing and reforming the criminal justice system to prevent future miscarriages of justice. The advent of DNA testing has proven that innocent individuals are, in fact, convicted. It has also enabled the Innocence Project to study the causes of these injustices and pursue legislative and administrative reform initiatives designed to enhance the truth-seeking functions of the criminal justice system - including preventing future wrongful convictions and identifying actual perpetrators. 1 In particular, the work of the Innocence Project has helped to expose the problem of mistaken identification as the leading cause of wrongful convictions, contributing to nearly 7 5 percent of such cases, including 1 In49 percent of the wrongful convictions proven by post-conviction DNA testing, our work also helped identify the real perpetrators of those crimes. Halfofthese true offenders are known to have committed a total of 139 additional violent crimes, including 76 rapes and 33 murders, following the arrest of the actually innocent person who was then erroneously prosecuted and convicted. Benjamin N. Cardozo School of Law, Yeshiva University

53 Innocence Project, Inc. January 24, 2014 Page 2 INNOCENCE PROJECT JJTiilf 1! fn!jjuhlilll!ijl!hjuflll lj t l ff[!uff Jll eight of the nine exoneration cases based on DNA evidence in Massachusetts. 2 The Innocence Project's extensive experience in mistaken eyewitness identification cases has led it to call for a variety of systemic reforms. These include improving police procedures so that officers are required to adhere to "best practices," proposing model legislation, and highlighting the need for expert testimony and expansive, science-based jury instructions to educate jurors about empirically-proven factors affecting the reliability of eyewitness identifications. The Innocence Project has served as amicus curiae in cases involving questions relating to the evaluation and use of eyewitness identification evidence around the country, most notably in State v. Henderson, 208 N.J. 208 (2011) and State v. Lawson, 352 Or. 724 (2012). The Innocence Project commends the Supreme Judicial Court for achieving a consensus among stakeholders in the criminal justice system that comprehensively addresses the reliability of eyewitness identification evidence at both the front end - its collection by law enforcement - as well as the back end - when and how it is used in courts. It is this type of comprehensive and interdependent reform that creates the best and most realistic conditions to improve the fairness and reliability of the criminal justice system. Such reform will protect innocents from wrongful conviction, enhance public safety, and increase public confidence in the accuracy of criminal trials. Accord Report at 11. The Study Group has carried out the Court's vision in a thoughtful, detailed and clear manner. The Report accurately describes and accounts for the findings of more than thirty years of peerreviewed scientific research and offers an important roadmap for reform that, if approved by the Supreme Judicial Court, will be of use not only in Massachusetts, but throughout the country as other jurisdictions follow Massachusetts' lead and engage in holistic eyewitness identification reform. Just as the Study Group built on the work of the supreme courts of New Jersey and Oregon, so too will courts, legislatures and law enforcement agencies look to Massachusetts to guide their reform efforts. The Innocence Project strongly supports the majority of the recommendations set forth in the Report, including those set forth in both minority reports. The Innocence Project also agrees with the comments and suggestions prepared by the Committee for Public Counsel Services (CPCS) and the Massachusetts Association of Criminal Defense Lawyers (MACDL). We appreciate the opportunity to offer the following comments and suggestions, which we hope will supplement the excellent work done by the Study Group. COMMENTS AND RECOMMENDATIONS The Innocence Project's greatest concern relates to the absence of sufficient guidance for courts evaluating the reliability of identification evidence once a pre-trial hearing is granted. While the Report offers detailed scientific findings concerning many aspects of eyewitness memory and perception, it does not offer courts specific guidance about how to apply those scientific findings to the task of evaluating identification evidence. We believe such guidance is necessary to ensure that the Report's recommendations are fully adopted. 2 Innocence Project, Know the Cases, (last visited Jan. 23, 2014).

54 Innocence Project, Inc. January 24, 2014 Page 3 INNOCENCE PROJECT So, for example, scientific research has established that, of the several flaws in the existing Manson balancing approach, none is more fundamental than the effect of suggestive procedures on the reliability factors. A suggestive identification procedure can artificially inflate a witness's self-reports regarding three of the five Manson factors: opportunity to view, attention paid, and certainty. See Henderson at 286 ("[W]hen [self-]reports are tainted by a suggestive process, they become poor measures in a balancing test designed to bar unreliable evidence.") Thus, despite the fact that suggestive identification procedures actually decrease reliability, suggestion can increase the likelihood that a court applying Manson will find that the identification was reliable. "The irony of the current test is that the more suggestive the procedure, the greater the chance eyewitnesses will seem confident and report better viewing conditions." Id While the Report identifies those estimator variables that can be inflated by suggestion (as well as those self-reported variables that witnesses tend to overestimate and those that are not well correlated with accuracy, such as certainty and description accuracy), we do not believe that the Report provides sufficient guidance to courts considering the effects of suggestion on a challenged identification. Scientific research has demonstrated that the amount of suggestion that can substantially contaminate memory is directly correlated to the strength of the original memory. 3 For example, if a witness observes a crime for a few seconds, in darkness, from a great distance, and while under great stress, his ability to encode a stranger's facial features will be very poor, resulting in a weak memory of the perpetrator. The witness's identification will be more easily affected by a relatively small amount of suggestion when compared with a witness who had an excellent opportunity to view a perpetrator (in daylight, for a long period of time, in the absence of stress). Consequently, courts evaluating whether suggestion by state or non-state actors in a particular case rises to the threshold level necessary to hold a hearing, or so corrupts an identification as to make it inadmissible, must also look to the strength of the original memory as reflected by the estimator variables present in the case. These principles should be clarified for courts applying the new legal framework. In addition, we believe it is critical for courts considering eyewitness identification evidence pretrial to consider primary evidence in the form of the witness's testimony (with the understanding, expressed in the Report, that witness's memories can be altered by suggestive feedback and information obtained after the fact), initial crime reports to emergency services, and reports by first responders. This primary evidence will reveal both the witness's statements at the point in time closest to the crime and will also contain potentially objective evidence about estimator variables (e.g., lighting conditions, distance, disguise, the witness's physiological conditions, etc.). Such a preference should be articulated in the Court's ultimate findings. Finally, as mentioned in the Report and elsewhere in our comments, courts should be encouraged to hear from experts at pretrial hearings in cases involving compromised identifications or whenever courts would find such testimony useful. The availability of witnesses and experts at pretrial hearings will enable courts to fashion more precise and useful intermediate remedies which will make trials more efficient and accurate. 3 See Nancy Steblay et al., The Eyewitness Post-Identification Feedback Effect 15 Years Later: Theoretical and Policy Implications, Psychology, Public Policy, and Law (in press) (Meta-analysis shows that the impact of feedback (suggestiveness) was less for accurate witnesses, who presumably had stronger initial memories, than for inaccurate witnesses, who had worse initial memories.)

55 Innocence Project, Inc. January 24, 2014 Page4 INNOCENCE PROJECT General Principles The Report is animated by certain general principles that the Innocence Project strongly supports: Perhaps most important to the future of eyewitness identification reform in Massachusetts is the Study Group's exhortation that the Report not be treated as a "definitive statement on the science of eyewitness identification, or on the police practices and criminal procedures most appropriate in light of the science." Report at 12. We agree that "[a]s a matter of justice, our courts must be able to respond to the science as it evolves rather than 'catch up' to advances in research after years of inaction." Id. We support the establishment of a Standing Committee on Eyewitness Evidence that would be responsible for periodic review of scientific and legal developments affecting eyewitness evidence and making appropriate recommendations to the Justices based on their findings. We urge the Court to create this committee and direct it to meet and report back to the Court on an annual basis. Relatedly, we strongly support the Report's emphasis on ongoing education and training of stakeholders. Indeed, we believe that ongoing education and training is essential to harnessing the full benefits of the Report's recommendations. Given the identified limitations of available resources, we urge the Court to identify creative and resourceefficient ways to ensure that meaningful education and training is provided on an ongoing basis to judges, attorneys and members of law enforcement. Of course, this training should also reflect any advances in scientific understanding identified by the Standing Committee on Eyewitness Evidence, described above. We therefore agree with the Study Group's suggestion that the Standing Committee on Eyewitness Evidence also be tasked with ensuring that the continuing education and training needs of stakeholders are identified and met. As noted in the Report, the Innocence Project is interested in working collaboratively with law enforcement to develop videotaped trainings that can be used during roll call. Report at 105. We support the Study Group's conclusion that judicial notice oflegislative facts, expansive jury instructions, and improved police procedures, while sure to improve the collection and adjudication of eyewitness identification evidence, are not substitutes for expert testimony concerning the factors that affect eyewitness memory and perception, either pre-trial or at trial. We concur specifically with MACDL's reasoning on this point and support MACDL's recommendation that the Supreme Judicial Court make explicit the principle that these system improvements are not permissible grounds to deny defense counsel funds for an expert under G.L. c. 261, 27C. MACDL Comments at 2. We agree with those members of the Study Group who felt ''the trace evidence analogy most accurately captures the nature of eyewitness evidence and forms the appropriate basis for analyzing reliability." Report at 45. Neuroscience has demonstrated that longterm memories exist as physical traces in the brain. Working from this fact, researchers have endorsed the trace evidence analogy in the legal context. Id. See also Lawson at 748. This analogy supports the uniform adoption of protocols for the handling and treatment of eyewitness identification evidence to minimize the possibility of

56 Innocence Project, Inc. January 24, 2014 Page 5 INNOCENCE PROJECT contamination and maximize reliability and accuracy. For adjudicatory purposes, when eyewitness evidence is treated as trace evidence, the burden of production shifts to the proponent of the evidence to establish its baseline reliability, as is the general rule for all physical evidence. The requirement that the defendant prove a negative to challenge the admissibility of the evidence is a difficult burden to bear because the information necessary to establish the evidence's reliability (including the circumstances under which it was obtained) is within the control of the prosecution, not the defendant. Like CPCS and MACDL, the Innocence Project reads the Report to permit challenges to the admissibility of eyewitness identification evidence either under a due process rubric or under traditional evidentiary grounds. Report at 39-40; Accord CPCS Comments at 6-7; MACDL Comments at 5. Given that the Minority Statement of Attorney James M. Doyle takes the view that the Report has dispensed with the evidentiary approach, we agree with CPCS and MACDL that the Court should clarify that trial judges continue to possess the authority to exclude identification evidence, or craft alternative remedies, based on the application of traditional evidentiary rules, even where such remedies may not be constitutionally required. Judicial Notice of Legislative Facts The Innocence Project strongly supports the Report's recommendation that courts take judicial notice as legislative facts of relevant scientific findings regarding eyewitness memory and perception, as set forth in Lawson app. at As Attorney James M. Doyle explained in his Minority Statement, judicial notice of scientific findings will increase accuracy and transparency in the adjudication of criminal matters. The Innocence Project concurs with the concerns raised by CPCS and agrees with CPCS' s recommendation that, in adopting this recommendation, the Supreme Judicial Court make clear that the principles set out in the Report constitute a baseline and not a ceiling for what can be drawn from the scientific literature. CPCS Comments at I-2. Further, and consistent with the Report's recommendation that courts recognize the evolving nature of the scientific literature, the principles set forth in the Report represent the research findings as they stand now and should not be viewed as stagnant. The Innocence Project submits one additional scientific finding for the Court's consideration: scientific research has shown that identifications by trained observers (including members oflaw enforcement) are equally affected by the factors that have been shown to affect eyewitness memory and perception. 4 Yet most people believe that trained observers, as a rule, make more reliable and accurate identifications. These findings should be included as part of the body of research that the Report has collected to serve as the basis for judicial notice in cases involving 4 See, e.g., Claudia J. Stanny & Thomas C. Johnson, Effects of Stress Induced by a Simulated Shooting on Recall by Police and Citizen Witnesses, 113 Am. J. Psycho!. 359 (2000) ("[SJeveral reviews of the eyewitness literature concluded that there is little evidence that police recall witnessed events any more accurately than citizens.") (Collecting research). This finding is borne out by DNA exonerations that involved police officers misidentifying actually innocent suspects. See Case of Stephan Cowans (served 5.5 years), available at: Case of Steven Barnes (served 19.5 years), available at: Case of Scott Fappiano (served 21 years), available at: _ Fappiano. php.

57 Innocence Project, Inc. January 24, 2014 Page 6 INNOCENCE PROJECT eyewitness identifications. Consequently, the Court should make clear that the mere fact that an identification is made by a police officer or other trained observer should not preclude defendants from raising challenges to that evidence or obtaining pre-trial hearings on the grounds set forth in the Report. Police Practices The Innocence Project's experience collaborating with members oflaw enforcement nationwide supports the Study Group's finding that a "well-trained detective who uses research-based techniques can decrease the likelihood of misidentification and preserve the witness's ability to recognize the offender later." Report at 85. Likewise, our experience supports the Study Group's conclusion that uniform, science-based practices that are memorialized in written policies on which members of law enforcement are regularly trained offers a substantial protection against the wrongful conviction of innocent suspects. In addition to those comments ofcpcs and MACDL with which we concur (see below), the Innocence Project would also urge the Court to amend the police practice recommendations contained in the Report as follows: All law enforcement agencies should be required to maintain a written policy, supported by the scientific research described in the Report, on eyewitness identification procedures. We expect- as with the decision in Com. v. DiGiambattista, 442 Mass. 423, (2004), which ultimately compelled changes in police practice to require the recording of custodial interrogations - local law enforcement agencies will craft policies mindful of the Report's recommendations. Indeed, we understand that efforts are already Wlderway within the law enforcement community to ensure police protocols at the local agency comport with the Report's recommendations guiding system variables. We recommend that the Police Practice Subcommittee work in coordination with the Study Group to define a mechanism to ensure that local law enforcement agencies comply uniformly- through the adoption of written policies-with the recommendations of the Court. The Report's recommendation that "[w)henever practicable, the police should videotape or audiotape a photo array or lineup" is a critical reform. Report at 88. The Innocence Project urges the Court to further require that, where recording is not practicable, police must document the reason(s) why recording was not practicable and find that reviewing courts may consider the proffered reasons in fashioning remedies for a failure to record. Comment I to the Police Practice Subcommittee's Recommendation raises concerns about the use of composites, sketches, and mug files. These concerns are well-founded. Composites or sketches were used in nearly 30 percent of the Innocence Project's DNA exoneration cases that involved eyewitness misidentification. While the best practices provide that "[t]he use of composites and sketches and the showing of mug files are disfavored," (Report at 87) we believe that a more specific recommendation, incorporating the findings of Comment I, should be included in the section "General Best Practices" (as opposed to "Best Practices for Showups"). This recommendation should specifically approve the use of mug files only where 1) the police investigation has

58 Innocence Project, Inc. January 24, 2014 Page 7 INNOCENCE PROJECT specific evidence to limit and guide the selection of suspects beyond general descriptors (age, race, and size), such as known affiliation with a particular gang; 2) after limiting mug files to a particular class based on evidence gathered in the investigation, police further limit the selected mug files, 3) police do not tell the witness or witnesses anything about the mug file classification (i.e., that it consists of known gang members or arrestees); and 4) all descriptors used to define a mug file search or collection, whether physical or electronic, should be documented and preserved. The Report provides for specific pre-lineup instructions that are designed to reduce the natural pressure a witness feels to make an identification, and to mitigate a witness's natural assumption that when he is called to make an identification the police have caught the perpetrator. Report at 106. Scientific research supports these instructions and we strongly urge the Court to adopt them as written, with one addition that will further the overarching goals of pre-lineup instructions: a standardized instruction that the witness should not feel compelled to make an identification. The Report provides for the possibility that a showup procedure could be used with more than one witness in the same case. Report at 89 (Recommending that "[i]f showups are to be conducted with multiple witnesses, they should be conducted in such a way that one witness cannot see or hear the procedure or results of another witness."). We agree with the recommendation that multiple witnesses must be separated and not permitted to witness another's identification procedure (including a showup) or to learn of the outcome of that procedure. However, due to the inherently suggestive nature of a showup, once one witness makes a positive identification of a suspect through a showup, resulting in probable cause for an arrest, any subsequent witnesses should be shown properly composed photo arrays or lineups. We urge the Court to amend the recommendation to incorporate this recommendation. We urge the Court to recommend that law enforcement consider utilizing hand held devices to conduct photo identification procedures instead of one-on-one show ups within two hours from the time a witness made observations of criminal activity. Being able to rapidly deploy a fairly constructed photo procedure instead of a one-on-one show up procedure can greatly reduce false show up identifications which are, all agree, inherently suggestive and particularly subject to what's known as a clothing bias. See Jennifer E. Dysart et al., Show-Ups: The Critical Issue of Clothing Bias, 20 Applied Cognitive Psychology 1009 (2006). Accord Lawson at 784. In addition to the above recommendations, the Innocence Project concurs with the comments of CPCS and MACDL concerning best practices for Massachusetts Police Departments: Consistent with the Minority Statement of Attorney Natarajan and Judges Blitzman and Gertner, the Report's distinction between "Best Practices" and "Specific Best Police Practices" is unjustified. Report at ; See also CPCS Comments at 2; MACDL Comments at 2. Instead, as the Minority Statement urges, the Court should adopt the single, unified set of "Best Police Practices" set out at pages of the Report.

59 Innocence Project, Inc. January 24, 2014 Page 8 INNOCENCE PROJECT Consistent with the scientific research, law enforcement should explicitly advise the witness, prior to obtaining a description, not to guess at any particular features of the perpetrator and to describe only those features that the witness clearly remembers. CPCS Comments at 3. There can be no rational basis to distinguish between photo arrays and other types of photographic identification procedures with respect to the need to document and preserve outcomes. The Report's recommendation that police preserve photographic arrays and document steps taken to preserve photographic arrays (Report at 86) should therefore be extended to include all forms of photographic identification procedures, including but not limited to mug books and electronic mugshot databases. In addition, all contemporaneous notes taken by police of the original description taken from any witness and any other statements by witnesses should be documented and preserved. CPCS Comments at 3-4; MACDL Comments at 3. The Report recommends that showups should only be conducted within two hours after the witness's observation of the suspect. Report at 87. For the reasons set forth in their letters, the Innocence Project strongly agrees with CPCS and MACDL that, in accepting the Report, the Supreme Judicial Court should also require that showups only be conducted where there is no probable cause to arrest the suspect independent of the crime under investigation. Accord State v. Dubose, 285 Wis. 2d 143, (2005) ("[E]vidence obtained from an out-of-court showup is inherently suggestive and will not be admissible unless, based on the totality of the circumstances, the procedure was necessary. A showup will not be necessary, however, unless the police lacked probable cause to make an arrest or, as a result of other exigent circumstances, could not have conducted a lineup or photo array.) See also CPCS Comments at 4; MACDL Comments at3. Pretrial Hearings In addition to the concern raised above regarding guidance to courts and the concerns raised below concerning jury instructions, we recommend that the Court revise the Report's recommendations on remedies available following pre-trial hearings: The Report now provides that "[w]here the court finds that police have failed to follow the Best Police Practices or failed to record the identification procedures where it was feasible to do so, it shall give appropriate jury instructions." The Innocence Project urges the Court to fashion strong cautionary language for courts to use where police have failed to follow the Best Police Practices or have failed to record the identification procedures despite the ability to do so. o When there is evidence that the police failed to follow the Best Police Practices, the following instruction should be given after the instructions on Identification Procedures set forth at : In this case, if you find that the police did not follow the [Best Police Practice not followed], you should evaluate the identification that resulted from these procedures with particular care. Compare

60 Innocence Project, Inc. January 24, 2014 Page 9 INNOCENCE PROJECT Commonwealth v. Ciampa, 406 Mass. 257, 264, 547 N.E.2d 314, 319 (1989) (accomplice testimony instruction). o When there is evidence that the police failed to record an identification procedure where it was feasible to do so, the following instruction should be included after the instructions on Identification Procedures set forth at : In this case, you heard evidence that the police did not make a video or audio recording of the (photo array or lineup) procedure. The Supreme Judicial Court-this state's highest court-has expressed a preference that such procedures be recorded whenever practicable. Since there is no recording of the (photo array or lineup) procedure in this case, you should weigh evidence of the alleged identification of the defendant that supposedly resulted from that procedure with great caution and care. The reason is that the Commonwealth may have had the ability to record the procedure that led to the alleged identification of the Defendant, which the Commonwealth is now asking you to find beyond a reasonable doubt was accurate, but instead is asking you to rely on a summary of those circumstances drawn from the possibly fallible or selective memory of its witness(es). The absence of the recording of this procedure allows you, but does not require you, to find that the Commonwealth has failed to prove the reliability of this alleged identification beyond a reasonable doubt. CPCS Comments at 8-9 (quoting Commonwealth v. DiGiambattista, 442 Mass. 423, (2004)). Additionally, we join in the concerns raised by CPCS and MACDL with respect to the Report's recommendations for pretrial hearings: We strongly agree with CPCS and MACDL's position that the Court should not require corroboration where defendants seek pretrial hearings based on the presence of estimator variables that cast doubt on the reliability of the identification. CPCS Comments at 4-5; MACDL Comments at 5. We concur with CPCS's explanation of the problems with this approach, including that such a requirement may run afoul of the Supreme Court's holding in Holmes v. South Carolina, 547 U.S. 319, 331 (2006), but disagree with CPCS's proposed remedy. CPCS Comments at 4-5. Because we believe that there will be only a small number of cases where estimator variables alone would require a hearing, and because we know that corroboration is common in wrongful conviction cases (and, in particular, where eyewitness misidentification occurs) we submit that there should be no corroboration requirement to obtain a pretrial hearing where the defendant has shown that estimator variables cast doubt on the reliability of the identification. Of the 227 DNA exonerations stemming from convictions involving mistaken identifications, 123 (54 percent) involved at least one other evidentiary error (forensics, false confessions, and/or informants) that corroborated the misidentification. Indeed, recent research suggests that not only are multiple errors common in wrongful conviction cases, but that although the multiple errors may appear "independent" - e.g., a misidentification and a false confession - they each have the power to corrupt other, seemingly unrelated evidence. 5 5 See Saul M. Kassin, Daniel Bogart, and Jacqueline Kerner, Confessions That Corrupt: Evidence from the DNA Exoneration Case Files, 23 Psycho!. Sci. 41, 45 (2012).