State v. Raynor

337 Conn. 527

• Court: Supreme Court of Connecticut
• Decided: December 4, 2020
• Docket: SC20183
• Status: Published

Opinion

August 17, 2021 CONNECTICUT LAW JOURNAL Page 3

337 Conn. 527 AUGUST, 2021 527 State v. Raynor

STATE OF CONNECTICUT v. DONALD RAYNOR (SC 20183) Robinson, C. J., and Palmer, McDonald, D’Auria, Mullins and Kahn, Js.*

Syllabus

Convicted, after a jury trial, of the crime of murder in connection with the shooting death of the victim, the defendant appealed. The defendant and the victim were members of rival street gangs in Hartford. On the day of the shooting, the defendant called R, another member of his gang, and told him that he wanted to test an assault rifle. R drove with the defendant through areas of Hartford frequented by members of the victim’s gang, and, as R drove, the defendant shot at the victim and killed him. Thirteen months later, the police recovered an assault rifle in connection with an unrelated investigation, and the state’s expert witness, S, a firearm and toolmark examiner, testified that several cas- ings recovered from the scene of the victim’s murder and the scene of a subsequent, unrelated shooting were positively identified as having been fired from the same assault rifle the police recovered. In affirming the defendant’s conviction, the Appellate Court concluded that the trial court properly denied the defendant’s motion to exclude or limit the scope of S’s testimony and that the trial court did not abuse its discretion in admitting evidence of uncharged misconduct related to the subsequent shooting. On the granting of certification, the defendant appealed to this court. Held: 1. The Appellate Court improperly upheld the trial court’s denial of the defendant’s motion for a hearing, pursuant to this court’s decision in State v. Porter (241 Conn. 57), on the reliability and accuracy of the methodology used by S in connection with his anticipated firearm and toolmark testimony: the trial court, having based its decision to deny the defendant’s motion soley on earlier Appellate Court precedent con- cluding that the science of firearm and toolmark identication is well established, abused its discretion by failing to determine whether the criticisms of firearm and toolmark analysis contained in certain reports cited by the defendant in his motion cast enough doubt on whether the science in that field remained well established so as to warrant a Porter hearing; moreover, this court lacked a fair assurance that the admission of S’s testimony did not substantially affect the verdict, and, thus, the trial court’s denial of the defendant’s motion for a Porter hearing was not harmless; accordingly, the defendant was entitled to a new trial. (Three justices concurring separately in one opinion)

* The listing of justices reflects their seniority status on this court as of the date of oral argument. Page 4 CONNECTICUT LAW JOURNAL August 17, 2021

528 AUGUST, 2021 337 Conn. 527 State v. Raynor 2. The Appellate Court properly upheld the trial court’s denial of the defen- dant’s motion to limit the scope of S’s conclusions regarding the ballistics evidence to a ‘‘more likely than not’’ standard; given that the trial court was asked to limit S’s testimony in a highly proscribed manner, and in light of the scant information and lack of case law provided in support of the defendant’s motion, the trial court’s denial of that motion was not an abuse of discretion. 3. The Appellate Court improperly upheld the trial court’s admission of uncharged misconduct evidence concerning a shooting in which the defendant allegedly was involved and that occurred subsequent to the shooting that formed the basis of the murder charge in the present case, as the prejudicial impact of that evidence unduly exceeded its probative value: the subsequent shooting was a less severe crime than the murder in the present case because neither of the victims of the subsequent shooting was struck by the shots fired, and both shootings shared com- mon characteristics, including individuals being shot at outside of their homes; moreover, evidence of the subsequent shooting was introduced through the testimony of one of the victims of that shooting and was not limited to the the fact that there was a shooting but consisted of details regarding the surrounding events that could have aroused the jurors’ emotions; furthermore, the subsequent shooting occurred eight months after the murder at issue in the present case, and no evidence suggested that the subsequent shooting was motivated by or related to the murder.

Argued February 21—officially released December 4, 2020**

Procedural History

Substitute information charging the defendant with the crime of murder, brought to the Superior Court in the judicial district of Hartford and tried to the jury before the court, Kwak, J.; verdict and judgment of guilty, from which the defendant appealed to this court, which transferred the appeal to the Appellate Court, Keller, Elgo and Eveleigh, Js., which affirmed the judg- ment of the trial court, and the defendant, on the grant- ing of certification, appealed to this court. Reversed; new trial. Andrew P. O’Shea, with whom was Damon A. R. Kirschbaum, for the appellant (defendant). ** December 4, 2020, the date that this decision was released as a slip opinion, is the operative date for all substantive and procedural purposes. August 17, 2021 CONNECTICUT LAW JOURNAL Page 5

337 Conn. 527 AUGUST, 2021 529 State v. Raynor

James A. Killen, senior assistant state’s attorney, with whom, on the brief, were Gail P. Hardy, former state’s attorney, and Patrick J. Griffin, state’s attorney, for the appellee (state). Maura Barry Grinalds and Darcy McGraw filed a brief for the Connecticut Innocence Project et al. as amici curiae. Lisa J. Steele, assigned counsel, filed a brief for the Connecticut Criminal Defense Lawyers Association as amicus curiae. Charles D. Ray, Angela M. Healey, and M. Chris Fabricant, pro hac vice, filed a brief for the Innocence Project, Inc., as amicus curiae. Opinion

KAHN, J. The defendant, Donald Raynor, appeals from the judgment of the Appellate Court, which affirmed the judgment of conviction, rendered after a jury trial, of the crime of murder in violation of General Statutes § 53a-54a (a).1 State v. Raynor, 181 Conn. App. 760, 778, 189 A.3d 652 (2018). The defendant claims that the Appellate Court incorrectly concluded that the trial court had properly (1) denied his motion for a Porter2 1 This court granted the defendant’s petition for certification to appeal, limited to the following issues: (1) ‘‘Did the Appellate Court correctly con- clude that the trial court had properly denied the defendant’s motion for a Porter hearing to determine the reliability of firearm and toolmark identifica- tion?’’ (2) ‘‘Did the Appellate Court correctly conclude that the trial court had properly denied the defendant’s motion in limine to limit the scope of the testimony of the state’s expert on firearm and toolmark analysis?’’ And (3) ‘‘[d]id the Appellate Court correctly conclude that the trial court had properly admitted the uncharged misconduct evidence?’’ State v. Raynor, 330 Conn. 910, 193 A.3d 49 (2018). 2 State v. Porter, 241 Conn. 57, 80–90, 698 A.2d 739 (1997), cert. denied, 523 U.S. 1058, 118 S. Ct. 1384, 140 L. Ed. 2d 645 (1998). ‘‘In Porter, we followed the United States Supreme Court’s decision in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993), and held that testimony based on scientific evidence should be subjected to a flexible test to determine the reliability of methods used to reach a particular conclusion. . . . A Porter analysis involves a two part inquiry that assesses the reliability and relevance of the witness’ methods. . . . First, the party offering the expert testimony must show that the Page 6 CONNECTICUT LAW JOURNAL August 17, 2021

530 AUGUST, 2021 337 Conn. 527 State v. Raynor

hearing on the reliability of ballistics evidence, (2) denied his motion in limine seeking to limit the scope of testimony from the state’s firearm and toolmark exam- iner, and (3) denied the defendant’s motion to exclude uncharged misconduct evidence related to a subsequent shooting. As to the first issue, the defendant claims that reports authored by the National Academy of Sciences (NAS)3 call into question the reliability of methodolo- expert’s methods for reaching his conclusion are reliable. A nonexhaustive list of factors for the court to consider include: general acceptance in the relevant scientific community; whether the methodology underlying the scientific evidence has been tested and subjected to peer review; the known or potential rate of error; the prestige and background of the expert witness supporting the evidence; the extent to which the technique at issue relies [on] subjective judgments made by the expert rather than on objectively verifiable criteria; whether the expert can present and explain the data and methodology underlying the testimony in a manner that assists the jury in drawing conclusions therefrom; and whether the technique or methodology was developed solely for purposes of litigation. . . . Second, the proposed scientific testimony must be demonstrably relevant to the facts of the particu- lar case in which it is offered, and not simply be valid in the abstract. . . . Put another way, the proponent of scientific evidence must establish that the specific scientific testimony at issue is, in fact, derived from and based [on] . . . [scientifically reliable] methodology. . . . ‘‘Additionally, we recognized in Porter that . . . [t]he actual operation of each [Porter] factor, as is the determination of which factors should be considered at all, depends greatly on the specific context of each case in which each particular [threshold admissibility] analysis is conducted. . . . There is, however, a critical postulate that underlies the Porter factors and indeed underlies the entire Porter analysis: in order for the trial court, in the performance of its role as the gatekeeper for scientific evidence, properly to assess the threshold admissibility of scientific evidence, the proponent of the evidence must provide a sufficient articulation of the methodology underlying the scientific evidence. Without such an articulation, the trial court is entirely ill-equipped to determine if the scientific evidence is reliable upon consideration of the various Porter factors. Furthermore, without a clear understanding as to the methodology and its workings, the trial court also cannot properly undertake its analysis under the fit requirement of Porter, ensuring that the proffered scientific evidence, in fact, is based upon the reliable methodology articulated.’’ (Internal quotation marks omitted.) State v. Edwards, 325 Conn. 97, 124–25, 156 A.3d 506 (2017). 3 The defendant cites two particular publications. See National Research Council of the National Academies, Strengthening Forensic Science in the United States: A Path Forward (2009) (2009 NAS report); National Research Council of the National Academies, Ballistic Imaging (2008) (2008 NAS report). For the sake of simplicity, we collectively refer to these publications as the NAS reports. August 17, 2021 CONNECTICUT LAW JOURNAL Page 7

337 Conn. 527 AUGUST, 2021 531 State v. Raynor

gies employed in firearm and toolmark examinations and that, as a result, a Porter hearing was necessary to determine if such evidence is admissible. Furthermore, the defendant argues that both the trial court and the Appellate Court construed State v. Legnani, 109 Conn. App. 399, 421, 951 A.2d 674, cert. denied, 289 Conn. 940, 959 A.2d 1007 (2008), too broadly by concluding that a Porter hearing on the reliability of firearm and toolmark examinations is never necessary because it is a well established and admissible science. As to the second issue, the defendant argues that, even if the firearm and toolmark examination evidence was admissible without a Porter hearing, the trial court improperly denied his motion in limine, which would have required the state’s expert, James Stephenson, to clarify that his conclu- sions that certain bullet casings were fired from a spe- cific firearm were not certainties but merely ‘‘more likely than not’’ to be correct. As to the third issue, the defen- dant claims that the probative value of evidence related to a subsequent shooting, which was admitted to estab- lish the defendant’s identity and to show that he had access to the firearm used in the present case, was out- weighed by its prejudicial effect. For the reasons that follow, we conclude that the Appellate Court (1) improp- erly upheld the trial court’s denial of the defendant’s motion for a Porter hearing on the reliability of ballistics evidence based solely on the holding in Legnani, (2) properly upheld the trial court’s denial of the defen- dant’s motion in limine, which sought to limit the scope of Stephenson’s conclusions, and (3) improperly upheld the trial court’s denial of the motion to exclude evidence of uncharged misconduct. We therefore conclude that the defendant is entitled to a new trial and, accordingly, reverse the judgment of the Appellate Court. The record reveals the following relevant facts and procedural history. The defendant was a member of the Money Green Bedrock (Bedrock) street gang in Hart- Page 8 CONNECTICUT LAW JOURNAL August 17, 2021

532 AUGUST, 2021 337 Conn. 527 State v. Raynor

ford, and the victim, Delano Gray, was a member of a rival street gang, The Avenue, also known as The Ave. Prior to the events giving rise to the present case, the defendant and the victim were involved in two incidents stemming from the rivalry between their gangs. The first incident, which occurred at an unspecified date prior to 2006, involved the victim’s firing shots at the defendant and another Bedrock member. The second incident, which occurred approximately one week prior to the events giving rise to the present case, occurred when the victim saw the defendant and another Bed- rock member, Jose Rivera, at a restaurant in The Ave- nue’s territory. As the defendant and Rivera were leav- ing the restaurant, Rivera noticed that the victim was taking a photograph of the defendant’s car. Rivera relayed this to the defendant, who responded that ‘‘[the victim] had to go,’’ which Rivera understood to mean that ‘‘[the victim] had to get killed for what he did.’’ During the early morning hours of June 18, 2007, the defendant called Rivera and told him that he wanted to ‘‘test out [a] .223 [caliber] assault rifle and that [the defendant] wanted to go see if [they] could find any Avenue guys,’’ which Rivera understood to mean they were ‘‘gonna go look for some Avenue guys to kill.’’ The defendant had owned that assault rifle for approxi- mately one month, and Rivera had been with the defen- dant when he purchased it. The defendant picked up Rivera and drove to the back of the defendant’s apart- ment building on Bedford Street, parking next to a non- functioning vehicle that belonged to Rivera and was used for ‘‘stashing drugs [and] guns . . . .’’ The defen- dant put on latex gloves, removed the .223 caliber assault rifle from a bag stored in the trunk of the non- functioning vehicle, and loaded the assault rifle with ‘‘a big magazine clip.’’ The defendant and Rivera then got back into the functioning vehicle; Rivera drove, and the defendant sat in the backseat with the assault rifle. Rivera drove the vehicle around certain areas in the north end of Hartford frequented by members of The August 17, 2021 CONNECTICUT LAW JOURNAL Page 9

337 Conn. 527 AUGUST, 2021 533 State v. Raynor

Avenue. While Rivera was driving on Enfield Street, he told the defendant that he saw the victim standing on the sidewalk engaged in conversation with a woman. At the defendant’s instruction, Rivera drove back around the block. As Rivera drove down Enfield Street for the second time, he rolled down the rear driver’s side win- dow and slowed the vehicle down to a roll. The defen- dant hung out of the window and started shooting the assault rifle at the victim. The victim and his female companion attempted to flee, running in different direc- tions, but the victim fell to the ground after taking only about three steps. The defendant kept shooting after the victim fell to the ground, firing at least ten to fifteen times, and then Rivera and the defendant drove away. The victim died as a result of gunshot wounds to his chest and neck. On July 16, 2008, thirteen months after the Enfield Street murder, the police recovered a .223 caliber Kel- Tec assault rifle in an unrelated investigation after receiving a tip from a confidential informant. In August, 2011, after being arrested for an unrelated homicide, Rivera gave a statement to the police in which he con- fessed to his involvement in the victim’s murder, impli- cated the defendant as the shooter, and he identified the .223 caliber Kel-Tec assault rifle recovered by the police in July, 2008, as the weapon that the defendant used to shoot the victim. Simultaneously, in August, 2011, the police met with and obtained a written state- ment from the victim of a shooting on Baltimore Street that occurred on February 16, 2008—a shooting at which Rivera was not present. That individual identified the defendant as the shooter in that crime and stated that he had fired a rifle at her and her partner. Stephen- son testified that casings recovered from the crime scenes of the victim’s murder on Enfield Street and the subsequent shooting on Baltimore Street were posi- tively identified as having been fired from the .223 cali- ber Kel-Tec assault rifle that had been recovered by the police. Page 10 CONNECTICUT LAW JOURNAL August 17, 2021

534 AUGUST, 2021 337 Conn. 527 State v. Raynor

In 2013, the defendant was charged with murder in violation of § 53a-54a (a), conspiracy to commit murder in violation of General Statutes §§ 53a-48 and 53a-54a (a), and criminal use of a firearm in violation of General Statutes § 53a-216 (a). The defendant’s first jury trial, conducted in 2014, ended in a mistrial after the jury was unable to agree on a verdict. At the second trial, conducted in 2015, the state charged the defendant with only one count of murder in violation of § 53a-54a (a). The second jury returned a guilty verdict, and the court sentenced the defendant to sixty years of incarceration with a twenty-five year mandatory minimum. The defendant subsequently appealed, claiming ‘‘that the trial court (1) improperly denied [his] motion in limine to exclude or limit the scope of the testimony of the state’s expert witness on firearm and toolmark identification, and (2) abused its discretion by granting the state’s motion for uncharged misconduct related to a shooting that occurred approximately eight months after the events of [the present] case.’’ State v. Raynor, supra, 181 Conn. App. 762. The Appellate Court concluded that the trial court ‘‘properly relied upon Legnani, and did not abuse its discretion by denying the defendant’s motion in limine to exclude or limit Stephenson’s testi- mony.’’ Id., 771. Furthermore, the Appellate Court con- cluded that the trial court ‘‘did not abuse its discretion by admitting the uncharged misconduct evidence related to [a subsequent shooting].’’ Id., 778. This appeal followed. Additional facts and procedural history will be set forth as necessary. I EXPERT BALLISTIC TESTIMONY We begin with the defendant’s claims challenging the admissibility and scope of Stephenson’s testimony relating to firearm and toolmark analysis. The following additional facts and procedural history are relevant to the resolution of these claims. In anticipation of testi- August 17, 2021 CONNECTICUT LAW JOURNAL Page 11

337 Conn. 527 AUGUST, 2021 535 State v. Raynor

mony by Stephenson at trial, the defendant filed (1) a motion for a Porter hearing on the admissibility of fire- arm and toolmark analysis, and, in the alternative, (2) a motion in limine to limit the scope of Stephenson’s conclusions. The defendant argued in his motion for a Porter hear- ing that the NAS reports called into question the relia- bility and accuracy of the methodology employed by Stephenson and that there was ‘‘relatively little legal accounting’’ for those reports.4 The defendant con- tended that a Porter hearing would demonstrate that the methodology used by Stephenson was not scientifically valid and was, therefore, inadmissible. The state opposed the defendant’s motion for a Porter hearing, arguing that ‘‘Connecticut state law firmly holds that the science of firearm and toolmark identification has been so well established that a trial court does not have to conduct a Porter hearing prior to admitting such evidence.’’ In the event that the motion for a Porter hearing was denied, or if one were held and resulted in the admission of Stephenson’s testimony, defense counsel argued dur- ing oral argument on the motion that Stephenson should be restricted to stating only that casings recovered from the crime scenes were ‘‘more likely than not’’ fired from the .223 caliber Kel-Tec assault rifle. Anticipating that Stephenson would testify that ‘‘the shell casings recov- ered from the shooting of [the victim] were fired from the same weapon as the shell casings recovered from the [Baltimore Street shooting] based upon his foren- sic toolmark analysis of those casings,’’ defense coun- sel cited to the NAS reports for the proposition that, 4 For example, in his principal brief to this court, the defendant interprets the NAS reports to stand for the proposition that ‘‘firearm toolmark identifi- cation has not been proven to be scientifically valid for three primary rea- sons: (1) the fundamental assumptions of the methodology are not reliable, (2) the standard for identifying matches to particular firearms is impermissi- bly vague and subjective, and (3) there is no quantification of the accuracy and error rates of identifications.’’ Page 12 CONNECTICUT LAW JOURNAL August 17, 2021

536 AUGUST, 2021 337 Conn. 527 State v. Raynor

‘‘[b]ecause not enough is known about the variabilities among individual tools and guns, we are not able to spec- ify how many points of similarity are necessary for a given level of confidence in the result. Sufficient studies have not been done to understand the reliability and repeatability of the methods.’’ (Emphasis omitted; inter- nal quotation marks omitted.) Defense counsel noted that the authors of the NAS reports agreed that ‘‘class characteristics are helpful in narrowing the pool of [fire- arms] that may have left a distinctive mark [on a casing]. Individual patterns from manufacture or from wear might, in some cases, be distinctive enough to suggest one particular source, but additional studies should be performed to make the process of individualization more precise and repeatable.’’5 (Emphasis omitted; internal quotation marks omitted.) On the basis of these observations, defense counsel asked that a ‘‘limiting order and instruction, similar to that in [United States v. Glynn, 578 F. Supp. 2d 567, 574–75 (S.D.N.Y. 2008)], be granted,’’ to permit Stephen- son to testify only that a firearms match was ‘‘more likely than not.’’ In response, the state claimed that ‘‘while the defendant has located a federal judge in the Southern District of New York who might agree with 5 At trial, Stephenson described the three sets of characteristics that fire- arm and toolmark examiners compare between casings recovered from a crime scene and a test fire from a suspected weapon to determine the existence of a match. Class characteristics are the general ammunition characteristics that are intentionally made during the manufacturing process, including, but not limited to, the size, shape, weight, diameter, grooves, and other features that indicate the manufacturer and caliber of that specific ammunition. Subclass characteristics are markings on the ammunition that are not made for the purpose of the ammunition itself but that are uninten- tionally made during the manufacturing process. These markings are made by the tools as the ammunition is manufactured, for example, the bore tool as it cuts the grooves. Individual characteristics are fine striated marks that are left on the ammunition by a specific firearm due to unique markings left by the actual manufacturing of a barrel or the repeated use of the firearm that has worn down or chipped that barrel. August 17, 2021 CONNECTICUT LAW JOURNAL Page 13

337 Conn. 527 AUGUST, 2021 537 State v. Raynor

his contention, the appellate courts of . . . Connecti- cut do not.’’ Following oral arguments, the trial court denied the defendant’s motion for a Porter hearing and motion in limine, concluding that the forensic science of firearms experts ‘‘has been well established, and we have a case, [Legnani], which stands for the proposition that this is not a new science. Therefore, a Porter hear- ing is not necessary.’’ The trial court expounded that it had ‘‘read the [2009 NAS] article. [The judge] under- stood [NAS] recommend[ed] some further studies or data collection. That’s their recommendation, but, again, the case law in Connecticut is what it is.’’ Stephenson subsequently testified before the jury that the Connecticut State Forensic Laboratory employed the Association of Firearm and Tool Mark Examiners’ (AFTE) theory of identification, which is generally accepted in the science of firearm and toolm- ark identification, and he explained the tenets of that theory. He explained that ‘‘we know that, through the theory of identification, that no two tools are—have left—will leave an examiner to the point where he would make a false identification based on his examina- tion.’’ Stephenson proceeded to testify that twelve of the fifteen casings recovered from the Enfield Street murder were ‘‘positively matched’’ to the .223 caliber Kel-Tec assault rifle and that the remaining three cas- ings had insufficient marks found for the purpose of identification. In addition, Stephenson testified that sev- enteen of the twenty-two casings recovered from the Baltimore Street shooting were ‘‘positively identified as being fired from the Kel-Tec rifle’’ and that the remaining five casings did not have sufficient marks to make a comparison for identification. During an exten- sive cross-examination, ‘‘[d]efense counsel . . . high- lighted the ways in which firearm and toolmark identi- fication does not follow precisely the scientific method— i.e., by not protecting against confirmation bias—and Page 14 CONNECTICUT LAW JOURNAL August 17, 2021

538 AUGUST, 2021 337 Conn. 527 State v. Raynor

that the [AFTE] theory of identification is not a com- pletely objective theory.’’ State v. Raynor, supra, 181 Conn. App. 768. Stephenson also acknowledged that he was aware of the 2009 NAS report and conceded that some—but not all—of the criticisms of firearm and tool- mark analysis were valid. On appeal to the Appellate Court, the defendant claimed that the trial court ‘‘abused its discretion by denying his motion in limine and request for a Porter hearing. The defendant argue[d] that the [NAS reports] establish that the methodology underlying firearm and toolmark identification is not reliable, and, as a result, the [trial] court should have precluded Stephenson from opining that particular cartridge casings positively matched the firearm in evidence. In the alternative, the defendant argue[d] that the [trial] court should have lim- ited Stephenson’s testimony so that he could opine only that his conclusions were ‘more likely than not . . . correct.’ ’’ State v. Raynor, supra, 181 Conn. App. 768. ‘‘The state argue[d] that the [trial] court properly relied upon [Legnani] in concluding that the admissibility of firearm and toolmark identification evidence is well established and, therefore, properly denied the defen- dant’s motion.’’ Id. The Appellate Court concluded that ‘‘Legnani is con- trolling precedent on the issue of whether the science of firearm and toolmark identification is well established, and thus binds our resolution of this claim.’’ 6 Id., 770. The Appellate Court acknowledged that Legnani pre- 6 The Appellate Court also noted that ‘‘policy dictates that one panel should not . . . reverse the ruling of a previous panel. The reversal may be accomplished only if the appeal is heard en banc.’’ (Emphasis in original; internal quotation marks omitted.) State v. Raynor, supra, 181 Conn. App. 770 n.4. ‘‘On November 27, 2017, the defendant filed a motion for consideration en banc, which [the Appellate Court] denied on January 10, 2018. Addition- ally, the entire [Appellate Court did not order] that [the] case be considered en banc pursuant to Practice Book § 70-7 (b), nor [was it] persuaded that en banc review [was] warranted. Therefore, [the Appellate Court would] not overrule Legnani.’’ Id. August 17, 2021 CONNECTICUT LAW JOURNAL Page 15

337 Conn. 527 AUGUST, 2021 539 State v. Raynor

dated the NAS reports but clarified that the reports ‘‘do not overrule or otherwise abrogate the existing case law in this state; nor do the [United States] District Court cases or the cases from other states that the defendant has cited in support of his claim. More importantly, the defendant did not proffer his own expert witness to testify that the science of firearm and toolmark identifi- cation is not reliable. The evidence admitted during the cross-examination of Stephenson included the flaws and criticisms of firearm and toolmark identification. The jury was free to give this evidence as much or as little weight as it saw fit.’’ Id., 771. For these reasons, the Appellate Court upheld the trial court’s denial of the defendant’s motion for a Porter hearing and motion in limine, holding that ‘‘[a] Porter hearing to determine the validity of firearm and toolmark identification was not required. The state had to establish only that the fire- arm and toolmark evidence was relevant, which it did.’’ Id. The Appellate Court acknowledged, however, ‘‘that there has been some evolvement in the field of firearm and toolmark identification since [it] decided Legnani.’’ Id., 770 n.4. Despite its familiarity with the NAS reports, the Appellate Court highlighted that ‘‘[d]efense counsel . . . extensively cross-examined Stephenson regarding the recent criticisms of firearm and toolmark identifica- tion, during which Stephenson acknowledged the valid- ity of at least some of those criticisms. Even if [the Appellate Court] were inclined to review the scientific validity of firearm and toolmark identification—and therefore [were] inclined to review the holding of Leg- nani—the circumstances of the [case did] not warrant a departure from [its] precedent. The defendant [had] not proffered his own expert to rebut the notion that firearm and toolmark evidence is sufficiently reliable as to be admitted without first requiring a Porter hearing. Therefore, [the Appellate Court] adhere[d] to [its] prec- Page 16 CONNECTICUT LAW JOURNAL August 17, 2021

540 AUGUST, 2021 337 Conn. 527 State v. Raynor

edent that holds that the admissibility of firearm and toolmark identification is well established.’’ Id., 770–71 n.4. This appeal followed. A Porter Hearing In the present case, the defendant claims that the Appel- late Court improperly upheld the trial court’s denial of his motion for a Porter hearing because both the trial court and the Appellate Court interpreted Legnani too broadly when each determined that it was bound by that precedent, notwithstanding the fact that the defen- dant had highlighted new evidence and case law that questioned the reliability of the methodology used in firearm and tookmark analysis. The defendant claims that such a broad interpretation of Legnani ‘‘would result in trial courts admitting false testimony merely on the basis that the methodologies supporting that testimony, which we now know to be unreliable and unvalidated, were admissible at some point in the past.’’ In response, the state claims that ‘‘[t]he trial court properly concluded that Legnani remained good law, even after the [NAS reports], because courts in Connecticut and throughout the nation, including those which have conducted [Por- ter] hearings, have overwhelmingly reaffirmed that expert testimony regarding firearm and toolmark identi- fication is admissible, notwithstanding the concerns expressed in that report.’’ We agree with the defendant that the trial court’s exclusive reliance on Legnani in assessing the request for a Porter hearing was erroneous. ‘‘It is axiomatic that [t]he trial court’s ruling on the admissibility of evidence is entitled to great deference. In this regard, the trial court is vested with wide discre- tion in determining the admissibility of evidence. . . . Accordingly, [t]he trial court’s ruling on evidentiary matters will be overturned only upon a showing of a clear abuse of the court’s discretion. . . . Because a August 17, 2021 CONNECTICUT LAW JOURNAL Page 17

337 Conn. 527 AUGUST, 2021 541 State v. Raynor

trial court’s ruling under Porter involves the admissibil- ity of evidence, we review that ruling on appeal for an abuse of discretion.’’ (Citation omitted; internal quota- tion marks omitted.) State v. Sorabella, 277 Conn. 155, 214, 891 A.2d 897, cert. denied, 549 U.S. 821, 127 S. Ct. 131, 166 L. Ed. 2d 36 (2006). Implicit in that well established principle, however, is the requirement that the trial court exercise its discretion. ‘‘Where . . . the trial court is properly called upon to exercise its discre- tion, its failure to do so is error.’’ (Internal quotation marks omitted.) Higgins v. Karp, 243 Conn. 495, 504, 706 A.2d 1 (1998). Therefore, ‘‘we must determine whether the trial court abused its discretion in deter-mining that a Porter hearing was not required and, if so, we must also determine whether this ruling was nevertheless harm- less.’’ State v. Martinez, 143 Conn. App. 541, 557, 69 A.3d 975 (2013), rev’d on other grounds, 319 Conn. 712, 127 A.3d 164 (2015). ‘‘In order to establish the harmfulness of a trial court ruling, the defendant must show that it is more probable than not that the improper action affected the result.’’ (Internal quotation marks omitted.) State v. Torres, 85 Conn. App. 303, 328, 858 A.2d 776, cert. denied, 271 Conn. 947, 861 A.2d 1179 (2004). In the present case, it is apparent from the record that the trial court failed to exercise its discretion when it denied the defendant’s motion for a Porter hearing. The trial court did not consider the NAS reports that the defendant cited in his motion; it noted that it had reviewed the reports but that it was bound by Legnani to find that the science of firearm and toolmark identifi- cation is well established. Similarly, the Appellate Court stated that ‘‘Legnani is controlling precedent on the issue of whether the science of firearm and toolmark identification is well established, and thus binds [its] resolution of [the defendant’s] claim.’’ State v. Raynor, supra, 181 Conn. App. 770. We conclude that the trial court failed to exercise—and, therefore, abused—its discretion to determine whether the criticisms of fire- Page 18 CONNECTICUT LAW JOURNAL August 17, 2021

542 AUGUST, 2021 337 Conn. 527 State v. Raynor

arm and toolmark analysis contained in the NAS reports and highlighted by the defendant cast substantial enough doubt on whether the science of that field remains well established to warrant a Porter hearing.7 A mere cursory look at the ramifications of a trial court’s being absolutely bound by Legnani illustrates why such an approach would be impractical. Trial court judges serve a gatekeeping function with respect to the admissibility of expert testimony, and, in performing that function, they assess the validity of the methodolo- gies underlying proffered scientific evidence. See State v. Porter, 241 Conn. 57, 80–90, 698 A.2d 739 (1997), cert. denied, 523 U.S. 1058, 118 S. Ct. 1384, 140 L. Ed. 2d 645 (1998); see also Daubert v. Merrell Dow Pharmaceuti- cals, Inc., 509 U.S. 579, 589–90, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993). Over two decades ago, this court held that ‘‘an admissibility test for scientific evidence premised solely on its ‘general acceptance’ is conceptu- ally flawed,’’ thereby rejecting the then applicable Frye8 standard and adopting the Daubert approach to the admissibility of scientific evidence.9 (Emphasis in origi- 7 We emphasize that the question of whether the evidence referenced in the defendant’s motion for a Porter hearing casts sufficient doubt on the reliability of the methodology employed by the firearm and toolmark expert to warrant a Porter hearing must be vested, in the first instance, in the sound discretion of the trial court. See, e.g., State v. Jackson, 304 Conn. 383, 412, 40 A.3d 290 (2012). We note, however, that various courts have considered the NAS reports and concluded that firearm and toolmark evi- dence continues to be both reliable and admissible. See, e.g., United States v. Otero, 849 F. Supp. 2d 425, 430, 437–38 (D.N.J. 2012), aff’d, 557 Fed. Appx. 146 (2014); State v. Terrell, Superior Court, judicial district of New Haven, Docket No. CR-XX-XXXXXXX-S (March 21, 2019) (68 Conn. L. Rptr. 323, 325, 327); Johnston v. State, 27 So. 3d 11, 20–22 (Fla.), cert. denied, 562 U.S. 964, 131 S. Ct. 459, 178 L. Ed. 2d 292 (2010); State v. Adams, Docket No. COA10-1363, 2011 WL 1938270, *6–7 (N.C. App. May 17, 2011). 8 Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923). 9 Porter hearings held by trial courts are synonymous with Daubert hear- ings held by federal district courts and state courts of other jurisdictions. The hearings, regardless of their title, involve the application of the principles articulated in Daubert. See footnote 2 of this opinion. August 17, 2021 CONNECTICUT LAW JOURNAL Page 19

337 Conn. 527 AUGUST, 2021 543 State v. Raynor

nal.) State v. Porter, supra, 75–76. This court noted, however, that ‘‘some scientific principles have become so well established than an explicit Daubert analysis is not necessary for admission of evidence thereunder. By this, we do not mean to reestablish the Frye general acceptance test. We do acknowledge, however . . . that a very few scientific principles are so firmly estab- lished as to have attained the status of scientific law, such as the laws of thermodynamics, [and that such principles] properly are subject to judicial notice . . . . Evidence derived from such principles would clearly withstand a Daubert analysis, and thus may be admitted simply on a showing of relevance.’’ (Citation omitted; internal quotation marks omitted.) Id., 85 n.30. The Appellate Court has held that a trial court did not abuse its discretion where it concluded that ballis- tics and firearms analysis fell into that category of scien- tific principles so firmly established as to negate the need for a Porter hearing. State v. Legnani, supra, 109 Conn. App. 419–21. Science, however, is not static. Meth- odologies are continually challenged and improved so that an approach once favored by the scientific commu- nity may later cede to a novel approach or simply fall out of favor in its entirety. See, e.g., Bone Shirt v. Hazeltine, 461 F.3d 1011, 1026 (8th Cir. 2006) (Gruen- der, J., concurring in the judgment) (‘‘Science evolves, and scientific methods that were once considered unas- sailable truths have been discarded over time. Unrelia- ble testimony based upon those outdated theories and methods must be discarded as well, lest scientific stare decisis ensure that such theories survive only in court.’’); cf. Upjohn Co. v. Finch, 422 F.2d 944, 951 (6th Cir. 1970). The gatekeeping function of the trial court requires, at a minimum, that judges consider any new evidence that a defendant presents when deciding whether to grant or deny a motion for a Porter hearing. To hold otherwise would transform the trial court’s gatekeeping function—which requires judges to regu- Page 20 CONNECTICUT LAW JOURNAL August 17, 2021

544 AUGUST, 2021 337 Conn. 527 State v. Raynor

late carefully which categories of scientific evidence are sufficiently reliable to present to the fact finders— into one of routine mandatory admission of such evi- dence, regardless of advances in a particular field and its continued reliability. Having concluded that it was an abuse of discretion for the trial court to deny the defendant’s motion for a Porter hearing without considering the proffered evi- dence challenging the methodology supporting tool- mark and firearm analysis, we must now determine whether that error was harmful. ‘‘When an improper evidentiary ruling is not constitutional in nature, the defendant bears the burden of demonstrating that the error was harmful. . . . [W]hether [an improper ruling] is harmless in a particular case depends upon a number of factors, such as the importance of the witness’ testi- mony in the prosecution’s case, whether the testimony was cumulative, the presence or absence of evidence corroborating or contradicting the testimony of the wit- ness on material points, the extent of cross-examination otherwise permitted, and, of course, the overall strength of the prosecution’s case. . . . Most importantly, we must examine the impact of the . . . evidence on the trier of fact and the result of the trial. . . . [T]he proper standard for determining whether an erroneous eviden- tiary ruling is harmless should be whether the jury’s verdict was substantially swayed by the error. . . . Accordingly, a nonconstitutional error is harmless when an appellate court has a fair assurance that the error did not substantially affect the verdict.’’ (Internal quotation marks omitted.) State v. Edwards, 325 Conn. 97, 133, 156 A.3d 506 (2017). After reviewing the evidence in the present case, we lack a fair assurance that the trial court’s admission of Stephenson’s testimony did not substantially affect the verdict. There is no doubt that Stephenson was impor- tant to the state’s case; his testimony was the only object- ive evidence that connected the casings found at the August 17, 2021 CONNECTICUT LAW JOURNAL Page 21

337 Conn. 527 AUGUST, 2021 545 State v. Raynor

Enfield Street murder with the .223 caliber Kel-Tec assault rifle recovered by the police. There can be little doubt that the jurors likely found Stephenson’s expert opinion highly convincing in light of the technical nature of his analysis and his various credentials.10 See, e.g., State v. Jackson, 334 Conn. 793, 819, 224 A.3d 886 (2020) (‘‘[t]here can be little doubt that jurors would have viewed as highly convincing [the] expert opinion; the testimony was presented in technical terms and used impressive visual displays to convey important information, and it came from a law enforcement officer unconnected to the department that investigated the crime’’). The exclusion of Stephenson’s expert testimony would have made the state’s overall case against the defendant much weaker because Stephenson corrobo- rated the testimony of Rivera, the only witness to iden- tify the defendant as the shooter. Without Stephenson’s expert testimony, the state would have relied primarily on Rivera’s testimony related to the Enfield Street mur- der.11 Rivera was also the sole witness to testify that 10 From 1994 to 2014, Stephenson was employed by the Connecticut State Forensic Laboratory’s Firearms Identification Unit as a firearm and toolmark examiner. From 2008 to 2014, he also served as a member of the Scientific Working Group for Firearms and Toolmarks, which ‘‘wrote protocol and procedures and adopted information that was disseminated to the firearms and toolmark examiners’’ throughout the world. Since 2014, Stephenson had worked as a private consultant for firearm and toolmark identification. In addition, Stephenson had attended numerous trainings offered by the AFTE, the Bureau of Alcohol, Tobacco, Firearms and Explosives, and the Federal Bureau of Investigation. Over the course of his career, Stephenson had examined tens of thousands of firearms and testified a total of 321 times in both state and federal courts. 11 The jury heard testimony from Alisha Stevens, who was standing with the victim on Enfield Street when he was murdered. Stevens, however, could not identify the shooter or the firearm. The jury also heard testimony from Deborah Parker and Stephenson about the shooting on Baltimore Street. Parker identified the defendant as having fired a rifle at her and her partner, but she could not positively identify the .223 caliber Kel-Tec assault rifle as the weapon used. As described in part II of this opinion, however, testimony related to the Baltimore Street shooting had extremely limited probative value. Page 22 CONNECTICUT LAW JOURNAL August 17, 2021

546 AUGUST, 2021 337 Conn. 527 State v. Raynor

the defendant and the victim had a confrontation in the week leading up to the murder, to identify the defendant as the shooter on Enfield Street, to confirm that the .223 caliber Kel-Tec assault rifle was the same one used by the defendant, and to acknowledge that the defen- dant knew the police had subsequently recovered the murder weapon. Rivera, however, had both a motive to testify falsely and credibility issues. See State v. Jack- son, supra, 334 Conn. 819–20. Rivera testified that he was involved with the victim’s murder on Enfield Street in June, 2007, but he did not provide information to the police about that murder until August, 2011, more than four years later. In addition, when Rivera finally did speak to the police about the Enfield Street murder, it was only after he had been arrested in connection with an incident that occurred on July 1, 2011. Rivera con- fessed his role in the July 1, 2011 incident and provided a written statement to the police. He also provided additional statements related to several other incidents in Hartford, one of which was the Enfield Street murder. At trial, Rivera testified that he had been sentenced as a result of the July 1, 2011 incident and was serving a total effective sentence of forty-two years of incarcera- tion for convictions of murder, conspiracy to commit murder, and a weapons charge. In addition, the jury heard Rivera testify that he was arrested pursuant to a warrant on November 5, 2013, for his involvement in the Enfield Street murder and had pending charges of accessory to commit murder, conspiracy to commit mur- der, and criminal possession of a firearm. Rivera also had a number of other pending charges that, when com- bined, added up to several decades of potential jail time.12 12 The jury heard Rivera testify to the following pending cases in addition to the charges related to the Enfield Street murder, all of which he was arrested for, pursuant to a warrant issued on November 5, 2013: (1) an incident on December 15, 2007, for which he was charged with attempted murder and conspiracy to commit murder; (2) an incident on July 22, 2008, for which he was charged with conspiracy to commit murder, two counts of attempt to commit murder, and two counts of accessory to commit assault in the first degree; (3) an incident on August, 9, 2008, for which he was August 17, 2021 CONNECTICUT LAW JOURNAL Page 23

337 Conn. 527 AUGUST, 2021 547 State v. Raynor

In order to testify at the defendant’s trial, Rivera entered into a written cooperation agreement with the state, in which he agreed to disclose truthfully any and all mat- ters related to his criminal activity, and the criminal activ- ity of others with whom he was involved. In exchange, the state agreed that, if Rivera did so when called upon, it would agree to consent to a hearing for sentence mod- ification before a judge of the Superior Court. In addi- tion, the state agreed that it would recommend that any sentences Rivera received for his pending charges run concurrently to the sentence of forty-two years that he was serving. Nor did the extensive cross-examination of the ballis- tics expert render the admission of his testimony harm- less. Defense counsel rigorously cross-examined Steph- enson on the methodology, critiques, and partial sub- jectivity of firearm and toolmark analysis, including questions about his knowledge of, and the conclusions from, the NAS reports. In this manner, the jury heard testimony that cast at least some doubt on the reliability of the methodology. Stephenson, however, consistently explained that while the NAS reports contained sugges- tions for improving the methodology—some of which he acknowledged were sound—the criticisms did not undermine its scientific validity. Throughout his cross- examination, Stephenson maintained that his conclu- sions were accurate. See, e.g., State v. Edwards, supra, 325 Conn. 134–35 (rigorous cross-examination of expert by defense counsel led to admission that expert could not guarantee accuracy of maps or determine exact charged with manslaughter in the first degree with a firearm, six counts of accessory to commit assault in the first degree, and weapons offenses; (4) an incident on August 9, 2008, for which he was charged with conspiracy to commit murder, two counts of accessory to commit murder, and criminal possession of a firearm; and (5) an incident on August 10, 2008, for which he was charged with conspiracy to commit murder, two counts of attempt to commit murder, two counts of assault in the first degree, and weapons offenses. In addition, Rivera was also arrested pursuant to a warrant issued in October, 2012, and charged with the sale of narcotics. Page 24 CONNECTICUT LAW JOURNAL August 17, 2021

548 AUGUST, 2021 337 Conn. 527 State v. Raynor

location of defendant’s cell phone, contributing to determination that trial court’s improper admission of certain testimony regarding cell phone data constituted harmless error). In addition, juries ‘‘tend to give great credence and weight to what . . . experts say,’’ and the defendant sought to have this expert testimony excluded, thereby preventing the jury from hearing Ste- phenson testify at all. ‘‘Symposium on Forensic Expert Testimony, Daubert, and Rule 702,’’ 86 Fordham L. Rev. 1463, 1508 (2018); see also D. McQuiston-Surrett & M. Saks, ‘‘Communicating Opinion Evidence in the Foren- sic Identification Sciences: Accuracy and Impact,’’ 59 Hastings L.J. 1159, 1189 (2007) (‘‘unfortunately, cross- examination and the use of opposing experts do not appear to effectively counter expert testimony, regard- less of the logical vulnerability of the initial expert testi- mony’’). We therefore cannot conclude that the rigorous cross-examination mounted by defense counsel so undercut Stephenson’s testimony that its admission was necessarily harmless. The state’s sole claim that any error by the trial court is harmless is restricted to an argument regarding moot- ness. Specifically, the state argues that ‘‘this issue already has been examined by a Connecticut Superior Court after a Porter hearing; [State v. Terrell, Superior Court, judicial district of New Haven, Docket No. CR- XX-XXXXXXX-S (March 21, 2019) (68 Conn. L. Rptr. 323)]; and its conclusions in support of admissibility mirror those reached following similar hearings by courts in other jurisdictions.’’13 Relying on State v. Balbi, 89 Conn. 13 In Terrell, the defendant ‘‘moved to preclude the [s]tate from presenting the testimony of . . . a firearm and toolmark examiner . . . to the jury . . . because the methodology of toolmark analysis is not scientifically valid. In the alternative, the defendant request[ed] that the [trial] court limit the scope of [the expert’s] testimony by prohibiting him from testifying that the shell casing found at the scene was fired from the firearm located there.’’ State v. Terrell, supra, 68 Conn. L. Rptr. 324. In that case, the court, Alander, J., granted the defendant’s request for a Porter hearing notwithstanding State v. Raynor, supra, 181 Conn. App. 760, and State v. Legnani, supra, 109 Conn. App. 399. See State v. Terrell, supra, 324. In so doing, the court August 17, 2021 CONNECTICUT LAW JOURNAL Page 25

337 Conn. 527 AUGUST, 2021 549 State v. Raynor

App. 567, 576–77, 874 A.2d 288, cert. denied, 275 Conn. 919, 883 A.2d 1246 (2005),14 the state argues that ‘‘[a] determination by one court that a methodology satisfies the Porter test renders [it] unnecessary for other courts to repeat the process. . . . There is no compelling rea- in Terrell noted that ‘‘both [Raynor and Legnani] held that the trial court did not abuse its discretion in refusing to conduct . . . a hearing on the issue of firearm analysis . . . implicitly leav[ing] a trial court the discretion to hold such a hearing.’’ (Citation omitted.) Id. Furthermore, it concluded that, ‘‘[g]iven recent national studies raising questions regarding the method- ology used in firearm and toolmark examination . . . a hearing on the validity of the methodology was warranted.’’ Id. The court in Terrell pro- ceeded to hold a Porter hearing and concluded that the state had established that ‘‘basic techniques employed by firearm and [toolmark] examiners are generally accepted in the relevant scientific community.’’ Id., 327. As part of the Porter hearing, the court in Terrell considered the NAS reports as well as a report issued by the President’s Council of Advisors on Science and Technology, titled ‘‘Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature Comparison Methods . . . .’’ (Inter- nal quotation marks omitted.) Id., 325–26. The court also denied the defen- dant’s motion to preclude the testimony of the expert in its entirety but prohibited the expert from testifying that ‘‘the likelihood that a firearm other than the [one] recovered at the crime scene could have fired the recovered [subject] casing is so remote as to be considered a practical impossibility.’’ (Internal quotation marks omitted.) Id., 327; see id., 328. 14 In Balbi, the trial court took judicial notice of its own decision in a separate case, as well as the decision of another Superior Court judge, that a horizontal gaze nystagmus test was a scientifically reliable and relevant test. State v. Balbi, supra, 89 Conn. App. 572. During the pendency of the appeal in Balbi, the Appellate Court considered horizontal gaze nystagmus evidence in State v. Commins, 83 Conn. App. 496, 501–502, 850 A.2d 1074, aff’d, 276 Conn. 503, 886 A.2d 824 (2005). ‘‘In Commins, the trial judge conducted a Porter hearing during which he heard extensive testimony . . . . At the conclusion of the Porter hearing, the court found that the horizontal gaze nystagmus test and its underlying methodology [are] gener- ally accepted in the scientific community—a conclusion that alone would likely suffice to establish a sufficient foundation for admission—but also that [they satisfy] many of the remaining Porter criteria. On the basis of that determination, the [trial judge] allowed evidence related to the test.’’ State v. Balbi, supra, 575–76. The Appellate Court affirmed the judgment of the trial court in Commins. State v. Commins, supra, 514. Returning to the appeal in Balbi, the Appellate Court held that its ‘‘determination in Commins that horizontal gaze nystagmus evidence satisfies the Porter test for the admission of scientific evidence rendered it unnecessary for the [trial] court in [Balbi] to conduct its own Porter hearing prior to admitting evidence about the test.’’ State v. Balbi, supra, 576. There is no indication that additional evidence challenging the horizontal gaze nystagmus test was brought to the attention of the trial court in Balbi that was not previously considered in Commins. Page 26 CONNECTICUT LAW JOURNAL August 17, 2021

550 AUGUST, 2021 337 Conn. 527 State v. Raynor

son to put the state to the burden of having to reestab- lish, in case after case, the same proposition. Requiring our trial judges to repeatedly hold Porter hearings would serve no legitimate purpose and would need- lessly squander judicial resources.’’ (Citations omitted; internal quotation marks omitted.) Although, as dis- cussed subsequently in this opinion, we agree that a Porter hearing is not necessary in every trial in which scientific evidence is presented, that fact does not mean that a Porter hearing held by one trial court is binding on another. See, e.g., In re Emma F., 315 Conn. 414, 432–33, 107 A.3d 947 (2015) (‘‘[A] trial court decision does not establish binding precedent. . . . Indeed, under the law of the case doctrine, the trial court’s decision need not even be followed by a judge making a subsequent decision in [that] very case.’’ (Citations omitted; internal quotation marks omitted.)). Although the Superior Court in Terrell considered the same evi- dence presented by the defendant in the present case in his motion for a Porter hearing; see footnote 13 of this opinion; Terrell was also not appealed to, or reviewed by, this court. The state’s argument that the result of the Porter hearing in Terrell renders the issue presented in this appeal moot must, therefore, fail. For the foregoing reasons, we conclude that the error was harmful and that the defendant is entitled to a new trial. This court’s conclusion that a trial court must exer- cise its discretion to at least consider evidence pre- sented by the defendant when deciding whether to grant a motion for a Porter hearing does not mean that a defendant’s challenge, no matter how slight, to an estab- lished methodology warrants a full Porter hearing. We provide the following examples to illustrate the gate- keeping function of the trial courts in light of this deci- sion. When a trial court considers a defendant’s motion for a Porter hearing, it may decide that the methodology prior to that point either (1) has been deemed so well established so as to not warrant a Porter hearing, as August 17, 2021 CONNECTICUT LAW JOURNAL Page 27

337 Conn. 527 AUGUST, 2021 551 State v. Raynor

was the case with firearm and toolmark analysis at the time of the defendant’s trial, or (2) has been subject to a Porter hearing by another trial court. Under both of these scenarios, the trial judge has several options depending on the strength of the evidence presented in a motion for a Porter hearing. Under the first scenario, the trial court begins from the premise that the methodology is well established and that, as a result, a Porter hearing is not necessary. It must then consider the evidence presented by the defendant to determine whether that well established methodology has been sufficiently challenged to war- rant a Porter hearing. The trial court has the discretion to deny the motion, concluding that the defendant has not presented sufficient evidence in his motion to dem- onstrate that the methodology may no longer be well established, or to grant the motion, concluding that the defendant has presented evidence sufficiently casting doubt on the continued reliability of the methodology and, therefore, that a full Porter hearing is necessary.15 Although, under this scenario, the defendant bears the heavy burden of challenging a potentially lengthy scien- tific and legal history of the reliability of the methodol- ogy without a Porter hearing, that burden is not insur- mountable. This was recently evidenced when, despite the Appellate Court’s decision in Legnani, a Superior Court granted a Porter hearing on firearm and toolmark analysis. State v. Terrell, supra, 68 Conn. L. Rptr. 324; see footnote 13 of this opinion. Under the second scenario, once any trial court has held a Porter hearing on a particular methodology, then judges have slightly different options when considering 15 Even when an appellate court has upheld a trial court’s denial of a Porter hearing, such a decision is binding only to the extent that a future challenge to the reliability of the methodology relies on the same evidence considered by the prior trial court. In the present case, the NAS reports submitted by the defendant postdated the decision in Legnani. Page 28 CONNECTICUT LAW JOURNAL August 17, 2021

552 AUGUST, 2021 337 Conn. 527 State v. Raynor

motions for subsequent Porter hearings. If a party high- lights the same evidence challenging a methodology as was evaluated in a previous Porter hearing, the trial court may—but is not required to—take judicial notice of the previous hearing and consider the prior court’s analysis of the methodology and conclusion as to its reliability when exercising its discretion to grant or deny the defendant’s motion for a subsequent Porter hearing.16 If a party highlights new evidence regarding the reliability of the methodology that was not evaluated in the previous Porter hearing, the trial court may— but, again, is not required to—take judicial notice of the previous hearing and consider the prior court’s anal- ysis of the methodology and conclusion in conjunction with the additional evidence presented in determining whether to grant or deny a subsequent motion for a Porter hearing.17 For the foregoing reasons, we conclude that the trial court improperly denied the defendant’s motion for a Porter hearing based solely on Legnani, without consid- ering new evidence offered by the defendant, and we do not have a fair assurance that this error was not harm- less. As a result, the defendant is entitled to a new trial. B Scope of Testimony We address the defendant’s second claim because it is likely to arise on remand. Specifically, the defendant 16 For example, a trial court may take judicial notice of the conclusions reached with respect to firearm and toolmark methodology following the Porter hearing in Terrell. See generally State v. Terrell, supra, 68 Conn. L. Rptr. 323. 17 The purpose of taking judicial notice of previous Porter hearings is to avoid redundancies created by holding successive hearings when another trial court has considered the same evidence challenging the methodology. This does not, however, mean that the reliability of the underlying methodol- ogy is insulated from appellate review. A trial court’s decision to deny a Porter hearing is still subject to appellate review, and taking judicial notice does not prevent review for abuse of discretion. August 17, 2021 CONNECTICUT LAW JOURNAL Page 29

337 Conn. 527 AUGUST, 2021 553 State v. Raynor

claims that the Appellate Court improperly upheld the trial court’s denial of his motion in limine to proscribe the scope of Stephenson’s testimony, highlighting that at least one Superior Court in this state, as well as other courts across the country, have limited the opinions of firearm and toolmark examiners in a variety of manners, ‘‘including precluding them from stating that expelled casings or bullets are matched to a firearm to the exclu- sion of all other firearms, requiring them to clarify that the likelihood of their conclusions being true is ‘more likely than not,’ or requiring them to clarify that the cer- tainty of their opinions was limited in some other man- ner.’’ (Footnotes omitted.) The defendant argues that such a limitation is particularly appropriate as recent decades have ushered in a ‘‘greater reliance on inter- changeable parts in manufacturing [that] has substan- tially altered the degree of unique features in the firing pins and other components of firearms,’’ thereby erod- ing the fundamental assumptions of firearm and toolm- ark examinations so as to render them insufficiently reliable to permit match statements. The defendant fur- ther requests that, even if this court were to conclude that it was permissible for firearm and tool-mark experts to testify that a particular casing was fired from a spe- cific firearm, we use our supervisory authority to limit the scope of such testimony in Connecticut courts. In response, the state claims that the specific restriction requested by the defendant—‘‘more likely than not’’— is arbitrary, inaccurate, and unsupported by the law generally applicable to expert testimony. The state con- cedes, however, that ‘‘it may be true that the methodol- ogy employed by firearm and toolmark identification experts would not currently support any representation that their conclusions are 100 percent infallible . . . .’’ Furthermore, if this court were to adopt a rule proscrib- ing the language that an expert must use in stating his opinion that a particular casing was fired from a specific Page 30 CONNECTICUT LAW JOURNAL August 17, 2021

554 AUGUST, 2021 337 Conn. 527 State v. Raynor

firearm, the state indicated that it would support a requirement that the expert phrase his opinion in terms of ‘‘a reasonable degree of certainty’’ or ‘‘a practical certainty . . . .’’ (Internal quotation marks omitted.) We agree with the state. We begin with the applicable standard of review. ‘‘[A] trial court retains broad discretion in ruling on the quali- fications of expert witnesses and [in] determining whether their opinions are relevant.’’ State v. Guilbert, 306 Conn. 218, 257, 49 A.3d 705 (2012). ‘‘[S]uch testimony is admis- sible if the trial court determines that the expert is qualified and that the proffered testimony is relevant and would aid the jury.’’ State v. Williams, 317 Conn. 691, 702, 119 A.3d 1194 (2015). When a party seeks to exclude or limit the scope of an expert’s testimony, the burden is on the party who files the motion in limine to show that the challenged remarks were prejudicial in light of the entire proceeding. Cf. State v. Binet, 192 Conn. 618, 628, 473 A.2d 1200 (1984). ‘‘We review a trial court’s decision to preclude expert testimony for an abuse of discretion. . . . We afford our trial courts wide discretion in determining whether to admit expert testimony and, unless the trial court’s decision is unrea- sonable, made on untenable grounds . . . or involves a clear misconception of the law, we will not disturb its decision.’’ (Internal quotation marks omitted.) State v. Williams, supra, 701–702. In the present case, defense counsel requested that the trial court restrict Stephenson to using very specific language that connoted a narrow scientific conclusion but did not provide the court with sufficient information to inform its decision. With respect to limiting the scope of Stephenson’s testimony, the defendant’s motion in limine simply requested that ‘‘a limiting order and instruction, similar to that in Glynn, be granted’’ with- out providing additional details as to what that instruc- tion would entail, why it would be appropriate for the August 17, 2021 CONNECTICUT LAW JOURNAL Page 31

337 Conn. 527 AUGUST, 2021 555 State v. Raynor

trial court to adopt that standard, or how failing to limit the scope of Stephenson’s testimony would prejudice the defendant. During pretrial argument on the motion, defense counsel urged the trial court to adopt the approach used in Glynn and to conclude that, due to the shortcomings of firearm and toolmark analysis described in the NAS reports, it should limit Stephenson to testifying only that it was ‘‘more likely than not’’ that the bullets were fired from the same gun. In addition, in both the motion in limine and oral arguments on the motion, the defense relied on a single United States District Court decision from the Southern District of New York to support its argument. Given the highly pro- scribed language requested, combined with the scant information and lack of case law provided in support of the defendant’s motion, we conclude that the trial court’s denial of the motion was not an abuse of discre- tion. See, e.g., State v. Binet, supra, 192 Conn. 624 (‘‘[t]he record before the court . . . could hardly provide it with a solid basis upon which to grant the defendant’s motion’’). We pause briefly to qualify our holding. Our conclu- sion that the trial court in the present case properly declined to limit the scope of Stephenson’s testimony to only ‘‘more likely than not’’ should not be taken as blanket approval of unlimited testimony from firearm and toolmark experts. As both the defendant and the state acknowledge, a substantial number of courts addressing this issue, including the United States Court of Appeals for the Second Circuit, have prohibited experts from testifying that a bullet or casing matched a specific firearm with absolute certainty or to the exclusion of all other firearms. See, e.g., United States v. Gil, 680 Fed. Appx. 11, 13–14 (2d Cir. 2017); United States v. Diaz, Docket No. CR-05-00167 (WHA), 2007 WL 485967, *1 (N.D. Cal. February 12, 2007). These courts, however, do not agree on what language is appropriate. Options include requiring an expert to state Page 32 CONNECTICUT LAW JOURNAL August 17, 2021

556 AUGUST, 2021 337 Conn. 527 State v. Raynor

that his degree of certainty is only ‘‘more likely than not’’; (internal quotation marks omitted) United States v. Glynn, supra, 578 F. Supp. 2d 574–75; that the identifi- cation is to ‘‘a reasonable degree of certainty’’; United States v. Monteiro, 407 F. Supp. 2d 351, 355 (D. Mass. 2006); that the identification is to ‘‘a practical certainty’’; (internal quotation marks omitted) United States v. McCluskey, Docket No. 10-2734 (JCH), 2013 WL 12335325, *10 (D.N.M. February 7, 2013); that the identi- fying characteristics on two items are ‘‘consistent with’’ each other (internal quotation marks omitted); United States v. Johnson, Docket No. 16 Cr. 281 (PGG), 2019 WL 1130258, *20 (S.D.N.Y. March 11, 2019); that the recovered firearm ‘‘cannot be excluded as the source’’ of the recovered casing; United States v. Shipp, 422 F. Supp. 3d 762, 783 (E.D.N.Y. 2019); or that the expert be requested to describe only similar and distinguishing features without characterizing a conclusion. United States v. Green, 405 F. Supp. 2d 104, 108–109 (D. Mass. 2005). At least one Connecticut trial court has prohib- ited a firearm and toolmark expert from testifying that ‘‘the likelihood that a firearm other than [the one] recov- ered at the crime scene could have fired the recovered [subject] casing is so remote as to be considered a practical impossibility.’’ (Internal quotation marks omit- ted.) State v. Terrell, supra, 68 Conn. L. Rptr. 327. In addi- tion, the state concedes that testifying to the certainty of a match ‘‘to the exclusion of all others’’ would not be appropriate and that it has no objection to a standard requiring an expert to limit his conclusions to either ‘‘a reasonable degree of certainty’’ or ‘‘a practical certainty . . . .’’ (Internal quotation marks omitted.) The defen- dant asks this court to exercise its supervisory author- ity18 to limit the scope of testimony from firearm and 18 ‘‘It is well settled that [a]ppellate courts possess an inherent supervisory authority over the administration of justice. . . . Under our supervisory authority, we have adopted rules intended to guide the lower courts in the administration of justice in all aspects of the criminal process. . . . Gener- ally, cases in which we have invoked our supervisory authority for rule August 17, 2021 CONNECTICUT LAW JOURNAL Page 33

337 Conn. 527 AUGUST, 2021 557 State v. Raynor

toolmark experts. Although we decline to do so, our deci- sion does not preclude trial courts from imposing appro- priate limits on such expert testimony when deemed necessary. For the foregoing reasons, we conclude that the trial court properly denied the defendant’s motion in limine to limit the scope of Stephenson’s testimony to a ‘‘more likely than not’’ standard. II UNCHARGED MISCONDUCT Although our conclusion in part I A of this opinion is dispositive of the present appeal, in the interest of judicial economy, we address the defendant’s claim that the trial court improperly admitted uncharged miscon- duct because we conclude the issue is likely to arise on remand. The following additional facts and procedural history are relevant to the resolution of this claim. Prior to the start of trial, the state filed a motion in which it sought permission to offer evidence of uncharged misconduct related to the Baltimore Street shooting in order to prove identity and means. Defense counsel opposed the motion, claiming that evidence of the Balti- more Street shooting was surplus and prejudicial in light of the fact that Rivera would provide in court identifications of both the defendant and the firearm used in the Enfield Street murder. Defense counsel fur- ther argued that testimony indicating that the defendant may have used the same gun in the subsequent Balti- making have fallen into two categories. . . . In the first category are cases wherein we have utilized our supervisory power to articulate a procedural rule as a matter of policy, either as [a] holding or dictum, but without reversing [the underlying judgment] or portions thereof. . . . In the second category are cases wherein we have utilized our supervisory powers to articulate a rule or otherwise take measures necessary to remedy a perceived injustice with respect to a preserved or unpreserved claim on appeal.’’ (Citation omitted; internal quotation marks omitted.) State v. Weatherspoon, 332 Conn. 531, 552–53, 212 A.3d 208 (2019). Page 34 CONNECTICUT LAW JOURNAL August 17, 2021

558 AUGUST, 2021 337 Conn. 527 State v. Raynor

more Street shooting was not relevant to the present case and served only to implicate him in a separate, unrelated crime. Following argument, the trial court granted the state’s motion for permission to offer evi- dence of uncharged misconduct, concluding that the evidence fell within the identity and means exceptions under § 4-5 (c) of the Connecticut Code of Evidence. At trial, the state proceeded to introduce evidence of the Baltimore Street shooting, primarily through the testimony of Deborah Parker, the victim of that crime. Parker testified that, at approximately 2:30 a.m. on February 16, 2008, she and her partner, Darryl Spence, returned to their residence on Baltimore Street in Hart- ford, where they lived with two of their sons. As Parker and Spence got out of their vehicle, which belonged to Parker’s oldest son, who did not live at the residence on Baltimore Street, Parker noticed two men walking on the street. As the men approached, Parker saw the taller of the two men fire a handgun in her direction. Then, the shorter of the two men fired a rifle in her direction, but she could not identify the specific weapon used. Parker saw the faces of both shooters illuminated by a streetlight as she took cover underneath a vehicle, and Spence ran away to hide elsewhere. Even though at least twenty-nine shots had been fired, neither Parker nor Spence was injured. The police responded to the Baltimore Street shoot- ing, but Parker declined to provide a written statement about the incident, and, at that time, she did not know the identity of the shooters and was not confident that she would be able to identify them in the future. Later that morning, Parker walked through her kitchen as her sons were looking at photographs on the computer from a concert they attended the night before and dis- cussing a fight they got into at that concert. Parker recognized the shooters in the photographs, and her sons provided the first name or nickname for each of August 17, 2021 CONNECTICUT LAW JOURNAL Page 35

337 Conn. 527 AUGUST, 2021 559 State v. Raynor

the men she identified, including the defendant. Parker called the police and told them where they could find the photographs, but indicated that she did not know the shooters’ full names and declined to file a written statement. The police never got back to Parker, and she did not follow up thereafter. In April, 2011, over three years after the Baltimore Street shooting, Parker’s oldest son was murdered in an unrelated incident. Then, on August 24, 2011, Parker was approached by the police to discuss the Baltimore Street shooting. At that time, Parker identified both shooters from a photographic array and learned the defendant’s last name from the police; she then signed a written statement regarding the Baltimore Street shooting. Parker was not aware of the victim’s murder on Enfield Street and had never heard his name. In addition, Stephenson testified regarding the cas- ings that were recovered from the Baltimore Street shooting. Of the twenty-two .223 caliber casings recov- ered from the crime scene, Stephenson positively identi- fied seventeen as having been fired from the recovered Kel-Tec assault rifle, the same weapon that Stephenson testified matched the casings from the Enfield Street murder and that Rivera identified as having been used by the defendant in that crime. See part I of this opinion. Stephenson’s testimony, combined with Parker’s, estab- lished that eight months following the Enfield Street murder, the defendant was identified as having been involved in the Baltimore Street shooting using a weapon that was matched through ballistics evidence to the weapon used in the Enfield Street murder, which was subsequently recovered five months later through an unrelated investigation. On appeal to the Appellate Court, the defendant argued that the probative value of the uncharged mis- conduct evidence was outweighed by the risk of unfair prejudice. State v. Raynor, supra, 181 Conn. App. 772. Page 36 CONNECTICUT LAW JOURNAL August 17, 2021

560 AUGUST, 2021 337 Conn. 527 State v. Raynor

Specifically, the defendant argued that ‘‘the evidence [was] more prejudicial than probative because Parker’s identification of the defendant was exceedingly unre- liable, that the similarities between the charged and uncharged conduct [rendered the] admission of the uncharged misconduct overly prejudicial, and that the uncharged misconduct painted the defendant as a deranged gunman.’’ (Internal quotation marks omitted.) Id., 774. The state argued that the trial court properly admitted the evidence to establish identity and means. Id., 772. The Appellate Court reasoned that, ‘‘[a]lthough the facts of the uncharged misconduct involved the defendant attempting to shoot Parker and Spence, [it was] much less severe than [that] of the charged con- duct, and, therefore, admission of the uncharged mis- conduct evidence cannot be said to have unduly aroused the jurors’ emotions.’’19 Id., 778. ‘‘Additionally, the [trial] court . . . gave the jury limiting instructions on three occasions . . . . These . . . instructions provided, inter alia, that the uncharged misconduct evi- dence was being admitted solely to show or establish [the] identity of the person who committed the crimes alleged . . . and the availability of the means to com- mit those crimes.’’ (Internal quotation marks omitted.) Id., 777. Accordingly, the Appellate Court concluded that the trial court had not abused its discretion in determining that the probative value of the uncharged misconduct was not outweighed by the risk of unfair prejudice. Id., 777–78. This appeal followed. 19 The Appellate Court also could not say that ‘‘admission of the uncharged misconduct evidence created a distracting side issue, as the evidence admit- ted linked the rifle and the perpetrator of the uncharged shooting to the murder at issue in [the present] case. Additionally, the presentation of evidence related to the Baltimore Street shooting did not take up an inordi- nate amount of time, as the presentation of the uncharged misconduct evidence comprised at most one and one-half days of a six day trial. Finally, the defendant was not unfairly surprised by the admission of this evidence, as it was admitted in the defendant’s first trial and [as] the state filed a pretrial motion for the admission of uncharged misconduct evidence.’’ (Footnote omitted.) State v. Raynor, supra, 181 Conn. App. 778. August 17, 2021 CONNECTICUT LAW JOURNAL Page 37

337 Conn. 527 AUGUST, 2021 561 State v. Raynor

In the present case, the defendant claims that the Appellate Court incorrectly upheld the trial court’s admission of uncharged misconduct related to the Balti- more Street shooting because the ‘‘tremendous prejudi- cial impact of the prior misconduct evidence over- whelmed its minimal probative value.’’ In response, the state claims that the trial court’s ruling that the proba- tive value of the Baltimore Street shooting outweighed its prejudicial impact was neither so arbitrary as to vitiate logic, nor based on improper or irrelevant fac- tors.20 Specifically, the state highlights that this court has recognized the probative value of evidence when a defendant used the same weapon in another crime, and that this court has observed that there is a reduced risk of unduly arousing the jurors’ emotions when the severity of the uncharged misconduct is less than the severity of the crime at issue. On the basis of the evidence contained within the record presently before us, we agree with the defendant that the trial court incorrectly admitted the challenged evidence because its prejudi- cial impact outweighed its probative value. ‘‘[A]s a general rule, evidence of prior misconduct is inadmissible to prove that a criminal defendant is guilty of the crime of which the defendant is accused. . . . Such evidence cannot be used to suggest that the defen- dant has a bad character or a propensity for criminal behavior.’’ (Internal quotation marks omitted.) State v. Collins, 299 Conn. 567, 582, 10 A.3d 1005, cert. denied, 565 U.S. 908, 132 S. Ct. 314, 181 L. Ed. 2d 193 (2011). ‘‘The well established exceptions to the general prohibition against the admission of uncharged misconduct are set forth in § 4-5 [c] of the Connecticut Code of Evidence, which provides in relevant part that ‘[e]vidence of other crimes, wrongs or acts of a person is admissible . . . 20 The state also contends that any error in this regard was harmless. Because we address this claim as an issue likely to arise on remand, we need not address questions of harmless error in the present appeal. Page 38 CONNECTICUT LAW JOURNAL August 17, 2021

562 AUGUST, 2021 337 Conn. 527 State v. Raynor

to prove intent, identity, malice, motive, common plan or scheme, absence of mistake or accident, knowledge, a system of criminal activity, or an element of the crime, or to corroborate crucial prosecution testimony.’ ’’ Id., 583. ‘‘We have developed a two part test to determine the admissibility of such evidence. First, the evidence must be relevant and material to at least one of the cir- cumstances encompassed by the exceptions [set forth in § 4-5 (c) of the Connecticut Code of Evidence].21 . . . Second, the probative value of the evidence must outweigh its prejudicial effect. . . . Because of the dif- ficulties inherent in this balancing process, the trial court’s decision will be reversed only whe[n] abuse of discretion is manifest or whe[n] an injustice appears to have been done. . . . On review by this court, there- fore, every reasonable presumption should be given in favor of the trial court’s ruling.’’ (Footnote added; internal quotation marks omitted.) Id., 582. ‘‘In determining whether the prejudicial effect of oth- erwise relevant evidence outweighs its probative value, we consider whether: (1) . . . the facts offered may unduly arouse the [jurors’] emotions, hostility or sympa- thy, (2) . . . the proof and answering evidence it pro- vokes may create a side issue that will unduly distract the jury from the main issues, (3) . . . the evidence offered and the counterproof will consume an undue amount of time, and (4) . . . the defendant, having no reasonable ground to anticipate the evidence, is unfairly surprised and unprepared to meet it.’’ (Internal quota- tion marks omitted.) Id., 586–87. This court has repeatedly held that ‘‘[t]he prejudicial impact of uncharged misconduct evidence is assessed in light of its relative ‘viciousness’ in comparison with the charged conduct.’’ State v. Campbell, 328 Conn. 444, 522–23, 180 A.3d 882 (2018); see also State v. Collins, 21 The defendant does not contest the relevancy of the evidence relating to the Baltimore Street shooting. August 17, 2021 CONNECTICUT LAW JOURNAL Page 39

337 Conn. 527 AUGUST, 2021 563 State v. Raynor

supra, 299 Conn. 588 (‘‘[u]ncharged misconduct evi- dence has been held not unduly prejudicial when the evidentiary substantiation of the vicious conduct, with which the defendant was charged, far outweighed, in severity, the character of his prior misconduct’’ (inter- nal quotation marks omitted)). The rationale behind this proposition is that the jurors’ emotions are already aroused by the more severe crime of murder, for which the defendant is charged, and, thus, a less severe, uncharged crime is unlikely to arouse their emotions beyond that point. The question of whether the evidence is unduly prejudicial, however, does not turn solely on the relative severity of the uncharged misconduct. Instead, prejudice is assessed on a continuum—on which severity is a factor—but whether that prejudice is undue can only be determined when it is weighed against the probative value of the evidence. In the present case, the Baltimore Street shooting was a less severe crime than the Enfield Street murder solely due to the fact that neither Parker nor Spence was hit by one of the dozens of shots fired. The Balti- more Street shooting and the Enfield Street murder, however, shared other common characteristics, includ- ing individuals being shot at by assailants outside of their own homes. Each incident involved two people, one male and one female, who were currently in, or had recently been in, a romantic relationship, as opposed to groups of friends or associates. These shootings each occurred in the middle of the night, involved dozens of shots being fired, and ended with the assailants fleeing the scene. While these two incidents were not identical, the similarities cannot be dismissed as irrelevant, and, together, they increase the risk of undue prejudice. See, e.g., State v. Artieri, 206 Conn. 81, 87, 536 A.2d 567 (1988) (‘‘[w]here the prior crime is quite similar to the offense being tried, a high degree of prejudice is created and a strong showing of probative value would be neces- Page 40 CONNECTICUT LAW JOURNAL August 17, 2021

564 AUGUST, 2021 337 Conn. 527 State v. Raynor

sary to warrant admissibility’’ (internal quotation marks omitted)). In addition, evidence of the Baltimore Street shooting was introduced through the testimony of Parker, the victim of that crime, and her testimony was not limited only to the fact that there was a shooting, with no other details regarding the surrounding events. See State v. Collins, supra, 299 Conn. 589 (‘‘we find significant the trial court’s efforts to have the prosecution admonish its witnesses that any testimony about the [previous uncharged misconduct] was to be limited only to the fact that there was a shooting’’). Instead, Parker testi- fied in detail about the shooting, including her feelings of being scared and her exact movements during the shooting, and she detailed her initial efforts to follow up with the police. Parker also described events beyond the Baltimore Street shooting. She suggested that her sons and the defendant had been involved in an alterca- tion at a concert the night before and revealed that her oldest son was murdered shortly before she spoke to the police again in August, 2011. While none of these details in isolation is determinative of whether the evi- dence is unduly prejudicial,22 when combined, they could arouse the jurors’ emotions and require a higher level of probative value to overcome the prejudicial impact.23 22 The judge at the defendant’s first trial noted the risk of admitting improper character evidence. During pretrial oral arguments, the trial court stated that it was ‘‘very concerned in this case that the . . . defendant not be tried . . . under the theory of a bad man. We don’t do that. So, I think both sides are going to have to be careful in not opening doors that—I don’t know if I’ve seen a case with so many doors.’’ 23 The state claims that the prejudicial impact is negated ‘‘in light of the care with which the trial court, on three separate occasions, cautioned the jury as to the limited use for which [it was] to consider the evidence.’’ We acknowledge that the trial court gave the jury limiting instructions that the uncharged misconduct evidence was being admitted ‘‘solely to show or establish [the] identity of the person who committed the crimes alleged in this information, and the availability of the means to commit those crimes’’ on the following three occasions: (1) prior to the state first presenting August 17, 2021 CONNECTICUT LAW JOURNAL Page 41

337 Conn. 527 AUGUST, 2021 565 State v. Raynor

The probative value of the Baltimore Street shooting was too low to overcome its prejudicial impact. The Baltimore Street shooting occurred eight months after the Enfield Street murder. There was no evidence to suggest that the Baltimore Street shooting was moti- vated by or related to the earlier Enfield Street murder. They were separate shootings and, with the exception of the defendant, involved different participants and unrelated victims. Parker’s testimony relating to the Baltimore Street shooting was admitted to prove that the defendant had been involved in this separate, subse- quent gun related crime, where the shell casings matched the .223 caliber Kel-Tec assault rifle. Evidence that the defendant was involved in a shooting in which he allegedly used the same weapon only minimally increased the probability that he was the shooter who used that weapon eight months prior during the Enfield Street murder. This connection is further eroded by the fact that the .223 caliber Kel-Tec assault rifle was not recovered at the scene of the Baltimore Street shooting but, instead, five months later from a different location following a lead provided by a confidential informant. Cf. State v. Collins, supra, 299 Conn. 570–76.24 The state evidence of the Baltimore Street shooting, (2) following Parker’s testimony, and (3) during its final charge to the jury. See State v. Beavers, 290 Conn. 386, 406, 963 A.2d 956 (2009) (‘‘the care with which the [trial] court weighed the evidence and devised measures for reducing its prejudicial effect mili- tates against a finding an abuse of discretion’’ (internal quotation marks omitted)). Although limiting instructions serve to ‘‘minimize any prejudice that might arise from the admission of . . . prior misconduct evidence’’; State v. Cutler, 293 Conn. 303, 314, 977 A.2d 209 (2009), overruled in part on other grounds by State v. Elson, 311 Conn. 726, 91 A.3d 862 (2014); the instructions would have needed to virtually eliminate the prejudice in the present case, given the very low probative value of the evidence of the Baltimore Street shooting. We cannot conclude that even multiple limiting instructions could have achieved that goal. In addition, limiting instructions may feature more prominently in a harmless error analysis. See footnote 20 of this opinion. 24 The Appellate Court held that Collins guided its resolution of the claim based on the proposition that uncharged misconduct is admissible when the ‘‘severity of the charged conduct outweigh[s] the severity of the uncharged Page 42 CONNECTICUT LAW JOURNAL August 17, 2021

566 AUGUST, 2021 337 Conn. 527 State v. Raynor

did not need to introduce evidence of the Baltimore Street shooting to connect the defendant to the .223 caliber Kel-Tec assault rifle used in the Enfield Street murder that the police subsequently recovered from a different location. The state presented direct evidence from Rivera connecting the defendant to that gun and the Enfield Street murder.25 Having reviewed the record in the present case, we conclude that the prejudicial effect of the uncharged misconduct unduly exceeded its probative value. The judgment of the Appellate Court is reversed and the case is remanded to that court with direction to reverse the judgment of the trial court and to remand the case to that court for a new trial. In this opinion the other justices concurred.

D’AURIA, J., with whom PALMER and MULLINS, Js., join, concurring. I agree with and join the majority opinion. In particular, I agree with the majority that conduct.’’ State v. Raynor, supra, 181 Conn. App. 777. The facts of Collins, however, are distinguishable from the facts of the present case. In Collins, the uncharged misconduct involved a shooting that not only occurred in closer temporal proximity to the charged murder than in the present case— four months as opposed to eight months—it occurred prior to the charged murder. See State v. Collins, supra, 299 Conn. 570–72. We also observe that the defendant in that case admitted to committing the uncharged misconduct with a chrome and black nine millimeter handgun. Id., 572. Further, in Collins, a witness testified to having seen the defendant with the same gun used in the uncharged misconduct several days before the charged murder. Id., 573–74. Thus, the facts of Collins made the uncharged misconduct highly probative. 25 Rivera provided eyewitness testimony that the defendant purchased the .223 caliber Kel-Tec assault rifle one to one and one-half months before the Enfield Street murder, and also testified that the defendant had called him for the purpose of finding and shooting members of The Avenue, the defen- dant used the .223 caliber Kel-Tec assault rifle as the shooter at the Enfield Street murder, and the defendant knew where the weapon was located prior to its being recovered by the police, because the defendant was the one who notified Rivera that it had been found. In addition, Stephenson testified that the casings recovered from the Enfield Street murder matched the recovered .223 caliber Kel-Tec assault rifle. August 17, 2021 CONNECTICUT LAW JOURNAL Page 43

337 Conn. 527 AUGUST, 2021 567 State v. Raynor

the trial court improperly refused to conduct a hearing pursuant to State v. Porter, 241 Conn. 57, 80–90, 698 A.2d 739 (1997), cert. denied, 523 U.S. 1058, 118 S. Ct. 1384, 140 L. Ed. 2d 645 (1998), and that the inclusion of the contested expert evidence substantially affected the verdict. Specifically, I agree that the trial court abused its discretion by relying on the holding of State v. Legnani, 109 Conn. App. 399, 421, 951 A.2d 674, cert. denied, 289 Conn. 940, 959 A.2d 1007 (2008), to deny the defendant, Donald Raynor, a Porter hearing. Moreover, I agree that a separate trial court’s ruling in State v. Terrell, Superior Court, judicial district of New Haven, Docket No. CR-XX-XXXXXXX-S (March 21, 2019) (68 Conn. L. Rptr. 323), which has not been subject to appellate scrutiny, cannot save the trial court’s ruling in the pres- ent case. See footnote 13 of the majority opinion and accompanying text. I write separately to raise an issue regarding the proper remedy in cases like this one in which the trial court improperly refuses to hold a Porter hearing. I believe there is an argument that this error can be cured by a limited remand for a Porter hearing, with the vacatur of the trial court’s judgment following only if the trial court ultimately finds the contested expert evidence inadmissible. This court previously has held that, when the trial court conducts a Porter hearing but abuses its discre- tion by admitting or precluding the expert evidence, the proper remedy is a new trial if the admission of the expert evidence substantially affected the verdict. See, e.g., Maher v. Quest Diagnostics, Inc., 269 Conn. 154, 157 n.4, 182–83, 847 A.2d 978 (2004). The reason for such a remedy is logical: the record establishes that the inadmissible evidence infected the trial court’s judg- ment. Likewise, this court also has remanded a case for a new trial when the trial court improperly refused to hold a Porter hearing at all, and the expert evidence substantially affected the verdict, although the court could not determine whether the evidence was inadmis- Page 44 CONNECTICUT LAW JOURNAL August 17, 2021

568 AUGUST, 2021 337 Conn. 527 State v. Raynor

sible. See Prentice v. Dalco Electric, Inc., 280 Conn. 336, 339, 907 A.2d 1204 (2006) (‘‘We conclude that the testimony in question was scientific evidence that required a validity assessment designed to ensure relia- bility pursuant to our analysis in Porter. Accordingly, we reverse the judgment of the trial court and remand the case for a new trial.’’), cert. denied, 549 U.S. 1266, 127 S. Ct. 1494, 167 L. Ed. 2d 230 (2007). The reason for a new trial in this second scenario, however, is not as clear to me. This court has not had the opportunity to determine whether a different remedy is required. In a future case, I would entertain an argument that, once an appellate court determines that the trial court improperly refused to conduct a Porter hearing and the contested expert evidence substantially affected the verdict, a new trial is not automatically the proper rem- edy but that, instead, we can direct the trial court on remand to hold the Porter hearing, even postjudgment. If the expert witness’ methodology does not pass mus- ter, then the trial court would have to vacate the convic- tion and order a new trial. If the methodology passes muster, however, the judgment of conviction would remain intact. After the Porter hearing, however, the defendant could appeal the trial court’s ruling confirm- ing the admissibility of the evidence. I recognize, however, that it is not appropriate in the present case to address or decide this issue for two reasons. First, under current law, which neither party contests or asks us to overrule, this court has deter- mined that a remand for a new trial is the proper remedy when the trial court improperly refused to hold a Porter hearing and the admission of the expert testimony sub- stantially affected the verdict. Second, neither party has raised or briefed whether a remand for a Porter hearing is the appropriate remedy, before this court or the Appellate Court. I therefore agree with the majority’s determination to remand the present case for a new trial. August 17, 2021 CONNECTICUT LAW JOURNAL Page 45

337 Conn. 527 AUGUST, 2021 569 State v. Raynor

I now turn to our case law. In Prentice v. Dalco Electric, Inc., supra, 280 Conn. 345–47, a civil case, this court held that the trial court incorrectly determined that expert testimony by a mechanical and forensic engineer regarding the effect of wind conditions was not scien- tific, and, thus, the court improperly refused to conduct a Porter hearing. After determining that this error sub- stantially affected the verdict, this court remanded the case for a new trial. Id., 359–61. My research has not turned up a criminal case from this court or the Appel- late Court in which the trial court improperly refused to hold a Porter hearing and that error substantially affected the verdict, but it would have to follow that a new trial would be the remedy in criminal cases as well.1

Additionally, a remand for a new trial is the majority rule under jurisprudence regarding Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993), the federal analogue to Porter. See, e.g., Dodge v. Cotter Corp., 328 F.3d 1212, 1229 (10th Cir.) (when trial court fails to conduct Daub- ert hearing, proper remedy is new trial, not remand for Daubert hearing), cert. denied, 540 U.S. 1003, 124 S. Ct. 533, 157 L. Ed. 2d 408 (2003); cf. Goebel v. Denver & 1 This is supported by this court’s recent decision in State v. Edwards, 325 Conn. 97, 133–34, 156 A.3d 506 (2017), in which we held that the trial court improperly refused to hold a Porter hearing but that this failure was harmless, as it did not substantially affect the verdict. Nevertheless, in discussing the applicable standard of review and legal principles, we stated that a new trial would be required only if the defendant showed that the error substantially affected the verdict, citing to cases involving improper evidentiary rulings that did not involve Porter. Id., 123 (‘‘ ‘[i]f we determine that a court acted improperly with respect to the admissibility of expert testimony, we will reverse the trial court’s judgment and grant a new trial only if the impropriety was harmful to the appealing party’ ’’); id. (quoting Weaver v. McKnight, 313 Conn. 393, 405, 97 A.3d 920 (2014), which involved trial court’s improper exclusion of expert testimony on ground that experts lacked sufficient qualifications). This at least suggests that a remand for a new trial would have been the proper remedy under current law if the defendant had proven harm. Page 46 CONNECTICUT LAW JOURNAL August 17, 2021

570 AUGUST, 2021 337 Conn. 527 State v. Raynor

Rio Grande Western Railroad Co., 215 F.3d 1083, 1089 (10th Cir. 2000) (explaining that, when trial court fails to conduct Daubert hearing, there are two possible remedies: new trial, or, if record is sufficient to deter- mine reliability, appeals court may undertake its own Daubert analysis). There is at least one court that has recognized that a limited remand for a Porter hearing may be the proper remedy in certain instances, however. See United States v. Bacon, 979 F.3d 766, 770 (9th Cir. 2020) (en banc) (overruling prior case law that required reviewing court to remand case for new trial if trial court improperly failed to hold Daubert hearing and holding instead that, ‘‘when a panel of [the Ninth Circuit] concludes that [a] district court has committed a [nonharmless] Daubert error, the panel has discretion to impose a remedy ‘as may be just under the circumstances,’ ’’ including order- ing limited remand); see United States v. Bacon, 829 Fed. Appx. 247, 248 (9th Cir. 2020) (directing District Court on remand to conduct whatever proceedings it deems appropriate to determine whether expert testi- mony was admissible pursuant to Daubert); see also Estate of Barabin v. AstenJohnson, Inc., 740 F.3d 457, 468–69 (9th Cir.) (Nguyen, J., concurring in part and dissenting in part) (questioning majority’s remedy of new trial and urging conditional vacating of judgment and remand to District Court to conduct Daubert deter- mination in first instance), cert. denied, 574 U.S. 815, 135 S. Ct. 55, 190 L. Ed. 2d 30 (2014); Estate of Barabin v. AstenJohnson, Inc., supra, 471, (Nguyen, J., concur- ring in part and dissenting in part) (remand for Daubert hearing is better remedy because it does not unreason- ably burden judicial system by requiring new trial and because, in absence of Daubert hearing, reviewing court cannot determine whether expert testimony was inad- missible and, thus, cannot conduct harmless error anal- ysis). In the present case, the trial court erred by failing August 17, 2021 CONNECTICUT LAW JOURNAL Page 47

337 Conn. 527 AUGUST, 2021 571 State v. Raynor

to exercise its discretion. A limited remand for a Porter hearing could remedy that error, requiring a new trial only if the expert witness’ methodology does not pass muster. This procedure may be a more efficient use of judicial resources than requiring new trials in all cases involving the improper refusal to conduct a Porter hear- ing. ‘‘If the disputed expert testimony was admissible pursuant to . . . [Porter], despite the [trial] court’s fail- ure to fulfill its gatekeeping function, then no harm, no foul. On the other hand, if the testimony was inadmissi- ble, then a . . . [new trial would be required].’’ (Emphasis added.) Estate of Barabin v. AstenJohnson, Inc., supra, 469 (Nguyen, J., concurring in part and dissenting in part). For similar efficiency reasons, this court has employed this exact procedure to remedy errors in other criminal cases that are at least arguably analogous. For example, in State v. Jarzbek, 204 Conn. 683, 684, 529 A.2d 1245 (1987), cert. denied, 484 U.S. 1061, 108 S. Ct. 1017, 98 L. Ed. 2d 982 (1988), a case involving sexual abuse of a minor victim, the state requested and was granted permission to exclude the defendant from the witness room during the videotaping of the minor victim’s tes- timony. After a jury found the defendant guilty, he appealed, arguing that this violated his constitutional right to confrontation. Id., 690. This court held that, ‘‘in criminal prosecutions involving the alleged sexual abuse of children of tender years, the practice of video- taping the testimony of a minor victim outside the physi- cal presence of the defendant is, in appropriate cir- cumstances, constitutionally permissible’’ but that the trial court first ‘‘must determine, at an evidentiary hear- ing, whether the state has demonstrated a compelling need for excluding the defendant from the witness room during the videotaping of a minor victim’s testimony.’’ Id., 704. Because the trial court did not conduct this hearing prior to granting the state’s request, this court Page 48 CONNECTICUT LAW JOURNAL August 17, 2021

572 AUGUST, 2021 337 Conn. 527 State v. Raynor

remanded the case for an evidentiary hearing, explain- ing that, ‘‘[i]f the [trial] court concludes that the state has not met its burden of proving such a need by clear and convincing evidence, the defendant is entitled to a new trial from which the videotaped testimony of [the victim] must be excluded. If the court concludes that the state has met its burden, the defendant’s conviction must stand, subject to any further appeal by the defen- dant to this court concerning the validity of the trial court’s ruling on this issue.’’ Id., 708. Similarly, when this court has determined that the trial court improperly failed to conduct or apply the proper standard for a competency hearing, this court has remanded the case for a competency hearing and required a new trial only if the trial court determines on remand that the defendant was incompetent to stand trial at the time of the trial. See State v. Connor, 292 Conn. 483, 528–29, 533, 973 A.2d 627 (2009). Likewise, this court has determined that, when the trial court improperly failed to conduct a hearing pursuant to Bat- son v. Kentucky, 476 U.S. 79, 106 S. Ct. 1712, 90 L. Ed. 2d 69 (1986), and the circumstances surrounding the voir dire can be reconstructed, a limited remand for a Batson hearing is the appropriate remedy, not a new trial. See State v. Rigual, 256 Conn. 1, 12–13, 771 A.2d 939 (2001). Additionally, this court has remanded cases for hearings pursuant to Brady v. Maryland, 373 U.S. 83, 83 S. Ct. 1194, 10 L. Ed. 2d 215 (1963), conditioning the need for any new trial on the outcome of the Brady hearing. See State v. Pollitt, 205 Conn. 132, 134–37, 531 A.2d 125 (1987).2 2 In State v. Snook, 210 Conn. 244, 555 A.2d 390, cert. denied, 492 U.S. 924, 109 S. Ct. 3258, 106 L. Ed. 2d 603 (1989), this court explained that it has ‘‘on several occasions remanded a case for the limited purpose of conducting an evidentiary hearing necessary for appellate review of a claim. See, e.g., State v. Badgett, 200 Conn. 412, 433–34, 512 A.2d 160 [(remanding case for factual hearing to determine whether illegally discovered evidence was admissible under recently articulated inevitable discovery rule), cert. denied, 479 U.S. 940, 107 S. Ct. 423, 93 L. Ed. 2d 373 (1986)]; State v. Pollitt, August 17, 2021 CONNECTICUT LAW JOURNAL Page 49

337 Conn. 527 AUGUST, 2021 573 State v. Raynor

In these cases, a reversal of the conviction followed only if the trial court on remand found that reversible error occurred. This is not unlike ordering a trial court to make findings by way of an articulation while an appeal is pending in our court, the only difference being that, procedurally, we would actually remand the matter rather than simply order an articulation. Even the rules for articulations prompted by a party’s motion contem- plate situations in which a trial court might take further evidence. See Practice Book § 66-5 (‘‘[i]f any party requests it and it is deemed necessary by the trial court, the trial court shall hold a hearing at which arguments may be heard, evidence taken or a stipulation of counsel received and approved’’). There might be contrary arguments I have not consid- ered. But the state and the public have an interest in pre- serving convictions otherwise fairly arrived at. In light of the case law previously discussed, I would entertain an argument in a future case that such a procedure should be undertaken when the trial court improperly refuses to conduct a Porter hearing. Until a Porter hearing is held, we do not know if the defendant’s trial was tainted and, thus, do not know whether a new trial is required. Nevertheless, in light of our current law and the fact that the appropriate scope of the remedy has not been argued by the parties, I agree with the majority that we must remand the present case for a new trial.

199 Conn. 399, 415–16, 508 A.2d 1 (1986) (remanding case for factual hearing to determine whether state suppressed exculpatory evidence); cf. State v. Garrison, 199 Conn. 383, 388–89, 507 A.2d 467 (1986) ([when] court [was] unable to determine from record whether state established defendant’s guilt beyond reasonable doubt, court ordered remand for further articulation of trial court’s grounds for rejecting defendant’s defenses).’’ State v. Snook, supra, 254.