State v. McFarland

353 Conn. 169

• Court: Supreme Court of Connecticut
• Decided: September 2, 2025
• Docket: SC20802
• Status: Published

Opinion

STATE OF CONNECTICUT v. WILLIE MCFARLAND (SC 20802) Mullins, C. J., and McDonald, D’Auria, Ecker, Alexander and Dannehy, Js.

Syllabus

Convicted of two counts of murder, the defendant appealed to this court. Although the murders occurred in 1987, the case remained unresolved until scientific advancements in DNA testing led to new findings that resulted in the defendant’s arrest in 2019. In a pretrial motion to dismiss the murder charges, which the trial court denied, the defendant claimed that the thirty- two year delay between the murders and his arrest violated his rights under the due process clauses of the federal and state constitutions. On appeal, the defendant renewed his constitutional claims and also challenged the trial court’s decision not to order a new competency hearing as well as an evidentiary ruling. Held: Page 4 CONNECTICUT LAW JOURNAL September 2, 2025

170 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland The trial court properly rejected the defendant’s federal and state due pro- cess claims arising from the prearrest delay, as the defendant failed to establish that his conviction offended the community’s sense of fair play and decency. With respect to its rejection of the defendant’s claim under the federal constitution, this court unanimously adhered to existing precedent applying a two-pronged test to prearrest delay claims pursuant to which a defendant, to establish a due process violation, must demonstrate that actual and substantial prejudice resulted from the delay and that the state delayed the defendant’s arrest to obtain an unfair tactical advantage or for other improper purposes. With respect to this court’s rejection of the defendant’s claim under the state constitution, a majority of this court adopted a balancing test similar to that endorsed by the trial court, pursuant to which the defendant must make a threshold showing of actual and substantial prejudice, the state then must establish the reasons for the delay, and, finally, the trial court balances the prejudice to the defendant against the state’s reasons for the delay. The trial court did not abuse its discretion when it declined to order a new competency evaluation of the defendant after finding him competent to stand trial. The trial court had already ordered four competency evaluations, and, in the absence of a substantial change in circumstances raising a reasonable doubt as to the defendant’s competency, the trial court properly declined to order a fifth competency evaluation. Moreover, contrary to the defendant’s claims, it was not improper for the trial court to rely on its own observations of the defendant’s behavior or to consider a prior competency report in denying the defendant’s request for another competency evaluation. The trial court did not abuse its discretion in declining to admit certain out- of-court statements by a deceased witness, S, under the residual exception to the hearsay rule because, although there was a reasonable necessity for their admission, the statements were disjointed, inconsistent, implausible and unreliable, S was never subject to cross-examination regarding the numerous inconsistencies in her statements, the statements contained multi- ple layers of hearsay, and the fact that S signed two of the statements under penalty of law and made handwritten edits to one of them did not otherwise render the statements trustworthy and reliable. (Six justices concurring separately in three opinions) Argued October 31, 2024—officially released September 2, 2025

Procedural History

Substitute information charging the defendant with two counts of the crime of murder, brought to the Supe- September 2, 2025 CONNECTICUT LAW JOURNAL Page 5

353 Conn. 169 SEPTEMBER, 2025 171 State v. McFarland

rior Court in the judicial district of New Haven, where the case was tried to the jury before Vitale, J.; verdict and judgment of guilty, from which the defendant appealed to this court. Affirmed. Lisa J. Steele, assigned counsel, for the appellee (defendant). Robert J. Scheinblum, special assistant state’s attor- ney, with whom, on the brief, were John Doyle, state’s attorney, Seth Garbarsky, supervisory assistant state’s attorney, and Lisa M. D’Angelo, executive assistant state’s attorney, for the appellee (state). Opinion

PER CURIAM. In August, 1987, the decomposed bod- ies of the victims, Fred Harris and his son, Gregory Harris, were found murdered in an apartment they shared in Hamden. The crime went unsolved until 2019, when scientific advancements in DNA testing led to the arrest of the defendant, Willie McFarland. The defen- dant now appeals from his conviction of both murders, following a jury trial in 2022. The primary issue on appeal is whether the thirty-two year delay between the 1987 murders and his 2019 arrest violated the defen- dant’s right to due process under the federal and state constitutions. The defendant also claims that the trial court abused its discretion by (1) declining to order a new competency evaluation after previously finding him competent to stand trial, and (2) finding that a deceased witness’ statements included in police reports were not admissible under the residual exception to the hearsay rule because they did not manifest the required guarantees of trustworthiness and reliability. We affirm the judgment of conviction. The jury reasonably could have found the following facts. On August 27, 1987, the Hamden police received a call from a relative of the victims indicating that they Page 6 CONNECTICUT LAW JOURNAL September 2, 2025

172 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

had not been seen or heard from since August 21. The police conducted a welfare check of the victims’ apart- ment and discovered their badly decomposed bodies lying side by side in an upstairs bedroom. Both victims had sustained fatal throat lacerations, and their hands and feet had been bound with cord or wire. Gregory Harris also had a nonfatal stab wound to his chest, and Fred Harris had two fatal stab wounds to his chest. There was no sign of forced entry, and the door was locked when the police arrived. The apartment had been ransacked, and a boom box in the bedroom had apparent stab marks. The police found a baking pan with melted butter and a butter wrapper next to the victims’ bodies, as well as a yellow work glove that matched another glove discovered outside of the apart- ment. A black handled kitchen knife, later identified as the murder weapon, was found in the bathroom sink, and Fred Harris’ brown leather wallet, containing his identification cards and a volunteer firefighter’s badge, was found on the living room floor. The defendant was released from prison on August 20, 1987, which was the day before the victims were last seen alive. In the early morning hours of August 22, the Hamden police arrested the defendant approxi- mately one mile away from the victims’ apartment for sexually assaulting a female acquaintance, C.1 During the sexual assault incident, the defendant had used a knife and was discovered with ‘‘a fair amount of blood’’ on him at the time of that arrest. The following week, after the discovery of the bodies of the victims in the present case, detectives interviewed the defendant about his location at the time of the murders. The defendant denied any involvement in the murders. He initially 1 In accordance with our policy of protecting the privacy interests of the victims of sexual abuse, we decline to identify C or others through whom her identity may be ascertained. See General Statutes § 54-86e. September 2, 2025 CONNECTICUT LAW JOURNAL Page 7

353 Conn. 169 SEPTEMBER, 2025 173 State v. McFarland

denied knowing the victims but later acknowledged that he might have known Gregory Harris from a car wash where they had worked together. The detectives initially accepted his explanation that he was covered in his own blood when he was arrested because C had accidentally stabbed him. The bloody clothing that he was wearing was not preserved or connected by the police to the victims’ murders. In the late 1980s and early 1990s, the police explored the possibility that Lee Copeland, a friend of Gregory Harris, had been involved in the murders. Donald Bruce Hankins gave a statement to the police that appeared to implicate Copeland. Copeland’s friend, Veronica Saars- Doyle, also gave a series of inconsistent statements to the police that implicated Copeland, Hankins, and, eventually, herself in the murders. In March, 1996, the defendant, still incarcerated after pleading guilty to sexually assaulting C in 1987, con- tacted the Hamden police to confess to the victims’ murders. He explained that he was motivated to confess because he had found religion. The defendant also had been diagnosed with a serious medical condition and believed that he would die in prison. Between that time and February, 1997, he provided a series of written and verbal confessions that evolved over time and contained many internal inconsistencies. Notably, although the defendant’s earlier confessions minimized his own involvement while attempting to implicate C, the Latin Kings street gang, and others against whom he harbored resentments, in his later confessions, he took sole responsibility for the murders. In his confessions, the defendant accurately described numerous aspects of the crime scene that only the perpetrator could have known. These included the exact locations and orienta- tions of the victims’ bodies in the apartment, the cord or wire used to bind their hands and feet, the injuries inflicted, the location of and damage to the boom box, Page 8 CONNECTICUT LAW JOURNAL September 2, 2025

174 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

the identification and location of the murder weapon in the upstairs bathroom sink, the presence of the melted butter and butter wrapper, and a description of Fred Harris’ wallet and badge. The defendant stated that he had socialized with Gregory Harris in the apartment on two or three prior occasions and that he had been welcomed into the apartment on the night of the mur- ders and had locked the door upon leaving. Law enforce- ment officials were unable to confirm other aspects of the defendant’s confessions that were not consistent with the crime scene evidence, and the state elected not to prosecute the defendant at that time. By 2006 or so, scientific advancements allowed the police to isolate and test small amounts of DNA, other- wise known as ‘‘touch DNA.’’ The police obtained a sample of the defendant’s DNA pursuant to a search warrant. In 2009, examiners at the state forensic science laboratory conducted an analysis and concluded that the defendant could be eliminated as a contributor to the DNA inside the yellow work glove found at the crime scene. The case remained unsolved for an additional decade. In 2018, after further scientific advancements in DNA testing, the police resubmitted the DNA to the state forensic science laboratory for analysis. Examiners used a newly developed test that was much more effec- tive at generating data from degraded samples of genetic material. On the basis of this new analysis, the examiners concluded that the DNA found inside the yellow work glove was ‘‘consistent with . . . being a mixture of four contributors, with at least one of them being male,’’ and that ‘‘the DNA profile . . . [would be] at least 1.5 million times more likely to occur if it originated from [the defendant] and three unknown individuals than if it originated from four unknown indi- viduals.’’ Around the same time, the police requested a DNA sample from Copeland, which he voluntarily September 2, 2025 CONNECTICUT LAW JOURNAL Page 9

353 Conn. 169 SEPTEMBER, 2025 175 State v. McFarland

provided. The match between the crime scene DNA and Copeland’s DNA sample was deemed ‘‘inconclusive.’’ Michael T. Bourke, a DNA analyst with the state forensic science laboratory, testified that the ‘‘inconclusive’’ result indicated that it was somewhere between 1 and 1000 times more likely that the crime scene sample was made up of DNA from Copeland and three unknown individuals than four unknown individuals. On the basis of this new DNA testing, in 2019, the state charged the defendant with the victims’ murders. Following his arrest, the defendant sought to represent himself but otherwise refused to participate in the pro- ceedings. Between January, 2020, and October, 2022, the trial court ordered several competency evaluations of the defendant. Initially, the court found the defendant not competent. Ultimately, the defendant was restored to competency, over defense counsel’s objections. The case was tried to a jury in 2022. At trial, the defendant sought to introduce, under the residual excep- tion to the hearsay rule, police reports containing Saars- Doyle’s statements implicating Copeland in the mur- ders. The trial court denied the defendant’s request and excluded Saars-Doyle’s statements from evidence. The jury returned a verdict of guilty on both counts of mur- der in violation of General Statutes § 53a-54a (a), and the trial court rendered judgment in accordance with the verdict and sentenced the defendant to a total effec- tive term of 120 years of imprisonment. This direct appeal followed. See General Statutes § 51-199 (b) (3). I The defendant first claims that the thirty-two year delay between the crimes and his arrest violated his right to a fair trial under the due process clauses of the federal and state constitutions.2 In State v. Morrill, 197 2 The defendant raised these constitutional claims in a pretrial motion to dismiss. Page 10 CONNECTICUT LAW JOURNAL September 2, 2025

176 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

Conn. 507, 522, 498 A.2d 76 (1985), and State v. Carri- one, 188 Conn. 681, 693–94, 453 A.2d 1137 (1982), cert. denied, 460 U.S. 1084, 103 S. Ct. 1775, 76 L. Ed. 2d 347 (1983), this court adopted the two-pronged approach to prearrest delay3 claims brought under the federal due process clauses, as construed by United States v. Marion, 404 U.S. 307, 92 S. Ct. 455, 30 L. Ed. 2d 468 (1971), and United States v. Lovasco, 431 U.S. 783, 97 S. Ct. 2044, 52 L. Ed. 2d 752 (1977). Under that two- pronged test, ‘‘[i]n order to establish a due process violation because of [preaccusation] delay, the defen- dant must show both that actual substantial prejudice resulted from the delay and that the reasons for the delay were wholly unjustifiable, as [when] the state seeks to gain a tactical advantage over the defendant.’’ State v. Morrill, supra, 522; see, e.g., State v. Roger B., 297 Conn. 607, 614–15, 999 A.2d 752 (2010); State v. Littlejohn, 199 Conn. 631, 645–47, 508 A.2d 1376 (1986). In the present case, the trial court concluded that the defendant could not prevail under the two-pronged test because there was no evidence that the state had delayed arresting him to obtain an unfair tactical advantage or for other improper purposes. The trial court then conducted a thorough analysis pursuant to State v. Geisler, 222 Conn. 672, 684–86, 610 A.2d 1225 (1992), of the defendant’s independent claims brought under the due process clauses of the Connecti- cut constitution. See Conn. Const., art. I, §§ 8 and 9. The trial court concluded that the Connecticut constitution confers broader protections under which courts apply a balancing test to prearrest delay claims brought under the state due process clauses. The trial court then con- 3 Courts generally use the terms ‘‘prearrest delay,’’ ‘‘preindictment delay,’’ and ‘‘preaccusation delay’’ interchangeably in the due process context. See, e.g., United States v. Sowa, 34 F.3d 447, 450 (7th Cir. 1994), cert. denied, 513 U.S. 1117, 115 S. Ct. 915, 130 L. Ed. 2d 796 (1995); State v. Police, 343 Conn. 274, 286 n.9, 273 A.3d 211 (2022). September 2, 2025 CONNECTICUT LAW JOURNAL Page 11

353 Conn. 169 SEPTEMBER, 2025 177 State v. McFarland

cluded that the defendant could not prevail under the balancing test. Even assuming that the defendant had suffered the required actual and substantial prejudice, the trial court found that the prearrest delay was wholly justified because the state promptly initiated the prose- cution once newly developed DNA testing technologies implicated the defendant in the murders. We unanimously conclude that the trial court prop- erly rejected the defendant’s federal and state due pro- cess claims arising from the prearrest delay in this case because he has failed to establish that his conviction offended ‘‘the community’s sense of fair play and decency . . . .’’ (Citation omitted; internal quotation marks omitted.) United States v. Lovasco, supra, 431 U.S. 790. With respect to the federal constitutional claims, we unanimously adhere to our existing prece- dent applying the two-pronged test. See, e.g., State v. Roger B., supra, 297 Conn. 614–15; State v. Morrill, supra, 198 Conn. 522. With respect to the defendant’s claims brought under the Connecticut constitution, a majority of this court adopts a balancing approach similar to that endorsed by the trial court in this case, which reflects the approach employed by the United States Courts of Appeals for the Fourth, Seventh and Ninth Circuits, and a minority of other states. See, e.g., United States v. Sowa, 34 F.3d 447, 451 (7th Cir. 1994), cert. denied, 513 U.S. 1117, 115 S. Ct. 915, 130 L. Ed. 2d 796 (1995); Howell v. Barker, 904 F.2d 889, 895 (4th Cir.), cert. denied, 498 U.S. 1016, 111 S. Ct. 590, 112 L. Ed. 2d 595 (1990); United States v. Moran, 759 F.2d 777, 782–83 (9th Cir. 1985), cert. denied, 474 U.S. 1102, 106 S. Ct. 885, 88 L. Ed. 2d 920 (1986). The specific reasoning of the majority of the court is explained in two separate concurring opinions. One opinion is a plurality opinion authored by Justice Alexander and joined by Chief Justice Mullins and Jus- tice Dannehy. The other opinion is authored by Justice Page 12 CONNECTICUT LAW JOURNAL September 2, 2025

178 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

Ecker and joined in part by Justice McDonald. Both opinions adopt the same balancing test under the Con- necticut constitution, pursuant to which a criminal defen- dant bears the burden of proving that a delay in prosecu- tion has caused actual and substantial prejudice; once the defendant has made this threshold showing, the burden shifts to the state to establish the reasons for the delay. Both concurring opinions conclude that the defendant’s prearrest delay claim fails under this bal- ancing test. In contrast, Justice D’Auria, in his concurring opin- ion, concludes that the Connecticut constitution pro- vides no greater protection against prearrest delay than the due process clause of the federal constitution. He, therefore, would follow the approach of a majority of the federal courts of appeals and other states, which adopt the two-pronged test set forth in Marion and Lovasco, as explained in State v. Morrill, supra, 197 Conn. 522, that requires the defendant to prove both actual and substantial prejudice, and that the state delayed arrest to obtain an unfair tactical advantage or for other improper purposes. See, e.g., United States v. Crouch, 84 F.3d 1497, 1514 (5th Cir. 1996), cert. denied, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997), and cert. denied sub nom. Frye v. United States, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997); see also id., 1511–12 (citing cases); D. Diethrich, Note, ‘‘If It Ain’t Broke, Don’t Fix It: Indictment Delays and Due Process,’’ 15 Elon L. Rev. 181, 193 n.83 (2023) (‘‘[o]f our nation’s federal appellate courts, the First, Second, Third, Sixth, Eighth, Tenth, and Eleventh Circuits have either firmly adopted the two-prong[ed] approach or have indicated that this was the correct application to a [preindictment] delay challenge’’). With no proof of delay for tactical purposes or bad faith, Justice D’Auria would also reject the defendant’s state constitutional claims. September 2, 2025 CONNECTICUT LAW JOURNAL Page 13

353 Conn. 169 SEPTEMBER, 2025 179 State v. McFarland

II The defendant next claims that the trial court abused its discretion by declining to order a new competency evaluation after finding him competent to stand trial. He argues that the trial court should not have relied on one of several earlier competency reports and his unwillingness to participate in the legal proceedings. We disagree and conclude that the trial court did not abuse its discretion when it denied defense counsel’s request for another competency evaluation. A While his case was pending in the trial court, the defendant consistently refused to participate in and requested to be excused from the proceedings unless he could represent himself. From the outset, he expressed his belief that the prosecution against him was brought in retaliation for an action he had filed against the state. The defendant claimed that he was seeking a restraining order against the state, but he refused to elaborate until ‘‘the media and the grand jury’’ were present. In January, 2020, the trial court found that the defen- dant was not competent to stand trial or to represent himself, but that he could be restored to competency with inpatient hospitalization and treatment. The court ordered that the defendant be transferred to Whiting Forensic Hospital (Whiting) for sixty days to restore his competency. In July, 2020, the trial court held another hearing on the defendant’s competency. Although the defendant continued to refuse to participate in the proceedings, the forensic monitor at Whiting, Susan McKinley, testi- fied that he had ‘‘certainly demonstrated the ability to cooperate and collaborate’’ with the evaluation team. On the basis of this testimony and other evidence, the court found that the defendant was competent to stand Page 14 CONNECTICUT LAW JOURNAL September 2, 2025

180 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

trial and scheduled a separate hearing to determine his competency to represent himself. In May, 2021, the trial court held that separate hearing to determine whether the defendant was competent to represent himself. After excusing the defendant at his request, the court heard the testimony of Evan Vitiello, a psychiatrist, and reviewed the competency reports that he had prepared. It was Vitiello’s opinion that the defendant was not competent to represent himself because he showed deficits in communication, atten- tion, and concentration, as well as an extreme distrust of the criminal justice system. The court credited Vitiello’s opinion and found that the defendant was not compe- tent to represent himself. In September, 2022, as jury selection was scheduled to begin, defense counsel expressed his belief that the defendant still was not competent to stand trial. The defendant interjected and requested to be excused, and the trial court granted his request. Defense counsel requested that the court order another competency evalu- ation because the defendant had refused to meet with him on numerous occasions over the course of two years and had not replied to his letters. Defense counsel indicated that the defendant continued to harbor delu- sions regarding the nature of the prosecution and that he expected that the case would be resolved by a nonex- istent civil action. The court ordered another compe- tency evaluation. The evaluators at the Connecticut Mental Health Cen- ter reported that they were unable to complete another competency evaluation because the defendant refused to participate. In October, 2022, the trial court held another competency hearing. The court admitted into evidence the prior competency reports, and the state presented testimony from a counselor who regularly interacted with the defendant at the correctional facility September 2, 2025 CONNECTICUT LAW JOURNAL Page 15

353 Conn. 169 SEPTEMBER, 2025 181 State v. McFarland

where the defendant was being held and had observed that the defendant appropriately interacted with him and other inmates. The defendant introduced testimony from Howard Zonana, a forensic psychiatrist at the Con- necticut Mental Health Center. Zonana testified that he had never met or interacted with the defendant and that his review of the prior competency evaluations provided the basis for his opinion. He testified that the defendant’s conduct and diagnoses were consistent with a person suffering from narrow, specific delusions that rendered his behavior nonvolitional. The trial court issued a comprehensive oral ruling, concluding that the defendant was competent to stand trial. The court, relying on its own observations, as well as the opinions of the evaluators, found that the defendant had the ability to participate in the proceed- ings but chose not to ‘‘because of what he believes to be his own self-interest, a position arguably borne of his lengthy past and presumably negative interactions with the criminal justice system and periods of incarcer- ations, which [have] created a mistrust of the legal system in him.’’ The court did not credit Zonana’s opinion that the defendant’s delusions rendered him incompetent. The court denied defense counsel’s motion for reconsid- eration and request that the staff at Whiting reevaluate the defendant’s competency. B The following well established legal principles govern our review of this claim. ‘‘[T]he due process clause of the fourteenth amendment to the United States consti- tution prohibits the criminal prosecution of a defendant who is not competent to stand trial.’’ (Internal quotation marks omitted.) State v. Dort, 315 Conn. 151, 162, 106 A.3d 277 (2014). ‘‘[T]he test for competency must be whether [the defendant] has sufficient present ability to consult with his lawyer with a reasonable degree of Page 16 CONNECTICUT LAW JOURNAL September 2, 2025

182 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

rational understanding—and whether he has a rational as well as factual understanding of the proceedings against him.’’ (Internal quotation marks omitted.) State v. Johnson, 253 Conn. 1, 21, 751 A.2d 298 (2000). This standard for competency to stand trial is ‘‘a relatively low one’’ and will not be satisfied by proof of ‘‘mental illness or reduced mental capacity’’ alone. State v. Con- nor, 292 Conn. 483, 524, 973 A.2d 627 (2009). This constitutional rule has been codified in General Statutes § 54-56d, which establishes the procedures and standards governing the determination of a defendant’s competence to stand trial. See State v. Campbell, 328 Conn. 444, 485–86, 180 A.3d 882 (2018). Under § 54- 56d, ‘‘[a]ny party before the court—including the court itself—may raise the issue of the defendant’s compe- tency at any time during a criminal proceeding by requesting that the court order a competency examina- tion.’’ State v. Dort, supra, 315 Conn. 163–64. Section 54- 56d compels a court to order a competency examination ‘‘any time a reasonable doubt is raised regarding the defendant’s competency. . . . To establish such rea- sonable doubt, the defendant must present substantial evidence, not merely allegations, that he is incompe- tent.’’ (Citation omitted; internal quotation marks omit- ted.) State v. Ross, 269 Conn. 213, 272, 849 A.2d 648 (2004). Given the trial court’s superior ability to observe the defendant’s conduct, we review the trial court’s decision whether to order a competency examination for abuse of discretion. See, e.g., State v. Campbell, supra, 486; State v. Ross, supra, 270. On this record, we conclude that the trial court did not abuse its discretion in denying defense counsel’s request that it order a fifth competency evaluation. The court already had ordered four such evaluations, the last of which delayed jury selection. Although the defen- dant chose not to cooperate with the fourth evaluation, the hearing provided defense counsel with the opportu- September 2, 2025 CONNECTICUT LAW JOURNAL Page 17

353 Conn. 169 SEPTEMBER, 2025 183 State v. McFarland

nity to present further evidence in support of the request. The primary evidence defense counsel offered was Zonana’s testimony, which the trial court found to be unpersuasive. The court also carefully considered, and did not find compelling, defense counsel’s claim that the defendant was unable to assist in his defense as a result of his delusions. The trial court instead relied on its own observations and credited expert testimony from the evaluators who had met and interacted with the defendant that the defendant was intentionally uncooperative. The court found that the defendant had the ability to interact with others when he chose to do so and that his actions constituted a strategic choice in order to delay the trial. In the absence of a substantial change in circum- stances raising a reasonable doubt as to the defendant’s competency, the trial court properly declined to order a fifth competency evaluation. See, e.g., State v. Ross, supra, 269 Conn. 273 (trial court properly declined to order competency evaluation because defendant failed to present substantial evidence to establish reasonable doubt as to his competency); State v. Norris, 213 Conn. App. 253, 275–76, 277 A.3d 839 (trial court properly declined to order new competency evaluation because defendant failed to present evidence that his condition had changed since prior competency evaluation), cert. denied, 345 Conn. 910, 283 A.3d 980 (2022). We are not persuaded by the defendant’s argument that the trial court erred in considering the July, 2020 competency evaluation in connection with its decision to deny the request for an evaluation in October, 2022. The court’s inquiry was properly focused on whether the defendant was competent at the later time, and it was not error for the court to consider the defendant’s prior mental health history. See, e.g., State v. Campbell, supra, 328 Conn. 493 (upholding trial court’s compe- tency finding that rejected Zonana’s testimony and con- Page 18 CONNECTICUT LAW JOURNAL September 2, 2025

184 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

sidered other competency reports and supporting testimony from evaluators); State v. Edwards, 158 Conn. App. 119, 137, 118 A.3d 615 (court did not abuse its discretion in relying on observations, input, and previ- ous competency evaluation as bases for not ordering additional competency evaluation), cert. denied, 318 Conn. 906, 122 A.3d 634 (2015). In fact, defense counsel asked the court to rely on the January, 2020 and May, 2021 evaluations finding the defendant incompetent to stand trial and to represent himself, and to credit Zona- na’s testimony that was founded on those reports. We also are not persuaded by the defendant’s argu- ment that the trial court improperly relied on its own observations of the defendant’s behavior. A trial court is uniquely situated to observe and assess a defendant’s competency. It is not improper for a trial court to rely on those observations as part of the competency deter- mination. See, e.g., State v. Glen S., 207 Conn. App. 56, 76, 261 A.3d 805 (defendant’s failure to cooperate with competency evaluators was relevant to trial court’s assessment of defendant’s competency), cert. denied, 340 Conn. 909, 264 A.3d 577 (2021), cert. denied, U.S. , 142 S. Ct. 2685, 212 L. Ed. 2d 768 (2022); State v. Hines, 165 Conn. App. 1, 16–17, 138 A.3d 994 (trial court properly considered defendant’s behavior during competency hearing), cert. denied, 321 Conn. 920, 137 A.3d 764 (2016); State v. Jordan, 151 Conn. App. 1, 35–36, 92 A.3d 1032 (trial court properly considered its own observation of defendant’s behavior, including his voluntarily leaving courtroom), cert. denied, 314 Conn. 909, 100 A.3d 402 (2014). Accordingly, we conclude that the trial court did not abuse its discretion in declining to order another competency evaluation. III The defendant’s final claim is that the trial court abused its discretion by excluding Saars-Doyle’s state- September 2, 2025 CONNECTICUT LAW JOURNAL Page 19

353 Conn. 169 SEPTEMBER, 2025 185 State v. McFarland

ments implicating Copeland in the murders.4 He argues that the trial court should have admitted those state- ments under the residual exception to the hearsay rule and challenges the court’s finding that Saars-Doyle’s multilayered hearsay statements were unreliable and untrustworthy. We conclude that the trial court did not abuse its discretion by excluding Saars-Doyle’s state- ments. A During the state’s case-in-chief, defense counsel cross-examined Sean Dolan, a detective employed by the Hamden Police Department since 1995. When defense counsel inquired about Dolan’s review of Saars- Doyle’s statements, the prosecutor objected. The trial court excused the jury, and defense counsel made an extensive proffer seeking to introduce several police reports containing Saars-Doyle’s statements, as well as Dolan’s related testimony. Some of those police reports contain verbatim transcriptions of Saars-Doyle’s recorded statements. Others are narrative summaries of her state- ments to the police and of informal conversations with her, only some of which were recorded fully or in part. In addition to Saars-Doyle’s own observations, the police reports contain numerous statements that she claimed were made by third parties. Defense counsel contended that Saars-Doyle’s statements were relevant to establish Copeland’s third-party culpability and were admissible 4 The defendant frames this evidentiary claim as having a constitutional dimension because the trial court’s exclusion of Saars-Doyle’s statements violated (1) his due process rights, causing him to suffer prejudice as a result of the prearrest delay, and (2) his constitutional right to present a defense. We address the claim as a purely evidentiary matter because of our conclusion that the court correctly determined that Saars-Doyle’s statements were not admissible under the residual exception to the hearsay rule. See, e.g., State v. Bennett, 324 Conn. 744, 764, 155 A.3d 188 (2017) (‘‘[b]ecause the trial court’s evidentiary ruling was not improper, the defendant’s claim of an infringement on his constitutional right to present a defense on this basis must fail’’). Page 20 CONNECTICUT LAW JOURNAL September 2, 2025

186 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

under the residual exception to the hearsay rule. See Conn. Code Evid. § 8-9. The proffered police reports reveal that, between 1987 and 1990, the Hamden police obtained a series of statements from Saars-Doyle. These statements described her relationships with Copeland and the victims, and provided increasing amounts of detail relevant to the murders. Her audio-recorded and transcribed July 23, 1990 statement asserted that her prior statements were inaccurate as a result of medical problems and sub- stance abuse. In that statement, she implicated Cope- land, Hankins, and a third person named ‘‘Gus’’ in the murders; she asserted that they committed the murders while she waited in a car. When Saars-Doyle met with the police the next day to sign the transcription of this statement, the police questioned her about some of the details, and she then provided a significantly different version of the events of the murders, which she then claimed to have witnessed from inside of the apartment. During the defendant’s proffer, Dolan testified that, although he had reviewed the reports, other officers had interviewed Saars-Doyle and transcribed her state- ments. The prosecutor elicited testimony from Dolan about the various inconsistencies in Saars-Doyle’s state- ments, as well as the indications in the reports that she suffered from memory issues and had been drinking heavily on the night of the murders. The trial court excluded Saars-Doyle’s statements on the ground that they were too unreliable and untrust- worthy to satisfy the residual exception to the hearsay rule. The court found that the statements lacked suffi- cient indicia of reliability because, among other things, they were ‘‘disjointed [and] largely incoherent,’’ and they ‘‘strain[ed] credulity’’ because they included vari- ous inconsistencies, had notable omissions, and contra- dicted the crime scene evidence. The court also noted that Saars-Doyle had never been subject to formal cross- September 2, 2025 CONNECTICUT LAW JOURNAL Page 21

353 Conn. 169 SEPTEMBER, 2025 187 State v. McFarland

examination and that the statements contained many instances of hearsay within hearsay. B It is undisputed that Saars-Doyle’s statements to the police, which were offered at trial for the truth of the matters asserted therein, constituted hearsay. See Conn. Code Evid. § 8-1 (3). The statements also con- tained double hearsay in the form of statements attrib- uted by Saars-Doyle to Copeland, among others, requiring ‘‘each part of the combined statements [to be] independently admissible under a hearsay exception.’’ Conn. Code Evid. § 8-7. As such, the statements in their entirety were inadmissible unless all or part of their contents fell within an exception to the hearsay rule, as set forth in §§ 8-3 through 8-10 of the Connecticut Code of Evidence. The defendant contends that the statements were admissible under § 8-9 of the Connecticut Code of Evi- dence, known as the residual, or catchall, exception to the hearsay rule. The residual exception ‘‘allows a trial court to admit hearsay evidence not admissible under any of the established exceptions if: (1) there is a rea- sonable necessity for the admission of the statement, and (2) the statement is supported by the equivalent guarantees of reliability and trustworthiness essential to other evidence admitted under the traditional hear- say exceptions. . . . We have [explained] that [t]he residual hearsay exceptions [should be] applied in the rarest of cases . . . .’’ (Citations omitted; internal quo- tation marks omitted.) State v. Bennett, 324 Conn. 744, 762, 155 A.3d 188 (2017). We review the trial court’s decision to exclude hearsay evidence for abuse of dis- cretion and make every reasonable presumption in favor of upholding the ruling. See, e.g., id., 761–62. C The defendant met the first requirement of the resid- ual exception to the hearsay rule, a reasonable necessity Page 22 CONNECTICUT LAW JOURNAL September 2, 2025

188 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

for the admission of the statements, because Saars- Doyle was deceased at the time of trial. We conclude, however, that the trial court did not abuse its discretion in declining to admit Saars-Doyle’s hearsay statements under the residual exception because her statements were disjointed, inconsistent, implausible, and not reli- able. We defer to the trial court’s factual findings that there were serious impediments to Saars-Doyle’s ability to accurately observe and recall the events in question. The record supports the trial court’s findings that Saars- Doyle had hearing and vision problems, medical and medication issues, heavy alcohol and drug use, and self-reported memory loss and shock. In light of these multiple reasons to doubt the trustworthiness and relia- bility of the statements, we also consider it significant that Saars-Doyle was never subjected to cross-examina- tion regarding the numerous inconsistencies in her statements. Although the police did press her at times during their questioning, they generally did not chal- lenge her to explain these inconsistencies. See, e.g., State v. Bennett, supra, 324 Conn. 763 (‘‘[a] declarant’s availability for cross-examination has been deemed par- ticularly significant in determining whether hearsay evi- dence is supported by guarantees of trustworthiness and reliability’’); State v. McClendon, 248 Conn. 572, 583–85, 730 A.2d 1107 (1999) (residual exception to hear- say rule was not satisfied when, among other things, declarant was not available for cross-examination), over- ruled in part on other grounds by State v. Guilbert, 306 Conn. 218, 49 A.3d 705 (2012). Although Saars-Doyle accurately reported certain details of the crime scene, such as the location of the victims’ bodies in an upstairs bedroom, the presence of butter and work gloves, and the fact that the victims had been bound with cord or wire, many of these facts either had been publicized by the police or were revealed to Saars-Doyle prior to or during the inter- September 2, 2025 CONNECTICUT LAW JOURNAL Page 23

353 Conn. 169 SEPTEMBER, 2025 189 State v. McFarland

views. Moreover, various other details that she provided were inconsistent with the crime scene or lacked speci- ficity and suggested a wide range of possibilities, such as the use of various materials to bind the victims. When Saars-Doyle stated that she was prepared to tell the police about the incident, she often appeared to be changing or supplementing her account of the murders in response to prompts by the police, or asking them for more information. See, e.g., State v. McClendon, supra, 248 Conn. 584 (statement in police report was properly excluded under residual exception as not reli- able and trustworthy because it ‘‘was contradicted by the evidence presented at the trial and was not corrobo- rated by any other evidence’’); State v. Burton, 191 Conn. App. 808, 840–41, 216 A.3d 734 (video recording of police interview was properly excluded under residual exception to hearsay rule because declarant was not subject to cross-examination as to critical factual uncer- tainties and evidence at trial failed to corroborate her statement), cert. denied, 333 Conn. 927, 217 A.3d 995 (2019). We further agree with the trial court that the multiple layers of hearsay rendered Saars-Doyle’s statements less reliable. The narrative reports required the police to recall, characterize, and summarize what she had told them. Some statements in the various police reports were insulated by additional levels of hearsay because they derived from different declarants, includ- ing the police, Saars-Doyle’s mother, her sister, and Copeland. See, e.g., State v. Bennett, supra, 324 Conn. 764 (multiple levels of hearsay undermine reliability for admissibility under residual exception to hearsay rule); State v. Rivera, 181 Conn. App. 215, 225, 186 A.3d 70 (same), cert. denied, 329 Conn. 907, 184 A.3d 1216 (2018). Finally, the fact that Saars-Doyle signed two of her statements under penalty of law and made handwritten Page 24 CONNECTICUT LAW JOURNAL September 2, 2025

190 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

edits to one does not render the statements trustworthy and reliable. That each successive statement contra- dicted her prior statements in multiple, material respects indicates either that Saars-Doyle was unable to accu- rately recall and to consistently recount the events of August 21, 1987, or that she misrepresented those events in violation of her oath. For these reasons, we conclude that the trial court did not abuse its discretion by exclud- ing Saars-Doyle’s statements. The judgment is affirmed.

D’AURIA, J., concurring in the judgment. As the court’s per curiam opinion indicates, all panel members in the present case agree to affirm the judgment of conviction of the defendant, Willie McFarland: six members concur in three separate opinions. My own opinion is a concur- rence in name only. I disagree with nearly all of the reasoning of the other two concurring opinions, authored by Justices Ecker and Alexander, which both accept the defendant’s request to create a new balancing test under the Connecticut constitution, aimed at more broadly protecting a defendant’s due process right against unreasonable prearrest delay. Unlike the other concur- ring justices, I am not convinced that the state constitu- tion affords greater protection from prearrest delays than does the federal constitution. The balancing test that the other concurrences adopt might have surface appeal, but the details and mechanics of how that test works create myriad problems for defendants, the state, and trial court judges. I would continue to apply the two-pronged test that the United States Supreme Court created as a matter of federal constitutional law to react to state action aimed at disadvantaging defendants, which this court, nine federal circuit courts, and thirty states’ courts have consistently applied for fifty years. Although I certainly agree with Justice Alexander that we must conduct a ‘‘more rigorous analysis than September 2, 2025 CONNECTICUT LAW JOURNAL Page 25

353 Conn. 169 SEPTEMBER, 2025 191 State v. McFarland

simply tallying holdings’’ of other jurisdictions; Doe v. Hartford Roman Catholic Diocesan Corp., 317 Conn. 357, 427, 119 A.3d 462 (2015); I do not find persuasive the opinions of the minority of jurisdictions on which the two other concurrences rely or the two concur- rences themselves. I find no compelling reason to depart from our consistent application of the test that a major- ity of jurisdictions apply, and I disagree with both con- currences, which announce a balancing test while providing almost no guidance to trial courts about how to fix what both concurrences view as broken. I As an initial matter, I do not understand why the rest of this court has concluded that it is necessary to address the need for a balancing test under our state constitution when the state constitutional test yields the defendant no relief in this case. I am aware that my view on the matter is suspect because I disagree with the balancing test that the court announces today. But, in my opinion, we should wait for a case in which the application of the new balancing test would make a difference to the defendant. Because all members of the panel agree that the defendant cannot prevail on any test compelled by the due process clauses of article first, §§ 8 and 9, of our state constitution, it would comport with the doctrine of constitutional avoidance for us, at most, to ‘‘assum[e] without deciding,’’ as my concurring colleagues say, that the balancing test the trial court applied or that the defendant urges on appeal, without actually holding as a constitutional matter, that it does apply. See, e.g., State v. Haynes, 352 Conn. 236, 260, 336 A.3d 1139 (2025); State v. Sayles, 348 Conn. 669, 679, 310 A.3d 929 (2024). We have made quite a few of these assumptions recently in our cases, includ- ing even when a constitutional issue was not in play. See, e.g., State v. Myers, 352 Conn. 770, 779, 338 A.3d 1088 (2025); State v. James K., 347 Conn. 648, 662, 299 Page 26 CONNECTICUT LAW JOURNAL September 2, 2025

192 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

A.3d 243 (2023); State v. Lanier, 347 Conn. 179, 193, 296 A.3d 770 (2023).1 II Until today, this court for a generation has consis- tently agreed with and applied the two-pronged test articulated by the United States Supreme Court in United States v. Marion, 404 U.S. 307, 92 S. Ct. 455, 30 L. Ed. 2d 468 (1971), and United States v. Lovasco, 431 U.S. 783, 97 S. Ct. 2044, 52 L. Ed. 2d 752 (1977), to assess defendants’ prearrest delay claims on federal due process grounds. This test requires defendants to establish ‘‘both that actual substantial prejudice resulted from the delay and that the reasons for the delay were wholly unjustifiable, as where the state seeks to gain a tactical advantage over the defendant.’’ (Emphasis in original; internal quotation marks omitted.) State v. Roger B., 297 Conn. 607, 614, 999 A.2d 752 (2010); see also State v. Robinson, 213 Conn. 243, 248, 567 A.2d 1173 (1989), overruled in part on other grounds by State v. Colon, 257 Conn. 587, 778 A.2d 875 (2001); State v. John, 210 Conn. 652, 685–86, 557 A.2d 93, cert. denied, 493 U.S. 824, 110 S. Ct. 84, 107 L. Ed. 2d 50 (1989), and cert. denied sub nom. Seebeck v. Connecticut, 493 U.S. 824, 110 S. Ct. 84, 107 L. Ed. 2d 50 (1989); State v. Littlejohn, 199 Conn. 631, 646, 508 A.2d 1376 (1986); State v. Gaston, 198 Conn. 435, 445, 503 A.2d 594 (1986); State v. Morrill, 197 Conn. 507, 522, 498 A.2d 76 (1985); State v. Carrione, 188 Conn. 681, 693–94, 453 A.2d 1137 (1982), cert. denied, 460 U.S. 1084, 103 S. Ct. 1775, 76 L. Ed. 2d 347 (1983); State v. Echols, 170 Conn. 11, 16–17, 364 A.2d 225 (1975). 1 Justice Alexander goes halfway in by ‘‘[a]ssuming without deciding that the defendant has established actual and substantial prejudice in this case’’ before turning to the state’s justification for delay. As I will explain in part IV of this opinion, because a majority of this court has determined that it is prudent in the present case to announce this state constitutional balancing test, this is a missed opportunity to provide our trial courts with some idea of what might satisfy the prejudice side of the balancing. September 2, 2025 CONNECTICUT LAW JOURNAL Page 27

353 Conn. 169 SEPTEMBER, 2025 193 State v. McFarland

In Marion, the court recognized that statutes of limi- tations are legislatively enacted limits on prosecutorial delay that provide ‘‘predictability’’ and are ‘‘the primary guarantee against bringing overly stale criminal charges.’’ (Internal quotation marks omitted.) United States v. Marion, supra, 404 U.S. 322. Nevertheless, the court stated that a ‘‘statute of limitations does not fully define [a defendant’s] rights with respect to the events occurring prior to indictment’’ and held that the federal due process clause ‘‘would require dismissal of the indictment if it were shown at trial that the [preindict- ment] delay . . . caused substantial prejudice to [a defendant’s] rights to a fair trial and that the delay was an intentional device to gain tactical advantage over the accused.’’ Id., 324. Later, in Lovasco, the court clarified that the federal due process clause ‘‘has a limited role to play in pro- tecting against oppressive delay,’’ requiring proof both that a defendant suffered prejudice by the delay and that the reasons for the delay were unjustifiable. United States v. Lovasco, supra, 431 U.S. 789–90. The court held that delay caused by the government’s investiga- tion may be justifiable because, ‘‘[r]ather than deviating from elementary standards of fair play and decency, a prosecutor abides by them if he refuses to seek indict- ments until he is completely satisfied that he should prosecute and will be able promptly to establish guilt beyond a reasonable doubt.’’ (Internal quotation marks omitted.) Id., 795. The court then concluded that, ‘‘to prosecute a defendant following investigative delay does not deprive him of due process, even if his defense might have been somewhat prejudiced by the lapse of time.’’ Id., 796; see also United States v. Gouveia, 467 U.S. 180, 192, 104 S. Ct. 2292, 81 L. Ed. 2d 146 (1984) (describing Lovasco and Marion test as requiring ‘‘dis- missal of an indictment, even if it is brought within the statute of limitations, if the defendant can prove that Page 28 CONNECTICUT LAW JOURNAL September 2, 2025

194 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

the [g]overnment’s delay in bringing the indictment was a deliberate device to gain an advantage over him and that it caused him actual prejudice in presenting his defense’’). The court in Lovasco articulated meaningful policy reasons in support of its two-pronged approach. First, and principally, the court recognized that it is ‘‘undesir- able’’ to adopt a rule that ‘‘would increase the likelihood of unwarranted charges being filed, and would add to the time during which defendants stand accused but untried. These costs are by no means insubstantial since . . . a formal accusation may interfere with the defen- dant’s liberty . . . disrupt his employment, drain his financial resources, curtail his associations, subject him to public obloquy, and create anxiety in him, his family and his friends. . . . From the perspective of law enforcement officials, a requirement of immediate pros- ecution upon probable cause is equally unacceptable because it could make obtaining proof of guilt beyond a reasonable doubt impossible by causing potentially fruitful sources of information to evaporate before they are fully exploited. And from the standpoint of the courts, such a requirement is unwise because it would cause scarce resources to be consumed on cases that prove to be insubstantial, or that involve only some of the responsible parties or some of the criminal acts. Thus, no one’s interests would be well served by com- pelling prosecutors to initiate prosecutions as soon as they are legally entitled to do so.’’ (Citations omitted; footnotes omitted; internal quotation marks omitted.) United States v. Lovasco, supra, 431 U.S. 791–92. The court articulated three other compelling reasons supporting its two-pronged test. ‘‘First, compelling a prosecutor to file public charges as soon as the requisite proof has been developed against one participant on one charge would cause numerous problems in those cases in which a criminal transaction involves more September 2, 2025 CONNECTICUT LAW JOURNAL Page 29

353 Conn. 169 SEPTEMBER, 2025 195 State v. McFarland

than one person or more than one illegal act. In some instances, an immediate arrest or indictment would impair the prosecutor’s ability to continue his investiga- tion, thereby preventing society from bringing law- breakers to justice. In other cases, the prosecutor would be able to obtain additional indictments despite an early prosecution, but the necessary result would be multiple trials involving a single set of facts. Such trials place needless burdens on defendants, law enforcement offi- cials, and courts. ‘‘Second, insisting on immediate prosecution once sufficient evidence is developed to obtain a conviction would pressure prosecutors into resolving doubtful cases in favor of early—and possibly unwarranted— prosecutions. The determination of when the evidence available to the prosecution is sufficient to obtain a conviction is seldom clear-cut, and reasonable persons often will reach conflicting conclusions. . . . The deci- sion whether to prosecute [sometimes] require[s] a nec- essarily subjective evaluation of the strength of the circumstantial evidence available and the credibility of [the defendant’s] denial. Even if a prosecutor concluded that the case was weak and further investigation appro- priate, he would have no assurance that a reviewing court would agree. To avoid the risk that a subsequent indictment would be dismissed for preindictment delay, the prosecutor might feel constrained to file premature charges, with all the disadvantages that would entail. ‘‘Finally, requiring the [g]overnment to make charging decisions immediately upon assembling evidence suffi- cient to establish guilt would preclude the [g]overnment from giving full consideration to the desirability of not prosecuting in particular cases. The decision to file criminal charges, with the awesome consequences it entails, requires consideration of a wide range of factors in addition to the strength of the [g]overnment’s case, in order to determine whether prosecution would be Page 30 CONNECTICUT LAW JOURNAL September 2, 2025

196 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

in the public interest. Prosecutors often need more information than proof of a suspect’s guilt, therefore, before deciding whether to seek an indictment. . . . Requiring prosecution once the evidence of guilt is clear, however, could prevent a prosecutor from awaiting the information necessary for such a decision.’’ (Footnotes omitted.) Id., 792–95. Even if the other concurrences in the present case disclaim that their rule is intended to require prosecu- tors to bring charges as soon as possible, as I will explain in greater detail, I am not persuaded that we should apply a different test as a matter of state consti- tutional law. The federal two-pronged rule allows prose- cutors the necessary time to investigate their cases, while at the same time prohibiting them from delaying arrest to obtain an advantage over the defendant by the passage of time. If a defendant can prove that he is substantially prejudiced by that unjustifiable delay, then the court should dismiss the prosecution. III The defendant has convinced a majority of this court to discard this precedent and implement a new balanc- ing test forged under State v. Geisler, 222 Conn. 672, 685–86, 610 A.2d 1225 (1992). I disagree with both of the other concurrences that the Geisler factors2 support this dramatic change in our law. Beginning with the first five Geisler factors—constitutional text, Connecticut case law, federal case law, other states’ case law, and 2 When determining the contours of the protections provided by our state constitution, this court examines ‘‘(1) the text of the relevant constitutional provisions; (2) related Connecticut precedents; (3) persuasive federal prece- dents; (4) persuasive precedents of other state courts; (5) historical insights into the intent of [the] constitutional [framers]; and (6) contemporary under- standings of applicable economic and sociological norms [otherwise described as public policies].’’ (Internal quotation marks omitted.) State v. Langston, 346 Conn. 605, 625, 294 A.3d 1002 (2023), cert. denied, U.S. , 144 S. Ct. 698, 217 L. Ed. 2d 391 (2024). September 2, 2025 CONNECTICUT LAW JOURNAL Page 31

353 Conn. 169 SEPTEMBER, 2025 197 State v. McFarland

historical insights—I would conclude that none favors construing our due process clauses to demand the level of judicial scrutiny of the efforts of police and prosecu- tors that a majority of this court holds that our state’s courts must now undertake. Finally, I conclude, as mat- ter of our state’s public policy, that the rule announced today does a good deal more harm than good. A First, in my view, no text or constitutional history supports a different rule under our state constitution. Unlike Justice Ecker, I cannot conclude, as a textual or historical matter, that our due process clauses manifest any intent to provide broader protection from prearrest delay than does their federal counterparts. This court has often observed that the text of the ‘‘due process clauses of the state and the federal constitutions are virtually identical.’’ (Footnote omitted.) State v. Lock- hart, 298 Conn. 537, 551, 4 A.3d 1176 (2010); see also State v. Purcell, 331 Conn. 318, 344–45, 203 A.3d 542 (2019). Like the language in the federal due process clauses, no language in our state due process clauses suggests a particular level of protection from prearrest delay. Cf. State v. Griffin, 339 Conn. 631, 692, 262 A.3d 44 (2021), cert. denied, U.S. , 142 S. Ct. 873, 211 L. Ed. 2d 575 (2022). Nor are there any historical insights suggesting that Connecticut’s constitutional framers intended our due process clauses to afford defendants the supposed protection from prearrest delay that a majority of this court holds today that they afford. See State v. Santiago, 318 Conn. 1, 16–17, 122 A.3d 1 (2015) (‘‘due process protections take as their hallmark princi- ples of fundamental fairness rooted in our state’s unique common law, statutory, and constitutional traditions’’). In particular, I am unaware of any common-law recogni- tion of a measure of protection from prearrest delay, akin to the equitable defense of laches. And, while it is likely that advances in science such as DNA testing Page 32 CONNECTICUT LAW JOURNAL September 2, 2025

198 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

will continue to result in the resolution of cold cases and in more prosecutions of crimes committed long ago, unlike scientific advances, delayed prosecutions do not pose an entirely new issue for our courts. See generally, e.g., State v. Honsch, 349 Conn. 783, 322 A.3d 1019 (2024); State v. Skakel, 276 Conn. 633, 888 A.2d 985, cert. denied, 549 U.S. 1030, 127 S. Ct. 578, 166 L. Ed. 2d 428 (2006). Next, unlike the other concurrences, I do not find that our own precedent supports, let alone compels, this court’s adoption of a new balancing test. Both of the other concurrences principally rely on this court’s decision in State v. Hodge, 153 Conn. 564, 219 A.2d 367 (1966), in which this court decided, prior to Marion and Lovasco, to apply a totality of the circumstances test to conclude that a three week delay between the alleged offense and the defendant’s arrest for posses- sion and sale of narcotics did not constitute an unrea- sonable seizure of the person under [the predecessor to article first, § 7, of the 1965 Connecticut constitution]. See id., 566–69. Although Hodge concerned a different state constitutional provision, we ultimately held that a reasonable delay caused by police investigation ‘‘should not undermine the legality of subsequent arrests and convictions except under unusual and clearly prejudi- cial circumstances.’’ Id., 569. We also stated that ‘‘[o]ften a lengthy delay is required because the identity of the offender is unknown to the authorities or because addi- tional time is needed to gather sufficient evidence to justify an official charge against one who is suspected of crime. Under such circumstances, the applicable stat- ute of limitations is the ultimate safeguard against a long-delayed arrest and prosecution.’’ Id., 567. To the extent Hodge is relevant to the defendant’s claim under our state constitution’s due process clauses, as described, the test that this court articulated hews closer to the federal due process standard of Lovasco and Marion than to any minority balancing test September 2, 2025 CONNECTICUT LAW JOURNAL Page 33

353 Conn. 169 SEPTEMBER, 2025 199 State v. McFarland

that the concurrences propose. In Hodge, we indicated that, for a constitutional argument concerning prearrest delay to succeed, the defendant would have to establish not only prejudice but that ‘‘the delay in arresting a defendant (or in otherwise apprising him of the charges against him) continue[d] long after all the evidence ha[d] been assembled, and [became] a product of mere convenience to the state . . . .’’ Id. The authority that this court relied on in Hodge expressly required defen- dants to demonstrate both prejudice and the govern- ment’s attempt to gain an advantage but did not involve a burden shifting standard requiring the state to demon- strate that the delay was justified. See id., 568, citing Jackson v. United States, 351 F.2d 821, 823 (D.C. Cir. 1965) (defendant established that delay was ‘‘clear responsibility of the [g]overnment and was arranged solely for its advantage,’’ but claim failed because he could not establish prejudice (emphasis added)), and Ross v. United States, 349 F.2d 210, 215 (D.C. Cir. 1965) (defendant established government’s ‘‘purposeful delay of seven months between offense and arrest’’ and plau- sible claim of prejudice (emphasis added)). Further, ‘‘all the circumstances’’ we mandated in Hodge that a trial court consider, including ‘‘the length of the delay, the reason for the delay, prejudice to the defendant, and a timely presentation of the claim to the trial court’’; State v. Hodge, supra, 153 Conn. 568; are no different from the considerations of the federal two-pronged test. It is clear, therefore, that we had never applied a balancing approach prior to Marion and Lovasco, and, since that time, we have consistently agreed with the United States Supreme Court, albeit in response to federal constitutional challenges, that due process protections are ‘‘ ‘limited’ ’’ with respect to pre- arrest delay. State v. Morrill, supra, 197 Conn. 521; see also id. (due process clause ‘‘ ‘does not permit courts to abort criminal prosecutions simply because they dis- Page 34 CONNECTICUT LAW JOURNAL September 2, 2025

200 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

agree with a prosecutor’s judgment as to when to seek an indictment’ ’’).3 Finally, as both concurrences acknowledge, the major- ity rule applied in courts across the country is in line with the two-pronged test outlined by Marion and Lovasco. See E. DuBosar, Note, ‘‘Pre-Accusation Delay: An Issue Ripe for Adjudication by the United States Supreme Court,’’ 40 Fla. St. U. L. Rev. 659, 671–76 (2013); D. Rang, Comment, ‘‘The Waiting Game: How Preindict- ment Delay Threatens Due Process and Fair Trials,’’ 66 S.D. L. Rev. 143, 154–55 (2021). Nine federal circuits agree with our interpretation of United States Supreme Court precedent and apply the two-pronged test requir- ing a defendant to establish actual prejudice and that the delay was the product of an unjustifiable motive. See E. DuBosar, supra, 669–70. Three federal circuits reject that approach in favor of a balancing test. See id., 669–70; D. Rang, supra, 152–53. Likewise, approxi- mately thirty states’ courts use the two-pronged test; D. Rang, supra, 154; thirteen states’ courts apply some form of a balancing test; id., 155; and courts in the remaining states have not addressed the issue or the test that they apply is not clear. See E. DuBosar, supra, 681–84; D. Rang, supra, 155. Although the other concur- rences discount this weight of authority as nothing more than quantitative, I do not find the case law on which the concurrences rely any more persuasive than the arguments that our state constitutional text, history or Connecticut precedent support their balancing test. B I also believe that public policy (the final Geisler factor) does not support the balancing test, which defen- 3 If the other concurrences truly believe that Hodge is a precedent that dictates how we must treat the defendant’s due process claim for prearrest delay under article first, §§ 8 and 9, of our state constitution, it would follow that a defendant should have to suffer only ‘‘[s]ome prejudice,’’ not actual and substantial prejudice, which the majority announces will be the standard going forward. State v. Hodge, supra, 153 Conn. 568. September 2, 2025 CONNECTICUT LAW JOURNAL Page 35

353 Conn. 169 SEPTEMBER, 2025 201 State v. McFarland

dants may invoke in unlimited circumstances, that a majority of this court adopts today.4 This court is, of course, free to arrive at its own conclusions about how public policy should inform our construction of the state constitution. It is well established that the federal constitution, as interpreted by the United States Supreme Court, provides merely ‘‘the floor’’ when it comes to protecting the constitutional rights of Connecticut resi- dents. (Internal quotation marks omitted.) State v. Haynes, supra, 352 Conn. 245. We have, therefore, at times, departed from that court’s precedent based solely on our disagreement with its assessment of the societal costs and benefits of a particular constitutional rule when interpreting and applying parallel provisions of our state constitution. See, e.g., State v. Marsala, 216 Conn. 150, 167, 579 A.2d 58 (1990) (‘‘[w]e simply cannot accept the conclusion reached by the [court in United States v. Leon, 468 U.S. 897, 104 S. Ct. 3405, 82 L. Ed. 2d 677 (1984)] as a result of its weighing of the relevant costs and benefits of excluding evidence obtained through police officers’ good faith reliance upon a warrant issued by a detached and neutral judicial official’’). 4 Claims invoking the new balancing test are likely to be asserted in a motion to dismiss for delays of any length, regardless of the alleged crime and whether it has a statute of limitations. The balancing test encourages defense counsel to raise a due process claim in unlimited numbers of cases, even if it is doomed to failure. Upon a showing of prejudice, the other concurrences place the burden on the state to establish its reasons for delay. Although defense counsel might have no indication before filing a motion to dismiss how the state will justify a particular delay, the chance (however slim) that the state will be unable to explain its delay, resulting in the dismissal of all charges, would call for the filing of a motion in innumerable situations. Defense counsel might conclude that it would be wise to raise such a claim to anticipate a subsequent claim in a habeas action that failure to do so constituted deficient performance. Moreover, a prepared defense counsel can raise this due process claim in every case to gain an important, pretrial tactical advantage: a full walk-through of the state’s entire investiga- tion and an explanation as to why the state made certain decisions. Resources spent assembling the justification for any delay are resources the state cannot commit to the actual prosecution. Page 36 CONNECTICUT LAW JOURNAL September 2, 2025

202 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

I certainly have been open to these arguments and do not subscribe to following a particular rule as a matter of state constitutional law merely because the nation’s highest court has established it.5 We are, after all, our own laboratory of democracy. In the present case, however, I am not persuaded by any of the legal arguments advanced, and I cannot conclude that we will be on the right side of public policy by adopting the balancing test that a majority of this court adopts today. Rather, I believe that any policy reasons that arguably support the rule that all my participating col- leagues have concluded is now constitutionally com- pelled are outweighed by significant policy reasons against adopting such a rule. I continue to believe that ‘‘[t]he statute of limitations, not the due process clause, defines the limits of protec- tion’’ from prearrest delay. State v. Echols, supra, 170 Conn. 18. ‘‘The [d]ue [p]rocess [c]lause has a limited role to play in protecting against oppressive delay.’’ (Emphasis added.) United States v. Lovasco, supra, 431 U.S. 789. The legislative purpose behind a ‘‘statute of limitations is to safeguard criminal defendants from the deprivation of their liberty through the use of stale evidence to secure a conviction’’ by having ‘‘the important and salutary effect of encouraging law enforcement officials promptly to investigate suspected criminal activity.’’ (Internal quotation marks omitted.) State v. Daren Y., 350 Conn. 393, 410–11, 324 A.3d 734 (2024); 5 See, e.g., State v. Purcell, supra, 331 Conn. 362 (adopting state constitu- tional prophylactic rule to protect accused’s invocation of right to counsel during custodial interrogation); State v. Harris, 330 Conn. 91, 96, 131, 191 A.3d 119 (2018) (due process clause in article first, § 8, of Connecticut constitution provides broader protection than federal constitution with respect to admissibility of eyewitness identification testimony); cf. State v. Castillo, 329 Conn. 311, 338–39, 186 A.3d 672 (2018) (D’Auria, J., dissenting) (custodial protections mandated by Miranda v. Arizona, 384 U.S. 436, 478– 79, 86 S. Ct. 1602, 16 L. Ed. 2d 694 (1966), should be afforded to juveniles regardless of whether they are in police custody). September 2, 2025 CONNECTICUT LAW JOURNAL Page 37

353 Conn. 169 SEPTEMBER, 2025 203 State v. McFarland

see also State v. Police, 343 Conn. 274, 292, 273 A.3d 211 (2022) (statutes of limitations ‘‘ ‘represent legislative assessments of relative interests of the [s]tate and the defendant in administering and receiving justice’ ’’). The legislature’s decision not to enact a statute of limitations for murder constitutes a policy declaration as well, and, in my view, it is just as firm a declaration of policy as when it enacts a statute of limitations—short or long— for certain crimes. This is not a situation in which it is difficult to discern intent from legislative silence. See, e.g., State v. Hurdle, 350 Conn. 770, 781, 326 A.3d 528 (2024). Rather, the legislature’s decision not to enact a statute of limitations for murder establishes that it is the policy of the state to allow police and prosecutors an unlimited amount of time to investigate and prose- cute the most serious crime: the taking of a life. See Drumm v. Freedom of Information Commission, 348 Conn. 565, 597, 308 A.3d 993 (2024) (‘‘the fact that there is no statute of limitations means that unsolved murder investigations may remain open, at least nominally, forever’’). I understand the other concurrences to contend essentially that, under the due process clauses of our state constitution, the legislature’s determination to have, or not to have, a statute of limitations for certain crimes, can be held to be unconstitutional as applied to the circum- stances of a potential case. To protect against this hypothet- ical future case, both concurrences propose a test for criminal defendants that is less stringent than that outlined by federal law and that augments the legislatively deter- mined limitation period. As Justice Ecker reminds us; see part I A 3 of Justice Ecker’s concurring opinion; when evaluating whether a particular additional or substitute procedural safeguard is necessary as a matter of state constitutional due process, we have ‘‘[b]orrow[ed] from the federal constitutional standard enunciated by the United States Supreme Court in Mathews v. Eldridge, 424 U.S. Page 38 CONNECTICUT LAW JOURNAL September 2, 2025

204 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

319, 334–35, 96 S. Ct. 893, 47 L. Ed. 2d 18 (1976), [and] we have held that due process under the Connecticut constitu- tion requires us to ‘consider three factors: First, the private interest that will be affected by the official action; second, the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards; and finally, the [g]overnment’s interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail.’ ’’ State v. Morales, 232 Conn. 707, 721, 657 A.2d 585 (1995). Based on the familiar Mathews analysis, I conclude that the balancing test that this court has adopted is not required (or, frankly, a good idea) as a matter of state constitutional law. I am not satisfied that our state constitu- tion, any more than the federal constitution, mandates that a trial court at a pretrial or posttrial hearing evaluate the reasons why the prosecution made specific tactical or legis- lative decisions during its investigation, including those made based on the availability of financial and personnel resources. I now consider the Mathews factors. 1. Private Interest and Risk of Erroneous Deprivation. The private interest at stake—the defendant’s liberty—is obviously high. This would be true in any given criminal prosecution in which a defendant claims he has been preju- diced by the state’s delay in prosecuting him, although it is depicted in dramatic relief in the present case, involving a murder prosecution brought thirty-two years after the death of the victim. But the facts of the present case do not make the private interest in favor of the balancing test adopted by a majority of this court any higher given that the other concurring justices do not suggest—nor could they—that their balanc- ing test will be confined only to (1) murder cases, (2) prosecutions with no statutes of limitations, (3) prosecu- tions delayed for at least thirty years, or (4) cases solved by DNA testing, in which the claim of prejudice involves September 2, 2025 CONNECTICUT LAW JOURNAL Page 39

353 Conn. 169 SEPTEMBER, 2025 205 State v. McFarland

lost alibi witnesses or witnesses, such as in the present case, who would have identified a third party as the perpe- trator. The balancing test adopted in this case permits defendants to demand, for even the shortest delays, a pre- trial inquiry into the investigative techniques and motiva- tions of the state and the police, regardless of whether the charged crimes are subject to a statute of limitations. See, e.g., State v. Echols, supra, 170 Conn. 16–17 (rejecting defen- dant’s claim of constitutional prejudice based on six month delay in bringing charge of sale of narcotics to undercover police agent); State v. Coleman, 199 Conn. App. 172, 180 n.6, 235 A.3d 655 (rejecting defendant’s due process claim that he was prejudiced by state’s three year delay in bring- ing charges of assault, robbery and possession of firearm, which were subject to five year statute of limitations), cert. denied, 335 Conn. 966, 240 A.3d 281 (2020). In fact, Hodge, the case that both of the other concurrences consider seminal to the constitutional question in this case, involved a prosecution for the sale of narcotics in which the defen- dant claimed that he was prejudiced by a three week delay before being arrested. See State v. Hodge, supra, 153 Conn. 566–67. Nor can the other concurrences credibly maintain that the ‘‘risk of an erroneous deprivation of [the defendant’s] interest through the procedures used’’; Mathews v. Eldridge, supra, 424 U.S. 335—the trial rights that defen- dants presently enjoy—is high without the newly minted due process balancing test. In fact, both concurrences admit the risk is low. Justice Ecker, for example, assures us of ‘‘how difficult it is for a defendant to establish a due process violation and how rarely prosecutions are dismissed on th[e] basis [of the balancing test].’’ Part I A 4 of Justice Ecker’s concurring opinion. Justice Alexander seconds that notion, suggesting to trial courts that it will be ‘‘an exceedingly rare case for a defendant to establish a due process violation when ‘the delay is legitimately investigative in nature.’ ’’ Because the cases in which the Page 40 CONNECTICUT LAW JOURNAL September 2, 2025

206 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

new test will be invoked are likely to be anything but rare, however, making an appropriate evaluation of the remaining two factors under Mathews is critical. I now turn to the government’s interest. 2. The government’s interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail. The burdens likely to accompany the stan- dard adopted by a majority of this court and explained by the other concurrences are substantial. The United States Supreme Court was able to foresee that increasing the pressure on state prosecutors with a rule that put prosecu- tions at risk before prosecutors are ‘‘completely satisfied that [they] should prosecute’’ would result in rushed investi- gations and even greater problems for defendants. ‘‘Penaliz- ing prosecutors who defer action for these reasons would subordinate the goal of ‘orderly expedition’ to that of ‘mere speed’ . . . .’’ United States v. Lovasco, supra, 431 U.S. 795. This rule would ‘‘increase the likelihood of unwar- ranted charges being filed, and would add to the time during which defendants stand accused but untried,’’ which is massively disruptive to an accused’s life, liberty, and family. Id., 791. The balancing test that a majority of this court makes mandatory today under our state constitution will require that authorities investigating a crime preemptively antici- pate how the passage of time might impact a suspect’s defense (if they have a suspect and happen to know his defense), including when witnesses or other evidence might become unavailable. These anticipatory forces work to discourage prosecutors from waiting to bring a case until they are ‘‘completely satisfied that [they] should prose- cute and will be able promptly to establish guilt beyond a reasonable doubt,’’ or risk losing the prosecution alto- gether. Id., 795. I would decline this invitation to undermine the legislature’s public policy determination—reflected in its determination of whether to enact a statute of limita- September 2, 2025 CONNECTICUT LAW JOURNAL Page 41

353 Conn. 169 SEPTEMBER, 2025 207 State v. McFarland

tions—by adopting a test authorizing a court to void a prosecution—before or after trial and in the name of due process—absent state action seeking to gain a tactical advantage. I am reluctant to permit defendants and trial courts to wade into and second-guess the prosecution’s assessment of when appropriately to bring its case. ‘‘Allowing inquiry into when the police could have arrested or when the prosecutor could have charged would raise difficult prob- lems of proof,’’ causing trial courts to ‘‘be engaged in lengthy hearings in every case to determine whether or not the prosecuting authorities had proceeded diligently or otherwise.’’ (Emphasis added; internal quotation marks omitted.) United States v. Marion, supra, 404 U.S. 321 n.13. Even if trial courts deny most motions, as my colleagues make clear should happen, the impact on the executive prerogative will have occurred, either by hastening prose- cutions that authorities were not ready to bring, or by having to explain their investigative decisions in court. But cf. State v. Corchado, 200 Conn. 453, 460, 512 A.2d 183 (1986) (although ‘‘not unlimited,’’ ‘‘[t]he criminal justice system gives prosecutors a wide latitude and broad discre- tion in determining when, who, why and whether to prose- cute for violations of the criminal law’’ (internal quotation marks omitted)). On its face, it might appear that the balancing test bene- fits defendants while the two-pronged federal test favors the state. In my view, this is not accurate. The fundamental basis supporting the United States Supreme Court’s adop- tion of the two-pronged approach is the fairness afforded to prospective defendants and the rejection of a rule that would require an arrest prematurely. See United States v. Lovasco, supra, 431 U.S. 791, 793. I fear that prudential concerns that presently discourage the state from arresting a defendant as soon as possible just because it can, includ- ing ‘‘ ‘the rights of the accused and . . . the ability of society to protect itself’ ’’; id., 791; will be undermined by Page 42 CONNECTICUT LAW JOURNAL September 2, 2025

208 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

the balancing test, which, indeed, will create its own significant burdens by (1) pressuring prosecutors into resolving doubtful cases in favor of early and possibly unwarranted prosecutions, (2) discouraging the state from fully considering factors in addition to the strength of the case before determining whether prosecution would be in the public interest, and (3) impairing a prosecutor’s ability to continue to investigate cases that involve more than one criminal transaction or actor. See id., 793. The intensive inquiry of the balancing test will require the prosecution, along with investigative and cold case units in police departments statewide, to change the way it operates. The other two concurrences would have the state and its investigative teams specifically document how and why they made decisions to allocate resources to one case rather than another so that they might—perhaps many years down the road—satisfy their burden under this test. Anticipating that its eviden- tiary burden will shift once the defendant manifests that he is sufficiently prejudiced, the state will likely have to have evidence ready to present to a court that goes beyond simply whom it spoke with and when, but also why certain leads were not pursued or why other matters were given priority. When deciding whether to arrest a potential defendant, or even whether to make certain cases a priority, the state might also need to consider the health of potential witnesses for the defen- dant, if known, as well as the health of the state’s own investigators, who could justify decisions made in the past. Some might be retired, deceased, or otherwise unavailable when the state ultimately decides to prose- cute. If the state does not take these burdensome con- siderations into account, decades of work might be for naught under the new balancing test. Finally, balancing tests such as those the other con- currences favor are notoriously difficult to administer. September 2, 2025 CONNECTICUT LAW JOURNAL Page 43

353 Conn. 169 SEPTEMBER, 2025 209 State v. McFarland

It might be true that some balancing tests, such as the test for whether a defendant has been deprived of a speedy trial, involve factors similar to those relevant to the test announced today. See, e.g., State v. DePas- tino, 228 Conn. 552, 560, 638 A.2d 578 (1994) (factors to balance relevant to claim of violation of speedy trial right include ‘‘ ‘[1] [l]ength of delay, [2] the reason for the delay, [3] the defendant’s assertion of his right, and [4] prejudice to the defendant’ ’’). That is where the resemblance ends. For example, the speedy trial test does not involve a shift of the burden of proof from the defendant to the state upon a showing of prejudice. But, even more distinguishable than the lack of burden shifting is that these factors ‘‘must be considered together with such other circumstances as may be rele- vant.’’ (Internal quotation marks omitted.) State v. Roman, 320 Conn. 400, 418, 133 A.3d 441 (2016). In contrast, the other concurrences call for the weighing of two values against one another—the state’s justification against the defendant’s prejudice—to reach an absolute answer as to whether the case can proceed.6 One court, the United States Court of Appeals for the Fifth Circuit, has described this exercise as ‘‘compar[ing] the incom- parable.’’ United States v. Crouch, 84 F.3d 1497, 1512 (5th Cir. 1996), cert. denied, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997), and cert. denied sub nom. Frye v. United States, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997). ‘‘The items to be placed on either side of the balance (imprecise in themselves) are wholly different from each other and have no possible common denominator that would allow determination of which ‘weighs’ the most.’’ Id. The Fifth Circuit, in its en banc ruling, continued: ‘‘Not only is there no scale or conver- 6 Justice Alexander calls for trial courts to ‘‘ ‘balance the defendant’s prejudice against the government’s justification for delay.’ ’’ Justice Ecker says it just a little differently: the court must ‘‘weigh the state’s justifications for the delay against the prejudice suffered by the defendant.’’ Part I A 2 of Justice Ecker’s concurring opinion. Page 44 CONNECTICUT LAW JOURNAL September 2, 2025

210 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

sion table to tell us whether eighty percent of minimally adequate prosecutorial and investigative staffing is out- weighed by a low-medium amount of actual prejudice, there are no recognized general standards or principles to aid us in making that determination and virtually no body of precedent or historic practice to look to for guidance.’’ Id. Therefore, the balancing test is impracti- cal and will likely lead to inconsistent results because ‘‘[t]he denominators on both sides of the equation are simply incommensurate, such that [i]t is more like judg- ing whether a particular line is longer than a particular rock is heavy.’’ (Internal quotation marks omitted.) D. Diethrich, Note, ‘‘If It Ain’t Broke, Don’t Fix It: Indict- ment Delays and Due Process,’’ 15 Elon L. Rev. 181, 202 (2023). Moreover, as I will explain in part IV of this opinion, trial courts will now be responsible for figuring out how to balance a rock and a line with almost no practical guidance as to how to accomplish this balance, save for being told that a defendant should rarely prevail.7 The administrative and financial burdens the balanc- ing test creates are not fictional or hypothetical. In fact, plagued by these problems, some jurisdictions (includ- ing the Fifth Circuit, which spawned the minority rule) have rejected their prior adoption of the balancing test, returning instead to the federal two-pronged approach. See, e.g., United States v. Crouch, supra, 84 F.3d 1508–11; see also Jackson v. State, 347 So. 3d 292, 306 (Fla. 2022) (rejecting balancing test under Florida law in favor of federal two-pronged test). The reasons that these jurisdictions have reversed course include both 7 For example, the other two concurrences have not ruled out, if theirs is a true balancing test, that a defendant’s claim of actual and substantial prejudice can outweigh, in a particular (but rare) case, even entirely defensi- ble and justified delays in bringing the prosecution. In that case, the prosecu- tion would be barred, as a matter of due process, notwithstanding the legislature’s determination not to enact a statute of limitations for the crime or to enact one of a certain length. September 2, 2025 CONNECTICUT LAW JOURNAL Page 45

353 Conn. 169 SEPTEMBER, 2025 211 State v. McFarland

the impracticality of balancing two very distinct values or considerations, and the separation of powers con- cerns implicated by requiring that courts second-guess decisions such as the staffing of certain investigations. See United States v. Crouch, supra, 1512–13; Jackson v. State, supra, 305.8 I would heed the warning of these jurisdictions. 3. The Value of Additional Safeguards. I understand my colleagues’ sincere concerns that delayed prosecu- tions have the potential to compromise a defendant’s due process right to a fair trial. I share those concerns. But I am not persuaded that it is sensible (let alone required) to ‘‘constitutionalize’’ the apparatus that a major- ity of this court erects in this case. Nor am I persuaded that it, in fact, brings with it significant additional safe- guards. Prosecutors already have significant incentives to ini- tiate an arrest as promptly as they can. Delays can significantly compromise the state’s case as much as it can a defendant’s case: evidence can decay, memories can fade, suspects may flee, and witnesses can become unavailable. Proving the defendant’s guilt beyond a rea- sonable doubt is a high burden as it is. It does not become easier to meet this standard with aging evi- dence and witnesses. Pressure from victims and their families can also work to prevent prosecutors from delaying the decision to make an arrest. And, of course, there are statutes of limitations for many crimes. For reasons already discussed, I do not believe that interpre- ting our due process clauses to require that the state 8 The United States Court of Appeals for the Fourth Circuit has entertained but rejected arguments that it should reject its use of the balancing test and instead adopt the two-pronged test on the ground that it is restricted from doing so. See Jones v. Angelone, 94 F.3d 900, 905 (4th Cir. 1996) (‘‘as the [c]ommonwealth is aware, we cannot, as a panel of the court, overrule the decision of another panel; only the en banc court may overrule a prior panel decision’’). Page 46 CONNECTICUT LAW JOURNAL September 2, 2025

212 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

hasten to make ‘‘[t]he decision to file criminal charges, with the awesome consequences it entails’’; United States v. Lovasco, supra, 431 U.S. 794; is a wise policy choice for either the state or for potential defendants. Moreover, our courts are already equipped with tools to ensure that a defendant’s rights are protected. One notable existing procedure available to defendants is embodied in General Statutes § 54-56, which provides trial courts with the authority to, ‘‘upon motion by the defendant, dismiss any information . . . if, in the opin- ion of the court, there is not sufficient . . . cause to justify the . . . continuing of such information or the placing of the person accused therein on trial.’’ We have observed that § 54-56 was enacted ‘‘to prevent unchecked power by a prosecuting attorney.’’ (Internal quotation marks omitted.) State v. Angel C., 245 Conn. 93, 116 n.26, 715 A.2d 652 (1998). We have also said that ‘‘the power to render a dismissal under § 54-56 for insufficient cause is to be sparingly exercised and then only with great caution and awareness of the probable consequences.’’ (Internal quotation marks omitted.) State v. Kinchen, 243 Conn. 690, 703–704, 707 A.2d 1255 (1998); see also State v. Corchado, supra, 200 Conn. 464. To ensure that this discretion is exercised in accordance with these principles, it is essential for ‘‘the court explic- itly to weigh all the competing factors and considera- tions of fundamental fairness to both sides—the defendant, the state and society, and presumably the victim. . . . This difficult and delicate process neces- sarily involves a careful consideration by the court of such factors as the strength of the state’s case, the likelihood of conviction, the severity of the crime, its effect on the victim, the strength of the defendant’s defense, the defendant’s personal situation, and all the other myriad factors that underlie a judgment regarding fundamental fairness.’’ (Citation omitted.) State v. Dills, 19 Conn. App. 495, 503–504, 563 A.2d 733 (1989); see September 2, 2025 CONNECTICUT LAW JOURNAL Page 47

353 Conn. 169 SEPTEMBER, 2025 213 State v. McFarland

also State v. Daniels, 209 Conn. 225, 238, 550 A.2d 885 (1988), cert. denied, 489 U.S. 1069, 109 S. Ct. 1349, 103 L. Ed. 2d 817 (1989); State v. Corchado, supra, 458. Thus, a trial court’s invocation of its authority to dismiss a case under the insufficient cause prong of § 54-56 can be justified only when (1) the court expressly and carefully has considered all of the relevant competing factors, and (2) dismissal is supported by overriding equitable considerations. See State v. Kinchen, supra, 704. There is not much case law outlining when the statute applies. But this court has made clear that it should be invoked ‘‘ ‘only in the most compelling of circumstances’ ’’; State v. Golodner, 305 Conn. 330, 339, 46 A.3d 71 (2012); meaning that trial court dismissals under this statute will be as ‘‘exceedingly rare’’ as Jus- tice Alexander suggests in her concurrence that dis- missals will be under the new constitutional balancing test adopted by a majority of this court. I would allow trial courts to employ this already existing statutory device rather than resorting to manufactured constitu- tional balancing tests that have little hope of being administered consistently.9 Furthermore, § 54-56 is but one of many other existing procedures available under our rules of practice that amply equip trial judges with authority to ensure that trials are conducted fairly in delayed prosecutions, just as in all prosecutions. For example, trial courts have ample discretion when making evidentiary rulings that they can—and should—exercise in favor of defendants if the passage of time has so compromised their ability 9 Notably, although a court undertaking an analysis under § 54-56 is called on to ‘‘weigh all the competing factors and considerations of fundamental fairness’’; State v. Dills, supra, 19 Conn. App. 504; there is no shift of the burden of proof and no weighing of one thing against another: prejudice and justification. Moreover, the fact that this balancing derives from case law interpreting and applying a statute means that the legislature can alter or eliminate the balancing, as opposed to a constitutional balancing that might be written in stone (unless overruled and abandoned). Page 48 CONNECTICUT LAW JOURNAL September 2, 2025

214 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

to defend themselves that their right to a fair trial could be jeopardized.10 Posttrial, the trial court also has both legal and discretionary authority to review a jury ver- dict, including by entertaining judicially recognized motions for a judgment of acquittal or for a new trial. See Practice Book §§ 42-51 and 42-53. Finally, this court has the authority and responsibility to review whether the evidence supported the jury’s finding of guilt beyond a reasonable doubt. Ultimately, I would not conclude, as my colleagues do, that the balancing test adopted by a minority of jurisdictions is necessary or appropriate as a matter of our state constitutional law. I do not believe that any- thing in the text or history of our state constitution, or in this court’s precedent, supports recognition of the additional procedural safeguard that a majority of this court today mandates, and, assisted by the Mathews test, I do not conclude that the state constitution other- wise compels it as a matter of policy. My analysis involves essentially the same reasons that both the Fifth Circuit and the Florida Supreme Court cited to when they both abandoned the balancing test after having initially started down that road. To use an idiom that, in my view, sums up the Mathews test, ‘‘the juice isn’t worth the squeeze.’’ In light of the procedures already avail- able to defendants and the ample discretion trial courts already possess to conduct trials fairly, I am not per- suaded that the value of the majority’s balancing test is sufficient to justify the burdens that it will impose on the state, the courts, and ultimately on defendants, especially given what the other concurrences admit is a low risk. I would decline the invitation that the other concurrences have accepted to go beyond the federal 10 For example, if the written statements of the deceased witness, Veronica Saars-Doyle, were not so inconsistent and unreliable, the trial court in the present case might have exercised its discretion to admit them so that the defendant could mount a defense. September 2, 2025 CONNECTICUT LAW JOURNAL Page 49

353 Conn. 169 SEPTEMBER, 2025 215 State v. McFarland

two-pronged standard, and instead would trust trial and appellate courts to ferret out that unicorn delayed pros- ecution that substantially prejudices a defendant, which the other concurrences are on a quest to prevent, and render relief more conventionally. IV The due process balancing test applicable to claims of unconstitutional prejudice from delayed prosecutions, announced today and explained in the two other con- currences, creates many questions, and the concur- rences provide little guidance to trial courts and the Appellate Court on how to administer it or to review its application, respectively. Assuring trial courts that it will be ‘‘exceedingly rare’’ for the test to be met does nothing to assist those courts in making the necessary findings or even to assist the parties in developing an appropriate record. The motivation for those concur- ring in the announcement of this test might be an antici- pated increase in erstwhile cold case prosecutions, as in the present case, in light of the revolutionary crime- solving capabilities of DNA testing. Those favoring today’s balancing test might be confident that, however our trial courts develop and apply the test ‘‘ ‘in the first instance,’ ’’ and however the ‘‘body of case law’’ develops applying this test as a question of law, as Justice Alexander states in her concurring opinion, sci- entific advances such as those in DNA testing, which might identify a suspect never before identified, or clinch a prosecution that perhaps involved a suspect but not sufficient proof to make an arrest, will always suffice to justify any delay and rebut any claim of actual and substantial prejudice. But delayed prosecution claims come in all varieties, and, as previously discussed, this state’s due process clauses (and therefore this balancing) are not limited to murder prosecutions, to crimes without statutes of Page 50 CONNECTICUT LAW JOURNAL September 2, 2025

216 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

limitations, or to particular cases finally solved by DNA testing. I do not believe it is wise to adopt a new test without a set framework as to how it is to work. Direct- ing our trial courts to several federal circuits that apply a version of the balancing test offers very little guidance because, as at least some of my colleagues acknowl- edge, there is no prevailing balancing test and the deci- sions in those circuits are inconsistent. Compare United States v. Avants, 367 F.3d 433, 442 (5th Cir. 2004) (addressing only bad faith prong and not prejudice prong), with United States v. Davis, 571 Fed. Appx. 248, 249 (4th Cir. 2014) (‘‘the [g]overnment did not commit prosecutorial [impropriety] in delaying [the] bringing [of] an indictment against [the defendant]’’). Left unre- solved is when trial courts are to apply the balancing (before or after trial), what evidence will suffice for the parties to meet their respective burdens, and how an appellate court will review the trial court’s determi- nation. It is predictable that the balancing test will not yield a consistent body of law. See, e.g., United States v. Wood, Docket Nos. 91-5186 and 92-5010, 1992 WL 301975, *1 (4th Cir. October 21, 1992) (decision without published opinion, 977 F.2d 575) (under balancing test, ‘‘a determination of whether such a delay has denied due process necessitates a case-by-case inquiry’’). The practicalities of implementing the balancing test that a majority of this court adopts present additional, and foreseeable, difficulties for the state, the defendant, and the courts. I highlight the most pressing of these issues in the hope of casting an eye forward to perhaps reconsidering whether the test is necessary or appro- priate the next time we have the issue before us.11 11 Unresolved, but likely to arise almost immediately, is whether the new balancing test announced today will apply retroactively to defendants who already have been convicted who might, by way of habeas corpus or other- wise, seek judicial review of the reasons why the state delayed their arrest. See, e.g., State v. Rogers, 344 Conn. 343, 364–65, 279 A.3d 184 (2022) (com- plete retroactive effect is most appropriate when new constitutional princi- ple is designed to enhance accuracy of criminal trials); State v. Turner, September 2, 2025 CONNECTICUT LAW JOURNAL Page 51

353 Conn. 169 SEPTEMBER, 2025 217 State v. McFarland

1. When the trial court should conduct this balancing is not clear. Presumably, the trial court must hold a hearing to entertain claims of prejudice from the defen- dant and justifications from the state, both of which are likely in many cases to require testimony. Justice Alexander suggests that the court could hold the hear- ing and make the determination either pretrial or post- trial. Justice Ecker does not say, but he articulates a view that one of the relevant factors in determining whether the defendant has established actual and sub- stantial prejudice from the delay includes the strength of the state’s case. See part I A 2 of Justice Ecker’s concurring opinion. If that is the case, I do not under- stand how this inquiry can be made pretrial in most cases, unless the court is to hear the entirety of the state’s case and to assess its strength (including, some- how, the credibility of witnesses) before the trial begins. And if this constitutional calculus is undertaken post- trial, and the jury has found the defendant guilty beyond a reasonable doubt, the trial court will be in the unenvi- able position of assessing, by gradations, how far on the spectrum of ‘‘beyond a reasonable doubt’’ the jury found that the state had proved its case so that factor can be balanced against the actual and substantial preju- dice the defendant has established. 2. Whenever this balancing is undertaken, the trial court must first determine whether the defendant has sufficiently made out a claim of harm from the delay that would result in ‘‘actual and substantial prejudice.’’ The two other concurrences provide examples of what does not satisfy the prejudice test but very little guid- 334 Conn. 660, 677 n.6, 224 A.3d 129 (2020) (same); see also Tatum v. Commissioner of Correction, 349 Conn. 733, 744–50, 322 A.3d 299 (2024) (determining whether this court’s creation of new watershed rule of criminal procedure applies on collateral review). Is the state required to dust off its decades old files to defend these challenges and offer reasons why it did not more rapidly arrest defendants who ultimately were convicted? Page 52 CONNECTICUT LAW JOURNAL September 2, 2025

218 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

ance about what would satisfy the test.12 In the present case, the trial court might have to make credibility determinations, including accounting for witnesses who might exculpate the defendant, inculpate another individual, or testify that the defendant had acted in self- defense, while intoxicated, or under extreme emotional distress. Even if the court decides to reject the claim of ‘‘actual and substantial prejudice,’’ it will then have to decide whether to complete the balancing test provi- sionally, and have the state present evidence justifying the delay, to perfect a complete record for appellate review. If the defendant succeeds in proving actual and sub- stantial prejudice, and the burden shifts to the state to justify the delay, the court today provides even less guidance about what happens next. Although Justice Alexander indicates that a defendant will rarely be able ‘‘to establish a due process violation when ‘the delay is legitimately investigative in nature,’ ’’ that inquiry, too, might require the trial court to assess the credibility of the witnesses offered in support of the legitimacy of the investigative delay. In a case that takes more than thirty years to solve, those who made choices in the investigation—for example, focusing on the wrong sus- pect, failing to interview certain witnesses, or putting the investigation on hold for lack of resources or for other priorities—might be long retired or deceased. Or will the court accept any reasonable explanation that is not contradicted by the record, including posthoc rationalizations, without regard to whether that was an actual reason for the delay? Cf. Johnson v. Commis- sioner of Correction, 330 Conn. 520, 550–51, 198 A.3d 12 Justice Ecker cites to one case in which a court held that the death of an alibi witness did not cause one of the codefendants to suffer actual and substantial prejudice because the witness’ testimony would have been cumulative of other witnesses’ testimony. See footnote 5 of Justice Ecker’s concurring opinion (citing United States v. McMutuary, 217 F.3d 477, 482 (7th Cir.), cert. denied, 531 U.S. 1001, 121 S. Ct. 502, 148 L. Ed. 2d 471 (2000)). September 2, 2025 CONNECTICUT LAW JOURNAL Page 53

353 Conn. 169 SEPTEMBER, 2025 219 State v. McFarland

52 (2019) (in reviewing performance of defense counsel, we are ‘‘required not simply to give [the] attorneys the benefit of the doubt . . . but to affirmatively entertain the range of possible reasons . . . counsel may have had for proceeding as they did’’ (internal quotation marks omitted)), quoting Cullen v. Pinholster, 563 U.S. 170, 196, 131 S. Ct. 1388, 179 L. Ed. 2d 557 (2011).13

3. We will have to settle on an appellate standard of review. Although Justice Alexander appears to promise plenary review of the outcome of the trial court’s bal- ancing,14 Justice Ecker does not say. Among the minor- ity of appellate courts that have applied the balancing test, uniformity is lacking in how they review trial court determinations of prejudice and justifications for delay. Some review for abuse of discretion, others for clear error, and yet others review de novo. See, e.g., United States v. Villa, 70 F.4th 704, 715 (4th Cir. 2023) (de novo review); United States v. Corona-Verbera, 509 F.3d 1105, 1114 (9th Cir. 2007) (abuse of discretion), cert. denied, 555 U.S. 865, 129 S. Ct. 149, 172 L. Ed. 2d 110 (2008). 13 Justice Alexander appears to favor the approach of accepting any reason- able justification, given her speculation—which might be a reasonable guess but is certainly not based on record evidence—that, in 1996, at the time of the defendant’s confessions in the present case, authorities did not pursue a prosecution of the defendant because doing so would have subjected him to the death penalty and they needed stronger evidence before charging him. It would be important for the state to know if it can simply proffer possibly plausible explanations to meet its burden once the defendant satis- fies his burden of proving prejudice. I am not certain whether a majority of this court would agree with Justice Alexander, however. 14 If, indeed, we conduct a de novo review, it might be advisable, especially in decades old murder cases, to avoid cases that were not dismissed (either pretrial or posttrial) for lack of sufficient prejudice and, therefore, in which no evidence was presented on the state’s justification for the delay and no balancing was undertaken. If an appellate court a few years later holds that the prejudice actually was sufficient and that the trial court should have conducted a balancing of that prejudice against the state’s purported reason for delay, the delay will only continue to mount. Page 54 CONNECTICUT LAW JOURNAL September 2, 2025

220 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

In sum, I am not persuaded that we should adopt the minority rule that a majority of this court indulges today. I would uphold the trial court’s denial of the defendant’s motion to dismiss on a ground different from that relied on by the rest of the court, namely, the defendant’s failure to satisfy the second prong of the federal two-pronged test: that the reasons for the delay were unjustifiable. Accordingly, I respectfully concur in the judgment.

ECKER, J., with whom McDONALD, J., joins as to part I, concurring. I concur in the opinion of the court. I write separately to elaborate on the analysis, con- tained in part I of the court’s opinion, used to adjudicate a claim that a defendant’s due process rights secured by the state constitution were violated by a prearrest delay.1 I also will briefly express my views regarding the proper treatment of the hearsay statements of the deceased witness in the context of this case. I PREARREST DELAY The defendant, Willie McFarland, claims that the thirty-two year delay between the commission of the murders of which he was convicted and his arrest vio- lated his right to a fair trial under the due process clauses of the federal and state constitutions. I begin my examination of this claim by reference to the analy- sis used by the trial court when it denied the defendant’s motion to dismiss based on the same claim. With respect to the federal constitution, the trial court recognized that there is a split of authority as to the governing test. 1 In the due process context, courts generally use the terms ‘‘prearrest delay,’’ ‘‘preindictment delay,’’ and ‘‘preaccusation delay’’ interchangeably to refer to the lapse in time between the occurrence of a crime and a formal charge being brought against the defendant for committing that crime. September 2, 2025 CONNECTICUT LAW JOURNAL Page 55

353 Conn. 169 SEPTEMBER, 2025 221 State v. McFarland

The differences include subtleties and variations that will be elaborated in the course of this opinion but can be readily summarized. A majority of federal courts of appeals, joined by many state courts when adjudicating federal due process claims, apply a stringent, two- pronged test: to establish that a prearrest delay offends due process, the defendant bears the burden of proving both that he suffered actual and substantial prejudice as a result of the delay and that the government engaged in the delay intentionally and for wholly unjustified or bad faith purposes, such as to gain an unfair tactical advantage or to harass the defendant. By contrast, a minority of federal courts of appeals and a significant number of other state courts deciding federal or state due process claims apply a more flexible balancing test under which the defendant still must establish that the delay caused actual and substantial prejudice at trial, but, once such a showing is made, the burden shifts to the government to explain the rea- sons for the delay. To decide the constitutional claim, the court then balances the prejudice to the defendant against the government’s proffered explanations and justifications for delay. The balancing test is less stringent than the two- pronged test in three ways. Although both tests require the defendant to demonstrate actual and substantial prejudice as a result of the prearrest delay, (1) the balancing approach imposes on the government the burden of explaining the reason for the delay in the arrest and prosecution of the defendant, (2) a delay may be deemed impermissible under the balancing approach if the proffered reasons for the delay are wholly unjustifiable, even in the absence of intentional bad faith on the part of the government, and (3) as with the closely related sixth amendment right to a speedy trial, the ultimate determination of whether the delay denied the defendant a fair trial involves balancing the Page 56 CONNECTICUT LAW JOURNAL September 2, 2025

222 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

government’s reasons for the delay against the preju- dice to the defendant. The trial court concluded that Connecticut has adopted the majority interpretation of the federal due process clause in State v. Morrill, 197 Conn. 507, 522, 498 A.2d 76 (1985), and that the defendant in the present case was unable to prevail under the two-pronged test because, regardless of any prejudice, he conceded that there was no evidence that the state had delayed in arresting him to obtain an unfair tactical advantage or for other improper purposes. After conducting a thor- ough Geisler analysis, however; see State v. Geisler, 222 Conn. 672, 684–85, 610 A.2d 1225 (1992); the trial court concluded that the state constitution confers broader protections in this regard and that Connecticut courts would apply a balancing test to claims brought under the due process clauses of the state constitution. Still, the trial court determined that the defendant could not prevail under the balancing test on this record. Even assuming, for the sake of argument, that the defendant had suffered prejudice, the court held that the prearrest delay was wholly justified because the state promptly initiated the prosecution once newly developed DNA technologies established that the defendant was involved in the murders of the victims, Fred Harris and Gregory Harris. On appeal, the defendant contends that we should confirm the trial court’s conclusion that a balancing test is necessary to satisfy due process under the state constitution or, alternatively, repudiate Morrill and its progeny and adopt the minority view as to the federal due process clause. Under either approach, he argues that he should prevail because the trial court misapplied the balancing test, and a proper analysis demonstrates that he was denied the right to a fair trial. This court agrees with the defendant that the due process clauses of the state constitution require the application of a September 2, 2025 CONNECTICUT LAW JOURNAL Page 57

353 Conn. 169 SEPTEMBER, 2025 223 State v. McFarland

balancing test that involves consideration of all of the circumstances relevant to both the causes and conse- quences of the prearrest delay.2 Applying that test, how- ever, this court also agrees with the trial court that the defendant failed to establish that he was denied a fair trial. A Scope of State Due Process Protections Relying on State v. Hodge, 153 Conn. 564, 568, 219 A.2d 367 (1966), the defendant contends that this court already has recognized that, to determine whether a prearrest delay transgresses due process limitations under our state constitution, we apply a totality of the circumstances test that is more flexible than the two- part test used by most federal courts. Although I agree in substance with this reading of Hodge, we also must acknowledge that Hodge was decided before State v. Geisler, supra, 222 Conn. 672, and it did not analyze the state constitutional issue in any depth or consider the factors that we subsequently identified as relevant in Geisler. See id., 684–85. It is therefore appropriate, in my view, ‘‘to engage in a more robust consideration’’ of the state constitutional claim in the present case. State v. Langston, 346 Conn. 605, 625, 294 A.3d 1002 (2023), cert. denied, U.S. , 144 S. Ct. 698, 217 L. Ed. 2d 391 (2024); see, e.g., id., 624–25. ‘‘In [Geisler], we identified six nonexclusive tools of analysis to be considered, to the extent applicable, whenever we are called on as a matter of first impres- sion to define the scope and parameters of the state constitution: (1) persuasive relevant federal precedents; (2) historical insights into the intent of our constitu- tional forebears; (3) the operative constitutional text; 2 Because this court concludes that the defendant is entitled to application of the balancing test under the state constitution, we do not revisit our interpretation of the proper legal analysis under the federal constitution. Page 58 CONNECTICUT LAW JOURNAL September 2, 2025

224 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

(4) related Connecticut precedents; (5) persuasive prec- edents of other states; and (6) contemporary under- standings of applicable economic and sociological norms, or, as otherwise described, relevant public poli- cies.’’ State v. Santiago, 318 Conn. 1, 17–18, 122 A.3d 1 (2015). We have also observed, however, that ‘‘[i]t is not critical to a proper Geisler analysis that we discuss the various factors in any particular order or even that we address each factor.’’ (Internal quotation marks omitted.) O’Shea v. Scherban, 339 Conn. 775, 797, 262 A.3d 776 (2021). I will focus on the four factors that I consider most useful to the resolution of the issue presented in this case. 1 Connecticut Precedent I agree with the defendant that Connecticut prece- dent carries substantial weight in the present case because it provides solid support for the trial court’s conclusion that excessive prearrest delay claims are assessed according to a balancing test, which takes all relevant circumstances into account, under the state constitution.3 In Hodge, the defendant contended that a three week delay between the commission of the narcotics offense and his arrest constituted an unreasonable seizure of his person, in violation of the predecessor to article first, § 7, of the 1965 Connecticut constitution. See State v. Hodge, supra, 153 Conn. 566–67. Although at trial the defendant in Hodge had challenged the legality of only the seizure, this court characterized the constitutional 3 The question under this Geisler factor is not whether there is controlling precedent establishing that broader protection has been found to exist under the state constitution, but whether there is any basis in the holdings or dicta of our decisions, or those of the Appellate Court, that supports adopting such broader protection. See, e.g., State v. Wilkins, 240 Conn. 489, 505, 692 A.2d 1233 (1997). September 2, 2025 CONNECTICUT LAW JOURNAL Page 59

353 Conn. 169 SEPTEMBER, 2025 225 State v. McFarland

issue more broadly: ‘‘[When] the delay in arresting a defendant (or in otherwise apprising him of the charges against him) continues long after all the evidence has been assembled, and becomes a product of mere conve- nience to the state, a question of an unreasonable sei- zure or lack of a fair trial may arise.’’ (Emphasis added.) Id., 567–68. Hodge found support for this proposition in Ross v. United States, 349 F.2d 210 (D.C. Cir. 1965), one of the first cases in the country to hold that a delay in bringing an indictment violated a defendant’s due process rights. See State v. Hodge, supra, 567–68; see also Ross v. United States, supra, 211–13. Ross assessed the due process claim according to a balancing test. See, e.g., Ross v. United States, supra, 213 (‘‘When inter- ests of this nature impinge [on] each other, as they have a way of doing, they must be accommodated. A balance must be struck, if one or the other is not to be sacrificed completely.’’). In Hodge, this court applied the due pro- cess analysis used in Ross as ‘‘to a claim of an unreason- able seizure brought under our state constitution.’’ State v. Hodge, supra, 568. Having thus framed the issue, Hodge adopted the following test for claims of prearrest delay under the state constitution: ‘‘The defendant’s rights under this claim must necessarily depend on all the circumstances, including the length of the delay, the reason for the delay, prejudice to the defendant, and a timely presenta- tion of the claim to the trial court. Some prejudice to the defendant’s case must be shown. . . . Such preju- dice might consist of the unavailability of alibi witnesses or the impaired memory of the defendant and others who vouch for his innocence.’’ (Citation omitted.) Id. This Hodge test is almost verbatim the test that this court uses to assess other constitutional claims arising from delayed prosecution, such as alleged violations of the sixth amendment right to a speedy trial. Compare id. (‘‘the length of the delay, the reason for the delay, Page 60 CONNECTICUT LAW JOURNAL September 2, 2025

226 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

prejudice to the defendant, and a timely presentation of the claim to the trial court’’), with Barker v. Wingo, 407 U.S. 514, 530, 92 S. Ct. 2182, 33 L. Ed. 2d 101 (1972) (considering ‘‘[l]ength of delay, the reason for the delay, the defendant’s assertion of his right, and prejudice to the defendant’’ to determine whether speedy trial right was violated). See also Gaines v. Manson, 194 Conn. 510, 527–28 and n.15, 481 A.2d 1084 (1984) (applying Wingo test to alleged violations of right to timely prose- cution of appeal under due process clause of state con- stitution). The United States Supreme Court repeatedly has referred to that nearly identical totality of the cir- cumstances test as a balancing test. See, e.g., Barker v. Wingo, supra, 530 (‘‘[t]he approach we accept is a balancing test’’). Hodge thus stands for the proposition that prearrest delay claims alleging the deprivation of the right to a fair trial under the state constitution are subject to a balancing test. While Hodge left open some questions about how the balancing test would apply, I think Hodge makes clear that intentional bad faith on the part of the state is not an absolute precondition to a due process violation, while also recognizing that ‘‘[a] reasonable delay . . . may be a proper adjunct to responsible police investigation and should not undermine the legal- ity of subsequent arrests and convictions except under unusual and clearly prejudicial circumstances.’’ State v. Hodge, supra, 153 Conn. 569. The court in Hodge noted that ‘‘[o]rdinarily a delay between the time of an offense and the time of making an arrest will not affect the legality of the arrest or of the criminal proceedings subsequent thereto’’ insofar as, ‘‘[o]ften a lengthy delay is required because the identity of the offender is unknown to the authorities or because additional time is needed to gather sufficient evidence to justify an official charge against one who is suspected of crime.’’ Id., 567. Hodge also recognized that ‘‘the applicable September 2, 2025 CONNECTICUT LAW JOURNAL Page 61

353 Conn. 169 SEPTEMBER, 2025 227 State v. McFarland

statute of limitations is the ultimate safeguard against a [long delayed] arrest and prosecution.’’ Id. After this court’s decision in Hodge, the United States Supreme Court issued a series of opinions clarifying that an unjustified delay in initiating a prosecution also can violate a defendant’s right to due process under the federal constitution. See United States v. Lovasco, 431 U.S. 783, 97 S. Ct. 2044, 52 L. Ed. 2d 752 (1977); United States v. Marion, 404 U.S. 307, 92 S. Ct. 455, 30 L. Ed. 2d 468 (1971). As I will discuss, a majority of the lower federal courts and state courts—including this court—have read Marion and Lovasco to require that, in order to successfully challenge a prearrest delay under the federal due process clauses, a criminal defen- dant must satisfy both prongs of the relatively stringent, two-pronged test. For example, in State v. Morrill, supra, 197 Conn. 522, this court construed Marion and Lovasco to mean that, ‘‘[i]n order to establish a due process violation because of [preaccusation] delay, the defendant must show both that actual substantial preju- dice resulted from the delay and that the reasons for the delay were wholly unjustifiable, as where the state seeks to gain a tactical advantage over the defendant.’’ The discussion in Morrill, however, was confined to the federal constitution; the court did not address whether the state constitution affords broader due pro- cess protections, and it did not mention Hodge, let alone purport to overrule it. The fact that this court already has articulated a rule that affords greater protection under the state constitution than does the prevailing federal rule strongly counsels in favor of upholding the trial court’s determination that a balancing test provides the proper analytical framework for consideration of the defendant’s state constitutional claim. 2 Persuasive Federal and State Court Authority There is significant disagreement among both federal and state courts regarding the correct legal standard Page 62 CONNECTICUT LAW JOURNAL September 2, 2025

228 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

applicable to due process claims arising from prearrest delay. Most of the federal courts of appeals have inter- preted Marion and Lovasco to mean that, under the federal constitution, the defendant bears the burden of establishing both that (1) he suffered actual and substantial prejudice as a result of the delayed prosecu- tion, and (2) the delay was tactical, intended to harass, or otherwise a purposeful manifestation of bad faith on the part of the government. However, three federal courts of appeals—the fourth, seventh, and ninth cir- cuits—have instead adopted some form of multifactor balancing test, more akin to the Hodge test. See, e.g., United States v. Sowa, 34 F.3d 447, 451 (7th Cir. 1994), cert. denied, 513 U.S. 1117, 115 S. Ct. 915, 130 L. Ed. 2d 796 (1995); Howell v. Barker, 904 F.2d 889, 895 (4th Cir.), cert. denied, 498 U.S. 1016, 111 S. Ct. 590, 112 L. Ed. 2d 595 (1990); United States v. Moran, 759 F.2d 777, 782–83 (9th Cir. 1985), cert. denied, 474 U.S. 1102, 106 S. Ct. 885, 88 L. Ed. 2d 920 (1986). This persistent circuit split is noteworthy.4 Although different panels of these circuits have artic- ulated and applied their respective tests somewhat dif- 4 Whereas the party arguing for broader protections under the state consti- tution typically is confined to arguing that United States Supreme Court concurring and dissenting opinions are more persuasive than the relevant majority opinions, in this case, even those federal courts that adhere to the majority approach acknowledge that the controlling opinions—Marion and Lovasco—are ambiguous. See, e.g., United States v. Crouch, 84 F.3d 1497, 1510 (5th Cir. 1996) (‘‘[w]e recognize that neither Marion nor Lovasco is crystal clear on this issue, and each opinion contains some language that can give comfort to either view’’), cert. denied, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997), and cert. denied sub nom. Frye v. United States, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997); Stoner v. Graddick, 751 F.2d 1535, 1542 (11th Cir. 1985) (Lovasco ‘‘strongly hinted’’ that bad faith is necessary component of successful due process challenge). The high court has declined to resolve the circuit split, allowing the federal courts of appeals to adopt different interpretations and to apply conflicting standards for one-half century without ever repudiating the minority approach. See, e.g., Barker v. Howell, 498 U.S. 1016, 111 S. Ct. 590, 112 L. Ed. 2d 595 (1990) (denying petition for writ of certiorari). September 2, 2025 CONNECTICUT LAW JOURNAL Page 63

353 Conn. 169 SEPTEMBER, 2025 229 State v. McFarland

ferently, they all tend to depart from the majority approach in three important respects. First, these cir- cuits hold that, once a defendant has established that he has been actually and substantially prejudiced by a delayed prosecution,5 the burden shifts to the govern- ment to explain and justify the delay. See, e.g., United States v. Sowa, supra, 34 F.3d 451. The rationale for this rule is that whichever party is privy to the relevant information should bear the burden of proof. See, e.g., id.; see also State v. Swebilius, 325 Conn. 793, 807–808, 159 A.3d 1099 (2017) (requiring that each party ‘‘bring forth evidence uniquely within its knowledge’’ ‘‘allo- cates burdens efficiently’’ and ‘‘encourages diligence by law enforcement’’). The defendant is best positioned to establish whether, and to what extent, the delayed prosecution caused prejudice, but only law enforce- ment and prosecutors know the reason for the delay. See, e.g., United States v. Sowa, supra, 451 (‘‘Once the defendant has cleared the monumental hurdle of prov- ing prejudice, the government should explain its rea- sons. How else is the defendant to know why the government waited so long to indict him?’’). To require the defendant to establish why individual public offi- cials chose to delay arresting him ‘‘would violate funda- mental conceptions of justice, as well as the community’s sense of fair play.’’ Howell v. Barker, supra, 904 F.2d 895; see also United States v. Sabath, 990 F. Supp. 1007, 1018 (N.D. Ill. 1998) (‘‘The circuits that require a showing of intentional bad faith delay to gain a tactical advantage over the defendant seem to set an impossible threshold for criminal defendants. No defendant can 5 Demonstrating actual and substantial prejudice requires the defendant to establish that the delay has compromised his ability to mount a defense, and not merely that certain evidence was lost to him. See, e.g., United States v. McMutuary, 217 F.3d 477, 482 (7th Cir.) (death of alibi witness did not cause actual and substantial prejudice because ‘‘[h]er testimony would have been largely cumulative of’’ other witness’ testimony), cert. denied, 531 U.S. 1001, 121 S. Ct. 502, 148 L. Ed. 2d 471 (2000). Page 64 CONNECTICUT LAW JOURNAL September 2, 2025

230 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

get into the mind of a prosecutor to determine why a case was delayed . . . .’’). Second, for a prearrest delay to violate due process under the minority rule, it is not necessary for the gov- ernment to have delayed in making the arrest for tactical reasons or otherwise to have demonstrated bad faith. A lengthy delay may be deemed unjustified when it results from recklessness, or (in some jurisdictions) even negligence, on the part of public officials. See, e.g., Howell v. Barker, supra, 904 F.2d 895; United States v. Moran, supra, 759 F.2d 781.6 Of course, the government is not obliged to bring a prosecution as quickly as possible; see, e.g., United States v. Lovasco, supra, 431 U.S. 790–91; and there should be a ‘‘safe harbor’’ for public officials as they investigate a crime, wait for evidence to ripen, and build a sufficiently strong case to warrant a prosecution. United States v. Sample, 565 F. Supp. 1166, 1183 (E.D. Va. 1983); see also United States v. Sowa, supra, 34 F.3d 451 (‘‘if the cause of the delay is legitimately invest- igative in nature, a court will not find a due process violation’’). It is highly desirable for the government to proceed carefully and cautiously, and there should be no pressure to bring charges out of fear that any delay will be subject to judicial second-guessing. Within rea- sonable limits, our doctrine on prearrest delay should reflect both the value of prudent caution and the need to honor prosecutorial discretion in this context. 6 In some of its more recent articulations of the rule under the federal constitution, the United States Supreme Court itself has indicated that reck- lessness on the part of the government, and not only intentional bad faith, also can be sufficient to satisfy the second prong of the strict, two-pronged test. See, e.g., United States v. Eight Thousand Eight Hundred & Fifty Dollars ($8,850) in United States Currency, 461 U.S. 555, 563, 103 S. Ct. 2005, 76 L. Ed. 2d 143 (1983) (‘‘such claims can prevail only upon a showing that the [g]overnment delayed seeking an indictment in a deliberate attempt to gain an unfair tactical advantage over the defendant or in reckless disre- gard of its probable prejudicial impact’’(emphasis added)). September 2, 2025 CONNECTICUT LAW JOURNAL Page 65

353 Conn. 169 SEPTEMBER, 2025 231 State v. McFarland

Third, to resolve the ultimate question of whether a delayed prosecution has deprived the defendant of the opportunity for a fair trial, the fourth, seventh, and ninth circuits balance the nature and extent of the actual and substantial prejudice against the causes and justifica- tions proffered by the government for the delay. As this court did in Hodge, these federal courts, in conducting this balancing, consider all of the relevant facts and circumstances of the particular case. See, e.g., Howell v. Barker, supra, 904 F.2d 895; United States v. Valona, 834 F.2d 1334, 1339 (7th Cir. 1987); United States v. Mays, 549 F.2d 670, 678 (9th Cir. 1977). To assess whether the defendant has established actual and sub- stantial prejudice, courts examine factors such as the length of the delay, the nature and seriousness of the prejudice, the strength of the government’s case, and whether the prejudice could be and was mitigated. See, e.g., J. Caplan, Note, ‘‘Better Never Than Late: Pre- Arrest Delay as a Violation of Due Process,’’ 1978 Duke L.J. 1041, 1048–53. On the justification side of the scale, these courts consider the motive for the delay, its rea- sonableness, any role that the defendant played in caus- ing the delay, whether the claim was timely raised, and any extraordinary factors supporting prosecution after the passage of a long period of time. See, e.g., id., 1043–57. The federal courts using this approach often frame this balancing test as one in which the greater the preju- dice to the defendant, the higher the burden on the government to establish a valid reason for the delay, and vice versa. See, e.g., United States v. Mays, supra, 549 F.2d 678 (‘‘[t]he greater the length of the delay and the more substantial the actual prejudice to the defendant becomes, the greater the reasonableness and the necessity for the delay will have to be to balance out the prejudice’’); see also United States v. Sample, supra, 565 F. Supp. 1183. The ultimate determination Page 66 CONNECTICUT LAW JOURNAL September 2, 2025

232 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

often is characterized, as it was in Lovasco, in terms of whether a delayed prosecution ‘‘violates those funda- mental conceptions of justice [that] lie at the base of our civil and political institutions . . . and [that] define the community’s sense of fair play and decency . . . .’’ (Citations omitted; internal quotation marks omitted.) United States v. Lovasco, supra, 431 U.S. 790; accord Howell v. Barker, supra, 904 F.2d 895; see also United States v. Moran, supra, 759 F.2d 782. There likewise are variations among the state courts in their respective approaches to due process limita- tions on prearrest delay. Although a majority apply some version of the strict, two-pronged test, under which a defendant bears the burden of establishing both prejudice and purposeful, ill-intentioned delay, more than one third of the states instead use a multifactor or balancing test under either the federal due process clauses or their respective state counterparts. See, e.g., E. DuBosar, Note, ‘‘Pre-Accusation Delay: An Issue Ripe for Adjudication by the United States Supreme Court,’’ 40 Fla. St. U. L. Rev. 659, 671–85 (2013). Despite the existence of some nuanced distinctions, the states that employ the balancing test all agree with the minority federal approach that (1) the state carries the burden of demonstrating the reasons for the delay because it is best positioned to explain and justify long delays in charging the defendant, (2) the prejudicial consequences of delay on the defense can render a criminal prosecu- tion fundamentally unjust if the state waits decades to initiate proceedings without any valid reason, and not only when the delay is tactical or malicious, and (3) the ultimate determination of whether a due process violation has occurred must involve consideration of all of the relevant circumstances surrounding the delay— including its causes and consequences—and weighing the state’s justifications for the delay against the preju- dice suffered by the defendant. See, e.g., People v. Nel- September 2, 2025 CONNECTICUT LAW JOURNAL Page 67

353 Conn. 169 SEPTEMBER, 2025 233 State v. McFarland

son, 43 Cal. 4th 1242, 1255, 185 P.3d 49, 78 Cal. Rptr. 3d 69 (‘‘The ultimate inquiry in determining a claim based [on] due process is whether the defendant will be denied a fair trial. If such deprivation results from unjustified delay by the prosecution coupled with preju- dice, it makes no difference whether [or not] the delay was deliberately designed to disadvantage the defen- dant . . . . [In either case], the defendant will be denied his right to a fair trial as a result of governmental conduct.’’), cert. denied, 555 U.S. 926, 129 S. Ct. 357, 172 L. Ed. 2d 219 (2008). The overwhelming majority of the courts that follow the stringent, two-pronged approach provide no mean- ingful explanation at all and offer no supporting analysis grounded in the language, history, or purpose of the federal due process clauses. They merely purport to apply what they take to be the better reading of the high court’s brief discussion of the issue in Marion and Lovasco. By contrast, those jurisdictions that use a totality of the circumstances or balancing test have explained at some length why applying the two-pronged test makes little sense. See, e.g., State v. Purcell, 331 Conn. 318, 348–51, 203 A.3d 542 (2019) (affording little weight to majority of state courts that simply endorsed majority opinion of United States Supreme Court with- out independently assessing its persuasiveness). Accordingly, these factors also counsel in favor of applying the Hodge test in the present case. 3 Constitutional Text and History Geisler also instructs that we examine the text and history of the relevant constitutional provisions. In the present case, these factors are consistent with, but pro- vide limited direct support for, a broader reading of the state constitution. Page 68 CONNECTICUT LAW JOURNAL September 2, 2025

234 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

The state constitution contains two due process clauses. Article first, § 8, of the Connecticut constitution provides in relevant part that ‘‘[n]o person shall . . . be deprived of life, liberty or property without due pro- cess of law . . . .’’ That language closely tracks the language of the due process clauses contained in the fifth and fourteenth amendments to the federal constitu- tion and, thus, does not provide any independent sup- port for the defendant’s position. At the same time, we have long recognized that article first, § 8, has a distinct meaning that, in various contexts, operates to confer broader due process protections than does the four- teenth amendment.7 ‘‘[W]e have long abandoned the notion that our state due process clause . . . has the same meaning and imposes the same limitations as its federal counterpart.’’ (Citation omitted; footnote omit- ted.) State v. Morales, 232 Conn. 707, 717, 657 A.2d 585 (1995). Article first, § 9, by contrast, has no direct federal counterpart. See W. Horton, The Connecticut State Con- stitution (2d Ed. 2012) p. 75. It provides: ‘‘No person shall be arrested, detained or punished, except in cases clearly warranted by law.’’ Conn. Const., art. I, § 9. It is well established that this provision confers distinct rights on the citizens of Connecticut. See, e.g., State v. Santiago, supra, 318 Conn. 14–15 (prohibiting capital punishment following prospective abolishment by legis- 7 See, e.g., State v. Purcell, supra, 331 Conn. 321 (police officers are required to clarify ambiguous request for counsel before continuing custodial interrogation); State v. Harris, 330 Conn. 91, 115, 191 A.3d 119 (2018) (adopt- ing heightened standard and burden shifting framework for reliability of eyewitness identification under due process clause of state constitution); State v. Morales, supra, 232 Conn. 726–27 (rejecting ‘‘the litmus test of bad faith on the part of the police,’’ which United States Supreme Court had adopted under federal constitution, to determine whether state’s failure to preserve evidence violates due process protections under state constitution); State v. Stoddard, 206 Conn. 157, 166, 537 A.2d 446 (1988) (‘‘a suspect must be informed promptly of timely efforts by counsel to render pertinent legal assistance’’). September 2, 2025 CONNECTICUT LAW JOURNAL Page 69

353 Conn. 169 SEPTEMBER, 2025 235 State v. McFarland

lature); see also State v. Boyd, 221 Conn. 685, 689–90, 607 A.2d 376, cert. denied, 506 U.S. 923, 113 S. Ct. 344, 121 L. Ed. 2d 259 (1992). But, again, nothing in the language of the provision directly addresses the issue of prearrest delays. I consider it relevant that the balancing test that the trial court applied in the present case is far from a novel interpretation of the ‘‘due process of law’’ language of article first, § 8. See, e.g., State v. Morales, supra, 232 Conn. 721 (citing Mathews v. Eldridge, 424 U.S. 319, 334–35, 96 S. Ct. 893, 47 L. Ed. 2d 18 (1976), under which procedural due process violations are subject to a test that balances (1) private interest at issue, (2) risk of erroneous deprivation and probable value of additional safeguards, and (3) government’s interests, including administrative and financial burdens of addi- tional safeguards). ‘‘Indeed,’’ as this court explained in Morales, ‘‘we frequently have rejected a litmus test that [requires the satisfaction of one particular element, such as bad faith], and instead have required courts to employ a balancing test, when ruling on issues that concern due process or fundamental fairness.’’ Id., 720. With respect to the historical analysis under Geisler, there is little guidance to be found, one way or the other, in our state’s preconstitutional practice and tradi- tions. The parties have not identified any instances of lengthy prearrest delays in colonial Connecticut, and the scenario presented by this case—a decades old cold case that suddenly came alive again in 2018 due to increasingly sensitive DNA technologies—is of dis- tinctly modern origin. That said, the underlying princi- ples are anything but new. The concept that criminal justice ought not be unnecessarily delayed traces to the Magna Carta, a charter of liberty that indelibly influ- enced our own notions of due process. See, e.g., State v. Pace, 129 Conn. 570, 572, 29 A.2d 755 (1943) (‘‘[t]he right of one accused of crime to have a fair and impartial Page 70 CONNECTICUT LAW JOURNAL September 2, 2025

236 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

trial has been the basis of Anglo-Saxon criminal jurisdic- tion ever since [the] Magna Carta’’). As one English jurist long ago expressed the unfairness associated with undue delay, ‘‘[i]t is monstrous to put a man on his trial after such a lapse of time. How can he account for his conduct so far back? . . . No man’s life would be safe if such a prosecution were permitted. It would be very unjust to put him on his trial.’’ Regina v. Robins, 1 Cox Crim. Cas. 114 (Somerset Winter Assizes 1844). Connecticut’s earliest legal codes likewise embraced the principle that justice should be delivered ‘‘without partiality or delay.’’ (Internal quotation marks omitted.) C. Collier, ‘‘The Common Law and Individual Rights in Connecticut Before the Federal Bill of Rights,’’ 76 Conn. B.J. 1, 12 (2002); see id., 12 n.23 (citing ‘‘the Code of 1672’’ of Connecticut Colony). And there is no doubt that, in the decade leading up to the adoption of our present constitution in 1965, prior to Marion and Lovasco, the prevailing approach to the issue, as in Hodge, was to consider and balance various factors, including the length of the delay, the reasons for the delay, and the prejudice to the defendant engendered by the delay. See, e.g., Ross v. United States, supra, 349 F.2d 211; Taylor v. United States, 238 F.2d 259, 260–62 (D.C. Cir. 1956); Territory v. Shito, 43 Haw. 203, 205– 206 (1959). 4 Relevant Public Policies I have explained why I find more persuasive the rea- soning of those courts that apply a balancing test to guide the due process analysis. In this part, I address the state’s primary arguments against adopting a balanc- ing test. One worry articulated by the state is that the balanc- ing test would allow perpetrators to go unpunished if they are fortunate enough to face a long delayed September 2, 2025 CONNECTICUT LAW JOURNAL Page 71

353 Conn. 169 SEPTEMBER, 2025 237 State v. McFarland

prosecution resulting from unexplained causes. That concern, while understandable, overlooks the demanding threshold prejudice requirement and has not been borne out in practice. In fact, the courts that apply the balanc- ing approach consistently emphasize both how difficult it is for a defendant to establish a due process violation and how rarely prosecutions are dismissed on this basis. See, e.g., United States v. Hagler, 700 F.3d 1091, 1099 (7th Cir. 2012) (‘‘[t]his burden is an exacting one; the showing must rest [on] more than mere speculative harm . . . and [the defendant] must present facts that are specific, concrete, and supported by evidence’’ (cita- tion omitted; internal quotation marks omitted)); United States v. Barken, 412 F.3d 1131, 1134 (9th Cir. 2005) (‘‘An indictment is rarely dismissed because delay by the prosecution rises to the level of a [f]ifth [a]mend- ment due process violation. . . . The defendant’s bur- den [of] show[ing] actual prejudice is heavy and is rarely met.’’ (Citation omitted.)); United States v. Suchecki, Docket No. 91-10127, 1993 WL 188368, *2 (9th Cir. June 2, 1993) (decision without published opinion, 995 F.2d 234) (‘‘These requirements are stringent and stringently enforced . . . . A recent Ninth Circuit decision found only two cases in the . . . seventeen years [between 1975 and 1992] in which any [federal] court of appeals upheld a [preindictment] delay due process claim.’’ (Citations omitted; emphasis in original.)); Pharm v. Hatcher, 984 F.2d 783, 786 (7th Cir.) (‘‘[i]n fact, we have never found [preaccusation] delay rising to the level of a constitutional violation’’), cert. denied, 510 U.S. 841, 114 S. Ct. 125, 126 L. Ed. 2d 89 (1993); United States v. Sample, supra, 565 F. Supp. 1175 (‘‘the defendant [usually] has difficulty sustaining the burden’’). The present matter is a case in point. With thirty-two years having elapsed between crime and arrest, this case involved a lengthy delay during which two third- party culpability witnesses had died, one of whom had Page 72 CONNECTICUT LAW JOURNAL September 2, 2025

238 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

specifically and repeatedly implicated different perpe- trators in the crimes, memories of an important alibi witness had faded, and key forensic evidence had degraded. And, yet, as I will explain, the length of the delay and the prejudice to the defendant, when bal- anced against the reasons for the delay, fall short of a due process violation. The test this court adopts today is not an easy one to satisfy. A second concern is that, if we put the onus on the state to justify its failure to expeditiously investigate and prosecute a crime, and we then scrutinize not only intentional but also wholly unjustifiable delays involv- ing something less than bad faith, we will place courts in the uncomfortable and unwelcome position of having to assess and, potentially, second-guess discretionary decisions by the police and prosecutors. In this case, for example, there may be a perfectly good reason why police officers did not follow up with witnesses who claimed to have seen the victims alive the weekend after the defendant was incarcerated, or perhaps they did, but records were lost or never made. And it is unreasonable, the argument goes, to expect the police to have documented those decisions or to anticipate having to account for them thirty-two years later, in 2022. Once again, however, while these concerns are under- standable, one-half century of practical experience has largely failed to bear them out. There is no indication that following the minority approach has hindered law enforcement or resulted in practical difficulties in the many jurisdictions that have adopted it. See, e.g., United States v. Sowa, supra, 34 F.3d 451 (‘‘this is not a heavy burden for the government to bear; [the] standard prac- tice in responding to motions to dismiss for [preindict- ment] delay is to explain the delay’’). If anything, the opposite has been true. See, e.g., D. Rang, Comment, ‘‘The Waiting Game: How Preindictment Delay Threat- September 2, 2025 CONNECTICUT LAW JOURNAL Page 73

353 Conn. 169 SEPTEMBER, 2025 239 State v. McFarland

ens Due Process and Fair Trials,’’ 66 S.D. L. Rev. 143, 166 (2021) (‘‘as time has moved on, many states have begun to recognize the problems associated with [the strict, two-pronged] test’’). The great majority of delays are investigative in nature, especially with respect to crimes for which there are no statute of limitations, and most jurisdictions recognize a ‘‘safe harbor’’ for credible investigative delays. United States v. Sample, supra, 565 F. Supp. 1183. Indeed, some of the jurisdictions that interpret the due process clause to require a balancing approach do not even count periods of investigative activity when assessing the existence and length of a delay. See, e.g., United States v. Sabath, supra, 990 F. Supp. 1019; People v. Grant, 82 Misc. 3d 991, 1000, 204 N.Y.S.3d 739 (2024). As a practical matter, when state officials believe that they have enough evidence to prosecute a serious crime, they typically proceed to do so expeditiously. When they do not initiate a prosecution, there typically is a good reason to wait. It is rare for the state simply to sit on a case for an extended period of time for no good reason. The relatively few cases nationwide in which charges have been dismissed or convictions reversed due to excessive prearrest delay have not involved courts nitpicking discretionary decisions by the police or pros- ecutors; rather, the lack of any justification for the delay has been palpable. A third argument the state offers for preferring a stricter rule is that the balancing approach infringes on the prerogative of the legislature. The theory is that, when the legislature enacts a statute of limitations for a particular crime, or decides that no statute of limita- tions is appropriate, it is, in effect, conducting a balanc- ing of interests, weighing the seriousness of the crime, the value of repose, society’s interest in seeing justice done after many years have passed, and the interests of a wrongly accused individual in having access to Page 74 CONNECTICUT LAW JOURNAL September 2, 2025

240 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

sufficiently fresh evidence to be able to mount a defense. Courts should not interfere in this legislative policymaking. This argument overlooks the fundamental purpose of judicial review, which is to enforce constitutional limitations on legislative enactments and on the execu- tive enforcement of those enactments. See, e.g., L. Tribe, American Constitutional Law (2d Ed. 1988) §§ 3-3 and 3-4, pp. 29–33 (judicial review extends to executive practice and enforcement, as well as to legislation). For example, with respect to our state constitutional protections against cruel and unusual punishment, which likewise reside within our dual due process clauses, we rejected the argument that we were com- pelled to accept ‘‘the judgment of the legislature—that the death penalty comports with contemporary stan- dards of decency and serves legitimate penological interests—and that to do otherwise [was] to usurp the proper role of the legislature . . . .’’ State v. Santiago, supra, 318 Conn. 133. We reasoned that the argument had long been rejected by this court because it ‘‘funda- mentally misunderstands the well established function and role of judicial review . . . .’’ Id. ‘‘If the fact that an elected legislature had authorized and enacted the punishment in question were enough to insulate it from judicial scrutiny,’’ we explained, then there would be no purpose for a constitutional provision barring the imposition of cruel and unusual punishment. Id., 135. Rather, our own review is both independent in spirit and distinct in nature from the legislative process by which politicians select punishments. See id. 5 Summary To summarize, an analysis of the relevant Geisler factors persuades me that §§ 8 and 9 of article first of the state constitution afford broader protections from September 2, 2025 CONNECTICUT LAW JOURNAL Page 75

353 Conn. 169 SEPTEMBER, 2025 241 State v. McFarland

oppressive prearrest delays than the strict, two-pronged approach that we have held is required under the United States constitution. Under this test, the defendant must first establish actual and substantial prejudice. Relevant factors include the length of the delay, the nature and seriousness of the prejudice, including whether the defendant was substantially prevented from raising or proving a meaningful defense, the strength of the state’s case, and whether any prejudice could be and was miti- gated. If the defendant establishes actual and substan- tial prejudice, the burden then shifts to the state to establish the reasons for any excessive delay. Finally, the court must balance the prejudice to the defendant against the justifications for the delay to determine whether the defendant has been deprived of a fair trial. B Application of the Balancing Test The remaining task is to apply this balancing test to the facts of the present case. Upon careful review of the totality of the circumstances, I agree with the trial court’s determination that the prosecution did not vio- late the defendant’s due process right to a fair trial under the Connecticut constitution. With respect to prejudice, the trial court assumed, for the sake of argument, that the defendant was preju- diced by the thirty-two year delay. Many courts, even those that apply a stricter due process standard, hold that actual and substantial prejudice can be established when a defendant demonstrates that a deceased witness would have testified in support of an alibi or third-party culpability defense that is not otherwise available, and there is at least a reasonable possibility that the jury would have credited that testimony. See, e.g., United States v. Rogers, 118 F.3d 466, 475 (6th Cir. 1997) (com- paring cases); cf. United States v. McMutuary, 217 F.3d 477, 482 (7th Cir.) (death of alibi witness did not cause Page 76 CONNECTICUT LAW JOURNAL September 2, 2025

242 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

actual and substantial prejudice because ‘‘[h]er testi- mony would have been largely cumulative of’’ other witness’ testimony), cert. denied, 531 U.S. 1001, 121 S. Ct. 502, 148 L. Ed. 2d 471 (2000). The present case is relatively rare in that we have the sworn statements of a deceased witness, Veronica Saars-Doyle, indicating firsthand knowledge that the murders were committed by Lee Copeland. Although we do not dispute the trial court’s determination that there were serious problems with her testimony,8 the defendant’s own confessions suffer from similar, if less extreme, defects, and Saars-Doyle’s testimony could have been bolstered by that of Bruce Hankins, another deceased witness. In addition, at trial, the defendant’s primary alibi witness, Louise Salvati, had no present memory of the events in question and could not confirm her 1987 statement to the police that she had seen Gregory Harris alive on August 22, 1987, after the defen- dant was incarcerated. Potentially important forensic evidence also had been discarded or degraded over the years, further hampering the defendant’s effort to develop a third-party culpability defense. The police apparently did not preserve the bloody shirt that the defendant was wearing when he was arrested on charges of sexual assault on August 22, 1987, and the match between the crime scene DNA and Copeland’s 2019 DNA sample was deemed ‘‘inconclusive’’—which could have meant that a person with his DNA profile was as much as 1000 times more likely to be implicated than an unknown individual, leaving the parties with no way to know whether a more accurate estimate would have been obtained when the evidence was fresher. Upon closer inspection, however, it must also be observed that this showing of prejudice contained sig- nificant weaknesses, particularly with respect to what 8 See part III of the per curiam opinion. September 2, 2025 CONNECTICUT LAW JOURNAL Page 77

353 Conn. 169 SEPTEMBER, 2025 243 State v. McFarland

would appear to be the defendant’s strongest claim, namely, the death of one or more witnesses who poten- tially would have testified in support of a third-party culpability defense. The evidence of prejudice suffered from three major deficiencies. First, it is well established that, for a prearrest delay to offend due process, the delay must be causally related to the prejudice. With respect to deceased wit- nesses, this means that the defendant must establish that the witness died during or after the period of delay, so that, but for the delay, the witness likely would have been available to testify. See, e.g., United States v. Crouch, 84 F.3d 1497, 1518 (5th Cir. 1996), cert. denied, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997), and cert. denied sub nom. Frye v. United States, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997); United States v. Cheung Kin Ping, 555 F.2d 1069, 1072–73 (2d Cir. 1977). In this case, the record does not reveal how long Saars-Doyle and Hankins lived after they gave their last police statements, respectively, in 1992 and 1993. It was the defendant’s burden to estab- lish these facts, and his failure to do so prevents him from establishing that the delay in prosecution probably caused the requisite actual and substantial prejudice to his defense. The earliest the state plausibly could have prosecuted the defendant was after he confessed to the murders in 1996; on this record, it is not known if Saars- Doyle and Hankins had died by that time. Second, to establish a viable third-party culpability defense, the defendant would have needed to prove not only that Copeland participated in the commission of the murders, but also that Copeland’s involvement meant that the defendant was not involved. See, e.g., United States v. Garcia, 74 F.4th 1073, 1101–1102 (10th Cir. 2023) (there was no prejudice when defendant and third party could have jointly committed crime), cert. denied sub nom. Troup v. United States, U.S. , Page 78 CONNECTICUT LAW JOURNAL September 2, 2025

244 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

144 S. Ct. 400, 217 L. Ed. 2d 216 (2023), and cert. denied sub nom. Gallegos v. United States, U.S. , 144 S. Ct. 608, 217 L. Ed. 2d 324 (2024); see also State v. Berger, 249 Conn. 218, 236 n.5, 733 A.2d 156 (1999) (third-party culpability doctrine requires acquittal only if jury deter- mines that third party alone was responsible for crime). In this case, Saars-Doyle’s primary value to the defense, as a potential third-party culpability witness, would have been diminished by the fact that her testimony was not necessarily inconsistent with the defendant’s guilt. In Saars-Doyle’s statement to the police on July 23, 1990, she indicated that, in addition to Copeland, Hankins, and herself, a fourth individual, identified only as ‘‘Gus,’’ was involved in the crimes. She described Gus as a short Black man9 who participated in beating and stabbing Fred Harris. Likewise, in his earlier confes- sions to the police, the defendant indicated that as many as five other individuals assisted him in the murders. The jury may well have found these early statements more credible than the defendant’s later statements that he committed the crimes on his own, which would have required him to single-handedly cut up various telephone wires and use them to bind two adult men hand and foot, all while simultaneously holding both men at bay with a steak knife. Accordingly, it is difficult to conclude that Saars-Doyle’s unavailability denied the defendant a fair trial because, even if the jury were inclined to believe Saars-Doyle’s inconsistent and implausible statements regarding Copeland’s culpabil- ity, it still could have found that the defendant collabo- rated with Copeland in committing the murders. Third, the strength of the state’s case, although not overwhelming, does not favor the defendant’s claim of actual and substantial prejudice. In most of the cases in which courts have found that a prearrest delay 9 The arrest report identifies the defendant as a five foot, six inch Black male. September 2, 2025 CONNECTICUT LAW JOURNAL Page 79

353 Conn. 169 SEPTEMBER, 2025 245 State v. McFarland

reached the level of a constitutional violation, the case against the defendant was weak and/or circumstantial, and the defendant had a potentially strong alibi or justi- fication defense that was actually and substantially compromised by the delay. In the present case, by con- trast, a DNA match placed the defendant at the murder scene, and he reached out repeatedly to the police to confess to the crimes, volunteering specific details that had not been made public. To believe that he had no involvement in the crimes, one would have to believe that (1) the defendant learned of those crime scene details by some other means while incarcerated, (2) he chose to use those details to falsely inculpate himself rather than to implicate either the actual perpetrators or some innocent third parties, and (3) the DNA match obtained in 2019 was a false positive or his DNA appeared in the glove by some innocuous means. For all of those propositions to be true seems extraordinarily unlikely. The exculpatory statements of Saars-Doyle, on the other hand, contain multiple, significant flaws that render her account insufficient to call the defen- dant’s guilt into doubt. See part III of the per curiam opinion.

We are left at the end of the prejudice analysis with an extremely long delay, Salvati’s lost memory, which precluded her from fleshing out her police report in a way that the jury might have found more compelling, degradation of key DNA evidence, and the loss of other evidence and the fading of other memories with the passage of time. It is a close case, in my view, whether these circumstances suffice to establish actual and sub- stantial prejudice on this record. Like the trial court, however, I will assume, without deciding, that, on this record, there was sufficient prejudice to reach the sec- ond step of the analysis and to consider the state’s proffered reasons for the delay. Page 80 CONNECTICUT LAW JOURNAL September 2, 2025

246 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

I agree that the state satisfied its burden of establish- ing that the lengthy delay in prosecuting the defendant was fully justified and, on balance, defeated the defen- dant’s claim of an unfair trial. The defendant concedes that the state did not act out of bad faith or with other impermissible motives. Nor is there any indication that the delay resulted from the state’s reckless disregard of the defendant’s rights. Rather, the trial court found that the state reasonably could have determined that it lacked sufficient evidence to establish the defendant’s guilt beyond a reasonable doubt until newly developed DNA technologies linked him to the crimes in 2019, substantiating his earlier confessions. I therefore can attribute no fault to the state for causing the delay. See, e.g., People v. Nelson, supra, 43 Cal. 4th 1256 (conclud- ing that lengthy delay in bringing murder charges had ‘‘strong’’ justification and balanced out prejudice to defendant, even though he was identified as suspect shortly after murder, because case was ‘‘not fully solve[d]’’ until comparison of defendant’s DNA to crime scene evidence twenty-six years later).

The defendant appears to claim that, at the very least, the state acted negligently, such as by failing to timely collect DNA samples from Hankins and Copeland, and by not keeping the yellow work glove, from which the DNA sample was obtained, in climate-controlled stor- age until 2009. These are not particularly strong claims, but, even if I were to assume for the sake of argument that some conduct of the state contributing to the delay was wholly unjustifiable, I cannot conclude, on balance, that the delay violated the defendant’s due process rights. The state’s delay was largely, if not exclusively, investigative in nature. The more sophisticated DNA technology necessary to identify the defendant as the perpetrator of the murders simply was unavailable for more than thirty years after the crimes. September 2, 2025 CONNECTICUT LAW JOURNAL Page 81

353 Conn. 169 SEPTEMBER, 2025 247 State v. McFarland

In light of the weaknesses in the defendant’s proof of actual and substantial prejudice when balanced against the state’s legitimate justification for the delay in his prosecution, I conclude under all of the circumstances that he has failed to establish that his conviction offended ‘‘the community’s sense of fair play and decency . . . .’’ (Citation omitted; internal quotation marks omitted.) United States v. Lovasco, supra. 431 U.S. 790. II SAARS-DOYLE’S HEARSAY STATEMENTS I agree with the court’s determination that the trial court did not abuse its discretion when it excluded from evidence the hearsay statements of the deceased witness Saars-Doyle; see part III C of the per curiam opinion; but with a caveat stemming from my agreement with the defendant’s contention that the prearrest delay and hearsay issues involving Saars-Doyle’s statements are not wholly disconnected. Although the trial court in the present case properly exercised its discretion to exclude the Saars-Doyle hearsay statements, trial courts faced with colorable due process concerns in this con- text should consider whether it is possible to exercise their discretion to admit evidence that would help miti- gate the prejudice caused by the delay. See, e.g., United States v. Sabath, supra, 990 F. Supp. 1010 (‘‘[the trial] [c]ourt attempted to explore the possibility of curing the prejudice from [the witness’] death by admitting his statements under the [catchall] hearsay exception’’); see also United States v. Barken, supra, 412 F.3d 1135 (loss of documents due to prearrest delay did not cause prejudice when other available evidence served as ‘‘ade- quate substitutes’’); Naposki v. Adams, Docket No. SA CV 16-00391 VBF (AFM), 2017 WL 907606, *9 (C.D. Cal. January 30, 2017) (‘‘[the] defense received the benefit of the trial court’s ruling permitting the defense to present [certain witnesses’] testimony, which otherwise would Page 82 CONNECTICUT LAW JOURNAL September 2, 2025

248 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

have been largely inadmissible hearsay, as a remedy for the [preindictment] delay’’); United States v. Santi- ago, 987 F. Supp. 2d 465, 486 (S.D.N.Y. 2013) (‘‘[t]o ameliorate the impact of [the witness’] unavailability, the [g]overnment . . . offered the defense the option of introducing all of [the witness’] statements—as long as all of them come in, in their entirety’’); Phillips v. Yates, Docket No. 2:10-cv-1368 KJM JFM (HC), 2012 WL 2995675, *11 (E.D. Cal. July 23, 2012) (‘‘the prejudice caused by the loss of . . . [the] witness was mitigated by the parties’ stipulation that [the] defendant’s previ- ous attorney would be permitted to testify regarding the [witness’] impeaching statements’’); People v. Wessels, Docket No. D058021, 2012 WL 1293803, *9 (Cal. App. April 17, 2012) (‘‘the prejudice to [the defendant] was mitigated because [the detective’s notes memorializing his interview of a forgetful witness] were recorded and used to refresh [the detective’s] recollection in many instances’’), review denied, California Supreme Court, Docket No. S202655 (July 11, 2012). To be clear, unless it is appropriate to do so as a remedial measure fashioned to cure prejudice other- wise amounting to a violation of due process, I do not suggest that plainly inadmissible evidence should none- theless be presented to the jury as a form of ‘‘rough justice.’’ My point is that when the evidence falls in a gray area of admissibility, one consideration that the trial court should include as part of its exercise of discretion to admit or exclude that evidence is whether, in a case in which a defendant can show prejudice resulting from prearrest delay, the interests of justice militate in favor of allowing the jury to decide for itself whether the proffered evidence, in combination with the other evidence, is sufficiently reliable to factor in its decisionmaking. After all, if Saars-Doyle had not died during the thirty-two year prearrest delay, the jury presumably would have heard her testimony, warts and September 2, 2025 CONNECTICUT LAW JOURNAL Page 83

353 Conn. 169 SEPTEMBER, 2025 249 State v. McFarland

all, and decided for itself to give her story whatever weight it deserved. As it happened, through no fault of the defendant, Saars-Doyle was dead by the time of trial. As a result of that fact and the trial court’s decision to exclude her statements, the jury never learned from any source that there was a witness claiming firsthand knowledge who told the police that a different person, not the defendant, had committed the murders. The due process clause did not compel a different result as a constitutional matter in this case, but a trial court operating within the scope of its discretion in making evidentiary rulings is free to take into account sub- constitutional considerations of fairness implicated when a defendant is prejudiced by the effects of a prear- rest delay.

ALEXANDER, J., with whom MULLINS, C. J., and DANNEHY, J., join, concurring. I concur in the opinion of the court affirming the conviction of the defendant, Willie McFarland, of two counts of murder in violation of General Statues § 53a-54a (a) in connection with his commission of a double homicide in 1987. I write separately with respect to the prearrest delay issue addressed in part I of the court’s opinion, which consid- ers whether the thirty-two year interval between the murders and the defendant’s arrest violated his due process rights under, among other provisions, article first, §§ 8 and 9, of the Connecticut constitution. In analyzing such prearrest delay claims under our state constitution, I reject the interpretation of the United States Supreme Court’s decisions in United States v. Lovasco, 431 U.S. 783, 795–96, 97 S. Ct. 2044, 52 L. Ed. 2d 752 (1977), and United States v. Marion, 404 U.S. 307, 324, 92 S. Ct. 455, 30 L. Ed. 2d 468 (1971), followed by a majority of the United States courts of appeals and other states. See, e.g., United States v. Crouch, 84 F.3d 1497, 1511–12 (5th Cir. 1996) (citing cases), cert. Page 84 CONNECTICUT LAW JOURNAL September 2, 2025

250 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

denied, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997), and cert. denied sub nom. Frye v. United States, 519 U.S. 1076, 117 S. Ct. 736, 136 L. Ed. 2d 676 (1997). Those jurisdictions have adopted a two-pronged test that requires ‘‘the defendant [to] show both that actual substantial prejudice resulted from the delay and that the reasons for the delay were wholly unjustifiable, as [when] the state seeks to gain a tactical advantage over the defendant.’’ (Emphasis in original; internal quota- tion marks omitted.) State v. Roger B., 297 Conn. 607, 614, 999 A.2d 752 (2010). Instead, like Justice Ecker, I would address due pro- cess challenges to prearrest delay brought under the Connecticut constitution through a balancing test that is consistent with the minority view followed by the United States Courts of Appeals for the Fourth, Seventh and Ninth Circuits. See, e.g., United States v. Sowa, 34 F.3d 447, 451 (7th Cir. 1994), cert. denied, 513 U.S. 1117, 115 S. Ct. 915, 130 L. Ed. 2d 796 (1995); Howell v. Barker, 904 F.2d 889, 895 (4th Cir.), cert. denied, 498 U.S. 1016, 111 S. Ct. 590, 112 L. Ed. 2d 595 (1990); United States v. Moran, 759 F.2d 777, 782–83 (9th Cir. 1985), cert. denied, 474 U.S. 1102, 106 S. Ct. 885, 88 L. Ed. 2d 920 (1986). Under this balancing test, the defendant must first establish actual and substantial prejudice, after which the burden shifts to the state to produce evidence of the reasons for the delay. See, e.g., United States v. Sowa, supra, 451. ‘‘[T]he court must [then] balance the defendant’s prejudice against the government’s justifi- cation for delay. . . . The basic inquiry then becomes whether the government’s action in prosecuting after substantial delay violates fundamental conceptions of justice or the community’s sense of fair play and decency.’’ (Citation omitted; internal quotation marks omitted.) Howell v. Barker, supra, 895. In my view, this allocation of the burden of proof is more equitable than that under the two-pronged test because it better reflects the evi- September 2, 2025 CONNECTICUT LAW JOURNAL Page 85

353 Conn. 169 SEPTEMBER, 2025 251 State v. McFarland

dence available to the defendant and the state. Assum- ing without deciding that the defendant has established actual and substantial prejudice in this case, I conclude that any prejudice was far outweighed by the legitimate investigative justifications for the delay. Accordingly, I conclude that the thirty-two year interval between the murders and the defendant’s arrest did not violate his due process rights under the Connecticut constitution. My rejection of the two-pronged test in favor of a balancing test under our state constitution is driven primarily by an analysis of three interrelated factors set forth in State v. Geisler, 222 Conn. 672, 684–86, 610 A.2d 1225 (1992).1 I find most persuasive case law from Connecticut, precedents of other state courts, and pub- lic policy considerations, and concentrate my analysis on them. See, e.g., State v. Jose A. B., 342 Conn. 489, 508, 270 A.3d 656 (2022) (court may focus on most persuasive Geisler factors). First, a prior decision of this court strongly supports the conclusion that a balancing test is required to address prearrest delays, namely, this court’s interpre- 1 In Geisler, ‘‘we enumerated the following six factors to be considered in construing the state constitution: (1) persuasive relevant federal prece- dents; (2) the text of the operative constitutional provisions; (3) historical insights into the intent of our constitutional forebears; (4) related Connecti- cut precedents; (5) persuasive precedents of other state courts; and (6) contemporary understandings of applicable economic and sociological norms, or as otherwise described, relevant public policies. . . . ‘‘The Geisler factors serve a dual purpose: they encourage the raising of state constitutional issues in a manner to which the opposing party . . . can respond; and they encourage a principled development of our state constitutional jurisprudence. Although in Geisler we compartmentalized the factors that should be considered in order to stress that a systematic analysis is required, we recognize that they may be inextricably interwoven. . . . [N]ot every Geisler factor is relevant in all cases. . . . Moreover, a proper Geisler analysis does not require us simply to tally and follow the decisions favoring one party’s state constitutional claim; a deeper review of those decisions’ underpinnings is required because we follow only persuasive decisions.’’ (Internal quotation marks omitted.) In re Ivory W., 342 Conn. 692, 719–20, 271 A.3d 633 (2022). Page 86 CONNECTICUT LAW JOURNAL September 2, 2025

252 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

tation of article first, § 7, of the Connecticut constitution in State v. Hodge, 153 Conn. 564, 567, 219 A.2d 367 (1966).2 Although the claim in Hodge was framed as a challenge to the reasonableness of the seizure of the defendant—who had been arrested on narcotics charges three weeks after the commission of the crimes—rather than a due process claim, this court’s analysis remains instructive. See id., 566–67. Indeed, in Hodge, this court cited federal due process decisions, including United States v. Holiday, 319 F.2d 775, 776 (2d Cir. 1963), and Ross v. United States, 349 F.2d 210, 213 (D.C. Cir. 1965), in concluding that, ‘‘[when] the delay in arresting a defendant (or in otherwise apprising him of the charges against him) continues long after all the evidence has been assembled, and becomes a product of mere conve- nience to the state, a question of an unreasonable sei- zure or lack of a fair trial may arise.’’ State v. Hodge, supra, 567–68. This court held that ‘‘[t]he defendant’s rights . . . must necessarily depend on all the circum- stances, including the length of the delay, the reason for the delay, prejudice to the defendant, and a timely presentation of the claim to the trial court.’’ Id., 568. We indicated that a defendant must show ‘‘[s]ome preju- dice’’ to his defense, such as ‘‘the unavailability of alibi witnesses or the impaired memory of the defendant and others who vouch for his innocence.’’ Id. Ultimately, we applied a balancing test to reject this claim because 2 Hodge was decided well before Geisler, which articulated this court’s now well established approach to state constitutional interpretation. See State v. Geisler, supra, 222 Conn. 684–86. Although pre-Geisler decisions may have great precedential significance; see, e.g., State v. Marsala, 216 Conn. 150, 159–61, 579 A.2d 58 (1990); State v. Stoddard, 206 Conn. 157, 164, 537 A.2d 446 (1988); I recognize that we have conducted a somewhat more searching review under Geisler when presented with a state constitu- tional issue that has been previously mentioned only in passing or without the benefit of a comprehensive, independent analysis. See, e.g., State v. Patel, 327 Conn. 932, 939–40, 171 A.3d 1037 (2017); see also State v. Haynes, 352 Conn. 236, 244–45, 336 A.3d 1139 (2025) (describing relationship between Geisler factors and doctrine of stare decisis with respect to pre-Geisler state constitutional decisions). September 2, 2025 CONNECTICUT LAW JOURNAL Page 87

353 Conn. 169 SEPTEMBER, 2025 253 State v. McFarland

the delay was ‘‘hardly . . . unreasonable’’ and resulted from the exigencies of the narcotics investigation, including protecting the identity of an undercover detective, and the defendant had not demonstrated any prejudice because ‘‘[h]is alibi witnesses displayed par- ticularly acute memories in recreating the events of the evening in question.’’ Id., 568–69. Second, I find persuasive the decisions of other state courts that have rejected the rigid, two-pronged test in favor of a balancing test. This Geisler factor ‘‘calls for a more rigorous analysis than simply tallying holdings . . . [and] requires us to determine which . . . state courts’ approaches provide a genuinely persuasive framework for resolving this state constitutional ques- tion.’’ Doe v. Hartford Roman Catholic Diocesan Corp., 317 Conn. 357, 427, 119 A.3d 462 (2015). The state courts that adhere to the two-pronged test do not provide much in the way of justification for doing so, other than conclusorily articulating agreement with it as the correct reading of Marion and Lovasco, particularly when that approach governs in the applicable federal court of appeals. See, e.g., Jackson v. State, 279 Ga. 449, 450–51, 614 S.E.2d 781 (2005). But see Jackson v. State, 347 So. 3d 292, 304–305 (Fla. 2022) (providing detailed analysis for following United States v. Crouch, supra, 84 F.3d 1497, and adopting two-pronged test as matter of federal constitutional law, including claimed difficulty in mathematical application of balancing test). State courts rejecting the two-pronged test in favor of a balancing test under their respective state constitu- tions, however, have generally grounded that choice in pragmatism and fairness concerns.3 See, e.g., People v. 3 Several other state courts have adopted the balancing test in considering claims brought solely under the federal constitution. See, e.g., People v. Lawson, 67 Ill. 2d 449, 458–59, 367 N.E.2d 1244 (1977); Fritz v. State, 811 P.2d 1353, 1366–67 (Okla. Crim. App. 1991); State v. Stokes, 350 Or. 44, 57, 64, 248 P.3d 953, cert. denied, 565 U.S. 920, 132 S. Ct. 343, 181 L. Ed. 2d 215 (2011); State v. Brazell, 325 S.C. 65, 72–73, 480 S.E.2d 64 (1997); State v. Oppelt, 172 Wn. 2d 285, 288–90, 257 P.3d 653 (2011). Page 88 CONNECTICUT LAW JOURNAL September 2, 2025

254 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

Nelson, 43 Cal. 4th 1242, 1254–55, 185 P.3d 49, 78 Cal. Rptr. 3d 69, cert. denied, 555 U.S. 926, 129 S. Ct. 357, 172 L. Ed. 2d 219 (2008); State v. Keliiheleua, 105 Haw. 174, 179 and n.5, 95 P.3d 605 (2004); State v. Malvo, 357 So. 2d 1084, 1086–87 (La. 1978); State v. Cyr, 588 A.2d 753, 756 (Me. 1991); State v. Passmore, 355 Mont. 187, 198–99 and n.5, 225 P.3d 1229 (2010); State v. Knicker- bocker, 152 N.H. 467, 469–70, 880 A.2d 419 (2005); State v. Luck, 15 Ohio St. 3d 150, 153–54, 472 N.E.2d 1097 (1984), cert. denied, 470 U.S. 1084, 105 S. Ct. 1845, 85 L. Ed. 2d 144 (1985); State ex rel. Knotts v. Facemire, 223 W. Va. 594, 603–604, 678 S.E.2d 847 (2009); see also People v. Regan, 39 N.Y.3d 459, 464–66, 212 N.E.3d 282, 191 N.Y.S.3d 265 (2023) (greater due process protection was afforded under New York constitution by treating preindictment and speedy trial claims similarly, with multifactor balancing applied to preindictment delay claims). When considering the proper allocation of the burden of proof, some state courts that utilize the balancing test point out that it is inequitable to require the defendant to prove the state’s reasons for delay when that informa- tion is in the hands of the state. See, e.g., State v. Lee, 375 S.C. 394, 400, 653 S.E.2d 259 (2007) (‘‘[r]equiring a higher burden of proof in proving improper motives on the part of the prosecution would put an almost impossible burden on defendants to maintain a . . . due process claim in [preindictment] delay cases’’); State ex rel. Knotts v. Facemire, supra, 223 W. Va. 603 (‘‘[o]nly by eliminating the burden imposed on a defendant to demonstrate that the [s]tate gained an advantage through preindictment delay, will the over- arching concern of fundamental fairness that under- girds the [d]ue [p]rocess [c]lause be furthered’’); see also Howell v. Barker, supra, 904 F.2d 895 (noting ‘‘the difficulty defendants either have encountered or will encounter in attempting to prove improper prosecu- September 2, 2025 CONNECTICUT LAW JOURNAL Page 89

353 Conn. 169 SEPTEMBER, 2025 255 State v. McFarland

torial motive’’). This is consistent with Connecticut case law regarding the allocation of the burden of proof, which considers which party is more likely to have access to the facts in question. See, e.g., State v. Swebi- lius, 325 Conn. 793, 807–808, 159 A.3d 1099 (2017); State v. Ray, 290 Conn. 602, 610, 966 A.2d 148 (2009). An additional pragmatic justification for the balanc- ing test identified by these state courts is that delay by the state need not be intentional to cause prejudice. See, e.g., People v. Nelson, supra, 43 Cal. 4th 1255–56; State ex rel. Knotts v. Facemire, supra, 223 W. Va. 602–603; see also Howell v. Barker, supra, 904 F.2d 895. For example, in State v. Luck, supra, 15 Ohio St. 3d 150, the Ohio Supreme Court held that a fifteen year preindictment delay was unjustifiable because the state had decided to commence its prosecution of the defen- dant ‘‘without one shred of new evidence—its case being substantially the same as it had been since 1968.’’ Id., 158–59. In Luck, the state’s delay was the result of an ‘‘ ‘error in judgment,’ ’’ which stopped the investiga- tion for fifteen years. Id. During that time, witnesses died, memories faded, evidence was lost, and the state acquired no new evidence. Id., 159. Balancing the ‘‘actual prejudice’’ to the defendant against the lack of a justifi- able reason for the delay in the commencement of pros- ecution, the court held that the delay violated ‘‘those fundamental conceptions of justice [that] lie at the base of our civil and political institutions . . . and [that] define the community’s sense of fair play and decency.’’ (Internal quotation marks omitted.) Id. Had the court in Luck instead adopted the strict two-pronged test, the defendant’s valid due process claim would have failed, despite the unjustifiable delay, because there was no evidence that the delay was intentional or otherwise the product of bad faith. Third, the public policy factor of the Geisler analysis also weighs heavily in favor of adopting a balancing Page 90 CONNECTICUT LAW JOURNAL September 2, 2025

256 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

test. I recognize that the statute of limitations is ‘‘the primary guarantee against bringing overly stale criminal charges. Such statutes represent legislative assess- ments of relative interests of the [s]tate and the defen- dant in administering and receiving justice . . . .’’ (Internal quotation marks omitted.) United States v. Marion, supra, 404 U.S. 322; see State v. Littlejohn, 199 Conn. 631, 646, 508 A.2d 1376 (1986) (recognizing relationship between due process protections and stat- ute of limitations, which ‘‘ordinarily . . . is the only protection afforded a defendant against the institution of stale criminal charges’’); State v. Hodge, supra, 153 Conn. 567 (statute of limitations is ‘‘ultimate safeguard against a [long delayed] arrest and prosecution’’). There is, however, a growing number of crimes beyond mur- der for which there is either no limitation period; see General Statutes § 54-193 (a); or a significantly extended limitation period, particularly for certain sex offenses. See General Statutes § 54-193 (b). This increase in the potential for prosecutions beyond the generally applica- ble limitation periods of five years for felony offenses and one year for most misdemeanor offenses; see Gen- eral Statutes § 54-193 (c) and (d); highlights the need for a balancing test to determine whether a delay in prosecution has so prejudiced the defendant that it constitutes a due process violation. Given the extraordi- narily high bar necessary for a defendant to prevail, both by having to establish actual and substantial preju- dice, and by demonstrating that such prejudice out- weighed the government’s justification for delay, I am confident that the adoption of a balancing test does not override a legislative decision not to have a statute of limitations for certain crimes but, rather, simply ensures a defendant’s right to a fair trial. It recognizes the state’s obligation to investigate a crime thoroughly before filing criminal charges, particularly in the most serious cases, and reflects the deference given to the state to rely September 2, 2025 CONNECTICUT LAW JOURNAL Page 91

353 Conn. 169 SEPTEMBER, 2025 257 State v. McFarland

on a limitation period in conducting its investigation, notwithstanding the potential effects of a delayed arrest and prosecution, such as the loss of documentary evi- dence, the death or disappearance of witnesses, and the impairment of memory. See, e.g., State v. A. B., 341 Conn. 47, 63–64, 266 A.3d 849 (2021); State v. Hodge, supra, 568 and n.3. Before turning to the record in this case, I outline the proper application of the balancing test. The defendant bears the initial burden of producing evidence that the delay in the prosecution has caused his defense to suffer ‘‘actual and substantial prejudice . . . .’’ United States v. Hagler, 700 F.3d 1091, 1099 (7th Cir. 2012). Allocation of this initial burden to the defendant is appropriate given his familiarity with the various strategic options for his defense and the effect of the delay on, for exam- ple, his ability to conduct effective cross-examination. Id., 1099–1100. ‘‘This burden is an exacting one; the showing must rest [on] more than mere speculative harm . . . and [the defendant] must present facts that are specific, concrete, and supported by evidence.’’ (Citation omitted; internal quotation marks omitted.) Id., 1099. Indeed, the Seventh Circuit has described the defendant’s burden as a ‘‘monumental hurdle . . . .’’ United States v. Sowa, supra, 34 F.3d 451. Actual prejudice is that which is nonspeculative in nature, and substantial prejudice is that which ‘‘mean- ingfully [impairs the defendant’s] ability to defend against the state’s charges to such an extent that the disposition of the criminal proceeding was likely affected.’’ Jones v. Angelone, 94 F.3d 900, 907 (4th Cir. 1996). Demon- strating actual and substantial prejudice requires the defen- dant to establish that the delay has largely deprived him of his ability to mount a defense through a specific showing of lost witnesses, or other evidence, whose information was not available from other sources. Id., 908; see, e.g., United States v. Hagler, supra, 700 F.3d Page 92 CONNECTICUT LAW JOURNAL September 2, 2025

258 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

1099 (‘‘the mere fact that memories have faded is not enough to establish excessive delay,’’ particularly when defendant is able to highlight discrepancies during cross-examination); United States v. Crouch, supra, 84 F.3d 1515 (‘‘mere loss of potential witnesses is insuffi- cient [without] a showing that their testimony would have actually aided the defense’’ (internal quotation marks omitted)). The court considers how the lost evi- dence would have affected the defense. See, e.g., United States v. McMutuary, 217 F.3d 477, 482 (7th Cir.) (death of alibi witness did not constitute actual and substantial prejudice because witness’ testimony was cumulative of other witnesses’ testimony), cert. denied, 531 U.S. 1001, 121 S. Ct. 502, 148 L. Ed. 2d 471 (2000); People v. Cowan, 50 Cal. 4th 401, 431–34, 236 P.3d 1074, 113 Cal. Rptr. 3d 850 (2010) (there was no actual and substantial prejudice because, despite fingerprint technician’s age- related inability to recompare prints or to consider dif- ferences, prints remained ‘‘readily available’’ for retesting and use by defense, and contemporaneous police reports and interview recordings documented conversations that had been subject of faded memory), cert. denied, 563 U.S. 905, 131 S. Ct. 1784, 179 L. Ed. 2d 657 (2011); cf. State v. Lee, supra, 375 S.C. 399 (defendant suffered actual and substantial prejudice because he lost access to contemporaneous documentary evidence that was likely exculpatory because it would have explained why state returned stepchildren to his home and did not prosecute him at time of initial allegations of sexual abuse of stepchildren). Finally, in determining whether there is actual and substantial prejudice, the court con- siders the overall strength of the state’s case. See, e.g., United States v. Benshop, 138 F.3d 1229, 1234–35 (8th Cir. 1998). Only after the defendant clears the high hurdle of demonstrating actual and substantial prejudice does the burden shift to the state to produce evidence of September 2, 2025 CONNECTICUT LAW JOURNAL Page 93

353 Conn. 169 SEPTEMBER, 2025 259 State v. McFarland

the reasons for the delay—evidence that the state is uniquely well suited to produce. See, e.g., United States v. Hagler, supra, 700 F.3d 1099; see also United States v. Sowa, supra, 34 F.3d 451 (observing that ‘‘this is not a heavy burden for the [state] to bear; [its] standard practice in responding to motions to dismiss for [prein- dictment] delay is to explain the delay’’). At that point, a court must balance the state’s ‘‘rea- sons and the defendant’s prejudice to determine whether the defendant was denied due process.’’ United States v. Hagler, supra, 700 F.3d 1099. Bearing in mind the prosecutorial obligations both to assemble a case that will establish a defendant’s guilt beyond a reason- able doubt and to do justice by not advancing dubious prosecutions, it will be an exceedingly rare case for a defendant to establish a due process violation when the cause of ‘‘the delay is legitimately investigative in nature.’’ United States v. Sowa, supra, 34 F.3d 451; see People v. Nelson, supra, 43 Cal. 4th 1256 (concluding that lengthy delay in bringing murder charge had ‘‘strong’’ justification and balanced out prejudice to defendant, despite fact that he was identified as suspect shortly after murder, because case was not ‘‘fully solve[d]’’ until comparison of defendant’s DNA to crime scene evidence twenty-six years later). The state is not obligated to file charges as soon as it has probable cause to do so. See, e.g., United States v. Lovasco, supra, 431 U.S. 795 (‘‘[r]ather than deviating from elementary standards of fair play and decency, a prosecutor abides by them if he refuses to seek indictments until he is completely satisfied that he should prosecute and will be able promptly to establish guilt beyond a reasonable doubt’’ (internal quotation marks omitted)). Because ‘‘[d]ue process is not reducible to a mathe- matical formula’’; Commonwealth v. Devlin, 460 Pa. 508, 516, 333 A.2d 888 (1975); see Gibbs v. Burke, 337 U.S. 773, 780, 69 S. Ct. 1247, 93 L. Ed. 2d 1686 (1949); Page 94 CONNECTICUT LAW JOURNAL September 2, 2025

260 SEPTEMBER, 2025 353 Conn. 169 State v. McFarland

the application of this balancing test will depend on the facts and circumstances of each case, as developed from the entire trial record, even when the matter is initially raised in a pretrial motion to dismiss,4 and will be informed by the development of a body of case law applying this test as a question of law, with guidance from other jurisdictions that apply a similar balancing test. See, e.g., United States v. Lovasco, supra, 431 U.S. 797 (‘‘[w]e therefore leave to the lower courts, in the first instance, the task of applying the settled principles of due process that we have discussed to the particular circumstances of individual cases’’); State v. Oppelt, 172 Wn. 2d 285, 290, 257 P.3d 653 (2011) (due process balancing in prearrest delay cases is question of law requiring examination of the entire record); see also In re Lukas K., 300 Conn. 463, 469, 14 A.3d 990 (2011) (due process balancing under Mathews v. Eldridge, 424 U.S. 319, 335, 96 S. Ct. 893, 47 L. Ed. 2d 18 (1976), is question of law subject to plenary review). In the present case, I assume without deciding that the defendant has carried his heavy burden of establish- ing actual and substantial prejudice. See, e.g., Howell v. Barker, supra, 904 F.2d 895. In balancing that prejudice against the state’s reasons for its delay in bringing the charges against the defendant, I conclude that the defen- 4 As a practical matter, motions to dismiss based on prearrest delay need not be conclusively decided before trial because they do not implicate a defendant’s right not to stand trial. See United States v. Crouch, supra, 84 F.3d 1516. As in the present case, a defendant may move to dismiss on the basis of a due process violation caused by prearrest delay, and a trial court may properly decide that motion as a pretrial matter. See id. Nevertheless, it will often be appropriate for the trial court to defer action on the motion, or for the defendant to renew that motion as a posttrial matter, because the actual trial record is likely to be determinative with respect to whether the defendant has established the requisite actual and substantial prejudice without resort to impermissible speculation. See, e.g., United States v. Mar- ion, supra, 404 U.S. 325–26; United States v. Crouch, supra, 1516–17; State v. Knickerbocker, supra, 152 N.H. 471; State v. Benson, 370 Or. 58, 67–68, 514 P.3d 491 (2022). September 2, 2025 CONNECTICUT LAW JOURNAL Page 95

353 Conn. 169 SEPTEMBER, 2025 261 State v. McFarland

dant has failed to demonstrate that his due process rights were violated by the thirty-two year interval between the murders and his arrest. There is no evi- dence in the record that the delay in prosecuting him was the result of anything other than legitimate investi- gative reasons, given that the defendant’s various con- fessions were not entirely consistent with the crime scene evidence. Moreover, I find one historic fact partic- ularly persuasive in the balancing analysis. When the defendant murdered Fred Harris and Gregory Harris in 1987, the ultimate penalty was available to the prosecu- tion during its investigation of this case.5 The caution exercised by the state, including not charging the defen- dant following his confessions in 1996 and then waiting for more conclusive DNA testing, was highly reasonable for purposes of the balancing test. I therefore conclude that, when balanced against the assumed prejudice, the purely investigative delay renders the defendant unable to establish that his conviction violated ‘‘fundamental [concepts] of justice or the community’s sense of fair play and decency.’’ (Internal quotation marks omitted.) Id., 895. Accordingly, I concur.

5 In 1987, Connecticut had a capital felony statute that made a double homicide an offense eligible for the imposition of the death penalty. See General Statutes (Rev. to 1987) §§ 53a-35a and 53a-54b (8). Although the death penalty has become a footnote in Connecticut’s legal history since the enactment of No. 12-5 of the 2012 Public Acts; see generally, e.g., State v. Peeler, 321 Conn. 375, 140 A.3d 811 (2016); State v. Santiago, 318 Conn. 1, 122 A.3d 1 (2015); it was a viable punishment during much of the lengthy investigation of this case.