Skakel v. Comm'r of Corr.

Full title: MICHAEL SKAKEL v. COMMISSIONER OF CORRECTION

Court: Supreme Court of Connecticut

Date published: May 4, 2018

329 Conn. 1 (Conn. 2018)

188 A.3d 1

Opinion

SC 19251

05-04-2018

Michael SKAKEL v. COMMISSIONER OF CORRECTION

Susann E. Gill, former supervisory assistant state's attorney, with whom were James A. Killen, senior assistant state's attorney, and, on the brief, Kevin T. Kane, chief state's attorney, John C. Smriga, state's attorney, Leonard C. Boyle, deputy chief state's attorney for operations, and Jonathan C. Benedict, former state's attorney, for the appellant-cross appellee (respondent). Hubert J. Santos, Hartford, with whom was Jessica M. Walker, for the appellee-cross appellant (petitioner).

PALMER, J.

Susann E. Gill, former supervisory assistant state's attorney, with whom were James A. Killen, senior assistant state's attorney, and, on the brief, Kevin T. Kane, chief state's attorney, John C. Smriga, state's attorney, Leonard C. Boyle, deputy chief state's attorney for operations, and Jonathan C. Benedict, former state's attorney, for the appellant-cross appellee (respondent).

Hubert J. Santos, Hartford, with whom was Jessica M. Walker, for the appellee-cross appellant (petitioner).

Palmer, Eveleigh, McDonald, Espinosa, Robinson, Vertefeuille and D'Auria, Js.^***

*** This appeal was originally argued before a panel of this court consisting of Justices Palmer, Zarella, McDonald, Espinosa, Robinson and Vertefeuille. Thereafter, Justice Eveleigh was added to the panel. Justice Eveleigh read the briefs and appendices, and listened to a recording of the oral argument prior to participating in this decision. After publication of our initial decision, in which we reversed the habeas court's judgment; see Skakel v. Commissioner of Correction , 325 Conn. 426, 531, 159 A.3d 109 (2016) ; Justice Zarella retired from the Judicial Branch. Thereafter, the petitioner filed a timely motion for reconsideration en banc, seeking review by a seven member panel of this court. Following a vote of the six remaining panel members, we granted the petitioner's request for a panel comprised of seven members. Justice D'Auria, who became a member of this court after the petitioner's motion for reconsideration en banc was filed, was then added to the panel, and he has read the briefs and appendices, and listened to a recording of the oral argument prior to participating in this decision. On reconsideration en banc, Justices McDonald and Robinson join Justice Palmer's majority opinion, Justice D'Auria joins all but part II of Justice Palmer's majority opinion, Justice D'Auria files a separate concurring opinion, Justices Espinosa and Vertefeuille join Justice Eveleigh's dissenting opinion, and Justice Espinosa files a separate dissenting opinion. Because of these changes in the panel and vote, this opinion supersedes in all respects our prior decision; see Skakel v. Commissioner of Correction , supra, at 426, 159 A.3d 109 ; on the issue for which reconsideration en banc was granted. See, e.g., Honulik v. Greenwich , 293 Conn. 698, 702 n.1, 980 A.2d 880 (2009).
The listing of justices reflects their seniority status on this court as of date of oral argument.

PALMER, J.

The sole issue now before us in this appeal by the respondent, the Commissioner of Correction, is whether the habeas court properly concluded that the petitioner, Michael Skakel, is entitled to a new trial because counsel in his murder case, Michael Sherman, rendered ineffective assistance by failing to obtain certain readily available evidence that he should have known was potentially critical to the petitioner's alibi defense, that is, the testimony of a disinterested alibi witness whom the habeas court found to be highly credible. Because we agree with the habeas court both that Sherman's failure to secure that evidence was constitutionally inexcusable and that that deficiency undermines confidence in the reliability of the petitioner's conviction—a conviction founded on a case, aptly characterized by the habeas court as far from overwhelming, that was devoid of any forensic evidence or eyewitness testimony linking the petitioner to the crime—we affirm the judgment of the habeas court ordering a new trial.

This case comes to this court again under the following circumstances. In 2002, a jury found the petitioner guilty of the brutal murder of his fifteen year old neighbor, Martha Moxley (victim), whose bludgeoned and partially unclothed body was discovered on October 31, 1975, behind her parents' home in the Belle Haven section of the town of Greenwich. This court affirmed his conviction; State v. Skakel , 276 Conn. 633, 770, 888 A.2d 985, cert. denied, 549 U.S. 1030, 127 S.Ct. 578, 166 L.Ed.2d 428 (2006) ; and, thereafter, the petitioner filed a petition for a writ of habeas corpus, principally claiming that his trial counsel, Sherman, had rendered ineffective assistance in numerous respects. The habeas court, Hon. Thomas A. Bishop , judge trial referee, agreed with several of the petitioner's claims, among them that Sherman had performed deficiently in investigating and presenting the petitioner's alibi defense by failing to adduce the testimony of a truthful and crucial alibi witness, Denis Ossorio. The habeas court further concluded that, in light of the relative weakness of the state's case and the powerful support that Ossorio's testimony provided for the petitioner's alibi, Sherman's deficient performance had so seriously prejudiced the petitioner that it is reasonably probable that the outcome of the petitioner's criminal trial would have been different if the jury had heard from Ossorio. The habeas court therefore rendered judgment granting the petition, ordering a new trial for the petitioner, and the respondent appealed. On appeal, in a closely divided decision, this court reversed the judgment of the habeas court, concluding that the petitioner had failed to prove any of his claims of ineffective assistance. See Skakel v. Commissioner of Correction , 325 Conn. 426, 430–31, 531, 159 A.3d 109 (2016).¹ The petitioner thereafter filed a timely motion for reconsideration en banc, limited to his claim that Sherman's performance with respect to the petitioner's alibi defense was constitutionally inadequate. We granted the petitioner's motion, and, upon reconsideration, we now conclude that the habeas court correctly determined that the petitioner is entitled to a new trial due to Sherman's failure to adequately investigate and present the petitioner's alibi defense, which rendered the petitioner's trial fundamentally unfair. Accordingly, we affirm the judgment of the habeas court.

1 Justice Zarella authored the majority opinion; Skakel v. Commissioner of Correction , supra, 325 Conn. at 430, 159 A.3d 109 ; which Justices Eveleigh, Espinosa and Vertefeuille joined. Id., at 531, 159 A.3d 109. Justice Palmer authored a dissenting opinion; id., at 534, 159 A.3d 109 (Palmer, J. , dissenting); which Justice McDonald joined; id. ; and with which Justice Robinson agreed in part; id., at 531, 159 A.3d 109 (Robinson, J. , concurring in part and dissenting in part).

I

FACTS AND PROCEDURAL HISTORY

The facts of this tragic case, which arises out of events that transpired more than forty years ago, are set forth in considerable detail in the habeas court's memorandum of decision and in this court's decision on the petitioner's direct appeal.² See State v. Skakel , supra, 276 Conn. at 640–53, 888 A.2d 985. For present purposes, we focus our attention on those facts and the procedural history that are most relevant to the respondent's claim that the habeas court incorrectly determined that Sherman rendered ineffective assistance of counsel in connection with his investigation and presentation of the petitioner's alibi defense. We more fully address all of the evidence presented in support of the conviction, however, later in this opinion when we consider whether Sherman's performance prejudiced the petitioner.

2 We note, in addition, that, because this opinion addresses the very same claim and reaches the very same conclusion in regard to that claim as the dissenting opinion in Skakel v. Commissioner of Correction , supra, 325 Conn. at 426, 159 A.3d 109, some of the discussion of the defendant's claim is taken verbatim from that dissenting opinion. See id., at 534–623, 159 A.3d 109 (Palmer, J. , dissenting).

Those facts are rather extensive but may be summarized as follows. In the early afternoon of October 31, 1975, the victim's body was discovered under a pine tree behind her parents' home. She had been severely and repeatedly beaten with a golf club, which was later determined to have belonged to the petitioner's then deceased mother. The victim had been last seen at approximately 9:30 p.m. on October 30, 1975, standing with the petitioner's seventeen year old brother, Thomas Skakel, in the Skakel family's driveway. The location where the police determined that the victim was attacked was along what would have been the most direct route between where she was last seen and her parents' home, indicating that she was likely murdered as she made her way home from the Skakel driveway.

Earlier that evening, at approximately 6:30 p.m., the victim had left her house to celebrate mischief night—the night before Halloween—with her friend, Helen Ix, and other children from the neighborhood. When the victim left her house, the petitioner, then fifteen years old, and his six siblings, Rushton Skakel, Jr., Julie Skakel, Thomas Skakel, John Skakel, David Skakel and Stephen Skakel, together with their cousin James Terrien, their tutor Kenneth Littleton, and Julie Skakel's friend, Andrea Shakespeare, were having dinner at the Belle Haven Club. This group returned home from dinner shortly before 9 p.m., at which time the victim, Ix, and eleven year old Geoffrey Byrne came by the Skakel house to visit.

The petitioner immediately led the visitors outside, where they all climbed into the Skakel family's Lincoln Continental to talk and listen to music. Shortly thereafter, Thomas Skakel joined them in the Lincoln. At approximately 9:25 p.m., the group was interrupted by Rushton Skakel, Jr., John Skakel and Terrien, who told them that they needed to use the car to take Terrien home, where they planned to watch Monty Python's Flying Circus, a television show, at 10 p.m. At that point, the victim, Ix, Byrne and Thomas Skakel got out of the car, while Rushton Skakel, Jr., John Skakel and Terrien got into the car with the petitioner. Upon exiting the car, Thomas Skakel and the victim began roughhousing in a flirtatious manner, which made Ix uncomfortable, prompting her, along with Byrne, to leave. According to Ix, the Lincoln was pulling out of the driveway as she and Byrne began walking home, leaving Thomas Skakel and the victim alone in the driveway. That was the last time any of the victim's friends reported seeing her.³ Ix arrived home at about 9:30 p.m. and telephoned a friend. At approximately 9:45 p.m., Ix' dog, an Australian shepherd named "Zock," began barking violently near the entrance to the victim's driveway, located directly across the street from the entrance to Ix' driveway. The barking was so incessant and agitated that Ix put down the telephone and opened the door to call the dog inside. Although, previously, Zock had always come when called, that night he refused to come no matter how fervently and repeatedly Ix called to him. In interviews with the police following the murder, and in testimony at the petitioner's criminal trial twenty-seven years later, Ix stated that, prior to that evening, she had never seen her dog behave in such an agitated manner. Ix explained that his barking that night, which she described as "scared" and "violent," was very different from the way he usually barked; he "was definitely disturbed by something that was going on," and he "was basically barking in the direction of the site where [the victim's] body was found [the next day]." The police later determined, on the basis of blood spatter found at the scene, that the victim was initially attacked in or near her driveway.⁴

3 We note that, when the police interviewed Thomas Skakel the next day, just hours after the victim's body was discovered, he stated that he went into his house for the evening immediately after Ix and Byrne departed, and that the victim also left at that time. Nearly twenty years later, however, Thomas Skakel disclosed that he had lied to the police and that, in fact, he and the victim had remained in his backyard for about twenty minutes, at which time they engaged in consensual sexual contact. In addition, on the eve of the petitioner's criminal trial, Thomas Skakel, in the presence of his attorney, repeated this version of the events to Sherman. The jury, however, never heard anything about Thomas Skakel's claimed sexual involvement with the victim that evening because Sherman was informed by counsel for Thomas Skakel that he would invoke his constitutional privilege against self-incrimination, and, accordingly, he was not called to testify at the petitioner's criminal trial. Sherman did not otherwise seek to introduce any evidence concerning that purported involvement. Consequently, for purposes of resolving the petitioner's motion for reconsideration en banc, we do not consider Thomas Skakel's statements either to the police or to Sherman.

4 The habeas court noted the following additional information concerning the unusual behavior of a dog between 9:30 and 10 p.m. on October 30, 1975. "This court is aware, from a review of the Greenwich police file, that ... [Robert] Bjork ... reported that he observed his dog, a springer spaniel, walk to the edge of his property, which bordered on the Moxley property, and 'then [walk] to the area of the pine tree where the body of the victim was subsequently found.... Bjork related that the dog would first walk to the area of the willow tree on the southwest corner of the Moxley property, where two large spots of blood were found, and then walk from this spot to the pine tree. At the time, [Bjork] related that he placed no significance in the action. He related that this was approximately 9:50 p.m.' This report was filed on April 8, 1976. Since the jury did not hear this evidence, and there is no claim by the petitioner that trial counsel was ineffective for not presenting this evidence, the court draws no conclusions from it regarding ineffectiveness. The court simply notes that such evidence, if presented, would have further buttressed the petitioner's claim regarding the approximate time of death and, by extension, the importance of his alibi defense." We, likewise, do not rely on this evidence in connection with our determination of the petitioner's motion for reconsideration en banc.

The victim's mother, Dorothy Moxley, reported hearing a similar commotion in her yard between 9:30 and 10 p.m. on the night of the murder. The victim's mother testified that the disturbance, which consisted of "excited voices" and "incessant barking," was so distracting that she stopped what she was doing to look out the window. According to the victim's mother, it was "very, very cold and very dark" outside on the night in question. When she could not see anything, she turned on the porch light but then immediately turned it off because she feared that whoever was there might steal the victim's new bike. The victim's mother grew worried when the victim had not returned by 11 p.m. and began calling all of the victim's friends, "everyone that [she] could think of," in an effort to locate the victim. When the victim still had not returned by 1 a.m., the victim's mother asked the victim's older brother, John Moxley, to go out and look for her. At 3:48 a.m., she finally called the Greenwich police to report the victim missing. During that telephone call, she stated that the victim had been "expected home at 9:30 p.m." She also stated that she had called several of the victim's friends before calling the police and had been told by one to "check with the ... Skakels ...." The victim's mother reported that she had "called the Skakel residence ... and spoke to Thomas" Skakel, who "informed [her] that he last saw [the victim] at approximately 9:30 the preceding night [and that the victim had] told him she was going home to do homework." At trial, the victim's mother testified that the victim was extremely reliable about coming home at a reasonable hour, which, as she recalled, was 9:30 p.m. on school nights and 10:30 p.m. on nonschool nights.

Following the murder, the Greenwich police retained Joseph Jachimczyk, then the chief medical examiner for Harris County, Texas, and a nationally renowned pathologist and criminalist, to assist them in their investigation. Jachimczyk, who testified as a defense witness at trial, determined that the victim's time of death was approximately 10 p.m. based on the contents of her stomach, the extent of rigor mortis, the time that she was expected home, and the report of dogs barking at the crime scene. Harold Wayne Carver II, who, at the time of trial, was the state's chief medical examiner, testified that he did not participate in the victim's autopsy, which had been performed by Elliot Gross, formerly the state's chief medical examiner, but that he was able to form an opinion about certain aspects of the victim's death on the basis of the record of the autopsy. In particular, Carver explained that, although he could not determine the time of death precisely on the basis of the condition of the victim's body, he believed that the victim died between 9:30 p.m. on October 30, 1975, and 12 or 1 a.m. the next morning. Carver further opined that it was likely that the victim died closer to 9:30 p.m. on October 30 than to when she was found the next day.

That time of death was also consistent with testimony adduced at trial establishing that the victim did not return home on the night of the murder when she was expected and that no one saw her alive after she was last seen in the Skakel driveway at 9:30 p.m. For example, one of the victim's friends, Jackie Wetenhall, who was with her that evening, testified that she returned home at 9 p.m. and that it was her understanding that the victim also had a curfew. As we previously noted, the victim's mother testified that the victim did not return home as expected and that her failure to do so was out of character. She further testified that, after the victim failed to return home by 11 p.m., she called every one of the victim's friends, and her own friends, as well, in an effort to locate the victim, but without success. Thomas G. Keegan, the detective in charge of the investigation for the Greenwich police, testified that the police canvassed the entire area after the murder in an effort to locate anyone who could shed light on the victim's time of death and whereabouts after 9:30 p.m., but also to no avail.⁵

5 Thus, as Keegan wrote in a letter to Jachimczyk shortly after the murder: "Our assumption is that death occurred about 10 p.m., October [30], as the investigation shows that two neighborhood dogs were highly agitated shortly before 10 p.m. We feel that, even though there was no school the next day, the [victim] left the Skakel house and was headed home because her friends were not going to remain out any longer that night. We have interviewed [400] people, and no one saw the [victim] after 9:30 p.m. on the night in question. It seems highly unlikely ... that a ... fifteen year old female would [wander the neighborhood alone] at night." Because this letter was never introduced into evidence at the petitioner's criminal trial, it plays no role in our analysis of the issues raised in this appeal.

Following the murder and for many years thereafter, the petitioner's brother, Thomas Skakel, was the prime suspect in the murder, not only because he was the last person to be seen with the victim, but also because, shortly after the murder, investigators determined that he had been untruthful about his activities on the night the victim was murdered.⁶ In 1976, the police sought permission from the state's attorney to apply for an arrest warrant for Thomas Skakel, but permission was denied because the state's attorney did not believe that the facts set forth in the warrant application were sufficient to establish probable cause to believe that Thomas Skakel had committed the murder. For a number of years, the Skakels' tutor, Littleton, was also a prime suspect, but, after considerable investigation, he was exonerated. Ultimately, the case went cold.

6 Immediately after the murder, Thomas Skakel told the police that he had not spent any time with the victim after his brothers and Terrien left the Skakel residence in the family's Lincoln because he had to complete a homework assignment that was due the next day. He also informed the police that he was in his bedroom working on that assignment at 10 p.m. Subsequently, however, the police learned from Thomas Skakel's teachers that there had been no homework assignment the next day. They also learned from Littleton that Thomas Skakel was not in his bedroom at 10 p.m., as he had claimed. Although this evidence was pertinent to claims raised in the respondent's appeal from the judgment of the habeas court, it does not bear on the outcome of the issues raised in the petitioner's motion for reconsideration en banc because the evidence was not introduced at the petitioner's criminal trial, and it is not otherwise the subject of any claim raised in the petitioner's motion for reconsideration en banc.

In the early 1990s, several events led to the reopening of the investigation and eventually resulted in the petitioner's arrest and conviction. Notable among them was the publication, in 1993, of "A Season in Purgatory," Dominick Dunne's best-selling novel in which Dunne depicted Thomas Skakel as the murderer. Because of the renewed scrutiny on his family, the petitioner's father, Rushton Skakel, Sr., hired a private security firm, Sutton Associates, to investigate the murder in the hope of exonerating his family. As part of that investigation, investigators from Sutton Associates interviewed Thomas Skakel and the petitioner, both of whom disclosed that they had lied to the police in 1975 about their activities on the night of the murder. Thomas Skakel told the investigators that, after his brothers left to take Terrien home at approximately 9:30 p.m., he and the victim had spent about twenty minutes in his backyard engaged in heavy petting and mutual masturbation. The petitioner, for his part, told the Sutton Associates investigators that, after returning from the Terrien home at around 11 p.m., he went back out, at around 12 a.m., to peep in the window of a woman who lived nearby. On the way home, he stopped at the victim's house, climbed a tree adjacent to the house, and masturbated. The petitioner later told the same story to Richard Hoffman, a ghost writer whom the petitioner hired in 1997 to assist him in writing his autobiography. In 1994, an employee of Sutton Associates stole the firm's files on the case, including detailed suspect profiles, and gave them to Dunne and to Leonard Levitt, a journalist who previously had written extensively about the case. On November 26, 1995, Levitt published the first of a series of newspaper articles in which he disclosed that the petitioner and Thomas Skakel had changed their stories with respect to their activities on the night of the murder. Dunne later gave the stolen Sutton Associates files to Mark Fuhrman, the former Los Angeles police detective who was notorious for his allegedly perjurious testimony at the Orenthal James (O.J.) Simpson murder trial. In 1998, Fuhrman published a book in which he accused the petitioner of the victim's murder and the victim's family of conspiring to cover it up. In his book, Fuhrman urged that a grand jury be empaneled immediately to investigate his theory.

Shortly thereafter, in September, 1998, a grand jury was convened to investigate the case, and the petitioner hired Sherman to represent him in connection with that proceeding. Numerous witnesses were called to testify before the grand jury, including the petitioner's cousin, Georgeann Dowdle, Terrien's sister, who, at the time of the murder, lived with Terrien in their mother's home. Significantly, Dowdle testified that she was at home with her young daughter and her "beau" on the night of the murder⁷ and heard her brother and Skakel cousins talking, but, given the passage of so much time, she could not recall whether she actually saw them. She further testified, however, that she gave a statement to the police several days after the victim's murder in which she explained that she had observed the petitioner and his two brothers at the Terrien home on the evening of October 30, 1975. Dowdle also testified that her statement to the police was truthful and accurate in all respects.

7 As we explain in greater detail hereinafter, neither the state nor Sherman ever asked Dowdle about the identity of her "beau," who was subsequently identified as Ossorio, until years after the petitioner's criminal trial.

Following the publication of the Levitt articles in 1995, and Fuhrman's book in 1998, several witnesses came forward claiming that, at various times and in different ways, the petitioner had incriminated himself in the victim's murder. Two such witnesses, Gregory Coleman and John Higgins, claimed that the petitioner had confessed to the murder in the late 1970s, while they were students with the petitioner at the Elan School (Elan), an alcohol and drug rehabilitation facility for troubled adolescents in Maine. Their testimony, which we discuss more fully in part V B 2 of this opinion, would become the cornerstone of the state's case against the petitioner.

At trial, the petitioner raised an alibi defense predicated on two separate but related factual assertions, first, that it was nearly certain that the victim was murdered between 9:30 and 10 p.m. on October 30, 1975, and, second, that he was at his cousin's house, a fifteen to twenty minute car ride from the scene of the crime, at the time of the murder. With respect to the time of the murder, the petitioner relied on the testimony of Jachimczyk, the forensic pathologist who had assisted the Greenwich police in their investigation and determined that the time of death was 10 p.m. He also relied on the testimony of the victim's mother concerning the time the victim was expected to be home, the fact that the victim invariably returned home when expected, and the loud commotion and incessant barking that she had heard in her yard between 9:30 and 10 p.m. In addition, the petitioner relied on Ix' testimony regarding her dog's bizarre behavior near the crime scene beginning at approximately 9:45 p.m., which, Sherman argued to the jury, effectively "time stamps when this crime occurred." Because no one reported seeing the victim alive after she left the company of Thomas Skakel at 9:30 p.m., Sherman also argued that it was unreasonable to think that the victim had remained outside in the cold, all alone, well after she was expected home and after all of her friends had gone home, until sometime after 11 p.m., when the petitioner returned from the Terrien home.

With respect to the petitioner's whereabouts when the victim was murdered, Sherman presented the testimony of the petitioner's brothers, Rushton Skakel, Jr., and John Skakel, and his cousins, Terrien and Dowdle. Consistent with their grand jury testimony and the statements that they had given to the police in 1975, Rushton Skakel, Jr., John Skakel and Terrien testified that, on the night of the murder, they and the petitioner left in the Skakels' Lincoln at approximately 9:30 p.m. and drove to the Terrien residence, where the petitioner and his two brothers remained until approximately 11 p.m., watching television and talking. Dowdle's trial testimony essentially mirrored her grand jury testimony. In particular, she again confirmed the accuracy and truthfulness of the statement that she had given to the police in 1975, namely, that she had "observed" the petitioner and his brothers at the Terrien home on the evening of October 30, 1975. When the state's attorney asked whether she previously had testified before the grand jury that she was at home with her "husband" on the evening in question, Dowdle responded that she actually was with a "friend" that evening, referring to the person she previously had characterized as her "beau" in her grand jury testimony. That portion of Dowdle's grand jury testimony was introduced into evidence at the petitioner's criminal trial.

To rebut the petitioner's alibi, the state's attorney aggressively sought to discredit each of the petitioner's alibi witnesses, arguing that, as members of the petitioner's family, they all had a strong motive to lie and that, in fact, they all were lying. Repeatedly, during closing argument, he implored the jury to "[c]onsider who the alibi witnesses are, all siblings or first cousins, not one single independent alibi witness ." (Emphasis added.) According to the state's attorney, the alibi was not only concocted, it was an integral part of a decades' long Skakel family conspiracy, orchestrated by the petitioner's father, Rushton Skakel, Sr., to protect the petitioner at all costs from the consequences of his heinous crime.

To support his contention that the petitioner did not go to the Terrien home on the night of the murder, the state's attorney presented the testimony of Shakespeare, Julie Skakel, and Ix. Shakespeare testified that she was at the Skakel residence that evening and that, to her recollection, the petitioner did not leave to go to Terrien house as claimed. Julie Skakel testified that, shortly after the murder, she informed the police that, on the night of the murder, at approximately 9:20 p.m., a person darted through her yard as she was getting into her car to drive Shakespeare home. Julie Skakel further acknowledged that she had told the police at the time of the murder that, although it was dark and she did not see the person's face, she assumed it was the petitioner and yelled, "Michael, come back here," but the person kept on running. When Julie Skakel was asked whether the Lincoln was still in the driveway when she saw this individual, she replied that she could not remember. Finally, the state's attorney elicited testimony from Ix, who stated that, although it was her impression that the petitioner was in the Lincoln when it pulled out of the driveway, she was not absolutely certain and would not have known if the petitioner had jumped out of the car after it left the driveway. In closing argument, the state's attorney relied on this testimony to underscore the point that, whereas "[n]o independent witness [could] say what happened once [the] Lincoln backed out of the driveway," Julie Skakel's testimony "certainly suggests" that the petitioner was not still in the Lincoln when it arrived at the Terrien residence.

The state's attorney further argued, however, that, even if the jury believed that the petitioner was at the Terrien home watching television that evening and did not return home until approximately 11 p.m., it could still find him guilty of the victim's murder because the forensic evidence did not rule out the possibility that the murder occurred as late as 5:30 a.m. on October 31, 1975. The state's attorney conceded, however, that the victim was likely murdered much earlier than that, and in no event later than 1 a.m., because, by then, her family was out searching for her. Despite raising the possibility that the victim was murdered after 11 p.m., the state's attorney never proffered any evidence to establish, or made any argument to explain, where or with whom the victim might have been from 9:30 p.m., when she was last seen by friends, until shortly after 11 p.m. Nor did the state's attorney offer an alternative explanation as to what had caused the agitated barking and other unusual noises in the victim's yard between 9:30 and 10 p.m.⁸ Instead, the state's attorney simply maintained that the jury was required to find the petitioner guilty even if some jurors rejected the petitioner's alibi and others accepted the alibi, as long as all twelve jurors concluded beyond a reasonable doubt that the petitioner had killed the victim, as the state alleged.

8 In fact, the state's attorney argued that the barking was additional proof that the petitioner did not go to the Terrien house as claimed because there was testimony suggesting that the petitioner was the only person in the neighborhood who could make Ix' dog behave in such a frenzied manner.

At the conclusion of the petitioner's criminal trial, the jury found the petitioner guilty of murder, and the court sentenced him to a term of imprisonment of twenty years to life. Following an unsuccessful appeal from the judgment of conviction; see State v. Skakel , supra, 276 Conn. at 640, 888 A.2d 985 ; and from the denial of his petition for a new trial based on newly discovered evidence; see Skakel v. State , 295 Conn. 447, 452, 991 A.2d 414 (2010) ; the petitioner commenced the present habeas action, claiming that Sherman's trial performance was deficient in myriad ways. After a two week trial, the habeas court granted the petition, concluding, in a comprehensive memorandum of decision, that Sherman's performance fell below the standard required by the sixth and fourteenth amendments to the federal constitution in ten separate and distinct respects. The habeas court further concluded that three of those deficiencies, one of which was Sherman's failure to adequately present the petitioner's alibi defense, were sufficiently prejudicial to render his trial fundamentally unfair, thereby requiring a new trial.⁹ The respondent appealed, and this court, in a four to three decision, reversed the judgment of the habeas court after concluding, contrary to the determination of that court, that the petitioner had failed as a matter of law to prove any of his ineffective assistance claims.¹⁰ See Skakel v. Commissioner of Correction , supra, 325 Conn. at 430–31, 159 A.3d 109. Immediately after the issuance of our opinion announcing the decision of the court, Justice Zarella, the author of the majority opinion, retired from the Judicial Branch. Thereafter, the petitioner filed a timely motion seeking en banc reconsideration, limited to our determination that Sherman's failure to identify and call Ossorio as an alibi witness did not violate the petitioner's right to the effective assistance of counsel.¹¹ Following a vote of the remaining panel members, we granted the petitioner's motion for reconsideration en banc, and, in accordance with that vote, Justice D'Auria, who, during the pendency of the petitioner's motion, had been appointed to fill the vacancy on this court created by Justice Zarella's retirement, was added to the panel. For the reasons set forth hereinafter, upon reconsideration en banc, we agree with the petitioner that the habeas court correctly concluded that he is entitled to a new trial due to Sherman's deficient performance in investigating and presenting the petitioner's alibi defense.¹² Before addressing the merits of this appeal, however, we turn first to the threshold issue raised by the petitioner's motion for reconsideration en banc, namely, the composition of the panel for purposes of deciding the motion.

9 More specifically, the habeas court found that Sherman had performed deficiently in failing (1) to competently present a third-party culpability claim regarding Littleton, (2) to investigate a third-party culpability claim concerning Adolph Hasbrouck and Burton Tinsley, two high school students from New York who, according to Gitano Bryant, a former Greenwich resident, were with him in Belle Haven on the night of the victim's murder and subsequently acknowledged to Bryant that they had murdered the victim, (3) to raise a third-party culpability claim implicating Thomas Skakel as the perpetrator, (4) to diligently pursue the petitioner's alibi defense by neglecting to locate and adduce testimony from an independent alibi witness, Ossorio, (5) to investigate and counter the testimony of Coleman regarding the petitioner's alleged confession while the petitioner was a resident at Elan, (6) to adequately rebut arguments made by the state's attorney to the jury, including but not limited to his claim that placing the petitioner in Elan was part of a family cover-up, (7) to employ and utilize expert testimony regarding the cruel and coercive treatment the petitioner experienced at Elan, (8) to undertake appropriate efforts to select an impartial jury, (9) to suppress audio recordings of the petitioner that were unlawfully seized from Hoffman by the state, and (10) to adequately prepare for and present a minimally effective closing argument. The three deficiencies that the habeas court found had so seriously prejudiced the petitioner as to require a new trial were (1) Sherman's failure to raise a third-party culpability claim implicating Thomas Skakel as the murderer, (2) his failure to identify, locate and call Ossorio as an alibi witness, and (3) his failure to adequately challenge the veracity of Coleman's testimony concerning the petitioner's alleged confession.
We also note that the petitioner raised several additional claims in support of his contention that he is entitled to a new trial due to Sherman's inadequate or otherwise flawed representation. The habeas court, however, rejected these claims.

10 The petitioner cross appealed, claiming, as alternative grounds for affirming the habeas court's judgment, that that court had improperly rejected several of his claims. See Skakel v. Commissioner of Correction , supra, 325 Conn. at 440, 159 A.3d 109. Some of the petitioner's claims on appeal pertained to challenges that the habeas court rejected on the ground that the petitioner had failed to establish deficient performance by Sherman, and others pertained to challenges that the habeas court rejected, despite its finding of inadequate performance, due to lack of proof of prejudice. This court rejected all of these claims, none of which is the subject of this opinion.

11 All three of the justices who either dissented or dissented in part from this court's majority opinion in Skakel v. Commissioner of Correction , 325 Conn. 426, 159 A.3d 109, would have affirmed the habeas court's judgment with respect to that issue. See Skakel v. Commissioner of Correction , supra, 325 Conn. at 533, 159 A.3d 109 (Robinson, J. , concurring in part and dissenting in part); id., at 534, 159 A.3d 109 (Palmer, J. , with whom McDonald, J. , joined, dissenting).

12 Because the petitioner's motion seeks reconsideration en banc only as to the petitioner's claim that Sherman rendered ineffective assistance in connection with the investigation and presentation of the petitioner's alibi defense, this opinion addresses that issue only.

II

MOTION FOR RECONSIDERATION EN BANC

In his motion for reconsideration en banc, the petitioner requested that the six remaining panel members add a seventh justice to the panel to replace Justice Zarella, who, as we have explained, retired from the Judicial Branch before that motion was filed. Upon consideration of the petitioner's request, as well as the respondent's objection, a majority of those six remaining panel members, namely, Justices Palmer, McDonald, Robinson and Vertefeuille, voted to grant the petitioner's request, and, consequently, Justice D'Auria was added to the panel. The decision to add Justice D'Auria reflects this court's strong and long-standing preference for resolving appeals and related motions, whenever possible, on their merits, a preference that is thwarted when, as in the present case, a tie vote on the merits would have resulted in a deadlock. The discretion vested in this court to add a seventh justice to the panel derives from the court's inherent supervisory authority over the administration of justice; see, e.g., State v. Baltas , 311 Conn. 786, 824, 91 A.3d 384 (2014) (appellate courts possess inherent supervisory authority over administration of justice); see also Practice Book § 60–2 ("[t]he supervision and control of the proceedings [on appeal] shall be in the court having appellate jurisdiction from the time the appeal is filed, or earlier, if appropriate"); authority that dates back to the seventeenth century. State v. DeJesus , 288 Conn. 418, 451, 953 A.2d 45 (2008). This discretion is undisputed.

Notably, as a general rule, this court's practice of undertaking steps to avoid a deadlock resulting from a tie vote in any appeal or related motion so that a decision on the merits may be reached is required by the plain language of General Statutes § 51–209. That statute provides in relevant part: "No ruling, judgment or decree of any court may be reversed, affirmed, sustained, modified or in any other manner affected by the Supreme Court or the Appellate Court unless a majority of the judges on the panel hearing the cause concur in the decision. No cause reserved, where no verdict has been rendered, judgment given or decree passed, shall be determined unless a majority of the judges on the panel hearing the cause concur in the decision. Whenever the Supreme Court is evenly divided as to the result, the court shall reconsider the case, with or without oral argument, with an odd number of judges...." General Statutes § 51–209 ; see also Practice Book § 71–5 ("[a] motion for reconsideration shall be treated as a motion for reconsideration en banc when any member of the court which decided the matter will not be available, within a reasonable amount of time, to act on the motion for reconsideration"). For present purposes, however, we need not decide whether § 51–209 also would mandate the addition of a seventh justice because this court, pursuant to its inherent supervisory authority, has the discretion to do so in the interests of justice. The addition of Justice D'Auria to the panel following Justice Zarella's retirement ensures a decision by the court on the merits of the petitioner's motion rather than an outcome predicated on an evenly divided vote that, because of the impasse, would not have resulted in a substantive resolution of the petitioner's claims.

We also observe that the decision to add Justice D'Auria to the panel in the present case is consistent with the practice of every other sister state court that has addressed the issue posed by the petitioner's motion, that is, whether to add a judge to a panel when an original panel member was unable to participate in the resolution of a timely filed motion for reconsideration. See, e.g., Commonwealth v. Franklin , Docket No. 12-P-569, 87 Mass.App.Ct. 1139, 2015 WL 4663516, *1 n.2 (2015) (adding judge to panel after retirement of judge who sat on original panel, and reversing decision of original panel to uphold trial court's denial of defendant's motion for new trial); University of Michigan Regents v. Titan Ins. Co. , 484 Mich. 852, 852, 769 N.W.2d 646 (2009) (granting reconsideration and vacating prior order denying leave to appeal in case in which justice on original panel had been replaced by another justice who participated in decision on motion for reconsideration); United States Fidelity Ins. & Guaranty Co. v. Michigan Catastrophic Claims Assn. , 484 Mich. 1, 11 and n.12, 795 N.W.2d 101 (2009) (granting motion for rehearing without further briefing or oral argument after change in composition of court and reversing earlier decision); Johnson v. Administrator, Ohio Bureau of Employment Services , 48 Ohio St. 3d 67, 69, 549 N.E.2d 153 (1990) (reversing prior judgment in same case after rehearing and change in composition of court); State v. Eriksen , 172 Wash. 2d 506, 509, 259 P.3d 1079 (2011) (twice granting reconsideration and withdrawing previous decisions, the second time after new justice replaced previous majority author on panel). No contrary authority has been brought to our attention.

In light of the nature of a motion for reconsideration—including, in particular, its corrective purpose and the broad discretion afforded this court in deciding such a motion—it is hardly surprising that our decisions have sometimes been modified or reversed upon reconsideration not only by larger panels, but also by the very same panels that initially issued the decisions. See, e.g., Pelletier v. Sordoni/Skanska Construction Co. , 264 Conn. 509, 511–12 and n.2, 825 A.2d 72 (2003) (superseding prior decision in part upon reconsideration); Gartrell v. Dept. of Correction , 259 Conn. 29, 44–45 and n.14, 787 A.2d 541 (2002) (superseding prior decision upon reconsideration); Williams v. Best Cleaners, Inc. , 237 Conn. 490, 492, 677 A.2d 1356 (1996) (concluding that prior decision in same case was incorrectly decided); see also W. Horton & K. Bartschi, Connecticut Practice Series: Connecticut Rules of Appellate Procedure (2017–2018 Ed.) § 71–5, p. 263, authors' comments (identifying, in addition, Cheshire Mortgage Service, Inc. v. Montes , 223 Conn. 80, 612 A.2d 1130 [1992], and Jaconski v. AMF, Inc. , 208 Conn. 230, 543 A.2d 728 [1988] ). In other cases, justices who have taken one position in the original decision have subsequently taken a different position upon reconsideration. See, e.g., Paige v. Saint Andrew's Roman Catholic Church Corp. , 250 Conn. 14, 35, 734 A.2d 85 (1999) (Palmer, J. , concurring) ("Upon reconsideration, I am persuaded that the evidence does not support the jury's verdict, and, therefore, I join the new majority opinion. I write separately to explain briefly why, despite my original vote to affirm the judgment of the trial court, I now vote to reverse that judgment."); State v. Chapman , 229 Conn. 529, 532, 643 A.2d 1213 (1994) (former Chief Justice Peters defected from three to two majority affirming Appellate Court's judgment in State v. Chapman , 227 Conn. 616, 632 A.2d 674 [1993], to join five to two majority reversing Appellate Court's judgment upon reconsideration en banc).

Finally, on multiple occasions, this court has reversed or substantially altered its original opinion in response to a motion for reconsideration following the retirement of a member of the court. See, e.g., State v. Santiago , 318 Conn. 1, 12–14, 122 A.3d 1 (2015) ; Groton v. United Steelworkers of America , 254 Conn. 35, 36 n.1, 757 A.2d 501 (2000) ( United Steelworkers ); State v. Washington , 182 Conn. 419, 420, 438 A.2d 1144 (1980) (substitute opinion for State v. Washington , 42 Conn. L.J., No. 1, p. 10A [July 1, 1980] ); McNamara v. Hamden , 176 Conn. 547, 398 A.2d 1161 (1978) (substitute opinion for McNamara v. Hamden , 39 Conn. L.J., No. 43, p. 6 [April 25, 1978] ). In United Steelworkers , a decision was issued by a five member panel on March 17, 2000; see Groton v. United Steelworkers of America , 252 Conn. 508, 508–509, 747 A.2d 1045 (2000) ; and, on March 21, 2000, Justice Peters reached her seventieth birthday and retired from this court. Groton v. United Steelworkers of America , supra, at 36 n.1, 757 A.2d 501. On April 28, 2000, the court granted a motion for reconsideration en banc, added three new panel members (two of whom had been on the court and a third who had recently joined the court), and reversed the original judgment. See id. Similarly, the original three to two decision in McNamara was overturned upon reconsideration following the retirement of one of the justices in the majority. See McNamara v. Hamden , supra, 176 Conn. at 548, 556, 398 A.2d 1161 ; McNamara v. Hamden , supra, 39 Conn. L.J., No. 43, pp. 6, 9. In Washington , the majority author, Justice Loiselle, was replaced by Justice Parskey for reconsideration. See State v. Washington , supra, 182 Conn. at 429, 438 A.2d 1144 ; State v. Washington , supra, 42 Conn. L.J., No. 1, p. 10A. In Washington , the court shifted directions, reaching the same result but deciding on constitutional grounds what had been initially decided on statutory grounds. See State v. Washington , supra, 182 Conn. at 421, 438 A.2d 1144 ; id., at 429–30, 438 A.2d 1144 (Bogdanski, J. , concurring). Most recently, a full panel of seven justices participated in the initial decision in Santiago , during consideration of which the legislature passed Public Acts 2012, No. 12–5, prospectively abolishing the death penalty in Connecticut. See State v. Santiago , 305 Conn. 101, 101, 307 n.167, 49 A.3d 566 (2012), superseded in part by State v. Santiago , 318 Conn. 1, 122 A.3d 1 (2015). We subsequently granted the defendant's motion for reconsideration and reheard the case with a panel that included three justices who had not participated in the prior decision, eventually superseding the judgment in our original decision.¹³ Compare State v. Santiago , supra, 318 Conn. at 1, 122 A.3d 1 (listing panel members), with State v. Santiago , supra, 305 Conn. at 101, 49 A.3d 566 (same).

13 Relatedly, the proper role of precedent following a recent shift in the composition of the court was addressed at some length by several members of this court in State v. Peeler , 321 Conn. 375, 140 A.3d 811 (2016), in which we rejected the state's claim that our holding in State v. Santiago , supra, 318 Conn. at 1, 122 A.3d 1, declaring the death penalty statute unconstitutional, was ill-advised and should be overruled. In separate concurring opinions in Peeler , former Chief Justice Rogers and Justice Robinson both emphasized that a panel of this court should exercise particular caution when it is asked to overrule a case, of very recent vintage, that previously had been decided by a different panel of the court. See State v. Peeler , supra, at 377–83, 140 A.3d 811 (Rogers, C.J. , concurring); id., at 413–16, 140 A.3d 811 (Robinson, J. , concurring). In a dissenting opinion, Justice Zarella maintained that the court in Peeler was obliged to overrule Santiago , explaining that, "[i]f this court now were to overturn Santiago , it would not be because Justice Robinson replaced Justice Norcott. Certainly, the change in court membership may be a circumstance under which the overruling occurs, but it is nothing more than pure happenstance. Instead, the actual reasons for overruling Santiago ... would be, one, a majority of the justices believes that decision is not supported by the law and, two, after weighing the benefit and costs of stare decisis, a majority of the justices concludes that Santiago is not deserving of stare decisis effect." Id., at 489–90, 140 A.3d 811 (Zarella, J. , dissenting). Justice Espinosa, who also dissented in Peeler on the ground that Santiago should be overruled, took the same position on this issue as Justice Zarella, stating that, "[a]s this court frequently has noted, [i]t is more important that the court should be right upon later and more elaborate consideration of the cases than consistent with previous declarations." (Internal quotation marks omitted.) Id., at 501–502, 140 A.3d 811 (Espinosa, J. , dissenting). "[T]he mere fact that a decision overruling Santiago would have occurred after the panel changed does not necessitate the conclusion that the panel change would have caused the court to overrule Santiago , and is nothing more than a logical fallacy, an example of post hoc, ergo propter hoc reasoning." (Emphasis omitted; internal quotation marks omitted.) Id., at 503, 140 A.3d 811 (Espinosa, J. , dissenting). We cannot understand how Justice Espinosa reasonably can conclude, on the one hand, that it would have been perfectly appropriate for an en banc panel of this court in Peeler to overrule an earlier en banc panel in Santiago —indeed, in Justice Espinosa's view, it was imperative that the court in Peeler overrule Santiago —and also conclude, in her dissenting opinion in this case, that it is improper for the reconstituted en banc panel in the present case, upon reconsideration, to overrule the original panel's decision.
It bears noting that Justice Espinosa's observations have even more force in the present case because, in Peeler , we were required to decide whether to follow the holding of recent prior precedent, namely, Santiago . Thus, whereas Peeler squarely implicated the doctrine of stare decisis, the present case does not because, as the Ohio Supreme Court has observed, the prudential concerns that animate the doctrine of stare decisis have no applicability in the context of a motion for reconsideration. See Rocky River v. State Employment Relations Board , 43 Ohio St. 3d 1, 5, 539 N.E.2d 103 (1989) ("Rocky [River] IV is not a different case than Rocky [River] I [39 Ohio St.3d 196, 530 N.E.2d 1 (1988) ]. It is the same case! Therefore, the doctrine [of stare decisis] cannot apply [to] Rocky [River] IV. "). A motion for reconsideration, unlike a challenge to an earlier, separate decision of the court, is designed "generally to point out errors or omissions in the original decision so that they may be corrected"; E. Prescott, Connecticut Appellate Practice and Procedure (5th Ed. 2016) § 8–5:9.2, p. 496; and such motions, "if meritorious ... may generate a revised opinion." Id., p. 497. The purpose of such motions, in other words, is to ensure the correctness, accuracy and consistency of our decisions when, as in the present case, no party seeks the overruling of prior precedent that otherwise would dictate the result of the case.

As we previously explained, these decisions are well within the broad mandate for reconsideration provided by Practice Book § 71–5. This court's review of a motion for reconsideration is, by design, a case-by-case inquiry intended not only to address unexpected developments in the law and jurisdictional errors, but also to serve as a check on the court's initial conclusions. Thus, when a justice of this court reviews his or her initial decision and finds a mistake, it is incumbent on that justice to change his or her vote accordingly, even if the motion for reconsideration fails to raise a new issue, a new line of reasoning, new facts, or new law. For the same reason, a justice who is added to a panel for purposes of a motion for reconsideration is not obligated to filter his or her consideration of the case through the lens of a predecessor panel member. Indeed, doing so would severely impair the effectiveness of motions for reconsideration as implements for ensuring the consistency and accuracy of our decisions.

Despite this wealth of authority, Justice Espinosa nevertheless contends in her dissenting opinion that, by adding Justice D'Auria to the panel in the present case, we have departed from this court's policy with respect to such motions. More specifically, Justice Espinosa claims that, ordinarily, when a member of this court has resigned from the bench after the issuance of an opinion but before any motion for reconsideration en banc in the case has been decided, we have not added a panel member. This argument is wholly unpersuasive, however, because the cases on which she relies are inapposite. None of them involved an evenly divided panel, as in the present case, and, consequently, no additional justice was needed in those cases to break a deadlock. See, e.g., Tomick v. United Parcel Service, Inc. , 324 Conn. 470, 472, 486, 153 A.3d 615 (2016) (original panel vote of four to two); Harris v. Bradley Memorial Hospital & Health Center, Inc. , 306 Conn. 304, 308, 339, 50 A.3d 841 (2012) (original panel vote was unanimous), cert. denied, 569 U.S. 918, 133 S.Ct. 1809, 185 L.Ed.2d 812 (2013) ; State v. Drupals , 306 Conn. 149, 151, 173, 49 A.3d 962 (2012) (same). We therefore reject Justice Espinosa's unfounded contention that it is somehow improper for the court to add a seventh panel member to decide the petitioner's motion.

Insofar as Justice Espinosa accuses the petitioner of judge shopping, we would simply note the obvious—namely, the petitioner had no role in Justice Zarella's decision to leave the court before the time period lapsed for filing a motion for reconsideration en banc, the petitioner timely filed that motion, and he had a legal right to file such a petition. The petitioner having garnered a favorable decision on his request to replace the seventh member of the en banc panel, we turn to the merits of the petitioner's motion.¹⁴ III

14 Justice Espinosa also contends that we should have resolved the petitioner's motion before Justice D'Auria became a member of this court. We disagree. As the preceding discussion reflects, substantial research and deliberation were necessary to reach a considered decision on this threshold issue of first impression for our court. Justice D'Auria was nominated and confirmed well before that process was completed. In other words, the remaining six panel members reasonably could not have considered and decided the issue prior to Justice D'Auria's confirmation as a member of this court.
We note, as well, that Justice Espinosa makes a number of other baseless claims and accusations. None of them warrants a response.

LEGAL PRINCIPLES GOVERNING CLAIMS OF INEFFECTIVE ASSISTANCE OF COUNSEL

Before addressing the merits of the petitioner's claim in the context of the habeas court's decision, we set forth the principles that guide our review of that claim. Under the sixth amendment to the United States constitution, a criminal defendant is guaranteed the right to the effective assistance of counsel. See, e.g., Strickland v. Washington , 466 U.S. 668, 687, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984) ; see also Evitts v. Lucey , 469 U.S. 387, 392, 105 S.Ct. 830, 83 L.Ed.2d 821 (1985) (sixth amendment right to effective assistance of counsel is made applicable to states through due process clause of fourteenth amendment to United States constitution). "The [s]ixth [a]mendment recognizes [this fundamental right] because it envisions [that counsel will play] a role that is critical to the ability of the adversarial system to produce just results." (Internal quotation marks omitted.) Kimmelman v. Morrison , 477 U.S. 365, 394, 106 S.Ct. 2574, 91 L.Ed.2d 305 (1986) ; see id., at 377, 106 S.Ct. 2574. That role is a vital one because "access to counsel's skill and knowledge is necessary to accord defendants the ample opportunity to meet the case of the prosecution to which they are entitled"; (internal quotation marks omitted) Strickland v. Washington , supra, at 685, 104 S.Ct. 2052 ; an opportunity that is essential if "the adversarial testing process [is to] work in the particular case." (Internal quotation marks omitted.) Kimmelman v. Morrison , supra, at 384, 106 S.Ct. 2574. Thus, because "[a]n accused is entitled to be assisted by an attorney, whether retained or appointed, who plays the role necessary to ensure that the trial is fair"; (internal quotation marks omitted) id., at 377, 106 S.Ct. 2574 ; "[t]he benchmark for judging any claim of ineffectiveness must be whether counsel's conduct so undermined the proper functioning of the adversarial process that the trial cannot be relied on as having produced a just result." Strickland v. Washington , supra, at 686, 104 S.Ct. 2052.

To determine whether a defendant is entitled to a new trial due to a breakdown in the adversarial process caused by counsel's inadequate representation, we apply the familiar two part test adopted by the court in Strickland . "A convicted defendant's claim that counsel's assistance was so defective as to require reversal of a conviction ... has two components. First, the defendant must show that counsel's performance was deficient. This requires [a] showing that counsel made errors so serious that counsel was not functioning as the 'counsel' guaranteed the defendant by the [s]ixth [a]mendment. Second, the defendant must show that the deficient performance prejudiced the defense. This requires [a] showing that counsel's errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable. Unless a defendant makes both showings, it cannot be said that the conviction ... resulted from a breakdown in the adversary process that renders the result unreliable." Id., at 687, 104 S.Ct. 2052. The sixth amendment, therefore, "does not guarantee perfect representation, only a reasonably competent attorney.... Representation is constitutionally ineffective only if it so undermined the proper functioning of the adversarial process that the defendant was denied a fair trial." (Citations omitted; internal quotation marks omitted.) Harrington v. Richter , 562 U.S. 86, 110, 131 S.Ct. 770, 178 L.Ed.2d 624 (2011).

A

The Performance Prong

With respect to the first component of the Strickland test, "the proper standard for attorney performance is that of reasonably effective assistance." Strickland v. Washington , supra, 466 U.S. at 687, 104 S.Ct. 2052. Consequently, to establish deficient performance by counsel, a defendant must show that, considering all of the circumstances, counsel's representation fell below an objective standard of reasonableness as measured by prevailing professional norms. Id., at 687–88, 104 S.Ct. 2052.

Moreover, strategic decisions of counsel, although not entirely immune from review, are entitled to substantial deference by the court. "Judicial scrutiny of counsel's performance must be highly deferential. It is all too tempting for a defendant to second-guess counsel's assistance after conviction or adverse sentence, and it is all too easy for a court, examining counsel's defense after it has proved unsuccessful, to conclude that a particular act or omission of counsel was unreasonable.... A fair assessment of attorney performance requires that every effort be made to eliminate the distorting effects of hindsight, to reconstruct the circumstances of counsel's challenged conduct, and to evaluate the conduct from counsel's perspective at the time. Because of the difficulties inherent in making the evaluation, a court must indulge a strong presumption that counsel's conduct falls within the wide range of reasonable professional assistance; that is, the defendant must overcome the presumption that, under the circumstances, the challenged action might be considered sound trial strategy." (Citation omitted; internal quotation marks omitted.) Id., at 689, 104 S.Ct. 2052. This is so because "[t]here are countless ways to provide effective assistance in any given case"; id. ; and "[e]ven the best criminal defense attorneys would not defend a particular client in the same way." Id.

The right to the effective assistance of counsel applies no less to the investigative stage of a criminal case than it does to the trial phase. Indeed, in Strickland , the court explained that the foregoing performance "standards require no special amplification in order to define counsel's duty to investigate .... [Simply stated] ... strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation. In other words, counsel has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary." Id., at 690–91, 104 S.Ct. 2052. That is, counsel's decision to forgo or truncate an investigation "must be directly assessed for reasonableness in all the circumstances ...." Id., at 691, 104 S.Ct. 2052. "In assessing the reasonableness of an attorney's investigation ... a court must consider not only the quantum of evidence already known to counsel, but also whether the known evidence would lead a reasonable attorney to investigate further." Wiggins v. Smith , 539 U.S. 510, 527, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003). In addition, in contrast to our evaluation of the constitutional adequacy of counsel's strategic decisions, which are entitled to deference, when the issue is whether "the investigation supporting counsel's [strategic] decision" to proceed in a certain manner "was itself reasonable"; (emphasis altered) id., at 523, 123 S.Ct. 2527 ; we "must conduct an objective review of [the reasonableness of counsel's] performance ...." (Citations omitted.) Id. Thus, "deference to counsel's strategic decisions does not excuse an inadequate investigation ...." (Citation omitted.) Williams v. Stephens , 575 Fed. Appx. 380, 386 (5th Cir.), cert. denied, ––– U.S. ––––, 135 S.Ct. 875, 190 L.Ed.2d 709 (2014). Although the reasonableness of any particular investigation necessarily depends on the unique facts of any given case; see Strickland v. Washington , supra, 466 U.S. at 688–89, 104 S.Ct. 2052 ; counsel has certain baseline investigative responsibilities that must be discharged in every criminal matter. "It is the duty of the [defense] lawyer to conduct a prompt investigation of the circumstances of the case and to explore all avenues leading to facts relevant to the merits of the case ...." (Internal quotation marks omitted.) Rompilla v. Beard , 545 U.S. 374, 387, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005). This duty exists irrespective of whether the defendant is helpful to counsel by providing information pertinent to his defense or whether he provides no such assistance. See, e.g., id., at 381, 125 S.Ct. 2456. Thus, "[a]n attorney's duty of investigation requires more than simply checking out the witnesses that the client himself identifies." Bigelow v. Haviland , 576 F.3d 284, 288 (6th Cir. 2009) ; see also id., at 288–89 ("[Defense counsel] had no reasonable basis for assuming that [the petitioner's] lack of information about still more witnesses meant that there were none to be found.... With every effort to view the facts as a defense lawyer would have [viewed them] at the time, it is difficult to see how [defense counsel] could have failed to realize that without seeking information that could either corroborate the alibi or contextualize it for the jury, he was seriously compromis[ing] [his] opportunity to present an alibi defense." [Citations omitted; internal quotation marks omitted.] ).

Of course, "the duty to investigate does not force defense lawyers to scour the globe on the off chance something will turn up; reasonably diligent counsel may draw a line when they have good reason to think further investigation would be a waste." Rompilla v. Beard , supra, 545 U.S. at 383, 125 S.Ct. 2456. In other words, counsel is not required to conduct an investigation that "promise[s] less than looking for a needle in a haystack, when a lawyer truly has reason to doubt there is any needle there." Id., at 389, 125 S.Ct. 2456. Because, however, "[p]retrial investigation and preparation are ... [key] to effective representation [by] counsel"; (internal quotation marks omitted) Daniels v. Woodford , 428 F.3d 1181, 1203 (9th Cir. 2005), cert. denied sub nom. Ayers v. Daniels , 550 U.S. 968, 127 S.Ct. 2876, 167 L.Ed.2d 1152 (2007) ; see also House v. Balkcom , 725 F.2d 608, 618 (11th Cir.) ("[p]retrial investigation, principally because it provides a basis [on] which most of the defense case must rest, is, perhaps, the most critical stage of a lawyer's preparation"), cert. denied, 469 U.S. 870, 105 S.Ct. 218, 83 L.Ed.2d 148 (1984) ; counsel is not free to simply ignore or disregard potential witnesses who might be able to provide exculpatory testimony. See, e.g., Blackmon v. Williams , 823 F.3d 1088, 1105 (7th Cir. 2016) ("Just one [potential] witness might have been able to give [the petitioner] a true alibi. At a minimum, all of [the potential witnesses] could have bolstered his [alibi] claim .... It is not reasonable strategy to leave such possible testimony unexplored [in such] circumstances."); Gersten v. Senkowski , 426 F.3d 588, 610 (2d Cir. 2005) (defense counsel rendered ineffective assistance in concluding investigation prematurely because he "never discovered any evidence to suggest one way or another whether [further investigation] would be counterproductive or such investigation fruitless, nor did counsel have any reasonable basis to conclude that such investigation would be wasteful"), cert. denied sub nom. Artus v. Gersten , 547 U.S. 1191, 126 S.Ct. 2882, 165 L.Ed.2d 894 (2006). Similarly, a decision by counsel to forgo an investigation into the possible testimony of a potentially significant witness is constitutionally impermissible unless counsel has a sound justification for doing so; speculation, guesswork or uninformed assumptions about the availability or import of that testimony will not suffice. Instead, counsel must seek to interview the witness to determine the value of any testimony that he may be able to provide. See, e.g., Ramonez v. Berghuis , 490 F.3d 482, 489 (6th Cir. 2007) ("[c]onstitutionally effective counsel must develop trial strategy in the true sense—not what bears a false label of 'strategy'—based on what investigation reveals witnesses will actually testify to, not based on what counsel guesses they might say in the absence of a full investigation"); Pavel v. Hollins , 261 F.3d 210, 221 (2d Cir. 2001) (defense counsel never contacted potentially favorable witness because counsel was "confident as to what [that] witness would say," but "counsel's anticipation [of that testimony] does not excuse the failure to find out" [internal quotation marks omitted] ). In other words, "counsel's anticipation of what a potential witness would say does not excuse the failure to find out; speculation cannot substitute for certainty." United States v. Moore , 554 F.2d 1086, 1093 (D.C. Cir. 1976). In the same vein, when counsel's failure to proceed with an investigation is due not to professional or strategic judgment but, instead, results from oversight, inattention or lack of thoroughness and preparation, no deference or presumption of reasonableness is warranted. See, e.g., Carter v. Duncan , 819 F.3d 931, 942 (7th Cir. 2016) ("[t]he consequences of inattention rather than reasoned strategic decisions are not entitled to the presumption of reasonableness" [internal quotation marks omitted] ); Wilson v. Mazzuca , 570 F.3d 490, 502 (2d Cir. 2009) (errors warranting determination of sixth amendment violation include "omissions [that] cannot be explained convincingly as resulting from a sound trial strategy, but [rather, that] arose from oversight, carelessness, ineptitude, or laziness" [internal quotation marks omitted] ).

With specific regard to the duty to investigate a defendant's alibi defense, counsel is obligated to make all reasonable efforts to identify and interview potential alibi witnesses. See, e.g., Towns v. Smith , 395 F.3d 251, 259 (6th Cir. 2005) ("Without even attempting to interview [the witness], counsel simply decided not to call him as a witness. That decision was objectively unreasonable because it was a decision made without undertaking a full investigation into whether [the witness] could assist in [the petitioner's] defense.... By failing even to contact [the witness] ... counsel abandoned his investigation at an unreasonable juncture, making a fully informed decision with respect to [whether to have the witness testify] impossible." [Citation omitted; internal quotation marks omitted.] ); Bryant v. Scott , 28 F.3d 1411, 1415 (5th Cir. 1994) ("[A]n attorney must engage in a reasonable amount of pretrial investigation and at a minimum ... interview potential witnesses and ... make an independent investigation of the facts and circumstances in the case.... [W]hen alibi witnesses are involved, it is unreasonable for counsel not to try to contact the witnesses and ascertain whether their testimony would aid the defense." [Citations omitted; internal quotation marks omitted.] ). Furthermore, a thorough investigation of an alibi defense is especially important when "the missing witness is disinterested in a case in which the other witnesses have a relationship to the defendant." Carter v. Duncan , supra, 819 F.3d at 943 ; see also Blackmon v. Williams , supra, 823 F.3d at 1104–1105 (explaining that unreasonableness of counsel's failure to investigate was compounded by "significant potential benefits of obtaining alibi testimony from witnesses unimpaired by family ties to [the petitioner]"); Montgomery v. Petersen , 846 F.2d 407, 413 (7th Cir. 1988) (characterizing disinterested alibi witness who defense counsel unreasonably failed to identify and locate as "extraordinarily significant" when all twelve alibi witnesses were either relatives or close friends of petitioner). Finally, we, like other courts, have identified several nonexclusive factors to be considered in determining whether counsel's failure to investigate and present the testimony of an additional alibi witness or witnesses was reasonable under the circumstances. They include (1) the importance of the alibi to the defense; see Gaines v. Commissioner of Correction , 306 Conn. 664, 674–75, 51 A.3d 948 (2012) ; (2) the significance of the witness' testimony to the alibi; see id., at 688, 51 A.3d 948 ; (3) the ease with which the witness could have been discovered; see id., at 685–86, 51 A.3d 948 ; and (4) the gravity of the criminal charges and the magnitude of the sentence that the petitioner faced. See id., at 684, 51 A.3d 948.¹⁵ B

15 See Blackmon v. Williams , supra, 823 F.3d at 1104–1105 (noting "significant potential benefits of obtaining alibi testimony from witnesses unimpaired by family ties" to petitioner); Bemore v. Chappell , 788 F.3d 1151, 1164 (9th Cir. 2015) ("[c]ounsel's duty to investigate and to prepare his client's [alibi] defense becomes especially pressing [when] ... the [alibi] witnesses and their credibility ... are crucial" [internal quotation marks omitted] ), cert. denied sub nom. Davis v. Bemore , ––– U.S. ––––, 136 S.Ct. 1173, 194 L.Ed.2d 241 (2016) ; Mosley v. Atchison , 689 F.3d 838, 848–49 (7th Cir. 2012) (counsel's failure to investigate additional alibi witnesses was unreasonable when petitioner's whereabouts at time of crime was central issue at criminal trial); Greiner v. Wells , 417 F.3d 305, 322 (2d Cir. 2005) ("[i]n nearly every case that concludes that counsel conducted a constitutionally deficient investigation, the courts point to readily available evidence neglected by counsel"), cert. denied sub nom. Wells v. Ercole , 546 U.S. 1184, 126 S.Ct. 1363, 164 L.Ed.2d 72 (2006) ; Huffington v. Nuth , 140 F.3d 572, 580–81 (4th Cir.) (courts are especially unsympathetic to counsel's failure to interview important, prospective witnesses when those witnesses were readily available), cert. denied, 525 U.S. 981, 119 S.Ct. 444, 142 L.Ed.2d 399 (1998) ; see also Gregg v. Rockview , 596 Fed. Appx. 72, 77 (3d Cir. 2015) ("[e]specially given the gravity of the criminal charges [the petitioner] was facing, counsel could not have reasonably elected to rely exclusively on [one witness] and forgo any investigation into [another]"); Raygoza v. Hulick , 474 F.3d 958, 964 (7th Cir.) ("[i]n a first-degree murder trial, it is almost impossible to see why a lawyer would not at least have investigated the alibi witnesses more thoroughly"), cert. denied sub nom. Randolph v. Raygoza , 552 U.S. 1033, 128 S.Ct. 613, 169 L.Ed.2d 413 (2007) ; Bryant v. Scott , supra, 28 F.3d at 1417–18 (noting importance of seriousness of offense and gravity of punishment in determining reasonableness); Coleman v. Brown , 802 F.2d 1227, 1234 (10th Cir. 1986) (noting gravity of punishment in determining reasonableness), cert. denied, 482 U.S. 909, 107 S.Ct. 2491, 96 L.Ed.2d 383 (1987).

The Prejudice Prong

When defense counsel's performance fails the reasonableness test, a new trial is required if there exists "a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Strickland v. Washington , supra, 466 U.S. at 694, 104 S.Ct. 2052. The question, therefore, "is whether there is a reasonable probability that, absent the errors, the [fact finder] would have had a reasonable doubt respecting guilt." Id., at 695, 104 S.Ct. 2052. "A reasonable probability is a probability sufficient to undermine confidence in the outcome." Id., at 694, 104 S.Ct. 2052. It is also clear, however, that "a defendant need not show that counsel's deficient conduct more likely than not altered the outcome of the case"; id., at 693, 104 S.Ct. 2052 ; because "[t]he result of a [criminal] proceeding can be rendered unreliable, and [thus] the proceeding itself unfair, even if errors of counsel cannot be shown by a preponderance of the evidence to have determined the outcome." Id., at 694, 104 S.Ct. 2052. The defendant must establish, instead, that counsel's constitutionally inadequate representation gives rise to a loss of confidence in the verdict. In evaluating such a claim, "the ultimate focus of inquiry must be on the fundamental fairness of the proceeding whose result is being challenged. In every case the court should be concerned with whether, despite the strong presumption of reliability, the result of the particular proceeding is unreliable because of a breakdown in the adversarial process that our system counts on to produce just results." Id., at 696, 104 S.Ct. 2052. Of course, a reviewing court does not conduct this inquiry in a vacuum. Rather, the court "must consider the totality of the evidence before the judge or jury. Some of the factual findings will have been unaffected by the errors, and factual findings that were affected will have been affected in different ways. Some errors will have had a pervasive effect on the inferences to be drawn from the evidence, altering the entire evidentiary picture, and some will have had an isolated, trivial effect. Moreover, a verdict or conclusion only weakly supported by the record is more likely to have been affected by errors than one with overwhelming record support. Taking the unaffected findings as a given, and taking due account of the effect of the errors on the remaining findings, a court making the prejudice inquiry must ask if the [petitioner] has met the burden of showing that the decision reached would reasonably likely have been different absent the errors." Id., at 695–96, 104 S.Ct. 2052. Furthermore, because our role in examining the state's case against the petitioner is to evaluate the strength of that evidence and not its sufficiency, we do not consider the evidence in the light most favorable to the state. See Lapointe v. Commissioner of Correction , 316 Conn. 225, 342 n.88, 112 A.3d 1 (2015) ; see also Tice v. Johnson , 647 F.3d 87, 110 (4th Cir. 2011) ("We are not bound ... to view the facts in the light most favorable to the prosecution. The familiar [evidentiary sufficiency] analysis centering on whether a reasonable jury could have [found] an adequately represented defendant [guilty] is considerably more deferential than the Strickland test for prejudice in an [ineffective assistance] case, which seeks only to discover whether the absence of error would have given rise to a reasonable probability of acquittal, such that confidence in the verdict is undermined."). Rather, we are required to undertake an objective review of the nature and strength of the state's case. See Lapointe v. Commissioner of Correction , supra, at 342 n.88, 112 A.3d 1 ;¹⁶ see also

16 Although we most recently made this observation in Lapointe v. Commissioner of Correction , supra, 316 Conn. at 342 n.88, 112 A.3d 1, a case involving our analysis of a claim that habeas counsel rendered ineffective assistance in failing to demonstrate that the state had improperly withheld exculpatory evidence in violation of Brady v. Maryland , 373 U.S. 83, 87, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963) ; see Lapointe v. Commissioner of Correction , supra, at 229, 112 A.3d 1 ; it is undisputed that the test for prejudice under Strickland is identical to the test for materiality under Brady . See, e.g., Lapointe v. Commissioner of Correction , supra, at 266–67, 112 A.3d 1 ; see also id., at 366, 112 A.3d 1 (Zarella, J. , dissenting) (same). Consequently, the nature of our review is precisely the same for both categories of cases.

Williams v. Taylor , 529 U.S. 362, 397–98, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000) (concluding that habeas court, in conducting its prejudice analysis, improperly failed to consider evidence favorable to petitioner); Elmore v. Ozmint , 661 F.3d 783, 868 (4th Cir. 2011) (habeas court "unreasonably broke from Strickland by considering less than the totality of the evidence, and [engaged in an analysis that] unreasonably discounted evidence favorable to [the petitioner] by unduly minimizing its import and evaluating it piecemeal"). "In assessing prejudice under Strickland , the question is not whether a court can be certain counsel's performance had no effect on the outcome or whether it is possible a reasonable doubt might have been established if counsel acted differently.... Instead, Strickland asks whether it is reasonably likely the result would have been different.... This does not require a showing that counsel's actions more likely than not altered the outcome, but the difference between Strickland 's prejudice standard and a more-probable-than-not standard is slight and matters only in the rarest case.... The likelihood of a different result must be substantial, not just conceivable." (Citations omitted; internal quotation marks omitted.) Harrington v. Richter , supra, 562 U.S. at 111–12, 131 S.Ct. 770.

C

Standard of Review

It is well established that "[t]he habeas court is afforded broad discretion in making its factual findings, and those findings will not be disturbed unless they are clearly erroneous.... Historical facts constitute a recital of external events and the credibility of their narrators.... Accordingly, [t]he habeas [court], as the trier of facts, is the sole arbiter of the credibility of witnesses and the weight to be given to their testimony.... The application of the habeas court's factual findings to the pertinent legal standard, however, presents a mixed question of law and fact, which is subject to plenary review." (Citations omitted; internal quotation marks omitted.) Gaines v. Commissioner of Correction , supra, 306 Conn. at 677, 51 A.3d 948.

IV

THE HABEAS TRIAL

A

Additional Facts

At the habeas trial, the petitioner sought to establish that Michael Sherman had performed deficiently because he made no effort to learn the identity of the "beau" who Georgeann Dowdle claimed had been with her on the night of October 30, 1975, and to ascertain whether her beau could provide disinterested corroboration of the petitioner's alibi. To that end, he presented the testimony of Denis Ossorio, a seventy-two year old retired psychologist at the time of the petitioner's habeas proceedings. Ossorio testified that, in 1975, he resided in Greenwich and operated an employment related program for women. Ossorio further testified that he was at the Terrien home on the evening of October 30, 1975, visiting Dowdle, with whom he had a personal relationship. According to Ossorio, the petitioner and two of his brothers were also there that evening, watching television with James Terrien, and he joined them in the television room periodically when Dowdle was otherwise occupied with her daughter. Ossorio recalled leaving the Terrien residence at about midnight and was not sure whether the Skakels had left before him. Ossorio further stated that he was living in Greenwich at the time of the petitioner's criminal trial and would have been available to testify, but no one from Sherman's office or the office of the state's attorney ever contacted him. Ossorio also explained that, although he was aware that the petitioner had been charged with the victim's murder, he did not pay close attention to the trial itself and was unaware that his recollection of the events of the evening of October 30, 1975, had any particular significance to the case.

The petitioner also presented the testimony of Michael Fitzpatrick, a prominent Connecticut attorney and past president of the Connecticut Criminal Defense Lawyers Association who specializes in criminal defense and civil litigation. Fitzpatrick testified that, on the basis of his expertise and experience in criminal law, it was his opinion that any reasonably competent criminal defense attorney, after receiving and reviewing Dowdle's grand jury testimony, "absolutely" would have ascertained Ossorio's identity and made reasonable efforts to locate and interview him. That investigation was required, according to Fitzpatrick, because it was incumbent on Sherman to confirm that Ossorio was present at the Terrien residence on October 30, 1975, and, if he in fact had been present, to ascertain whether his recollection of events would strengthen the petitioner's alibi defense. In particular, Fitzpatrick explained that, if Ossorio recalled that the petitioner was present at the Terrien home that evening, that testimony would have "[made] it impossible for the state to argue in summation that there [was] not a single independent [alibi] witness in the case, which was one of the chief grounds the state asserted for rejecting the alibi." Fitzpatrick further testified that Sherman's failure to identify and interview Ossorio "absolutely prejudiced" the petitioner because "it deprived [him] ... of the opportunity to present an independent alibi witness" who would have significantly enhanced the credibility of the petitioner's defense. On cross-examination, the respondent challenged Fitzpatrick's opinion that Sherman had rendered ineffective assistance by failing to present Ossorio's testimony, but adduced no expert testimony of its own on that issue.

Jason Throne, who served as Sherman's cocounsel at trial, also was a witness at the habeas trial. Throne testified that the petitioner's alibi was "extremely important" to the defense. When Thorne was asked if he and Sherman were "eager to find anyone who could corroborate [the alibi]," he responded, "[a]bsolutely, without question." Throne further stated that, "even more importantly," he and Sherman were "especially eager to find a nonfamily member who could corroborate it" because of the "obvious concern" that, because all of the alibi witnesses were family members, "the jury would perceive all of [them] as having bias and a motivation to lie or distort facts or truth, which wasn't the case.... I wish that we had even a single witness that wasn't blood related to include in that group [who] could have testified to the same facts that everyone else testified to, to establish that [the petitioner] was not there the night of the murder."

Sherman testified at the habeas trial, as well. When Sherman was asked whether the alibi was the petitioner's "principal defense" at trial, he responded, "[a]bsolutely ...." He also stated that it would have been "very important" to have an alibi witness who was not related to the petitioner and that, if he had located one, he would have had him testify, "[w]ithout a doubt." Sherman also acknowledged reading Dowdle's grand jury testimony prior to trial, including her statement that her beau was with her at the Terrien home the evening of October 30, 1975. When Sherman was asked why he had never inquired into the identity of Dowdle's beau, he responded: "I had no reason to suspect that he, in fact, would be helpful in that he saw [the petitioner] and the rest of the boys." In response to questioning from the respondent's counsel, Sherman indicated that, because Dowdle had testified before the grand jury that she "really didn't venture out" of the room on the evening of October 30, 1975, Ossorio, her guest, might well have stayed in the library, as well. Sherman also acknowledged that, because Dowdle recalled hearing but not seeing her Skakel cousins that evening, Ossorio also may not have seen the four boys.

B

Findings and Conclusions of the Habeas Court

On the basis of the foregoing evidence, the habeas court concluded that Sherman's performance was constitutionally deficient in that Sherman failed to identify Ossorio and to present his testimony to the jury. According to the habeas court, "Ossorio's testimony supported the petitioner's claim that, during the likely time of the murder, he was away from Belle Haven, as he indicated. To the [habeas] court, Ossorio was a disinterested and credible witness with a clear recollection of seeing the petitioner at the Terrien home on the evening in question. He testified credibly that not only was he present in the home with Dowdle and that he saw the petitioner there, but that he lived in the area throughout the time of the trial and would have readily been available to testify if asked." The habeas court further concluded that Sherman "was on notice from Dowdle's grand jury testimony that she was in the company of another person at the Terrien home, and she had identified this person as her beau.... Had ... Sherman made reasonable inquiry, he would have discovered Ossorio and gleaned that Ossorio was prepared to testify that the petitioner was present at the Terrien home during the evening in question. He would have learned, as well, that Ossorio was a disinterested and credible witness."¹⁷ (Internal quotation marks omitted.) The habeas court further concluded that the petitioner's defense was prejudiced by Sherman's failure to call Ossorio because, if the jury had heard his testimony, there was a reasonable probability that the outcome of the trial would have been different. The habeas court based that determination, in part, on the fact that the state's attorney had "vigorously contested the petitioner's claimed absence from the area [of the murder] between the hours of 9:15 ... and 11:15 p.m. Indeed, a fair reading of [the state's attorney's] closing argument suggests that he, too, acknowledged the strength of evidence that the victim likely had died at approximately 10 p.m. For example, while [the state's attorney] argued to the jury that the time of death was not integral to the charging document and that the [jurors] could find the petitioner guilty even if they believed his alibi, [the state's attorney] strenuously argued that the petitioner had not, in fact, gone to the Terrien residence as [he] claimed, and that it was the [petitioner's] presence at the crime scene at approximately 10 p.m. that likely caused ... [Helen Ix'] dog to bark in such an unusually disturbed manner. Additionally, even though the [state's attorney] adduced evidence that the time of death could have been any time between 9:30 p.m. [on October 30] and 1 a.m. ... the next day, there was weighty evidence that the murder took place while the petitioner claimed to have been absent from the Belle Haven area."

17 We fully agree with all aspects of Justice D'Auria's concurring opinion, in which he underscores, among other things, that the habeas court did not credit Sherman's habeas trial testimony that, in light of Dowdle's grand jury testimony that she could not recall actually observing the petitioner at the Terrien home on the night of the murder, Sherman had made a conscious decision not to investigate Dowdle's beau because he believed it also was unlikely that her beau would have observed the petitioner. In concluding, rather, that "Sherman's failure to investigate in this regard cannot be attributed to any strategic decision" and, therefore, was entitled to no deference, the habeas court undoubtedly attributed that failure to oversight or inattention. Indeed, as Justice D'Auria aptly notes, the habeas court declined to credit Sherman's explanation for failing to investigate Ossorio because that explanation is belied by the trial strategy Sherman actually employed at the petitioner's criminal trial, as reflected in the record of that trial. As Justice D'Auria notes, "Sherman came to the criminal trial equipped with a 1975 police report reflecting that, merely nine days after the murder, [Dowdle] ... indicated that she had indeed 'observed her brother [Terrien] and the Skakel brothers, [Rushton Skakel, Jr., John Skakel and the petitioner], return to [the Terrien] house sometime around 10 [p.m. on the night of the murder.]' Sherman partially succeeded in getting Dowdle to recount to the jury what the report said she had told [the] police in 1975 and that her memory of the night of the murder would have been better in 1975 than in 1998." Thus, it was Sherman's position at trial that Dowdle did observe the petitioner at the Terrien home that night, thereby defeating the respondent's belated claim in the habeas court that Sherman's failure to investigate Ossorio was founded on his understanding that Dowdle did not observe the petitioner there.
It is likely for the same reason that the habeas court never made an express finding crediting Sherman's testimony that he actually had read Dowdle's grand jury testimony. With respect to that issue, the habeas court stated only that "[Sherman] was on notice from Dowdle's grand jury testimony that she was in the company of another person at the Terrien home, and she had identified this person as her beau.... Had ... Sherman read and considered Dowdle's grand jury testimony, which was made available to him before the [petitioner's criminal] trial, he would have learned of the presence of an unrelated person in the Terrien household." (Internal quotation marks omitted.) It is apparent from this statement that, in the habeas court's view, regardless of whether Sherman read Dowdle's grand jury testimony, he was on notice of that testimony, and, consequently, he was required to make reasonable efforts to follow up on it. We fully agree with the habeas court's analysis in this regard. We note, further, that, if Sherman had not read Dowdle's grand jury testimony, or read it but failed to note the potential significance of that testimony with respect to her beau due to inattention or otherwise, Sherman's failure to take any further action with respect to that testimony cannot possibly be considered reasonable. In any event, for present purposes, we assume that Sherman did read Dowdle's testimony and, as he testified, elected not to pursue it. We make that assumption, however, only because it is the position that the respondent and the majority in Skakel v. Commissioner of Correction , supra, 325 Conn. at 426, 159 A.3d 109, have taken. As we explain hereinafter, that view of the habeas court's decision, even if credited, does not support that position.

Finally, with respect to the issue of prejudice, the habeas court noted "that the jury deliberated for four days, beginning on June 4, 2002, and reach[ed] a verdict on June 7, 2002. During the jury's deliberations, on June 5, 2002, the jury asked to have read back the testimony of Julie Skakel, Andrea Shakespeare and ... Ix. With this request, the jury also provided a note, which stated in [relevant] part: 'We would like to limit ... Ix' testimony to the discussion of who was in the driveway and who left in the car.' Significantly, the focus of the testimony of each of these witnesses was whether the petitioner had left the Belle Haven area at approximately 9:15 p.m. in the Lincoln [Continental] to go to the Terrien residence. Thus, even though the [trial] court charged the jury that [it] need not fix the time of death in order to find the petitioner guilty, the jury showed particular interest in the petitioner's whereabouts between 9:15 ... and 11:15 p.m.

"Given the weighty evidence that the victim was murdered in the time range of 9:30 ... to 10 p.m. on October 30, 1975, the importance of the petitioner's alibi defense to the fact finders cannot fairly be discounted. And, given the importance of the petitioner's alibi defense, its persuasiveness would have been greatly enhanced by the testimony of Ossorio, an independent and credible witness to the petitioner's presence at the Terrien household during the relevant evening hours of October 30, 1975." On the basis of these and other related findings, the habeas court concluded that Sherman's failure to call Ossorio as a witness entitled the petitioner to a new trial.

V

ANALYSIS

A

Michael Sherman's Deficient Performance

As we previously explained, the reasonableness of Sherman's decision not to investigate whether Georgeann Dowdle's beau could provide testimony favorable to the petitioner's alibi defense turns on the facts of the case and, more particularly, the circumstances pertaining to that defense and the potential witness. In light of the various relevant factors—the importance of the petitioner's alibi defense, the significance of Denis Ossorio's testimony to that defense, the ease with which Ossorio could have been located, and the gravity of the charges and potential punishment that the petitioner faced—it is abundantly clear that Sherman's decision to disregard Dowdle's grand jury testimony about her beau, a decision based solely on Sherman's belief that any inquiry into that subject matter would have been fruitless, was unreasonable.

First, as Sherman testified, and the state conceded at trial, the petitioner's alibi was his primary defense to the state's case against him. This is because, although the state contended that it was possible that the victim was murdered as late as 1 a.m. on October 31, 1975, the substantial weight of the evidence indicated that the murder most likely was committed between 9:30 and 10 p.m. on October 30. Consequently, because the state was required to disprove the petitioner's alibi beyond a reasonable doubt; see, e.g., State v. Butler , 207 Conn. 619, 631, 543 A.2d 270 (1988) (defendant in criminal case is entitled to instruction that state must rebut alibi defense beyond reasonable doubt); if the jury believed the petitioner's alibi witnesses—indeed, even if the petitioner's witnesses merely raised a reasonable doubt in the jurors' minds as to the petitioner's whereabouts between 9:30 and 10 p.m.—there is a good likelihood that the petitioner would have been acquitted.

The importance of the petitioner's alibi defense is underscored by how vigorously the state sought to discredit it. The state's attorney claimed that it had been concocted by the Skakel family and founded on the perjurious testimony of the petitioner's alibi witnesses. The state's attorney spent a considerable amount of time, both in adducing testimony from the state's witnesses and in cross-examining the petitioner's witnesses, as well as during closing argument, attempting to demonstrate that the petitioner's alibi had been fabricated. It is likely that the state's attorney challenged the petitioner's alibi so aggressively because, as the Supreme Court of New Jersey has observed, "few defenses have greater potential for creating reasonable doubt as to a defendant's guilt in the minds of the [jurors than an alibi]." (Internal quotation marks omitted.) State v. Porter , 216 N.J. 343, 353, 80 A.3d 732 (2013). Conversely, the state's attorney made no effort to establish any narrative to explain how the victim could have been murdered after 11 p.m. In particular, the state's attorney never presented any evidence as to why the victim would have remained out past her curfew in cold weather, where she could have been between 9:30 and 11 p.m., with whom she could have been during that period of time, or why the extensive police investigation into her whereabouts never yielded a single credible piece of information relating to those matters. Because the state's attorney adduced no such evidence, his closing argument contained no mention of any scenario pursuant to which the murder could have occurred as late as 11 p.m.

Second, it could hardly have been easier for Sherman to have ascertained that Ossorio had critical alibi testimony to offer, such that even the most rudimentary of inquiries would have led Sherman directly and immediately to Ossorio. See, e.g., Rompilla v. Beard , supra, 545 U.S. at 389, 125 S.Ct. 2456 (explaining that "[t]he unreasonableness of attempting no more than [counsel] did was heightened by the easy availability of the [material evidence]"). Upon reading Dowdle's grand jury testimony and learning that her beau was with her at the Terrien residence on the evening of October 30, 1975, all Sherman had to do was pick up the telephone and ask Dowdle—one of the petitioner's own alibi witnesses—to identify her beau. And, then, after learning that her beau was Ossorio, it would have been easy for Sherman to locate and speak to him—indeed, a look in the telephone listings and another telephone call would have sufficed—because he lived just a few miles from Sherman's office. As in all criminal cases that involve the issue of defense counsel's failure to interview a potential witness to ascertain what he or she has to say, counsel has no absolute obligation "to actually track down" the witness, "only that he put in a reasonable effort to do so." Avery v. Prelesnik , 548 F.3d 434, 438 (6th Cir. 2008), cert. denied, 558 U.S. 932, 130 S.Ct. 80, 175 L.Ed.2d 234 (2009) ; see also id. ("There is no reason based on professional judgment why [trial counsel] would not have pursued speaking to [the potential alibi witness]. The [trial] court correctly concluded that [trial counsel] was under a duty to reasonably investigate, which entails, at the bare minimum, asking for [the potential alibi witness' telephone] number or address and reasonably attempting to contact him." [Internal quotation marks omitted.] ). In the present case, the most elementary and obvious of inquiries by Sherman or his investigator would have revealed that Ossorio was a critical alibi witness, and Sherman's unwillingness to take even those modest steps unreasonably deprived the petitioner of Ossorio's crucial trial testimony.

Consequently, this is not a case that required Sherman to devise a plan "to balance limited resources in accord with effective trial tactics and strategies." Harrington v. Richter , supra, 562 U.S. at 89, 131 S.Ct. 770 ; see also Rogers v. Zant , 13 F.3d 384, 387 (11th Cir.) ("[the] correct approach toward investigation reflects the reality that lawyers do not enjoy the benefit of endless time, energy or financial resources"), cert. denied, 513 U.S. 899, 115 S.Ct. 255, 130 L.Ed.2d 175 (1994). Taking his investigation into Ossorio's identity, whereabouts and possible testimony one step at a time, Sherman would have been able to successfully complete the investigation in two easy steps and at negligible expense. But, even if that were not so painfully apparent, the petitioner paid Sherman more than $1.5 million in legal fees, and so the cost of undertaking reasonable steps to locate Ossorio, a potentially critical witness, certainly was not an issue.

Third, the significance of Ossorio's testimony to the petitioner's alibi cannot be overstated: unquestionably, it was essential to the defense. That testimony, which the habeas court expressly credited, placed the petitioner at the Terrien residence during the relevant time frame on the evening of October 30, 1975, thereby fully corroborating the testimony of the petitioner's other alibi witnesses. But Ossorio's testimony, while corroborative, certainly was not cumulative, because the petitioner's other alibi witnesses were either siblings or cousins of the petitioner. Although Ossorio was friendly with Dowdle in the mid–1970s, there is no indication that he had maintained any ties to her or the Skakel family over the years, and, thus, he would have been an independent and unbiased witness with no motive to lie about seeing the petitioner at the Terrien home on the evening of October 30. The state's attorney emphatically and persistently maintained that the jury should not credit the petitioner's alibi because all of the alibi witnesses were closely related to the petitioner and were lying to protect him. In light of this contention by the state, credible testimony from Ossorio would have been absolutely critical, both to establish the credibility of the alibi generally and to demonstrate the credibility of the petitioner's witnesses more specifically. Indeed, if believed, Ossorio's testimony would have disproved the state's attorney's contention that the Skakel family had created the fictitious alibi to protect the petitioner and then continually lied, under oath and otherwise, in furtherance of the fraudulent scheme. Thus, with respect to the petitioner's alibi defense, the quantum of evidence already known to Sherman—evidence marked by the weakness inherent in any alibi defense comprised solely of the testimony of family members—should have prompted Sherman to investigate the lead provided by Dowdle. See, e.g., Wiggins v. Smith , supra, 539 U.S. at 527, 123 S.Ct. 2527 ("[i]n assessing the reasonableness of an attorney's investigation ... a court must consider not only the quantum of evidence already known to counsel, but also whether the known evidence would lead a reasonable attorney to investigate further").

In addition, as we discussed previously, the state adduced testimony from Helen Ix, Andrea Shakespeare, and Julie Skakel in an effort to discredit the petitioner's alibi defense. Testimony from a neutral, objective and credible witness like Ossorio would have refuted the testimony of those state witnesses, testimony that undoubtedly appeared far more significant in light of the state's contention that the petitioner's alibi witnesses all were lying. In fact, it seems likely that the jury was influenced by the testimony of Ix, Shakespeare and Julie Skakel because the jury, during its deliberations, asked that the testimony of those witnesses, insofar as it related to the petitioner's alibi, be read back.¹⁸

18 As we previously noted, each of these witnesses' recollection of the petitioner's presence was equivocal in some respect.

Along the same lines, Ossorio's testimony also would have refuted the state's attorney's claim that the alibi was an integral part of a broader Skakel family scheme to cover up for the petitioner. According to the state's attorney, this scheme was hatched immediately after the victim's murder and began with the disposal of incriminating evidence and the trip to Windham, New York, continued with the petitioner's enrollment at Elan, and, thereafter, was exemplified by his allegedly self-serving statements to Richard Hoffman, the ghostwriter assisting the petitioner with his book, and finally culminated in the perjurious grand jury and trial testimony of the petitioner's alibi witnesses. Because the allegedly fraudulent alibi provided the foundation for the state's attorney's claim of a grand family scheme, Ossorio's credible testimony demonstrating the validity of the alibi also would have debunked the state's attorney's broader conspiracy theory.

Finally, as a general matter, an adequate pretrial investigation is required in all criminal cases. But common sense dictates that, when the stakes are highest—when the criminal charges are most serious, exposing the defendant to the most lengthy of prison terms—the importance of a thorough pretrial investigation is that much greater. In the present case, both the gravity of the charged offense—murder—and the magnitude of the potential maximum sentence—life imprisonment—are obvious. In such circumstances, the responsibilities of defense counsel are especially great, commensurate with the heightened exposure, concerns and expectations of the defendant. Defense counsel must be particularly attentive to detail, because the defendant's life is on the line. Of course, the gravity of the murder charge placed Sherman on notice that he needed to put appropriate time, thought and effort into the case. He clearly did not live up to professional norms, however, in failing even to contact Dowdle after reading her grand jury testimony and learning that her beau was at the Terrien home, with her, on the evening of October 30, 1975.

In light of the foregoing, we agree with the habeas court that Sherman failed by a considerable margin to satisfy Strickland 's requirement that a decision to forgo or truncate a particular pretrial investigation must flow from an informed, professional judgment. That standard cannot possibly be met when counsel fails to undertake any steps to investigate evidence relating to the very matter that he has identified as the critical flaw in his primary defense. Accordingly, the habeas court properly reached the only conclusion that the facts and law support: Sherman could not reasonably have elected simply to ignore Dowdle's testimony and do nothing to contact her former beau, because all of the other alibi witnesses were close relatives of the petitioner, and Sherman knew both that the state would argue that those witnesses were all lying to protect the petitioner, and that an independent alibi witness, with no ties to the petitioner or his family, would have enhanced the credibility of the alibi immeasurably.

The respondent nonetheless makes several arguments as to why it was reasonable for Sherman not to investigate the identity of Dowdle's beau. The respondent contends that none of the petitioner's alibi witnesses ever mentioned seeing him at the Terrien house on the night of the murder, either in their statements to the police, in their grand jury testimony, or to Sherman directly. The respondent further asserts that, "even when Sherman asked if there was anyone else who could verify the alibi, [the] petitioner, his cousins, and his brothers essentially told Sherman there was no need to look into the beau." (Internal quotation marks omitted.) An examination of these arguments readily demonstrates that they depend on facts that were not found by the habeas court, are immaterial to Sherman's professional obligations even if factually based, or are otherwise groundless.

Contrary to the respondent's assertions, the habeas court made no finding as to whether the petitioner, Rushton Skakel, Jr., John Skakel or James Terrien told Sherman about the presence of another person at the Terrien home on the night of the murder, or whether Sherman even asked those witnesses about the presence of another person at the Terrien home that evening. The petitioner testified at the habeas trial that he had informed Sherman about Dowdle's boyfriend being present, whereas Sherman testified that the petitioner did not tell him about Ossorio's presence there. The habeas court made no finding either way, explaining, instead, that it made no difference whether the petitioner had informed Sherman about Ossorio because Sherman was on notice three years before trial, by virtue of Dowdle's grand jury testimony, that her beau was, in fact, at the Terrien residence. See footnote 17 of this opinion. Courts have consistently recognized that counsel reasonably cannot limit the pretrial investigation of a case to only those leads offered by the client himself. See, e.g., Rompilla v. Beard , supra, 545 U.S. at 381–83, 125 S.Ct. 2456 (although petitioner was unwilling to assist counsel in pretrial preparation and "was even actively obstructive by sending counsel off on false leads," counsel nevertheless had independent obligation to conduct thorough investigation); Daniels v. Woodford , supra, 428 F.3d at 1202–1203 ("[e]ven though [the petitioner] refused to speak to his counsel, [counsel] still had an independent duty to investigate [and prepare]" because "[p]retrial investigation and preparation are the keys to effective representation of counsel" [internal quotation marks omitted] ). Rather, counsel has an independent duty to investigate potentially important witnesses not suggested by the client, including, of course, potentially important alibi witnesses. See, e.g., Bigelow v. Haviland , supra, 576 F.3d at 288–89 (defense counsel could not reasonably assume that merely because petitioner had provided counsel with identity of alibi witness that there were no other such witnesses).

As to the other witnesses, Sherman's testimony at the habeas trial was equivocal: when questioned whether he had asked Rushton Skakel, Jr., and John Skakel about the presence of anyone else at the Terrien home that evening, Sherman responded, "[p]robably," and when asked the same question about Terrien, he responded, "I would assume I did." Thus, Sherman himself could not testify with any certainty that he had questioned those witnesses about the presence of another person at the Terrien home. In fact, the habeas court's findings are crystal clear that Sherman did not ask Dowdle this most basic of questions. When queried during the habeas trial whether he had asked Dowdle whether another person was present, Sherman responded, "I would assume I did," the same response he gave to the same question posed to him about Terrien. The habeas court expressly found, however, that, if Sherman had made such an inquiry of Dowdle, she would have told him about Ossorio. This finding by the habeas court necessarily means that the habeas court found that Sherman did not ask Dowdle about the presence of another person, even though Sherman testified that he "would assume" he did so. In other words, contrary to the respondent's assertions, the habeas court specifically found that, although Dowdle testified under oath, both during the grand jury proceedings and again at the petitioner's criminal trial, that her boyfriend was at the Terrien home with her on the evening in question, Sherman never bothered to ask her about that person's identity.

Furthermore, even if it is assumed that the petitioner's alibi witnesses did not tell Sherman about Ossorio's presence at the Terrien home on the night in question on their own initiative, their failure to do so is both readily explainable and irrelevant to the question at hand. By the time these witnesses were asked, twenty-seven years later, to recall the details of the events of October 30, 1975, they simply may have forgotten about Ossorio's presence at the Terrien home that evening. Indeed, Ossorio testified that he was in the television room only intermittently, while Dowdle was putting her child to bed. Cf. Bigelow v. Haviland , supra, 576 F.3d at 288 (counsel's duty to look beyond witnesses identified by client is especially significant when client may have trouble remembering them himself). And, even if one or more of the petitioner's witnesses did recall Ossorio being there that evening, there is no reason to believe that those witnesses appreciated the potential import of that information. In fact, it is obvious that even Dowdle, who knew that Ossorio was with her on the night in question, did not appreciate the potential significance of his presence at the Terrien home: she referred to her beau in her grand jury testimony and again at trial by happenstance, without any apparent awareness of his possible importance as a witness. Evidently, as far as Dowdle and the other alibi witnesses were concerned, their testimony placing the petitioner at the Terrien home that evening was sufficient to establish that he was present there, and, as nonlawyers, they had no reason to know that Ossorio potentially was a critically important witness because of the credibility issues inherent in an alibi predicated solely on the testimony of family members. In sum, the onus was not on the petitioner's alibi witnesses to divine what would have been important for Sherman to know; rather, it was Sherman's responsibility to elicit such information from them—or at the very least to recognize the significance of such information when witnesses like Dowdle divulged it of their own accord.

The respondent next contends that, because the petitioner's "siblings ... and his cousins supplied the police with the [petitioner's] alibi shortly after the murder, without ever mentioning Ossorio, and testified at the grand jury [proceedings] without mentioning Ossorio," Sherman "had no reason to go looking for other alibi witnesses when those who claimed to be there never gave any indication anyone else could verify [the petitioner's] presence at Terrien's [home] that evening." First, as we have explained, it is simply incorrect to assert that none of the petitioner's alibi witnesses mentioned Ossorio in their grand jury testimony. When asked about her recollection of the night in question, Dowdle mentioned Ossorio immediately, although not by name. Later, at the petitioner's criminal trial, the state's attorney himself asked Dowdle whether she previously had testified before the grand jury that she was at home with her "husband" on the night of the murder. Dowdle corrected the state's attorney, noting that the person she was with was merely a "friend." For reasons we cannot fathom, neither the state's attorney nor Sherman ever saw fit to question Dowdle as to the identity of the person whom she was with that evening.

Moreover, we do not agree with the respondent that Sherman had no reason to investigate Ossorio's identity simply because his name was not mentioned in any of the police reports prepared or witness statements taken in 1975. As we previously indicated, it is undisputed that the petitioner never was considered a suspect in the victim's murder before the mid–1990s but, rather, was only a potential witness. Indeed, the state's attorney acknowledged this fact at trial, pointing out that, until the 1990s, no witness ever had been asked about the petitioner's whereabouts or movements on the night of the murder because the police never suspected his involvement in the crime. Nor is there any evidence to suggest that anyone else who was at the Terrien home on the night of the murder was ever considered a suspect. Consequently, there was never any reason for the police to seek a complete accounting of all individuals who were present at the Terrien home on the evening of October 30, 1975. In fact, the respondent does not identify a single police report suggesting that such a question had been asked.¹⁹ In such circumstances, therefore, the fact that Ossorio's name did not surface until decades after the victim's murder, more or less unexpectedly, is entirely understandable.

19 If such a question had been asked, it would seem likely that, at the very least, one of the members of the staff serving at the large Terrien home would have been mentioned.

In sum, the fact that Ossorio's identity came to light for the first time during and exclusively from Dowdle's grand jury testimony reasonably could have given Sherman a reason to question whether Dowdle's testimony was accurate in this regard. Nevertheless, his obligation was to undertake some effort to answer that question rather than to dismiss it out of hand, given its potential significance.

In addition to the late timing of Ossorio's identity coming to light, the respondent argues that, even after it did, it was reasonable for Sherman to infer that Ossorio either saw nothing or would remember nothing about events that had occurred decades earlier. Specifically, the respondent argues that, because Dowdle indicated in her grand jury testimony that she mostly stayed in the library that evening and did not recall seeing the Skakel brothers herself, Ossorio, too, did not have occasion to see who was watching television in an adjacent room. This argument is unavailing for two reasons. First, it is based on sheer speculation that Ossorio stayed in the library all evening, even when Dowdle was out of the library putting her daughter to bed. The fact is that Sherman had no idea whether Ossorio stayed in the library, wandered around the house, spent time in the television room or otherwise ran into the petitioner or his brothers during the hour and one half or so that they were all together at the Terrien residence. Of course, the only way for Sherman to have found out is to have asked Dowdle or Ossorio, but, inexplicably, he made no effort to do so.

More fundamentally, however, the respondent's argument is unavailing because it is factually inaccurate. See, e.g., Wiggins v. Smith , supra, 539 U.S. at 526–27, 123 S.Ct. 2527 (noting that rationale utilized by state "to justify counsel's [failure to pursue] mitigating evidence resembles more a post hoc rationalization of counsel's conduct than an accurate description of [counsel's] deliberations prior to [trial]"). At the petitioner's criminal trial, Sherman vigorously disputed the state's contention that Dowdle had only heard her Skakel cousins on the night in question, and had not seen them. When Dowdle could not recall whether she had seen them, Sherman presented Dowdle with a copy of a 1975 police report indicating that, when she was interviewed by the police shortly after the murder, she told them that she had "observed" three of her Skakel cousins, including the petitioner, at the Terrien home that evening. Although the police report did not refresh her memory with respect to this issue, Dowdle testified that whatever she had told the police in 1975 would have been the truth. Accordingly, Sherman's position at trial—that Dowdle did, in fact, see the petitioner on the night in question—belies the respondent's assertion that Sherman's failure to investigate Ossorio's identity was based on his reasoned belief at the time of trial that, like Dowdle, Ossorio did not see the Skakel brothers on the night in question.

Thus, although Sherman reasonably could have questioned whether Ossorio would remember whether the petitioner was at the Terrien home on October 30, 1975, Sherman could not reasonably rule out that possibility without making some inquiries. Indeed, from its inception, this case concerned events long in the past, forcing both the state and the defense to do their best to develop facts based largely on distant memory and recall. It is not an exaggeration to say that this case could not have been brought but for the state's ability to locate witnesses who could remember and testify about events that had occurred decades earlier. Sherman's task in defending the petitioner necessarily required him to undertake the same investigation. His failure to do so with respect to Ossorio, merely because he did not think that Ossorio would be able to provide any useful information, was plainly deficient by any reasonable measure.²⁰

20 In his dissenting opinion, which we address in part VI of this opinion, Justice Eveleigh does not address Strickland 's performance prong except to note that he agrees with the majority's analysis of that issue in Skakel v. Commissioner of Correction , supra, 325 Conn. at 426, 159 A.3d 109. In that opinion, the majority concluded that Sherman's decision not to ascertain Ossorio's identity and potential testimony was not constitutionally deficient because Sherman reasonably, albeit wrongly, presumed, on the basis of Dowdle's grand jury testimony, that Dowdle did not see the petitioner on the night of the murder, and, therefore, it was unlikely that Ossorio had seen him either. See id., at 478, 159 A.3d 109. As we previously explained, however; see footnote 17 of this opinion; the habeas court did not credit Sherman's testimony as to why he failed to investigate Ossorio's identity, undoubtedly because, as Justice D'Auria explains in his concurring opinion, it so clearly smacked of the sort of post hoc rationalization that Strickland forbids.
As explained at length in the dissenting opinion in Skakel v. Commissioner of Correction , 325 Conn. at 558–87, 159 A.3d 109 (Palmer, J. , dissenting), the majority's conclusion in that decision with respect Strickland 's performance prong was also fundamentally flawed for two additional reasons: first, because it applied an unprecedented and manifestly incorrect legal standard; see id., at 559–62, 159 A.3d 109 (Palmer, J. , dissenting) (majority's analysis and conclusion were improperly predicated on its determination that Sherman reasonably believed that Ossorio likely would not be able to provide testimony material to petitioner's alibi defense, and correct legal standard is whether, under all relevant circumstances, competent attorney would have taken reasonable efforts to determine whether Ossorio had information that was useful to establish petitioner's alibi defense); and, second, because it failed to take into account any of the considerations relevant to determining whether Sherman's failure to present Ossorio's testimony was reasonable. See id., at 563–64, 159 A.3d 109 (Palmer, J. , dissenting) (majority improperly failed to consider factors most relevant to ascertaining reasonableness of Sherman's failure to locate Ossorio and to call him as witness, including, inter alia, importance of petitioner's alibi defense, significance of Ossorio's testimony to that defense, import of Ossorio's testimony to rebut state's claim of long-standing family cover-up, ease with which Sherman could have discovered that Ossorio could provide critically important alibi testimony, and gravity of criminal charges and magnitude of sentence that petitioner faced).

B

Prejudice

We fully agree with the habeas court that, if Sherman had located Ossorio and called him as a witness at the petitioner's criminal trial, there is a reasonable probability of a different outcome, that is, a probability sufficient to undermine confidence in the result. As we discussed, throughout the criminal trial, the state's attorney forcefully and persistently argued that the family alibi witnesses were all lying to protect the petitioner. As the petitioner himself aptly explained on direct appeal from his criminal conviction, "[t]his devastating 'cover-up' theme not only conveyed a familial verdict of guilt, it also gutted the credibility of all alibi witnesses in one argumentative thrust, and appealed to the jury's sense of outrage that a wealthy family thought it was able to trick the police by concocting a false alibi." Ossorio's testimony, however, necessarily would have bolstered the credibility of those family alibi witnesses substantially because, if the jury credited him—an independent alibi witness with no apparent reason to lie—it would have had no reason not to credit the other alibi witnesses, as well. See, e.g., Montgomery v. Petersen , supra, 846 F.2d at 415 ("the jury might well have viewed the otherwise impeachable testimony of the [family alibi] witnesses who were presented at the ... trial in a different light had the jury also heard the testimony of this disinterested witness").

Indeed, as we also discussed, the state seized on the purportedly contrived alibi defense not only to discredit the alibi itself, but also to support its broader theme of a long-standing Skakel family conspiracy designed to conceal the petitioner's involvement in the victim's murder. This theory of a family conspiracy, which was repeatedly articulated by the state's attorney during his closing argument, related the involvement of numerous members of the petitioner's family who, according to the state's attorney, conspired over a period of decades to thwart the state's investigation into the victim's murder.²¹ Independent and objective testimony by Ossorio, however, would have enabled the petitioner to refute this central thesis of the state's case against him. See, e.g., Kyles v. Whitley , 514 U.S. 419, 441–45, 115 S.Ct. 1555, 131 L.Ed. 2d 490 (1995) (concluding that evidence withheld from petitioner would have undermined state's central thesis concerning commission of crime, and, therefore, petitioner was prejudiced by state's failure to disclose that evidence); see also Thomas v. Chappell , 678 F.3d 1086, 1105–1106 (9th Cir. 2012) (when prosecutor criticized and mocked defense witness as unworthy of belief, petitioner was prejudiced by defense counsel's failure to present witnesses who would have corroborated that witness' testimony), cert. denied, 568 U.S. 1186, 133 S.Ct. 1239, 185 L.Ed. 2d 231 (2013) ; Raygoza v. Hulick , 474 F.3d 958, 961, 965 (7th Cir.) (when prosecutor "hammered on the skimpiness" of alibi supported solely by petitioner's girlfriend, petitioner was prejudiced by counsel's failure to present independent alibi witness who would have corroborated girlfriend's testimony), cert. denied sub nom. Randolph v. Raygoza , 552 U.S. 1033, 128 S.Ct. 613, 169 L.Ed. 2d 413 (2007).

21 For example, the state's attorney argued that, "[i]f this case had come to trial when perhaps it should have some twenty years ago, if the Skakels hadn't managed to keep things under wraps for so long, the jury's task would have been a simple one of just determining the credibility of this interesting alibi ...." He also argued that, "[w]here you are really going to find the truth in this case is in determining what the [petitioner] and his greater family support group have done in this case sometimes with words, sometimes without." According to the state's attorney, the cover-up commenced just hours after the murder "with the disappearance of the golf club, the shaft, and any other evidence ... [of] the crime," and continued in the days and weeks immediately following the crime with a trip to the Skakel family's hunting lodge in Windham, New York. The state's attorney argued: "[W]hat did the Skakel family do to put this together? Someone seeing the police all over the place ... had the sense to get the players out of the area. The oldest brother [Rushton Skakel, Jr.] had already gone off to [Washington] D.C., so the first thing the next morning, [Kenneth] Littleton was ordered to take the four players, [the petitioner], John [Skakel], Thomas [Skakel] and ... Terrien, out of the way for awhile, for a short trip upstate. Now, clearly, that wasn't decided for the sake of protection of these kids .... The importance of that sudden, brief, one-night trip is that the alibi didn't begin to take shape until sometime after the return from Windham." The state's attorney also argued: "Not until after their return from Windham did the alibi begin to come up." "And then you have the additional fact of two weeks after the murder ... father Rushton Skakel [Sr.], escorting the entire family together plus ... Terrien, almost like leading the von Trapp family over the [Swiss] Alps to the police station to give their recorded but unsworn statements." Two years later, the state's attorney maintained, the petitioner was sent to Elan as part of the family cover-up. The state's attorney argued: "One thing that I submit helps tie all this together, particularly on the subject of Elan, and really see the truth, is the [petitioner's] very presence at that place. The defense scoffs at the idea despite I think such clear evidence of a cover-up. Why was the [petitioner] at Elan? This is really not a matter of seeing the forest [for] the trees. It is genuinely transparent." The state's attorney further maintained that the conspiracy lasted through the grand jury investigation with false and misleading testimony before the grand jury: "What the evidence says the Skakels and Terriens have done under oath before you and some even previously before a grand jury is intentionally suppress their memories and claim a lack of recall. Why? Because in their actual recall lies the truth." And, finally, of course, the state's attorney claimed that various members of the Skakel family—including Rushton Skakel, Jr., John Skakel, David Skakel, Julie Skakel, Terrien and Dowdle—all had lied at the petitioner's criminal trial, both in connection with the allegedly concocted alibi and otherwise. The state's attorney argued: "Let's stay with the alibi. Why is it so suspect? How was it produced? ... [W]hat did the Skakel family do to put this together? ... Consider who the alibi witnesses are, all siblings or first cousins, not one single independent alibi witness."

Furthermore, in presenting a far weaker alibi defense than would have been put forward by competent counsel—one that left the door wide open for the state to argue that the alibi was predicated solely on the testimony of close family members, all of whom were lying to protect the petitioner—Sherman's performance harmed the petitioner in yet another way, "for it is generally acknowledged that an attempt to create a false alibi constitutes evidence of the defendant's consciousness of guilt." (Internal quotation marks omitted.) Henry v. Poole , 409 F.3d 48, 65 (2d Cir. 2005), cert. denied, 547 U.S. 1040, 126 S.Ct. 1622, 164 L.Ed. 2d 334 (2006). Indeed, it reasonably may be argued that "[t]here is nothing as dangerous as a poorly investigated alibi. An attorney who is not thoroughly prepared does a disservice to his client and runs the risk of having his client convicted even [when] the prosecution's case is weak." (Internal quotation marks omitted.) Id. Finally, contrary to the contention of the respondent, which we address more fully hereinafter, this is not a case in which the evidence of guilt was so overwhelming that Sherman's serious errors can be discounted as trivial. To the contrary, as the habeas court observed, "[i]t would be an understatement to say that the state did not possess overwhelming evidence of the petitioner's guilt. An unsolved crime for more than two decades, there was evidence that, initially, the Greenwich police sought the arrest of [Thomas] Skakel without success and then focused on [Kenneth] Littleton to no avail before, finally, turning to the petitioner. The evidence adduced at trial was entirely circumstantial, consisting ... [primarily] of testimony from witnesses of assailable credibility who asserted that, at one time or another and in one form or another, the petitioner made inculpatory statements ... [and of] consciousness of guilt evidence ... [indicating] that the petitioner changed his initial account to the police of his movements on the evening of the murder."

Not only was there no forensic evidence or eyewitness testimony linking the petitioner to the crime, the state's primary witnesses came forward with incriminating evidence more than twenty years after the crime and did so only after either learning of the sizeable reward being offered in the case, reading Mark Fuhrman's 1998 book, Murder in Greenwich: Who Killed Martha Moxley, inculpating the petitioner,²² or both. Indeed, one key witness for the state, Shakespeare, the only person to testify that the petitioner did not go to the Terrien home on the night of the murder, completely changed her account of that evening after reading Fuhrman's book.²³ Thus, as the habeas court concluded, all of the state's witnesses were eminently impeachable. In sum, although the state's evidence was sufficient to convict the petitioner, that evidence was far from strong—and most certainly not strong enough such that it confidently can be said that Ossorio's critical alibi testimony simply would not have mattered to the jury.

22 In his book, Fuhrman asserted that the petitioner and the victim were boyfriend and girlfriend and that the petitioner flew into a jealous rage upon seeing the victim having a sexual encounter with Thomas Skakel. See M. Fuhrman, Murder In Greenwich (HarperCollins 1998) p. 215. Fuhrman claimed that he had learned about the relationship between the victim and the petitioner from unnamed sources and, further, that the victim's diary "clearly stated" that the petitioner was interested in her romantically. Id. At the petitioner's criminal trial, however, Fuhrman's claims both with respect to the victim's diary and the victim's purported relationship with the petitioner were debunked. Indeed, according to the victim's diary, the victim did not even become acquainted with the petitioner until two months before her death, and, during the entire time that they were acquainted, she was in a steady relationship with a boy from her high school. Fuhrman's false claims nevertheless appear to have formed the basis for the state's theory that the petitioner murdered the victim in a jealous rage; indeed, the only two witnesses that the state's attorney has identified as providing testimony in support of that theory—Elizabeth Arnold and Geranne Ridge—both admitted that Fuhrman's book, or tabloid accounts about the book, were instrumental to the substance of their testimony concerning the petitioner's purported motive. At trial, Arnold testified that, while she and the petitioner were students at Elan, the petitioner told her "that his brother [fucked] his girlfriend ... well, they didn't really have sex, but they were fooling around." On cross-examination, Arnold was asked why, when testifying before the grand jury or talking to the police, she never mentioned that the petitioner had told her that his brother had "fooled around" with his girlfriend. Arnold responded that she did not remember it at the time but that reading Fuhrman's book afterward had refreshed her recollection. Ridge testified that much of which she knew about the murder came from the tabloids.

23 Indeed, there is perhaps no better example of the seemingly corruptive effect of Fuhrman's book than the testimony of Shakespeare. At trial, Shakespeare testified that she was absolutely certain that the petitioner did not go to the Terrien home on the night of the murder and that, in fact, she had observed the boys who did go as they were leaving the driveway. Because Shakespeare was the only witness whose testimony placed the petitioner at home at the likely time of the murder, her testimony was extremely important, and, indeed, the jury asked to have it read back during its deliberations.
But Shakespeare's story at trial bore little resemblance to statements that she had given to the police in 1991. At that time, she told investigators that she had no independent recollection of any of the events in question because she was in the kitchen at the Skakel house with Julie Skakel the entire time. What little she knew, she explained, she had learned secondhand. Specifically, Shakespeare stated that she always had assumed that the petitioner did not go to the Terrien home because, after the murder, she had been told that there were four people in the driveway after the Lincoln departed, and she always had assumed that one of them was the petitioner. Shakespeare stated: "I thought that I heard ... that there were four of them back in the backyard, saying goodbye to each other.... It was my assumption, and it's a total assumption, that ... it was [Thomas Skakel], [the petitioner], [Ix], and [the victim].... I don't know where the information came from." Shakespeare's continued: "None of [it] I ... saw with my own two eyes. It's the tales I've heard over the years .... Did I see it? No. Do I know it for a fact? No." Shakespeare also acknowledged that she "thought, because I've heard, because of what's been, you know, told over the years ... that [the petitioner] and [Thomas Skakel] and [the victim] and [Ix] were out back ... hanging out, chitchatting. And then the girls went to go home, and the boys came in the house. That's what I'd always assumed." When informed by investigators that there were four children in the backyard after the Lincoln Continental left but that it was undisputed that the fourth child was not the petitioner but an eleven year old boy name Geoffrey Byrne, Shakespeare responded that she had never even heard the name Byrne before and that, of course, the petitioner could have gone to the Terrien residence because "I didn't see them leave, so I can't tell you who was in that car." Shakespeare also stated: "I didn't see anybody after a certain point [because] I was sitting in the kitchen ...."
At the petitioner's criminal trial eleven years later, however, Shakespeare testified that, in the twenty-seven years since the murder, she had never once doubted that the petitioner "was home after [the] car left ...." Indeed, Shakespeare insisted that she "was there ... when the boys left in the car to take [Terrien] home" and that she was "sure" she "saw them leave." Shakespeare was asked how it was possible that, in 1991, she had no such recollection, and whether she had read any books about the case in the intervening years. Shakespeare responded that she had "read Mark Fuhrman's book."

The respondent nevertheless makes two primary arguments as to why, in his view, there is no reasonable probability of a different result, even if Ossorio had testified. Specifically, the respondent claims that the petitioner's alibi was only a partial one, and the state's case was so strong that Ossorio's testimony would have made no difference in terms of the outcome. We address each of these contentions in turn.

1

Partial Alibi

The respondent claims that, because the petitioner's alibi was only a partial one, even if the jury had credited that alibi, it nonetheless would have found the petitioner guilty. In support of this contention, the respondent observes that, at trial, the state's attorney argued, and the jury was instructed, that the jury could accept the petitioner's alibi and still find the petitioner guilty. Although perhaps superficially appealing, this argument does not hold up upon closer examination.

We agree that, as a general rule, partial alibis are unconvincing. Indeed, it has been argued that a partial or incomplete alibi is not really an alibi in the truest sense; see, e.g., Williams v. State , 185 So.3d 1270, 1271 (Fla. App. 2016) ("a partial alibi is no alibi at all" [internal quotation marks omitted] ); because it fails to account for a defendant's whereabouts for at least some period of time during which the crime reasonably could have been committed by the defendant. Thus, when a true partial alibi is at issue, it is invariably the case that the defendant just as likely could have committed the crime during a period of time not covered by the alibi. Notably, each and every one of the cases on which the respondent relies falls squarely into this category.

As the habeas court explained, however, in the present case, the petitioner's alibi, if believed, establishes that he was not at the crime scene when the substantial weight of the evidence indicates that the victim was murdered. The respondent has identified no case in which a partial alibi was found to exist and in which the state's primary theory of the case, and the only one toward which its evidence was geared, was that the crime most likely occurred during the period of time covered by the defendant's alibi. Accordingly, this case simply does not involve the kind of alibi that courts treat as partial or incomplete.

The thin evidentiary reed on which the respondent's partial alibi theory rests is the trial testimony of Harold Wayne Carver II, then the state's chief medical examiner, who reviewed the 1975 autopsy report and opined that it was within the realm of scientific possibility that the victim died any time between 9:30 p.m. on October 30, 1975, and "many hours before she was found" the next afternoon. Carver's testimony establishing this broad time frame, however, does nothing to establish when within that time period the murder actually occurred. Indeed, the state's attorney discounted part of that window of time by conceding that the crime must have been committed no later than 1 a.m. the next morning because, by that time, the victim's family was out looking for her. Insofar as Carver offered any opinion as to when the murder actually occurred within the scientifically possible time frame, he opined that the victim probably was murdered "closer to 9:30 p.m." than when she was found the next day.

To be sure, the state's attorney observed during closing argument that the state did not have to disprove the petitioner's alibi for the jury to find him guilty, insofar as the autopsy report did not rule out the possibility that the victim was alive as late as 5:30 a.m. on October 31, 1975. However, the state's attorney made no effort to explain to the jury the victim's whereabouts in the one hour and forty-five minutes or so between the time her friends left her to return to their homes and the time the petitioner's alibi established his return home. It is reasonable to conclude that no such effort was made because, as we previously discussed, the substantial weight of the evidence indicated that the victim was murdered between 9:30 and 10 p.m. Indeed, for more than twenty years, that was the state's own theory of when the crime was committed, and no evidence has ever surfaced to undermine that theory. In light of the convincing evidence supporting the theory that the victim was murdered between 9:30 and 10 p.m. and the complete absence of evidence that she was alive but otherwise unseen after that time frame, there is little wonder that neither the state nor the respondent has ever articulated a plausible theory to support the possibility that she was murdered after 11 p.m. Accordingly, the respondent's attempt to negate the significance of the alibi under a partial alibi theory is unavailing.

2

Strength of the State's Case

We therefore turn to the question of whether, in light of the theory the state advanced at trial, there is a reasonable probability of a different result if Ossorio's credible testimony regarding the alibi had been presented to the jury. As we previously indicated, this is not a case in which there was any forensic evidence or eyewitness testimony connecting the petitioner to the crime. Nonetheless, the respondent argues that there is no reasonable probability of a different result because the evidence presented at trial, considered as a whole, was overwhelming. Specifically, the respondent contends that the present case, "unlike most murder cases, contained evidence of three explicit confessions of guilt" and a "multitude of other incriminatory statements" that the petitioner purportedly made over the years, including the petitioner's statement to Hoffman "placing himself at the crime scene on the night of the murder ...."

Before addressing the nature and strength of the evidence adduced by the state at the petitioner's criminal trial, it bears emphasis that our research has not revealed a single case, and the respondent has cited none, in which the failure to present the testimony of a credible, noncumulative, independent alibi witness was determined not to have prejudiced a petitioner under Strickland 's second prong. There are many cases, however, in which counsel's failure to present the testimony of even a questionable or cumulative alibi witness was deemed prejudicial in view of the critical importance of an alibi defense. See, e.g., Caldwell v. Lewis , 414 Fed. Appx. 809, 818 (6th Cir. 2011) ("[The] [c]ourt has recognized that when trial counsel fails to present an alibi witness, [t]he difference between the case that was and the case that should have been is undeniable.... [The] [c]ourt has held that the failure to produce an alibi witness at trial was prejudicial under Strickland , even [when] the ... [habeas] court said [that] the alibi witnesses would have been unconvincing, and there were other alibi witnesses presented at trial." [Citation omitted; internal quotation marks omitted.] ); Brown v. Myers , 137 F.3d 1154, 1155–56, 1157–58 (9th Cir. 1998) (petitioner suffered prejudice from counsel's failure to present alibi witnesses, even though their testimony "was vague with regard to time," and three eyewitnesses identified petitioner as shooter); see also Davis v. Lafler , 658 F.3d 525, 541 (6th Cir. 2011) ("[the] court has repeatedly found prejudice resulting from trial [counsel's] fail[ure] to investigate or present favorable witnesses"), cert. denied, 566 U.S. 947, 132 S.Ct. 1927, 182 L.Ed. 2d 788 (2012) ; Bigelow v. Haviland , supra, 576 F.3d at 291 (when case turned on credibility of state's witnesses, failure to produce alibi witness was prejudicial); Avery v. Prelesnik , supra, 548 F.3d at 439 ("[The] potential alibi witnesses coupled with an otherwise weak case render[ed] the failure to investigate the testimony sufficient to undermine confidence in the outcome of the jury verdict.... [T]he jury was deprived of the right to hear testimony that could have supplied ... reasonable doubt." [Internal quotation marks omitted.] ); Harrison v. Quarterman , 496 F.3d 419, 427 (5th Cir. 2007) ("[o]ur sister circuits have held that counsel prejudices his client's defense when [he] fails to call a witness who is central to establishing the defense's [theory of the case]"); Raygoza v. Hulick supra, 474 F.3d at 960, 964–65 (petitioner was prejudiced by counsel's failure to present independent alibi witness who would have corroborated testimony of petitioner's girlfriend that petitioner was thirty-five miles from crime scene at time of murder); Stewart v. Wolfenbarger , 468 F.3d 338, 359–61 (6th Cir. 2006) (petitioner was prejudiced by counsel's failure to call independent alibi witness to corroborate another alibi witness whose testimony was subject to impeachment); Alcala v. Woodford , 334 F.3d 862, 872–73 (9th Cir. 2003) (petitioner was prejudiced by counsel's failure to investigate and corroborate petitioner's alibi, insofar as prosecution's evidence was "far from compelling" and eyewitness made "confident" but "not unimpeachable" identification of petitioner); Lindstadt v. Keane , 239 F.3d 191, 204–205 (2d Cir. 2001) (petitioner was prejudiced when trial counsel failed to present evidence that could have corroborated petitioner's alibi claims); Montgomery v. Petersen , supra, 846 F.2d at 415 (petitioner was prejudiced by counsel's failure to call additional, disinterested alibi witnesses not subject to same impeachment as petitioner's other alibi witnesses, all of whom were family members); Syed v. State , 236 Md.App. 183, 196–98, 275–85, 181 A.3d 860, 2018 WL 1530300, *3, *45–49 (2018) (when state's case rested in part on testimony of witness who claimed to have helped petitioner dispose of victim's body, petitioner was prejudiced by counsel's failure to present testimony of independent alibi witness, inasmuch as "potential alibi witnesses coupled with an otherwise weak case render[ed] the failure to investigate the [alibi] testimony sufficient to undermine confidence in the outcome of the jury verdict" [internal quotation marks omitted] ). The only cases to the contrary are ones in which the exculpatory evidence was found not to be credible, or, in addition to such a finding, there was conclusive physical evidence linking the petitioner to the crime. See, e.g., Moore v. New York , 357 Fed. Appx. 398, 401 (2d Cir. 2009) (undercover police officers observed petitioner committing crime); Hemstreet v. Greiner , 491 F.3d 84, 92 (2d Cir. 2007) (petitioner was not prejudiced by counsel's failure to present additional alibi witness "of questionable veracity" when state's case was based on "[o]verwhelming items of forensic evidence connect[ing] [the petitioner and his accomplice] to the murder ... including [the victim's] blood in [the petitioner's] car"), cert. denied sub nom. Hemstreet v. Ercole , 552 U.S. 1119, 128 S.Ct. 962, 169 L.Ed. 2d 763 (2008).

In the present case, there was no unassailable evidence establishing the petitioner's guilt to the exclusion of others, and the habeas court found Ossorio's account credible. Accordingly, the foregoing authority, and the logic underlying it, compels us to conclude that Sherman's deficient performance in failing to investigate the independent alibi testimony of Ossorio was inherently or necessarily prejudicial. Nonetheless, we explain why the evidence on which the respondent relies—comprised almost exclusively of incriminating statements purportedly made by the petitioner—would not, in any event, compel a different conclusion. Specifically, we conclude that, although it certainly was within the province of the jury to credit some or all of those incriminating statements in reaching its guilty verdict, there is a reasonable likelihood that the already substantial impeachment evidence pertaining to those statements would have been afforded considerably more weight by the jury if the petitioner had presented credible, independent alibi testimony persuasively demonstrating that he was at the Terrien home when the murder likely occurred.

We begin with the observation that the statements deemed most incriminating by the respondent, which we discuss more fully hereinafter, were made during the petitioner's stay at Elan, where, according to all reports, he was sadistically interrogated about the victim's murder over a period of months and brutally beaten whenever he proclaimed his innocence. Indeed, in his closing argument, the state's attorney described Elan as having a "concentration camp-type atmosphere" that was "equivalent to the lower circles of hell." This court also acknowledged in the petitioner's direct criminal appeal the "extremely harsh and oppressive" atmosphere at Elan, under which residents were subjected to a program "predicated on ridicule and fear." State v. Skakel , supra, 276 Conn. at 717, 888 A.2d 985. Accordingly, some additional detail about the conditions at Elan and the petitioner's treatment there is useful to understand the context underlying the petitioner's statements, both at Elan and thereafter.²⁴

24 Apart from two confessions that the petitioner purportedly made while attending Elan, the respondent relies on a third "confession" that the petitioner is alleged to have made to a total stranger, Geranne Ridge, at a cocktail party in the 1990s, and a number of other statements, which run the gamut from odd to suspicious, that he reportedly made in the presence of a barber, and to the family chauffeur, to a ghostwriter whom he hired in the late 1990s to write his autobiography, and to Michael Meredith, another former Elan student. As we explain hereinafter in this opinion, an examination of this evidence reveals that it, too, was either readily impeachable, subject to differing interpretations, or, in some cases, both.

Trial testimony about Elan and the petitioner's cruel and inhumane treatment there has previously been briefly summarized by this court. "[Certain Elan] residents testified to the brutal and abusive treatment of residents, including the petitioner. The witnesses explained that school staff frequently accused the petitioner of the [victim's] murder and urged him to admit his involvement. When he refused to take responsibility, he was paddled, assaulted in a boxing ring, and forced to wear a sign that had written on it something to the effect of 'please confront me on the murder of my friend, Martha Moxley ....' These witnesses also stated that the petitioner denied involvement in the victim's murder, and, when the abuse continued, he parried their accusations by stating that he either did not know or could not recall what happened; they never heard the petitioner confess to the crime." Skakel v. Commissioner of Correction , supra, 325 Conn. at 438, 159 A.3d 109.

More specifically, every witness who attended Elan—with the notable exception of Gregory Coleman and John Higgins, the only Elan witnesses who claimed to have heard the petitioner confess and whose testimony constitutes two of the three confessions on which the respondent relies—testified that Joseph Ricci, the executive director of Elan, liked to taunt the petitioner about the victim's murder, constantly accusing him either of having committed the crime or of knowing who did. At one point, after the petitioner ran away from Elan, Ricci convened a general meeting, which typically was attended by 100 or more Elan residents and staff, and was called for the purpose of focusing on one or two residents who had violated Elan rules. At this meeting, the petitioner was singled out for his attempt to run away.

All of the witnesses gave similar accounts of the general meeting and certain other related events. Alice Dunn, a former student at Elan, testified that, for three days before the general meeting, the petitioner had been forced to stand in the corner of the school's dining room without any sleep. On the third day, at the general meeting itself, the petitioner was placed against the wall, and at least 150 residents confronted him by yelling and spitting in his face. After a while, the petitioner was placed in a boxing ring and questioned by Ricci about the victim's murder. According to Dunn, this was the first time that anyone at Elan ever had heard about the victim's murder. Ricci tried to get the petitioner to confess, but the petitioner insisted numerous times that he didn't do it. Each time the petitioner denied involvement in the crime, Ricci put him in the boxing ring, and students would "pummel" him until he was "physically ... wiped out ...."

According to Sarah Petersen, another former Elan student, the petitioner cried "uncontrollably" during the beatings. She stated that Ricci often "liked to pull [the petitioner] out [of the crowd at general meetings and] emotionally pound on him," saying things like, "we know you did this ...." When Ricci did not get the response that he was looking for, he would place the petitioner in the boxing ring or strike him with a paddle. Petersen testified that the petitioner always denied any involvement in the murder, but, after "long hours of torture," he would say that he did not remember just to "get them to lay off him for a little while."

Another former Elan student, Michael Wiggins, remembered the general meetings as pure "mayhem," with students hitting the petitioner as hard as they could while others screamed "hit him, hit him hard, hit him harder ...." Wiggins recalled that the petitioner always denied any involvement in the victim's murder until he was beaten down and extremely fatigued, at which point he would say, "I don't remember ...." The beatings would stop as soon as the petitioner expressed some doubt. According to Wiggins, the beatings would stop for everyone as soon as they told Ricci what Ricci wanted to hear, even if it was not true. According to Elizabeth Arnold, two days after the petitioner's first boxing ring incident, Ricci tried to reassure the petitioner at a group therapy session that he did not really think that the petitioner had murdered the victim, only that the petitioner knew who did and that he probably was covering up for his brother. The petitioner responded that "[h]e didn't know" and "had no recollection" about the night of the victim's murder. When the petitioner was asked about the murder, he sometimes would respond that he was drunk that evening and that he must have blacked out. Other times, he would say that he did not know if he or his brother was involved in the murder because "he had no memory of the incident whatsoever."

It is with this backdrop that the petitioner made his allegedly incriminating statements to Elan students Coleman and Higgins. The fact that all of those statements were either made at Elan or in the aftermath of his experience there places them in a light that a jury would be far less likely to disregard in the face of a credible alibi. More specifically, the treatment that the petitioner received at Elan as an adolescent was so brutal and coercive, and so directly related to his alleged involvement in the victim's murder, that the jury reasonably would question how that treatment affected the way the petitioner thought about the murder and how he responded to questions about it.²⁵ Indeed, because the petitioner's alleged involvement in the victim's murder was a constant topic of conversation at Elan, the jury also reasonably could have questioned whether witnesses like Coleman and Higgins, either because they were young and impressionable at the time or due to the passage of so much time, had simply conflated in their minds an accusation with a confession. As one Elan witness stated, it was "common knowledge" that the petitioner "was there because he had murdered somebody." "It was not a secret.... As far as in my fourteen year old head, that was [the petitioner's] punishment, going to Elan."

25 We note in this regard that Sherman never sought to explain to the jury that an innocent person—particularly an emotionally troubled adolescent who had been subjected to appalling physical and psychological coercion—could convince himself that he may have killed someone in a drunken stupor but have no recollection of doing so. Thus, Sherman offered no rebuttal to the state's attorney's assertion that only someone who had committed murder would express uncertainty when asked about his involvement in the crime. Because defense counsel had not raised certain challenges to the admission of these purported statements, the record on the petitioner's direct appeal lacked sufficient factual findings for us to "assume that the atmosphere at Elan was so coercive that any incriminating statement by the [petitioner] necessarily was the product of that coercive environment." State v. Skakel , supra, 276 Conn. at 723, 888 A.2d 985. Although this court was precluded under the circumstances from making any such assumption, the jury was free to draw such an inference. This court never questioned the brutality of the petitioner's treatment at Elan.

As the habeas court observed, however, of all the former Elan students who testified at the petitioner's criminal trial, only Coleman claimed that the petitioner had provided him with anything resembling a detailed account of the victim's murder. A twenty-five bag a day heroin addict, Coleman contacted a television station to tell his story in 1998 after seeing a tabloid news show based on Fuhrman's book and learning of the sizeable reward being offered in the case. It is conceded that, during the grand jury proceedings, Coleman testified under the influence of heroin. Thereafter, Coleman testified at the petitioner's probable cause hearing and explained that he met the petitioner for the first time when he was assigned to guard him at Elan, following the petitioner's attempt to escape. According to Coleman, the first thing that the petitioner ever said to him was, "I am going to get away with murder because I am a Kennedy ...." Coleman also testified that the petitioner told him that he had beaten a girl's head in with a golf club and, two days later, had gone back to the body and masturbated on it. Coleman died of a drug overdose before the petitioner's criminal trial, but his probable cause hearing testimony was admitted into evidence and read to the jury at that trial.

Coleman's account of what the petitioner allegedly told him, however, flew in the face of established facts, forcing the state's attorney to acknowledge in closing argument that, "[c]learly [Coleman] has some facts kind of backwards ...." Although the state's attorney urged the jury to attribute Coleman's "backwards" facts to "the fog of time," a fact finder also reasonably could have questioned whether Coleman's confusion had resulted from his inability to accurately recall the information he had gleaned about the murder from the television shows and magazine articles that had prompted him to come forward in the first place. A fact finder also could have believed that Coleman's testimony was merely the product of his obvious interest, fueled by his heroin addiction, in the reward. To be sure, Coleman conveyed nothing about the murder that was not already in the public domain when he first told the authorities about the petitioner's alleged confession.²⁶

26 Although we consider the issue of prejudice only in the context of the evidence actually presented at the criminal trial, the habeas court observed that additional evidence existed to impeach Coleman's account. Specifically, John Simpson, a former Elan resident who was present when the petitioner allegedly confessed to Coleman, testified unequivocally at the hearing on the petitioner's new trial petition that he remembered the conversation between Coleman and the petitioner in which Coleman claims the petitioner confessed and that, contrary to Coleman's testimony, the petitioner made no such incriminating statements.

The respondent argues that Coleman's testimony was nevertheless reliable because it was corroborated by Coleman's wife, Elizabeth Coleman, who testified that Coleman told her about the petitioner's confession in 1986, and by Jennifer Pease, who testified that Coleman told her, while they were students at Elan, that the petitioner had told him "that he bashed [the victim's] head in with a golf club." As with Coleman himself, however, the jury reasonably could have questioned whether his wife had a similar motive to fabricate, namely, to collect the reward money. The jury also reasonably could have questioned Pease's testimony in view of the fact that she waited until the final days of the trial to come forward, and then did so, it appears, for reasons unrelated to any information she claimed to have had concerning Coleman and the petitioner.²⁷ Finally, as we previously indicated, the jury also reasonably could have questioned whether Coleman actually believed that the petitioner had confessed to him, insofar as his youth or impairments may have caused his perception in that regard to be wrong. The state also introduced evidence that the petitioner had confessed to a second Elan student, Higgins, who testified that, on one occasion, when he and the petitioner were on guard duty at the school, the petitioner told him "about a murder that he was somehow involved in" and that "he remembered that there was a party going on ... at his house." The petitioner also remembered "going through some golf clubs" and "running through some woods." According to Higgins, the petitioner "was sobbing and crying," just "releasing emotion[s]" and "bleeding out." Higgins testified that the petitioner, through a progression of statements, said that "he didn't know whether he did it, that he may have done it, [that] he didn't know what happened, [and that], eventually, he came to the point that he [thought he] did do it, [that] he must have done it ...."

27 More specifically, it appears that Pease's belated decision to come forward was motivated by her intense dislike of another former Elan witness, Dunn. For example, on cross-examination, Pease admitted that, days before her testimony, using the screen name "Betty," she had posted a fairly lengthy screed on the "Crime News 2000" website in which she stated, "what the prosecution needs [are] some people who can testify to what ... Dunn really is, a monster."

Like Coleman, however, Higgins was far from unimpeachable. For example, on cross-examination, he acknowledged that, when he was initially contacted by the police, he told them repeatedly that the petitioner had never confessed in his presence. Higgins also acknowledged that he changed his initial story after the state's lead investigator in the case informed him that the reward had been increased to $100,000, and after the victim's mother, after receiving a phone call from Higgins, asked him to testify against the petitioner. Higgins also claimed that approximately twenty-five to thirty people were with him and the petitioner when the petitioner made his admissions but provided few names of these alleged witnesses, and no witness came forward to corroborate Higgins' testimony. Finally, Higgins claimed that his conversation with the petitioner was the first and only time that he had ever heard about the victim's murder, until he read about it in People Magazine in the 1990s. As we previously indicated, however, every other Elan witness—Petersen, Wiggins, Charles Seigen, Dorothy Rogers, Arnold and Dunn—testified unequivocally that the murder was a regular topic of conversation at the school, so much so that, for weeks on end, the petitioner was forced to wear an enormous sign around his neck inviting students to question him about his involvement in the victim's murder.

The third "confession" introduced by the state at trial came from a 2002 telephone conversation between two people unknown to the petitioner, Geranne Ridge and her friend, Matthew Attanian, in which Ridge claimed to have heard such a confession. During that conversation, which Attanian secretly recorded for Frank Garr, an inspector with the state's attorney's office, Ridge claimed to have met the petitioner at a party in 1997, and to have heard the petitioner confess, in front of everyone there, to murdering the victim, apparently because the victim had had sex with his brother and because the petitioner was "doing LSD and acid and really big-time drugs, mind, you know, altering drugs."

When under oath at the petitioner's criminal trial, however, Ridge testified, consistent with her previous statements to investigators, that nothing she had said to Attanian was true. Ridge explained that, although she had seen the petitioner at a party once, they were never introduced and never spoke. Ridge testified that she told Attanian that the petitioner had confessed to her because Attanian "was always bragging about who he knew, and [Ridge] had done some modeling, and [Attanian] is a part-time photographer, and he was talking about famous models he knew and so forth." Ridge just wanted to seem "more knowledgeable than [she] was" about the petitioner's case. Ridge testified that everything that she told Attanian had been gleaned from "magazines, newspapers and from [the tabloids]," like the "Star, Globe, [National] Enquirer, those kinds of things ...." In light of her testimony under oath and her credible explanation for her earlier statement, there is strong reason to question whether, even without a solid alibi for the petitioner, the jury would have found the statement credible.

In addition, the respondent relies on a number of other statements the jury heard, which the petitioner made or purportedly made throughout the years, that fell well short of an actual confession but, depending on one's view, could be suggestive of a consciousness of guilt. One such statement was made to Lawrence Zicarelli, who worked as the Skakels' chauffeur from 1976 through 1977. According to Zicarelli, the petitioner, following a fight with his father earlier in the day, stated that, if Zicarelli "knew what he had done, [Zicarelli] would never talk to him again," and that "he either had to kill himself or get out of the country." According to Zicarelli, later that same day, the petitioner jumped out of the family's car on the Triboro Bridge in New York while the car was stuck in traffic. Zicarelli further testified, however, that these incidents occurred approximately two years after the murder and that he had no idea what the petitioner was referring to at the time or what the petitioner had been fighting about with his father. Zicarelli also noted that he never mentioned the petitioner's statement to the police, even though detectives from the Greenwich police department regularly visited him in the late 1970s in an effort to obtain incriminating information about the Skakel family. According to Matthew Tucharoni, a Greenwich barber, three people who he believes were the petitioner, Julie Skakel, and one of the petitioner's brothers, came into his barber shop in the late 1970s, and the petitioner purportedly told Julie Skakel, while getting his hair trimmed, "I am going to get a gun, and I am going to kill him." Julie Skakel purportedly replied: "[Y]ou can't do that." The petitioner purportedly responded: "Why not? I did it before, I killed before." Julie Skakel then responded: "Shut up, Michael." Despite the passage of more than twenty-five years, Tucharoni also recalled that, when he finished cutting the hair of the person he believed to be the petitioner, "the total was $8 because I didn't wash it and blow dry it, so they [gave] me $10, and I figured $2 was my tip for the haircut." Tucharoni further testified that he never spoke about the petitioner's purported admissions to anyone until reading about the petitioner's trial in 2002, at which point he went to the state's attorney's office and was shown a picture of the petitioner taken in the mid to late 1970s. From that picture, Tucharoni identified the petitioner as the teen who, twenty-five years earlier, in the middle of a barber shop, purportedly claimed to have killed before.²⁸

28 In his closing argument, the state's attorney also cited the testimony of Shakespeare that, when she and Julie Skakel arrived at the Skakel home on the afternoon of October 31, 1975, following the discovery of the victim's body, the petitioner approached their car and informed them that "[the victim] had been killed and that he and [Thomas Skakel] were the last to see [the victim] that night." This is hardly inculpatory evidence, however, given that it was common knowledge, from the earliest moments of the investigation, that the victim was last seen alive in the Skakel driveway, in the company of the petitioner, Thomas Skakel, Helen Ix and Byrne. Additionally, at trial, the state presented the grand jury testimony of Mildred Ix, Helen Ix' mother, who testified that, sometime in the early 1980s, the petitioner's father [Rushton Skakel, Sr.] had told her that the petitioner "had come up to him and ... said, you know, I had a lot ... to drink that night, and I would like to see—I would like to see if—if I could have had so much to drink that I would have forgotten something, and I could have murdered [the victim] .... So he asked to go under Sodium Pentothal or whatever it was." At trial, however, Mildred Ix testified that her recollection of her conversation with the petitioner's father was incorrect and that the petitioner had not told his father that he wanted to find out if he could have murdered the victim.

Finally, in the 1990s, the petitioner informed writer Richard Hoffman and childhood friend Andrew Pugh that, after he returned home from the Terrien residence on the night of the murder, he went back out to peep in neighbors' windows and masturbated in a tree on the Moxley property. Although he told Hoffman that the tree was adjacent to the front of the victim's house, Pugh testified that he had always assumed that the tree that the petitioner was referring to was the one under which the victim's body was found. In 1987, the petitioner purportedly told Michael Meredith, another former Elan student, a similar story. According to Meredith, the petitioner told him that he could see the victim undressing and showering from the tree. Notably, when the victim's mother was asked whether there were any climbable trees next to her house, she replied that there were none because the branches had been trimmed "off very high ...." When asked specifically whether a person could climb any of the trees behind the house, next to the victim's third floor bedroom windows, she replied, only if the person "were like a monkey" because "[t]here were no branches, no branches. I mean ... they [were] all trimmed away." Similarly, when John Moxley, the victim's brother, was asked whether the trees adjacent to the front of the house, which the petitioner told Hoffman he had climbed on the night of the murder, were climbable, he replied that he thought they "were hemlocks, and the branches were like pencils."²⁹ Despite the admittedly suspicious nature of some of this evidence, some of which reasonably could be construed as demonstrating a consciousness of guilt, the state's case clearly cannot be described as strong or overwhelming. As the habeas court noted, the victim's murder remained unsolved for more than two decades, and, initially, the Greenwich police sought to arrest Thomas Skakel in connection with the victim's murder, without success, and then focused their attention on a second suspect, Skakel tutor Littleton, before turning, finally, to the petitioner. There were no eyewitnesses and no physical evidence connecting the petitioner to the crime, except for the murder weapon, to which many people had access prior to the crime. There also was no motive except for a highly dubious one devised by Mark Fuhrman, seemingly out of whole cloth. See footnote 22 of this opinion. Suffice it to say that courts routinely have held that defense counsel's failure to present exculpatory evidence was prejudicial in cases involving far stronger evidence. See, e.g., Thomas v. Chappell , supra, 678 F.3d at 1102–1103 ("The prosecution's evidence certainly goes a long way toward implicating [the] [p]etitioner. [The] [p]etitioner was present; was the last person seen with the victims by those who testified at trial, at a location near the murder site; had access to what could have been the murder weapon; told a bizarre, mostly uncorroborated tale of where he had been; identified [the victim's] body in a potentially suspicious manner; gave the police and acquaintances somewhat conflicting descriptions of his activities on the night of the murders; acted oddly after the murders; and owned a distinctive pipe that was found [in the vicinity of the] murder site .... Nevertheless, in [the court's] view, the case against [the] [p]etitioner was not overwhelmingly strong. The prosecution presented circumstantial evidence only: no motive, no murder weapon, no witness to the crime, no fingerprint evidence, and no blood or other bodily fluid evidence." [Emphasis omitted.] ); Raygoza v. Hulick , supra, 474 F.3d at 964–65 (even though state's evidence included several eyewitness identifications and petitioner's self-incriminating statement to friend on night of murder, petitioner was prejudiced by counsel's failure to present alibi witness who would have corroborated testimony of petitioner's girlfriend that petitioner was thirty-five miles away from crime scene at time of murder); Anderson v. Johnson , 338 F.3d 382, 393–94 (5th Cir. 2003) (describing as "weak," for purposes of Strickland , case relying primarily on eyewitness testimony); Wright v. Gramley , 125 F.3d 1038, 1043 (7th Cir. 1997) (same); United States ex rel. Freeman v. Lane , Docket No. 89 C 4642, 1990 WL 70558, *6 (N.D. Ill. May 16, 1990) (evidence of guilt was "not overwhelming" when conviction was based on testimony of eyewitness and no physical evidence corroborated witness' testimony), aff'd sub nom. Freeman v. Lane , 962 F.2d 1252 (7th Cir. 1992) ; Syed v. State , supra, 236 Md.App. at 283–85, 181 A.3d 860, 2018 WL 1530300, *49 (describing as weak, for purposes of Strickland , case predicated entirely on circumstantial evidence, which included testimony of witness who claimed to have helped petitioner dispose of victim's body).

29 At the petitioner's criminal trial, but outside the presence of the jury, the state's attorney argued that, with the advent of DNA testing in the early 1990s, the petitioner invented the masturbation story out of fear that his DNA might one day be found at the crime scene. As we previously indicated, however, Meredith testified that the petitioner had told him the masturbation story in 1987, years before DNA was used as an investigative tool in Connecticut. Accordingly, although the petitioner's masturbation story was sufficiently bizarre that the jury reasonably could have viewed it as consciousness of guilt evidence, the state's only theory with respect to that evidence was not supported by the evidence.

Thus, despite the respondent's efforts to depict the state's evidence as strong for purposes of applying Strickland 's prejudice prong, it demonstrably was not—a point further illustrated, as the habeas court noted, by the jury's four days of deliberations and request to have read back the only testimony that supported the state's theory that the petitioner did not go to the Terrien residence on the night of the murder, as he claimed. Cf. Thomas v. Chappell , supra, 678 F.3d at 1102–1103 (in evaluating strength of case under Strickland , "almost five full days" of deliberations and jury's request for read back of testimony supporting petitioner's defense were indicative of close case, and "the jury struggled with precisely the theory that adequate representation would have bolstered"); Daniels v. Woodford , supra, 428 F.3d at 1209–10 ("[that] [t]he jury deliberated for two days before returning a verdict ... suggests that [it] may have been influenced by [additional] mitigation evidence had it been offered," and "[t]his alone [was] sufficient for a finding of prejudice"); Dallago v. United States , 427 F.2d 546, 559 (D.C. Cir. 1969) ("[t]he jury deliberated for five days, and one would expect that if the evidence of guilt was overwhelming the jury would have succumbed much sooner").

In light of the foregoing, it is apparent that Sherman's deficient performance in failing to adduce the testimony of Ossorio resulted in prejudice to the petitioner. As we have explained, without Ossorio's testimony, the state was able to attack the petitioner's alibi—a complete alibi for the time period during which it is highly likely that the victim was murdered—as part of a Skakel family conspiracy to cover up the petitioner's involvement in the victim's murder. According to the state, this scheme began with the removal and destruction of incriminating evidence immediately after the victim's murder and continued in the months and years thereafter, culminating some twenty-five years later in the false alibi advanced by the petitioner's close family members both during the grand jury proceedings and at trial. Ossorio's disinterested testimony, if credited by the jury, would have defeated the state's theory of a fraudulent family conspiracy, with the false alibi as its centerpiece, thereby requiring the state to prove that the murder occurred sometime after 11 p.m. on October 30, 1975, when the petitioner returned home from the Terrien residence—a nearly impossible burden in view of the fact that the state has never proffered any explanation as to where the victim may have been or what she may have been doing from 9:30 p.m., when she was last seen alive, until at least 11 p.m. Moreover, as we have explained, an objective review of the state's evidence reveals that it was highly impeachable and far from strong. Under any reasonable view of the state's case, therefore, and considering the gravity of the prejudice flowing from Sherman's failure to call Ossorio as a witness, that failure seriously undermines confidence in the verdict. In such circumstances, the sixth amendment requires that the petitioner be afforded a new trial at which he will have the benefit of Ossorio's important exculpatory testimony.

VI

JUSTICE EVELEIGH's DISSENTING OPINION

In his dissenting opinion, Justice Eveleigh repeatedly charges the majority with minimizing the import of or overlooking the evidence and case law that do not "comport with its narrative of the case ...." As we explain hereinafter, these accusations are baseless.

A

The Facts

Throughout his dissenting opinion, Justice Eveleigh claims that the majority "consistently downplays or ignores evidence and arguments that contradict or fail to support its own theory of the case ...." Footnote 8 of Justice Eveleigh's dissenting opinion. To the contrary, we have scrutinized every line of testimony in this case, and carefully evaluated each and every exhibit, affording due consideration to the entire record in light of the parties' claims and arguments. Upon review of that record, we strongly disagree with Justice Eveleigh as to the strength and import of much of the evidence. In large part, that disagreement stems from the fact that Justice Eveleigh consistently construes the evidence in the light most favorable to the state, scarcely acknowledging any weakness in the state's case, rather than viewing the evidence objectively, as Strickland requires.³⁰ Justice Eveleigh's flawed methodology is compounded by his reliance on arguments that are either inconsistent with the state's theory of the case at trial—and thus were never made by the state—or are so speculative or tenuous that they have not been made by the respondent on appeal.

30 A good example of Justice Eveleigh's one-sided approach to reviewing the evidence is reflected in his extensive parsing of every inculpatory inference that the jury possibly could have drawn from Andrea Shakespeare's testimony that, when she and Julie Skakel arrived at the Skakel home on the afternoon of October 31, 1975, to what one state witness described as a "chaotic" scene of police, press, neighbors and children running in and out of the house, the petitioner approached their car and excitedly informed them that the victim had been murdered and that he and Thomas Skakel were the last ones to have seen her alive. Even though what the petitioner told his sister and Shakespeare was true; see footnote 28 of this opinion; Justice Eveleigh nonetheless construes the petitioner's statement to his sister and Shakespeare as tantamount to a confession, a conclusion that we believe is completely unsupported.

Rather than attempt to identify and explicate the numerous occasions on which Justice Eveleigh resorts to this methodology, we turn to one such instance that exemplifies it, namely, his treatment of the testimony of Michael Meredith. Justice Eveleigh roundly criticizes the majority for undervaluing Meredith's testimony. As Justice Eveleigh notes, Meredith testified that the petitioner had told him that he climbed a tree next to the Moxley house on the night of the murder and masturbated. Meredith further testified that, throughout his conversation with the petitioner, "[he] got the feeling like it was something that [the petitioner] had done before because he said ... [he] could see her when she was getting dressed or undressed or coming out of the shower ...." Justice Eveleigh contends that Meredith's testimony "conclusively demonstrated that, if the petitioner did go to the Terrien home, then the victim must have been murdered after [11 p.m.] ...." According to Justice Eveleigh, "[because] the jury reasonably could have credited [Meredith's] testimony indicating that the petitioner ... [saw] the victim alive [after 11 p.m.]," the petitioner's alibi story was immaterial, and Sherman's failure to present a stronger alibi could not have been prejudicial.

Justice Eveleigh's arguments with respect to Meredith, however, cannot be squared with the state's express theory of the case at trial, which, as the state's attorney explained to the jury, is that the victim never returned home on the night in question .³¹ Indeed, as we previously indicated, the state's attorney argued that the petitioner concocted the story about masturbating in a tree out of concern that his DNA might one day be discovered on or near the victim's body.³² See footnote 29 of this opinion. In considering the prejudicial impact of Michael Sherman's deficient performance, this court must consider the case as it was actually presented to the jury. See, e.g., Weeden v. Johnson , 854 F.3d 1063, 1072 (9th Cir. 2017) ("[i]n determining how omitted evidence would have altered the trial, [courts] may not invent arguments the prosecution could have made" [internal quotation marks omitted] ); Hardy v. Chappell , 849 F.3d 803, 823 (9th Cir. 2016) (" Strickland does not permit the court to reimagine the entire trial. [The court] must leave undisturbed the prosecution's case. [The court] only envision[s] what [counsel] should have presented in [the petitioner's] defense and determine[s] how that would have altered the trial. In doing so, [the court] may not invent arguments the prosecution could have made if it had known its theory of the case would be disproved."); Syed v. State , supra, 236 Md.App. at 283–87, 181 A.3d 860, 2018 WL 1530300, *49–50 (rejecting state's attempt to alter its theory of when murder occurred in light of credible alibi testimony adduced at petitioner's habeas trial).

31 In his opening statement to the jury, the state's attorney argued, "Martha Moxley, then and forever fifteen years of age, went out that evening and ... never made it home, resulting in an all-night effort by her mother, Dorothy [Moxley], to learn [of] her daughter's whereabouts, which weren't ... discovered until around noon the next day ...." During his closing argument, the state's attorney similarly argued: "[W]e realize that Martha [Moxley] didn't get home as expected by 10 or 10:30 [p.m.], and we could pretty much conclude that, by 1 [a.m.] ... she was never coming home." He further argued: "Martha [Moxley] ... wasn't supposed to be in until about 10:30 or so that night. Of course, she never got there." Finally, he stated: "Dorothy [Moxley] didn't become concerned until after 11 [p.m.] or so. Needless to say, Martha [Moxley] never did make it home."

32 Specifically, the state's attorney argued: "You didn't have to be a fly on the wall when ... Sutton Associates came into the picture in 1992 to understand why the [petitioner] soon was serving up his bizarre tale of masturbation in a tree .... He had masturbated, not in that cedar tree by John Moxley's room and not in that monkey tree that's on the side of the house but, rather, in the vicinity of [the victim's ] body ."

That the state's attorney did not argue, on the basis of Meredith's testimony, that the victim went home after leaving the Skakel driveway and proceeded to her room, where she remained until after the petitioner returned home from the Terrien house sometime after 11 p.m., all the while avoiding the notice of her worried mother, is undoubtedly because such an argument would be flatly contradicted by the testimony of two key witnesses for the state, the victim's mother and John Moxley, the victim's brother, both of whom testified that the victim was not at home at 11 p.m. They, of course, were two of only a handful of witnesses in the entire case with firsthand knowledge of the events in question. Justice Eveleigh has not cited a single case—because there is none—in which this or any other reviewing court has deemed itself free to adopt a theory of the case that was expressly rejected by the state at trial, and then assume that the jury could have found the defendant guilty on the basis of that theory.³³ Recognizing that the respondent's failure to point to any evidence that supports a finding that the victim was alive when the petitioner returned home from the Terrien house is fatal to the respondent's partial alibi claim, Justice Eveleigh purports to identify such evidence. For example, Justice Eveleigh contends that the jury reasonably could have concluded, on the basis of the testimony of the victim's mother, either that the victim was with friends from 9:30 to 11 p.m., because the victim's mother testified that the victim "really had no formal curfew," or that the victim was at home from 9:30 to 11 p.m., unbeknownst to her family, because her mother also testified "that it was possible ... that the victim had returned home during that time and then [had] gone out again without her knowledge." Justice Eveleigh maintains, in fact, that "there are countless, plausible explanations for where the victim could have been during the alibi period. In an age before cellphone communications, [she] could have been walking around the neighborhood looking for her friends. She could have been engaging in mischief night festivities with her [eleven year old] friend [Geoffrey] Byrne, who died a few years after the [victim's] murder, or [she could have been] hanging out at the Skakel residence with [Thomas Skakel], [who did not testify] at [the petitioner's criminal] trial. She could have been out with some other young man who, presumably, would not have been especially eager to come forward after the murder and inform law enforcement that he had been the last person to see her alive." Text accompanying footnote 37 of Justice Eveleigh's dissenting opinion. As we previously noted, these possibilities are so remote and conjectural that neither the state nor the respondent has ever seen fit to mention them.

33 We also reject Justice Eveleigh's repeated assertion that "the majority rel[ies] on ... facts and evidence that were not part of the trial record and could not have been considered by the jury [during] its deliberations." As we explained, we have relegated to footnotes certain evidence or information made known in court proceedings that occurred subsequent to the petitioner's criminal trial solely for the purpose of providing context for certain claims raised in this appeal. As we have made crystal clear on each such occasion, however, we have not relied on that evidence or information in resolving any of the issues presently before this court.

Even if we put aside the highly speculative nature of Justice Eveleigh's hypothetical scenarios, they are simply irrelevant because the issue before this court is not whether an argument can be made—however tenuous—that the victim was murdered after 11 p.m. The issue, rather, is whether the jury reasonably could have concluded that she was murdered prior to 11 p.m. For the reasons previously set forth in this opinion, it is readily apparent that the jury very well could have found—in fact, it is highly probable that it did find, on the basis of the evidence and arguments presented at trial—that she was murdered before 11 p.m.

Not only do they miss the point, Justice Eveleigh's speculative scenarios are also contrary to the arguments that the state's attorney made at trial. As we discussed, the state's attorney did not argue that the victim, unbeknownst to her mother and brother, returned home at 9:30 p.m. and remained there, unnoticed, until sneaking out sometime after 11 p.m. On the contrary, in his closing argument to the jury, the state's attorney asserted repeatedly and unequivocally that the victim was not home during that time period and that, in fact, the victim never made it home on the night in question .³⁴ Nor did the state's attorney argue that the victim was "hanging out at the Skakel residence with" Thomas Skakel or engaging in activities with Byrne, or "out with some other young man who, presumably, would not have been especially eager to come forward after the murder and [to] inform law enforcement that he had been the last person to see her alive." Id. Undoubtedly, the state's attorney did not make the arguments that Justice Eveleigh now asserts on its behalf because there was no evidence in the record to support them. E.g. State v. Lopez , 280 Conn. 779, 803, 911 A.2d 1099 (2007) ("Counsel may comment [on] facts properly in evidence and [on] reasonable inferences to be drawn from them.... Counsel may not, however, comment on or suggest an inference from facts not in evidence." [Internal quotation marks omitted.] ). Indeed, the evidence in the state's possession at the time of trial, which is part of the record in this appeal, indicates that the police interviewed hundreds of people at the time of the murder, including Byrne and the victim's other friends and neighbors, subjecting many of them to multiple lie detector tests; and yet not one of them professed any knowledge of the victim's whereabouts after 9:30 p.m. Justice Eveleigh also overlooks the fact that the state's attorney argued that Thomas Skakel had an alibi for the entire evening after the victim reportedly left him, at approximately 9:30 p.m., by his back door.

34 We note that, in arguing that the jury reasonably could have found that the victim went home after leaving the Skakel driveway, where she was observed by the petitioner after his return from the Terrien residence sometime after 11 p.m., Justice Eveleigh omits several facts that are inconsistent with this argument. For example, in support of his contention that the victim could have been at home from 9:30 until 11 p.m., Justice Eveleigh cites the testimony of the victim's mother "that it was possible ... that the victim had returned home during that time and then [had] gone out again without her knowledge." This was hardly the import of that testimony. Rather, when the victim's mother was asked whether such a scenario was possible, she responded: "Yes, I suppose it is possible, but, you know, I didn't know that it had ever ... happened before. [The victim] was very good at telling me everything that was going on. I mean, she talked to me all the time, and, you know, I don't think she did [that], but, you know, there is always a chance she could have." In any event, in stating unequivocally to the jury in closing argument that the victim never returned home that evening, the state's attorney expressly disavowed the theory that Justice Eveleigh posits—for the first time in the long history of this case—in his dissenting opinion.

Justice Eveleigh makes several additional factual arguments, purportedly to demonstrate why the jury reasonably could have found that the murder did not occur until after 11 p.m. and, therefore, why Ossorio's testimony was immaterial. For example, Justice Eveleigh argues that the jury may not have attached any significance to the violent barking by Helen Ix' dog, Zock, near the crime scene because (1) "[d]ogs, of course, are wont to bark, and the jury heard undisputed testimony from multiple witnesses that ... the Skakel family's German Shepherd ... and ... Zock ... as well as other neighborhood dogs, were chronic barkers," and (2) Ix testified "that [Zock's] barking was unusual more for its duration than its intensity" and that Zock's behavior could have been explained by the fact that many teenagers were out celebrating mischief night. We agree wholeheartedly that the jury was not required to attach significance to Zock's behavior, even though the Greenwich police and Joseph Jachimczyk, the medical examiner from Texas, did so for the better part of twenty-five years, both believing that it was a reliable indicator of the victim's time of death. As we have explained, however, the issue we must decide, and the issue that Justice Eveleigh does not address, is whether the jury reasonably could have found that the victim was murdered between 9:30 and 11 p.m.

Nevertheless, we take issue with Justice Eveleigh's assertion that Ix testified that Zock's behavior that evening "was unusual more for its duration than its intensity," and that Zock could have been reacting to teenagers out on mischief night rather than the assault on the victim. In fact, as we previously indicated, Ix testified that she had never seen her dog behave as he did on the night in question, that "[h]e always barked but not like that," and that "[t]here was really a difference ...." Indeed, Ix explained that Zock was so "disturbed by something that was going on," that he refused to come for the very first time in his life. She also stated that she never saw Zock behave in the same manner again after that evening. Nor is it accurate to say that Ix testified that Zock's behavior could just as easily have been explained by the fact that many teens were out celebrating mischief night. In fact, when asked that question, Ix replied, only if the teenagers had been "doing something destructive ...." Justice Eveleigh does not identify any evidence, and we are aware of none, that anyone other than the victim's killer was engaged in destructive behavior in the vicinity where Zock was observed in an extremely distressed state, barking in the direction of the victim's body.

B

The Law

Justice Eveleigh asserts that we have ignored or misapplied the relevant law in a number of respects, but he appears to treat two such assertions as most consequential. He contends, first, that we have improperly failed to acknowledge that the petitioner's alibi was a partial one and, therefore, of no consequence for purposes of either of Strickland 's two prongs, and, second, that we have substituted our judgment for that of the jury's in evaluating the strength of the state's case. In fact, it is Justice Eveleigh, not the majority, who has misapplied governing legal principles.

With respect to Justice Eveleigh's first contention, he states that, for purposes of evaluating prejudice, "[the] cases almost universally hold that defense counsel's failure to investigate, identify, or present an alibi witness either does not constitute deficient performance or is not prejudicial when that alibi would cover only a portion—even a substantial portion—of the time period during which the crime could have been, or was alleged to have been, committed." We do not disagree with this assertion as a general proposition. According to Justice Eveleigh, however, this general rule applies without exception in all circumstances in which, as in the present case, an alibi does not cover the entire time frame within which the crime could have occurred. As the following two hypothetical scenarios reveal, Justice Eveleigh is demonstrably wrong in applying broadly applicable partial alibi principles to the particular circumstances of the present case. Under hypothetical number one, which mirrors the factual scenario of the vast majority of cases in which an alibi does not span the full time period within which the crime could have been committed, the defendant is alleged to have committed that crime on any of days one through ten, and the defendant has an alibi for days one through nine only. In such circumstances, there is no less reason for a jury to find that the crime was committed on day ten than on days one through nine, and the alibi is considered a partial one, which courts frequently do not consider material for purposes of the performance or prejudice prongs of Strickland .

Under the second hypothetical, the facts are the same as those of the first hypothetical except that, in addition, both the defendant and the state have presented evidence that establishes or purports to establish that it is highly likely that the crime was committed on days one through nine, and, thus, it is highly unlikely that the crime was committed on day ten. In those circumstances, the potential significance of the alibi is manifest because, if the jury is persuaded by the parties' evidence that the crime likely was committed on one of the first nine days, the defendant's alibi for those days, if credited, would exonerate him. Of course, this second hypothetical mirrors the facts of the present case.³⁵ Justice Eveleigh's failure to acknowledge the critical distinction between these two scenarios leads him to the erroneous conclusion that the alibi advanced by the petitioner in the present case is no different from the ordinary case in which it is no more or less likely that the defendant committed the crime at any particular point in time within the period alleged by the state. Indeed, under Justice Eveleigh's flawed analysis, as long as there is even the remotest possibility that the defendant could have committed the crime at a time not covered by the alibi, that alibi would be deemed partial and therefore immaterial for Strickland purposes. In fact, this apparently would be the result under the analysis employed by Justice Eveleigh even if, for example, the evidence established to a near certainty that the crime was committed during the period covered by the defendant's alibi. We reject this conclusion because it so clearly defies reason and common sense.

35 Justice Eveleigh argues that the distinction the majority draws between a case in which there is evidence that the crime was committed during the alibi period and a case in which there is no such evidence "is inconsistent with the very concept of an alibi," which "has long been understood to mean that it is impossible for a person to have committed a crime because he or she was at another location when the crime was ... committed." (Emphasis omitted.) According to Justice Eveleigh, as long as "there is a realistic possibility that the crime was committed outside of the alibi period," then the petitioner cannot possibly have been prejudiced by counsel's failure to present an alibi witness. Contrary to Justice Eveleigh's contention, our determination in the present case is not remotely inconsistent with the concept of an "alibi," namely, "[a] defense based on the physical impossibility of a defendant's guilt ...." Black's Law Dictionary (10th Ed. 2014) p. 87. As we explained, if the jury had credited the petitioner's alibi and had concluded that the murder occurred at approximately 10 p.m.—as the evidence clearly suggested—then the jury necessarily would also have had to conclude that it was impossible for the petitioner to have committed the crime.

Ignoring this distinction, Justice Eveleigh argues that the majority is of the view that an alibi that covers only part of the time period during which a crime could have been committed is partial only if the probability that the crime was committed during the time period covered by the alibi is equal to the probability that it was committed during the alibi period. Justice Eveleigh then asserts that, in many partial alibi cases, "the evidence indicate[s] that it would have been extremely difficult for the petitioner to have committed the crime outside of the period during which he had a potential alibi. Because the evidence [leaves] open some realistic possibility that the crime [was] committed outside of the alibi period, however, the court applie[s] the partial alibi rule." Justice Eveleigh incorrectly equates those cases in which the alibi covers a majority of the time during which the crime could have been committed with a case, like the present one, in which the evidence demonstrates that it is more likely that the crime was committed during that alibi period.³⁶ As we explained, however, for present purposes, the two scenarios are vastly different. It is Justice Eveleigh's reliance on the false equivalence between the two scenarios that leads him to the wrong result.³⁷

36 Justice Eveleigh cites to just three cases, namely, Fargo v. Phillips , 58 Fed. Appx. 603 (6th Cir.), cert. denied, 539 U.S. 932, 123 S.Ct. 2585, 156 L.Ed. 2d 613 (2003), Spearman v. Commissioner of Correction , 164 Conn. App. 530, 138 A.3d 378, cert. denied, 321 Conn. 923, 138 A.3d 284 (2016), and Tinsley v. Commonwealth , Docket No. 1026-11-2, 2012 WL 1499352 (Va. App. May 1, 2012) ; for the proposition that courts have applied the so-called partial alibi rule in cases involving facts resembling those of the present case. Suffice it to say that those cases bear no legal or factual resemblance to the present case for numerous reasons, but most significantly because they simply do not involve a factual scenario, like the present one, in which the jury reasonably could find, on the basis of the evidence, that it is more likely that the crime was committed during a particular portion of the alibi period than during the remainder of that period.

37 Justice Eveleigh's reliance on this false equivalence also leads him repeatedly to misstate "[t]he principal question to be resolved" as one that requires a determination of "whether it is highly likely that the victim was [murdered] prior to ... 11:15 p.m. ... or whether the jury reasonably could have found that she was [murdered] later that night .... If the former, then Ossorio's testimony would have supported a full alibi, and we must determine whether presenting that testimony to the jury would have been reasonably likely to result in a different outcome. If the latter, however, then the trip to the Terrien home represented merely a partial or incomplete alibi, and, as a matter of law, defense counsel's failure to present one additional witness in support of that partial alibi was, at worst, a harmless error." (Emphasis omitted.) This statement of the "principal question" reflects Justice Eveleigh's fundamental misapprehension of the issue actually presented by this appeal. As we explained, because Sherman performed deficiently in failing to present the alibi testimony of Ossorio, the question is not "whether it is highly likely that the victim was [murdered] prior to ... 11:15 p.m. ... or whether the jury reasonably could have found that she was [murdered] later [than] that ...." The question is whether there is a reasonable probability of a different result because the jury reasonably could have found that the victim was murdered prior to 11:15 p.m. Under Justice Eveleigh's incorrect statement of the issue presented, the petitioner is required to prove that the victim necessarily was murdered prior to 11:15 p.m. in order to demonstrate prejudice, which is clearly not the applicable standard. See, e.g., Lapointe v. Commissioner of Correction , supra, 316 Conn. at 263, 112 A.3d 1 ("[t]he question is not whether the defendant would more likely than not have received a different verdict with the evidence, but whether in its absence he received a fair trial, understood as a trial resulting in a verdict worthy of confidence" [internal quotation marks omitted] ); see also Avery v. Prelesnik , supra, 548 F.3d at 439 ("[w]e do not ask whether [the petitioner] was ultimately innocent, but, rather, whether he was deprived [of] a reasonable shot of acquittal").

Second, Justice Eveleigh argues that the majority misapplies Strickland 's prejudice prong by considering the potential impact of Ossorio's testimony on the testimony of those witnesses who claimed that the petitioner had either confessed to, or otherwise implicated himself in, the victim's murder, because the latter testimony was not "linked to any particular time of death [and did not require] that the petitioner be present at the crime scene during the purported alibi period," and, therefore, the testimony is not directly affected by Ossorio's testimony. According to Justice Eveleigh, the only evidence that would be directly affected by the testimony of an alibi witness is evidence placing the petitioner at the crime scene at the exact moment when the crime was committed, such as the testimony of an eyewitness. Justice Eveleigh argues that, because there are no such witnesses in this case, and because the petitioner's purported admissions are vague as to time, the petitioner cannot possibly have been prejudiced by the absence of Ossorio's testimony. Justice Eveleigh is unable to cite a single case—because there is no such case—in which a court, in deciding a Strickland claim, has concluded that a credible alibi does not call into question the credibility of an alleged confession merely because the alleged confession is devoid of specifics. There is no such case because it is self-evident that the persuasive force of a disputed confession—particularly one lacking in crucial details, such as when the crime was committed—may well be undermined if that confession is placed in the context of an alibi that itself is persuasive.

Thus, in Gaines v. Commissioner of Correction , supra, 306 Conn. 664, 51 A.3d 948, a case very much on point, this court concluded that counsel's failure to present the testimony of two credible alibi witnesses "cast appreciable doubt on the state's case against the petitioner and ... undermined this court's confidence in the outcome of his criminal trial." (Internal quotation marks omitted.) Id., at 691, 51 A.3d 948. In Gaines , "the only evidence implicating the petitioner in the murders was the testimony of [two witnesses who claimed that the petitioner had confessed to or otherwise implicated himself in the murder] and, to a lesser extent, the testimony of [a third witness, who claimed to have seen the petitioner with the same type of gun used to commit the crime]." Id. Nevertheless, this court concluded that "[the] testimony [of the state's witnesses] was, itself, subject to substantial impeachment evidence that they had only implicated the petitioner to serve their own needs .... The alibi defense, [however ] ... likely would have permeated, to some degree, every aspect of the petitioner's criminal trial and raised a reasonable doubt in the minds of the jurors as to the petitioner's guilt ." (Emphasis added.) Id. In Lapointe v. Commissioner of Correction , supra, 316 Conn. at 225, 112 A.3d 1, this court similarly concluded that counsel's failure to present exculpatory forensic evidence, which established a narrower window in which a fire could have been started, prejudiced the petitioner because there was a reasonable probability that, if the jury had credited that evidence, it would have been less inclined to credit the petitioner's multiple confessions, which the petitioner claimed were false, and more likely to credit his alibi. See id., at 348–49, 112 A.3d 1.

Indeed, Justice Eveleigh's assertions notwithstanding, courts uniformly have held that a reviewing court cannot determine whether the omission of exculpatory evidence prejudiced a defendant without considering the impact of that evidence on every aspect of the state's case. See, e.g., Strickland v. Washington , supra, 466 U.S. at 695–96, 104 S.Ct. 2052 ("[s]ome errors will have had a pervasive effect on the inferences to be drawn from the evidence, altering the entire evidentiary picture"); see also, e.g., Kyles v. Whitley , supra, 514 U.S. at 445, 115 S.Ct. 1555 (considering impact, on every aspect of state's case, of state's failure to disclose evidence that could have been used to impeach several eyewitnesses and concluding that "[the] [d]amage to the prosecution's case would not have been confined to evidence of the eyewitnesses, for [the undisclosed evidence] would have raised opportunities to attack not only the probative value of crucial physical evidence and the circumstances in which it was found, but the thoroughness and even the good faith of the investigation, as well"); United States v. Agurs , 427 U.S. 97, 112, 113, 96 S.Ct. 2392, 49 L.Ed. 2d 342 (1976) (in assessing prejudice under Strickland and Brady , omitted exculpatory evidence "must be evaluated in the context of the entire record," and exculpatory evidence of even "minor importance" may well be "sufficient to create a reasonable doubt" when original case was not particularly strong); Lapointe v. Commissioner of Correction , supra, 316 Conn. at 294, 112 A.3d 1 (examining entirety of state's case and concluding that "the state's less than compelling case against the petitioner was such that any new evidence tending to cast doubt on the petitioner's responsibility for the charged crimes could well have led to an acquittal").

Accordingly, Justice Eveleigh is also manifestly incorrect in asserting that the habeas court and the majority, merely by pointing out the weaknesses in the state's evidence, have engaged in improper appellate fact finding or have made improper credibility determinations.³⁸ See, e.g., Gaines v. Commissioner of Correction , supra, 306 Conn. at 691, 51 A.3d 948 (concluding that petitioner was prejudiced by counsel's failure to present alibi defense because "[the] testimony [of the state's witnesses] was ... subject to substantial impeachment evidence that they had only implicated the petitioner to serve their own needs"). Contrary to Justice Eveleigh's contentions, the habeas court did not find—nor do we conclude—that the state's witnesses "lacked credibility" in the eyes of the jurors. Rather, the habeas court determined, and we fully agree, that, based on an objective evaluation of the testimony, the state's case was not strong, in that, as in Gaines , the credibility of all of the state's witnesses was subject to question. Contrary to Justice Eveleigh's claim, the fact that the jury may have resolved issues of witness credibility in favor of the state at trial is not at issue in this appeal; the question, rather, is whether the jury could have viewed the testimony of those witnesses differently if Sherman had presented the petitioner's alibi defense in a manner consistent with his obligations under the sixth amendment. To support his mistaken view of the law, Justice Eveleigh employs language from cases that simply are inapposite to the issue presented. For example, he quotes from Hope v. Cartledge , 857 F.3d 518, 525 (4th Cir. 2017), cert. denied, ––– U.S. ––––, 138 S.Ct. 646, 199 L.Ed. 2d 530 (2018), for the proposition that " '[t]he guilty verdict necessarily establishes that the jury found the [s]tate's witnesses to be credible and believed the [s]tate's version of events.' " Relying on this language, Justice Eveleigh then asserts that "[t]he argument that the state's case was weak because ... witnesses lacked credibility is, therefore, generally without merit because the jury necessarily resolved those questions in favor of the state." Even a cursory review of Hope reveals that it does not support the proposition for which it is cited because it does not involve counsel's failure to present exculpatory evidence. In Hope , the claimed ineffectiveness was counsel's failure to request a jury instruction indicating that the state was required to disprove the petitioner's alibi beyond a reasonable doubt. See Hope v. Cartledge , supra, at 523. The court ultimately determined that the petitioner was not prejudiced by counsel's deficient performance because the jury had been instructed fifteen times that the state was required to prove the petitioner's guilt beyond a reasonable doubt. Id., at 524–25. In reaching its conclusion, the court also observed that, at trial, the petitioner and the state had presented two mutually exclusive versions of the facts and that it was clear that the jury had credited the state's witnesses, who identified the petitioner as the perpetrator, over the petitioner's alibi witnesses, who claimed that the petitioner was with them at a party during the robbery. Id., at 525. The court concluded, therefore, that there was no reasonable probability of a different result at a new trial because the jury had already found the state's witnesses to be more credible than the petitioner's alibi witnesses, and no additional jury instruction regarding the state's burden of proof would have affected that finding. See id. Thus, Hope does not support the principle, espoused by Justice Eveleigh, that, because a jury may have credited the testimony of some or all of the state's witnesses, a petitioner is somehow foreclosed from demonstrating that the state's case nevertheless was not strong for purposes of the prejudice prong of Strickland .

38 See, e.g., text accompanying footnote 51 of Justice Eveleigh's dissenting opinion ("[a]lso of concern is the apparent willingness of both the habeas court and the majority to substitute their own credibility determinations for those of the jury"); footnote 51 of Justice Eveleigh's dissenting opinion ("the habeas court appears to have determined that the state's trial witnesses lacked credibility solely on the basis of its review of the cold trial record, [even though] the jury, which had the opportunity to observe [their] demeanor ... firsthand, clearly credited at least some of their testimony" [emphasis omitted] ); see also part III A 2 of Justice Eveleigh's dissenting opinion ("The majority examines each of the state's witnesses, explaining—from a cold trial record—why it does not find their testimony to be believable and, therefore, why a jury also conceivably might not credit them. In so doing, the majority ... relies on speculative arguments, which the petitioner himself has never made, requiring credibility determinations best left to the trier of fact. Such a review is, in my view, simply inappropriate in the context of a Strickland analysis.").

VII

CONCLUSION

Upon reconsideration of our original decision in this case, we agree with the petitioner that the habeas court correctly concluded that Sherman's failure to identify and call Ossorio as an alibi witness constituted deficient performance under the first prong of Strickland and that, under Strickland 's second prong, that inadequate performance resulted in prejudice to the petitioner sufficient to undermine confidence in the outcome of his criminal trial. Consequently, the habeas court also correctly determined that the petitioner, having been deprived of a fair trial, is entitled to a new trial at which he will have the benefit of Ossorio's alibi testimony.

The judgment is affirmed.

In this opinion McDONALD and ROBINSON, Js., concurred, and D'AURIA, J., concurred in all but part II.

D'AURIA, J., concurring in part. I agree with and join in the majority opinion, with the exception of part II of that opinion, about which I express no view.

I

On the last business day of 2016, a majority of an en banc panel of this court officially released its decision in this case, reversing the habeas court's judgment and thereby reinstating the petitioner's conviction of murder. The vote was four to three. The next day, the author of the majority opinion retired, leaving judicial service before his term of office expired. Six days later, the petitioner, Michael Skakel, filed a motion for reconsideration en banc, as our rules of practice permit. See Practice Book § 71–5. In that motion, he argues that, because of the authoring justice's retirement, Practice Book § 71–5 and General Statutes §§ 51–207 and 51–209 require that the court provide a "replacement" seventh panel member so that on reconsideration he will "enjoy a panel of the same size as that which heard the case." The respondent, the Commissioner of Correction, has objected to the petitioner's request for an en banc court to decide his motion for reconsideration.

The dispute about the proper composition of the panel deciding this motion to reconsider was foreseeable. In the past several years, other members of this court have retired and left judicial service just after the expedited official release of decisions in which they had participated, but before the deadline for postjudgment filings had passed (including cases heard en banc and split decisions). The uncertainty the petitioner's motion has created for the parties, the attorneys, the families and the public, who all seek finality in this matter, was therefore matched only by the certainty that such a motion would be filed days after the release of this court's decision. Although I am entirely comfortable with my own vote on the petitioner's motion for reconsideration, having been summoned by the remaining members of the original panel to rule on the motion, the dilemma created by the petitioner's motion and the subsequent delay in resolving this matter have been unfortunate.^{1 1 1}

1 As other opinions released today note, a majority of the remaining six members of the original panel voted to add a seventh judge to the panel deciding the motion for reconsideration.

1 It is undisputed that the deficient performance alleged to have occurred in the present case was not among the limited class of errors for which prejudice may be presumed. See Strickland v. Washington , supra, 466 U.S. at 692, 104 S.Ct. 2052.

1 Practice Book § 71–5 provides: "A motion for reconsideration will not be entertained unless filed with the appellate clerk within ten days from the date when the decision or any order being challenged is officially released. Any required fees shall be paid in accordance with the provisions of Sections 60–7 or 60–8.
"The motion for reconsideration shall state briefly the grounds for requesting reconsideration.
"A party may also request reconsideration en banc by placing 'en banc' in the caption of the motion and requesting such relief as an alternative to reconsideration by the panel.
"Whenever reconsideration en banc is sought, the motion shall state briefly why reconsideration en banc is necessary (for example, to secure or maintain uniformity of decision or because of the importance of the decision) and shall also state the names of the decisions, if any, with which the decision conflicts. A motion for reconsideration shall be treated as a motion for reconsideration en banc when any member of the court which decided the matter will not be available, within a reasonable time, to act on the motion for reconsideration."

II

The parties in this case have already received a considered decision of an en banc panel of this court. That decision is due respect, and a motion for reconsideration ordinarily should not be used merely as an "opportunity to have a second bite of the apple ...." (Emphasis omitted; internal quotation marks omitted.) Chapman Lumber, Inc. v. Tager , 288 Conn. 69, 94 n.28, 952 A.2d 1 (2008) ; see also C. R. Klewin Northeast, LLC v. Bridgeport , 282 Conn. 54, 101 n.39, 919 A.2d 1002 (2007).

Even so, "a motion for reconsideration is nothing more than an invitation to the court to consider exercising its inherent power to vacate or modify its own judgment ...." 56 Am. Jur. 2d 58, Motions § 40 (2010). As this court has noted, and as "the United States Supreme Court has said: 'It is a power inherent in every court of justice so long as it retains control of the subject matter and of the parties, to correct that which has been wrongfully done by virtue of its process.' ... United States v. Morgan , 307 U.S. 183, 197, 59 S.Ct. 795, 83 L.Ed. 1211 (1939)...." (Citation omitted.) Steele v. Stonington , 225 Conn. 217, 219 n.4, 622 A.2d 551 (1993). The usual grounds courts consider when deciding such a motion include whether "there is some decision or some principle of law which would have a controlling effect, and which has been overlooked, or that there has been a misapprehension of facts." (Internal quotation marks omitted.) Chapman Lumber, Inc. v. Tager , supra, 288 Conn. at 94 n.28, 952 A.2d 1. Ultimately, however, in exercising this inherent authority, "[t]he granting of a motion for reconsideration ... is within the sound discretion of the court." (Internal quotation marks omitted.) Mangiante v. Niemiec , 98 Conn. App. 567, 575, 910 A.2d 235 (2006).

Today's majority notes that it is not unprecedented for a state's highest court to reconsider—and then alter—the outcome of a case when a change in the court's membership has occurred between the announcement of the original decision and the court's ruling on a motion for reconsideration. See, e.g., United States Fidelity & Guaranty Co. v. Michigan Catastrophic Claims Assn. , 484 Mich. 1, 5–6, 795 N.W.2d 101 (2009) ; Johnson v. Administrator, Ohio Bureau of Employment Services , 48 Ohio St. 3d 67, 68–69, 549 N.E.2d 153 (1990). Predictably and understandably, in those cases, reconsideration was met with protest by those who considered it inappropriate for the court to alter the outcome of a case simply because, through intervening circumstances, membership on the court had changed. United States Fidelity & Guaranty Co. v. Michigan Catastrophic Claims Assn. , supra, at 27–30, 795 N.W.2d 101 (Young, J., dissenting); Johnson v. Administrator, Ohio Bureau of Employment Services , supra, at 71, 549 N.E.2d 153 (Holmes, J., dissenting).

However, the potential for a change in court membership during the pendency of any case—whether through death, resignation or the expiration of a judge's term—is a fact of life in our constitutional system. And in this case, the circumstances leading to this conundrum were not of the parties' creation but were created by this court. Moreover, as in any case, when this court exercises its discretionary authority to reconsider a decision, the outcome must turn ultimately not on the court's membership but on the strength of the parties' legal positions. Therefore, regardless of the route by which this matter came before me, as a current member of this court called to rule upon this case, I have undertaken to assess the strength of those positions.

Today, a majority of the court has decided to grant the petitioner's motion for reconsideration and to affirm the habeas court's judgment granting his petition. I concur in the determination that this court's earlier decision warrants reconsideration, and I join in all but part II of the majority's opinion. III

I share in the concern that this case has received so much judicial attention. But what ultimately distinguishes this petition for a writ of habeas corpus from so many others that come through our court system is that, after hearing testimony, taking evidence and finding facts, a habeas judge granted the petition.

It was the habeas judge—a veteran of both the habeas court and the Appellate Court—who listened to the explanation of the petitioner's counsel, Michael Sherman, for why he did not investigate the identity of Georgeann Dowdle's "beau." As today's majority notes well, the habeas judge did not credit this post hoc rationalization. See footnotes 17 and 20 of the majority opinion. The habeas judge's credibility determinations and findings of historical fact are entitled to deference from this court and cannot be disregarded unless clearly erroneous. Small v. Commissioner of Correction , 286 Conn. 707, 716, 946 A.2d 1203, cert. denied sub nom. Small v. Lantz , 555 U.S. 975, 129 S.Ct. 481, 172 L.Ed. 2d 336 (2008) ; see also Carr v. Schofield , 364 F.3d 1246, 1264–65 (11th Cir. 2004) ("[t]he determination of credibility, including an attorney's testimony regarding decisions of tactic and strategy, is within the province of the [habeas] court, which has the opportunity to observe and study the witness"), citing Cave v. Singletary , 971 F.2d 1513, 1518 (11th Cir. 1992) ("[c]onclusions regarding credibility are within the province of the [habeas] court judge, who has the opportunity to observe and analyze witnesses that we, as an appellate tribunal, lack").

The habeas judge also observed the testimony of Dowdle's "beau," Denis Ossorio, and, on the basis of his conduct, demeanor, and attitude, found him to be a "powerful witness in support of the petitioner's alibi claim."^{2 2 2} This finding is also entitled to deference from this court. Sanchez v. Commissioner of Correction , 314 Conn. 585, 604, 103 A.3d 954 (2014).

2 The habeas court found specifically with respect to Ossorio's testimony: "To the court, Ossorio was a disinterested and credible witness with a clear recollection of seeing the petitioner at the Terrien home on the evening in question. He testified credibly that not only was he present in the home with Dowdle and that he saw the petitioner there, but that he lived in the area throughout the time of the trial and would have readily been available to testify if asked." (Emphasis added.) The habeas court found Ossorio to be a "powerful" and "credible" witness, notwithstanding how long it took to locate him to tell his story. The record in this case is rife with witnesses called by both parties who never came forward until many years after the murder, or whose recollections changed—sometimes markedly—over time.

2 Of course, the Connecticut constitution may, in theory, afford broader protections with respect to the right to counsel than does the sixth amendment to the United States constitution. Neither the petitioner nor the majority argues, however, that the state constitution confers broader protection under the circumstances presented in this appeal.

2 General Statutes § 51–209 provides: "No ruling, judgment or decree of any court may be reversed, affirmed, sustained, modified or in any other manner affected by the Supreme Court or the Appellate Court unless a majority of the judges on the panel hearing the cause concur in the decision. No cause reserved, where no verdict has been rendered, judgment given or decree passed, shall be determined unless a majority of the judges on the panel hearing the cause concur in the decision. Whenever the Supreme Court is evenly divided as to the result, the court shall reconsider the case, with or without oral argument, with an odd number of judges. If the court reconsiders the case without oral argument, the judges who did not hear oral argument shall have available to them the electronic recording or transcript of the oral argument before participating in the decision. If a judge who is a member of a panel is not present for oral argument, the judge shall have available to him or her the electronic recording or transcript of the oral argument." See also Practice Book § 70–6 ("[w]hen the court is evenly divided as to the result, the court shall reconsider the case, with or without oral argument, with an odd number of justices or judges").

And it was the habeas judge who, after finding Ossorio credible, weighed Ossorio's testimony against the record of the petitioner's criminal trial. Independent of other claims made in the case, the habeas judge determined that, on the basis of the strength of Ossorio's credibility, had Sherman presented the testimony of this "disinterested and credible witness" to the jury "there is a reasonable likelihood that the jury would have been persuaded by his testimony" and that "there is a reasonable probability [the] outcome [of the trial] would have been different." See Shabazz v. State , 259 Conn. 811, 827–28, 792 A.2d 797 (2002) (trial court should grant petition for new trial if, among other things, "the evidence is sufficiently credible so that, if a second jury were to consider it together with all of the original trial evidence, it probably would yield a different result" [emphasis added] ).

That is the record confronting me. I have undertaken my task mindful that "[a]ppellate courts do not examine the record to determine whether the trier of fact could have reached a different conclusion. Instead, we examine the trial court's conclusion in order to determine whether it was legally correct and factually supported.... This distinction accords with our duty as an appellate tribunal to review, and not to retry, the proceedings of the trial court." (Internal quotation marks omitted.) O'Connor v. Larocque , 302 Conn. 562, 575, 31 A.3d 1 (2011). My review of the record and the governing law persuades me that the motion for reconsideration and the relief requested therein should be granted.

IV

I fully agree with today's majority that any " 'strategic' " decision by Sherman to disregard Dowdle's grand jury testimony about her "beau" was unreasonable under the circumstances and prejudicial to the petitioner's defense. See Gaines v. Commissioner of Correction , 306 Conn. 664, 674–76, 51 A.3d 948 (2012). An additional point made by the majority also bears emphasizing: although Sherman offered a justification for his disregard, it was not credited by the habeas court, and, in fact, when Sherman's actions are considered "as of the time of counsel's conduct"; (internal quotation marks omitted) id., at 688, 51 A.3d 948 ; it is apparent that any decision not to pursue the "beau" would have been contrary to the strategy Sherman actually employed at trial. Thus, any reliance on Sherman's stated reasons for not pursuing this lead to justify his conduct would resemble more of a post hoc rationalization than an actual strategic basis for his actions.

A

At the habeas trial, Sherman defended his failure to investigate as the product of a reasoned decision. He testified that he did not pursue the "beau" reference because Dowdle had told the grand jury that she had not seen her cousins, the Skakels, on the night in question and so he inferred that the "beau" likely had not either.^{3 3 3} Essentially, Sherman testified, tracking down the "beau" would be fruitless. This justification, if true, was unreasonable for all the reasons given by today's majority. See part V A of the majority opinion.

3 At the habeas trial, after being reminded of Dowdle's grand jury testimony, in which she first mentioned the "beau," Sherman testified as follows in response questions from the petitioner's counsel:
"Q. Did you ever try to find out who the beau was?
"A. No, because [Dowdle ] clearly testified [before the grand jury ] that she didn't know who was there ....
"Q. Did you ever try to find out who the beau was?
"A. He—he—I had no reason to suspect that he, in fact, would be helpful in that he saw [the petitioner ] and the rest of the boys. " (Emphasis added.) Days later in the habeas trial, the respondent's counsel gave Sherman another opportunity to explain his rationale for not even asking Dowdle who had been with her that night, and the following exchange occurred:
"Q. From that passage [in the grand jury testimony], did that give you any indication to think that [Dowdle's] beau would have heard or seen anything that night?
"A. Well, she said that she didn't venture out and that her beau was in [her] mother's library.
"Q. Okay. And would that give you any indication that, even if he heard voices, he'd be able to identify them.
"A. Well, it's her beau. I don't know how close they were, but she couldn't identify her own cousins' voices or her brother's voice necessarily, so how would the beau be able to do anything—be able to do anything—be any more accurate, especially, if [it was ] only occasionally that he went out there. " (Emphasis added.)

3 The jury was fully aware that the time of the offense was not an essential element of the charged crime and that the state was under no obligation to narrow the time of the commission of the offense more than the available evidence warranted. Indeed, the trial court charged the jury to that effect on two separate occasions.

3 For reasons that are not a matter of public record, the motion for reconsideration in Taylor was not decided until July 19, 2017.

As today's majority notes, however, the habeas court did not credit this testimony and instead found that "Sherman's failure to investigate in this regard cannot be attributed to any strategic decision under these circumstances." "The question of whether a decision was a tactical one is a question of fact." Porter v. Singletary , 14 F.3d 554, 558 (11th Cir.), cert. denied, 513 U.S. 1009, 115 S.Ct. 532, 130 L.Ed. 2d 435 (1994). Although courts often apply "a 'strong presumption' that counsel's attention to certain issues to the exclusion of others reflects trial tactics rather than 'sheer neglect' "; Harrington v. Richter , 562 U.S. 86, 109, 131 S.Ct. 770, 178 L.Ed. 2d 624 (2011) ; accord Yarborough v. Gentry , 540 U.S. 1, 8, 124 S.Ct. 1, 157 L.Ed. 2d 1 (2003) ; the habeas court in the present case ruled out trial tactics as a justification for Sherman's failure to investigate in this respect.^{4 4 4} In the light of this factual record, Sherman's testimony, which the original majority opinion accepted at face value, "resembles more a post hoc rationalization of counsel's conduct than an accurate description of [his] deliberations ...." Wiggins v. Smith , 539 U.S. 510, 527, 123 S.Ct. 2527, 156 L.Ed. 2d 471 (2003). A fair reading of the record therefore suggests that Sherman simply overlooked the significance of the "beau" referred to in Dowdle's grand jury testimony. See Wiggins v. Smith , supra, at 526, 123 S.Ct. 2527. (counsel's "failure to investigate thoroughly resulted from inattention, not reasoned strategic judgment").^{5 5}

4 As Justice Robinson expressed in his original concurring and dissenting opinion, "I cannot think of a single reasonable, strategic reason why Sherman would not at least attempt to track down Ossorio ...." Skakel v. Commissioner of Correction , 325 Conn. 426, 533, 159 A.3d 109 (2016) (Robinson, J. , concurring in part and dissenting in part).

4 The basic principle here is that several different things would have to happen for the petitioner to prevail, and he has the burden of proving that it is reasonably likely that all of them, collectively, would happen. For there to be nearly even odds that three different things will happen, it must be overwhelmingly likely that each will. For example, one would not place a wager that Boston teams will win the Super Bowl, the World Series, and the National Basketball Association Championship in a given year unless the New England Patriots, the Boston Red Sox, and the Boston Celtics were each prohibitive favorites. If each team was just a slight favorite, wagering on all three winning would be a sucker's bet.

4 Justice D'Auria concludes that "what ultimately distinguishes this petition for a writ of habeas corpus from so many others that come through our court system is that, after hearing testimony, taking evidence and finding facts, a habeas judge granted the petition." (Emphasis in original.) If the fact that the habeas court reached a particular result compelled this court to reach the same result, however, appellate review of such decisions would be pointless. I leave it to the reader to decide who makes the better and more justified arguments based on the record and the law. If the reader concludes, as I do, that there is no factual or legal support for the majority's decision, the only conclusion that can be drawn is that the majority decision is result oriented—a conclusion that is reinforced by the glaring inconsistencies between the various opinions issued by the author of the majority opinion in this case over the years and the majority decision here.

5 This reading of the record is consistent with the testimony of Jason Throne, Sherman's cocounsel, called to testify by the respondent at the habeas trial. Throne did not suggest that the defense team made a decision that the "beau" would not be helpful, or even that the "beau" was the topic of discussion. Rather, when asked by the petitioner's counsel, "[w]as there any discussion among the defense team to try and find out who this beau was?" Throne replied: "I don't recall any specific discussions about her beau; I just don't recall, I don't remember."

5 In the present case, the habeas court found prejudice largely on the basis of what the court described as its "fair reading" of the cold trial record, including the state's closing argument, the trial testimony of two medical experts, and several jury requests.

B

In considering the reasonableness of counsel's actions, Strickland v. Washington , 466 U.S. 668, 690, 104 S.Ct. 2052, 80 L.Ed. 2d 674 (1984), and its progeny admonish courts to review the challenged actions "as of the time of counsel's conduct." Neither the petitioner nor the respondent may benefit by later reconstructing the petitioner's criminal trial. This means at least two things.

First, "[i]t is all too tempting for a defendant to second-guess counsel's assistance after conviction or adverse sentence, and it is all too easy for a court, examining counsel's defense after it has proved unsuccessful, to conclude that a particular act or omission of counsel was unreasonable." Id., at 689, 104 S.Ct. 2052. Therefore, a "fair assessment of attorney performance requires that every effort be made to eliminate the distorting effects of hindsight ...." Id.

Second, but equally vital, the United States Supreme Court has instructed that courts are not to "indulge 'post hoc rationalization' for counsel's decisionmaking that contradicts the available evidence of counsel's actions ...." Harrington v. Richter , supra, 562 U.S. at 109, 131 S.Ct. 770, quoting Wiggins v. Smith , supra, 539 U.S. at 526–27, 123 S.Ct. 2527. The original majority did not mention this important Strickland principle. See Skakel v. Commissioner of Correction , 325 Conn. 426, 443–44, 159 A.3d 109 (2016).

C

It was reasonable for the habeas court not to credit Sherman's testimony about why he decided not to follow up on the "beau" reference because that testimony was contrary to Sherman's actual strategy at trial. As mentioned, Sherman testified that he did not investigate Dowdle's "beau" because he assumed that the "beau" had most likely not seen the Skakel brothers at the Terrien home given that, during her grand jury testimony, Dowdle could not remember seeing them. See footnote 3 of this concurring opinion. Recalling, however, that Strickland directs courts to review an attorney's challenged actions "as of the time of counsel's conduct"; Strickland v. Washington , supra, 466 U.S. at 690, 104 S.Ct. 2052 ; I agree with the majority's conclusion that Sherman's after-the-fact explanation for his failure to investigate is undermined by the fact that, at the time of the petitioner's criminal trial, Sherman himself did not accept Dowdle's grand jury testimony that she "[didn't] know who was there" at the Terrien's that night and that she could not identify the voices she heard that night with specificity. See part V A of the majority opinion. Although, in 1998, Dowdle had testified before the grand jury that she heard the voices of "cousins" or "Skakels" that night but "[didn't] know who was there at the time," Sherman came to the criminal trial equipped with a 1975 police report reflecting that, merely nine days after the murder, she had indicated that she had indeed "observed her brother and the Skakel brothers, Rushton–John–Michael, return to her house sometime around 10 [p.m.]" Sherman partially succeeded in getting Dowdle to recount to the jury what the report said she had told police in 1975 and that her memory of the night of the murder would have been better in 1975 than in 1998. Although Dowdle's own memory might have faltered years later, it would have been important for Sherman to determine if anyone else at her house that night also might have "observed" the petitioner. Sherman's conduct thus cannot be defended on the ground that he had presumed from Dowdle's grand jury testimony that neither Dowdle nor her "beau" had seen the petitioner that night.

Even if Sherman had decided to disregard Dowdle's 1998 grand jury testimony concerning her "beau," the identity of this other person (who turned out to be Ossorio) came up again—more conspicuously—at the criminal trial in 2002, thereby bolstering the finding that Sherman had acted unreasonably by failing to pursue this lead. On the heels of Sherman's attempt to get Dowdle to recollect what she had said to the police in 1975, the prosecutor, Jonathan Benedict, asked whether the "beau" referred to in the grand jury transcripts as being with her that night was her "husband." Dowdle corrected Benedict, testifying that the person was a "friend" of hers.^{6 6} Thus, Sherman was not confronted with just a single reference to the "beau" in the grand jury transcripts; his identity was probed at trial , including shortly after Sherman had attempted to refresh Dowdle's recollection so that she could recall seeing the petitioner at the Terrien home on the night in question.^{7 7}

6 The following colloquy between Benedict and Dowdle took place during the criminal trial:
"Q. Isn't it true that you and [your brother] James were never in the same room [on the night of the murder]?
"A. I don't remember.
"Q. Do you remember, you have indicated that you were with your husband?
"A. No, I didn't.
"Q. You were with somebody?
"A. A friend of mine.
"Q. I wouldn't suggest you misspoke, I misheard. You were with someone? "A. A friend of mine.
"Q. A friend. And where were you—where in the house were you when [James and the Skakels] entered?
"A. I was in the library."

6 I find it difficult, therefore, to understand the majority's sweeping statement that it is not aware of "a single case ... in which the failure to present the testimony of a credible, noncumulative, independent alibi witness was determined not to have prejudiced a petitioner under Strickland 's second prong" and, therefore, that defense counsel's "deficient performance in failing to investigate the independent alibi testimony of Ossorio was inherently or necessarily prejudicial." Many of the cases I have cited hold precisely that.
Indeed, even if we were to limit the discussion to cases in which the alibi at issue was undisputedly a complete one, the failure to identify or present a noncumulative alibi witness is not per se prejudicial. See, e.g., United States v. Turuseta , 853 F.Supp. 416, 422 (S.D. Fla. 1994), aff'd, 59 F.3d 1246 (11th Cir. 1995) (no prejudice because alibi testimony would not have directly contradicted confession); Torres–Arboleda v. Dugger , 636 So.2d 1321, 1324 (Fla. 1994) (no prejudice in light of eyewitness testimony placing petitioner at murder scene); see also United States ex rel. Kleba v. McGinnis , 796 F.2d 947, 959 (7th Cir. 1986) (Cudahy, J., concurring and dissenting) (although "complete alibi should ordinarily meet the prejudice requirement of Strickland ... if the evidence against a defendant is 'overwhelming,' a deficient performance of counsel that would otherwise be deemed prejudicial might fail to produce a reasonable probability of prejudice"); cf. Ford v. State , 314 S.C. 245, 248, 442 S.E.2d 604 (1994) (failure to seek alibi charge not prejudicial in light of overwhelming evidence of guilt).

7 In fact, at this time during the criminal trial, the court also allowed a transcript of the relevant portion of Dowdle's grand jury testimony into evidence as a full exhibit. While Sherman was present, Benedict read that portion of the transcript into the record, beginning with Dowdle's testimony that she "was in [the library] with [her] beau at the time ...."

7 It appears that the majority may have confused the legal standard governing partial alibis with the Strickland prejudice standard. Although it is true that the petitioner need only show a reasonable probability of a different result to prevail under Strickland , he nevertheless cannot prevail, as a matter of law, unless his alibi argument is legally sound. See King v. State , supra, 505 S.W.3d at 426.
In any event, the argument of the majority founders on another shoal. If his alibi was not a complete one, then the petitioner also cannot demonstrate that defense counsel performed deficiently. It is not deficient performance to fail to identify a witness who, at best, could have testified to a partial alibi. See Spearman v. Commissioner of Correction , supra, 164 Conn. App. at 546, 138 A.3d 378 ; Beasley v. State , supra, 18 So. 3d at 492. It is undisputed that, under the first prong of Strickland , the petitioner has the burden of establishing something more than the fact that the crime might have been committed during the alibi period.

It is no wonder then that the habeas court did not credit Sherman's after-the-fact justification for his failure to investigate. For many of the same reasons that lead today's majority to conclude that Sherman's failure to investigate the "beau" was not reasonable under the circumstances, the factual record, and logical inferences drawn from it, amply support the habeas court's finding that Sherman's failure to investigate the "beau" was not attributable to strategy.

At any rate, I agree with today's majority that, even if Sherman had made a strategic judgment about whether to investigate the "beau," that judgment was not reasonable under the circumstances, and the failure to follow this lead prejudiced the petitioner's defense at trial.

For the foregoing reasons, I concur in part with the majority's opinion.

EVELEIGH, J., with whom ESPINOSA and VERTEFEUILLE, Js., join, dissenting.

For the reasons articulated in the original majority opinion in this appeal; see Skakel v. Commissioner of Correction , 325 Conn. 426, 467–84, 159 A.3d 109 (2016) ; I continue to believe that the performance of defense counsel, Michael Sherman, was not deficient and, therefore, that the petitioner, Michael Skakel, was not denied his constitutional right to the effective assistance of counsel. Moreover, even if defense counsel's performance in failing to identify one additional alibi witness, Denis Ossorio, was so deficient as to warrant reconsideration of that decision, I am not persuaded that this alleged shortcoming prejudiced the defense.

As I explain more fully hereinafter, I could not disagree more strongly with the legal analysis of prejudice set forth in the new majority opinion. In particular, I believe that the majority fails to consider the well established rule that, as a matter of law, an alibi defense is no defense at all when it is reasonably possible that the crime was committed outside of the alibi period. That is certainly the case here.

Of perhaps equal concern is the majority's characterization of the factual record in this case. Although it recognizes that we are required to engage in a comprehensive, objective review of the factual record, the majority repeatedly minimizes or overlooks evidence and inferences that fail to comport with its narrative of the case, while exaggerating or overstating the strength of the petitioner's arguments. Barely acknowledging that the state's evidence was sufficiently compelling to persuade twelve jurors beyond a reasonable doubt of the petitioner's guilt, the majority takes it upon itself to make credibility determinations with regard to trial witnesses whose testimony bore no direct relationship to the alleged deficient performance of defense counsel. Perhaps most worrisome, the majority refers throughout its opinion to evidence from outside of the trial record, much of which is not even arguably a proper subject of judicial notice. Accordingly, I must respectfully dissent. I

Familiarity with the extensive factual and procedural background of the present case and the underlying trial regarding the murder of the victim, Martha Moxley, is presumed. See generally id. ; Skakel v. State , 295 Conn. 447, 991 A.2d 414 (2010) ; 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). Although the fact intensive nature of the question presently before this court will require a fairly extensive discussion of the evidence presented at trial, I defer that discussion to the relevant substantive parts of this dissenting opinion. Because the outcome of the present appeal hinges in no small part on the applicable legal standards, I begin by setting forth in some detail the rules that govern our review of a decision granting postconviction relief, through a writ of habeas corpus, on a claim of ineffective assistance of counsel. I consider, first, the law that a habeas court must apply in evaluating an ineffective assistance of counsel claim and, second, the standards by which an appellate tribunal reviews such a determination.

A

Ineffective Assistance of Counsel

The United States Supreme Court first articulated the two part test governing ineffective assistance of counsel claims in Strickland v. Washington , 466 U.S. 668, 687, 104 S.Ct. 2052, 80 L.Ed. 2d 674 (1984). "A convicted defendant's claim that counsel's assistance was so defective as to require reversal of a conviction ... has two components. First, the defendant must show that counsel's performance was deficient.... Second, the defendant must show that the deficient performance prejudiced the defense. This requires [a] showing that counsel's errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable. Unless a defendant makes both showings, it cannot be said that the conviction ... resulted from a breakdown in the adversary process that renders the result unreliable." Id. With respect to the second prong of the test, prejudice, the court in Strickland observed that "[a]ttorney errors come in an infinite variety and are as likely to be utterly harmless in a particular case as they are to be prejudicial." Id., at 693, 104 S.Ct. 2052. For this reason, the Supreme Court explained, "ineffectiveness claims alleging a deficiency in attorney performance are subject to a general requirement that the defendant affirmatively prove prejudice." Id.¹

In Strickland , the Supreme Court, having initially set forth the relatively nebulous fair trial/reliable result standard governing the prejudice prong, proceeded to try to articulate a more "workable principle" by which courts could assess whether any particular error made by defense counsel was harmless. Id. The court offered the following additional guidance: "Even if a defendant shows that particular errors of counsel were unreasonable ... the defendant must show that they actually had an adverse effect on the defense. It is not enough for the defendant to show that the errors had some conceivable effect on the outcome of the proceeding. Virtually every act or omission of counsel would meet that test ... and not every error that conceivably could have influenced the outcome undermines the reliability of the result of the proceeding.... [A]ny error, if it is indeed an error, impairs the presentation of the defense .... On the other hand ... a defendant need not show that counsel's deficient conduct more likely than not altered the outcome in the case. This outcome-determinative standard ... is not quite appropriate.... [Rather, under] the appropriate test for prejudice ... [t]he defendant must show that there is a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome." (Citations omitted; internal quotation marks omitted.) Id., at 693–94.

"In making the determination whether the specified errors resulted in the required prejudice, a court should presume ... that the judge or jury acted according to law.... The assessment of prejudice should proceed on the assumption that the [decision maker] is reasonably, conscientiously, and impartially applying the standards that govern the decision....

"In making this determination, a court hearing an ineffectiveness claim must consider the totality of the evidence before the judge or jury. Some of the factual findings will have been unaffected by the errors, and factual findings that were affected will have been affected in different ways. Some errors will have had a pervasive effect on the inferences to be drawn from the evidence, altering the entire evidentiary picture, and some will have had an isolated, trivial effect. Moreover, a verdict or conclusion only weakly supported by the record is more likely to have been affected by errors than one with overwhelming record support. Taking the unaffected findings as a given , and taking due account of the effect of the errors on the remaining findings, a court making the prejudice inquiry must ask if the defendant has met the burden of showing that the decision reached would reasonably likely have been different absent the errors ." (Emphasis added.) Id., at 694–96, 104 S.Ct. 2052.

Since it decided Strickland , the United States Supreme Court has emphasized the "highly demanding and heavy burden" that a petitioner must overcome in order to satisfy the prejudice prong. (Internal quotation marks omitted.) Williams v. Taylor , 529 U.S. 362, 394, 120 S.Ct. 1495, 146 L.Ed. 2d 389 (2000) ; see also Powell v. Warden , 272 Va. 217, 234, 634 S.E.2d 289 (2006) (emphasizing that prejudice prong of Strickland test imposes "highly demanding standard"), cert. denied, 551 U.S. 1118, 127 S.Ct. 2942, 168 L.Ed. 2d 269 (2007). For this reason, "cases in which habeas petitioners can properly prevail on the ground of ineffective assistance of counsel are few and far between." Rogers v. Zant , 13 F.3d 384, 386 (11th Cir.), cert. denied, 513 U.S. 899, 115 S.Ct. 255, 130 L.Ed. 2d 175 (1994) ; see also J. Cook, Constitutional Rights of the Accused (3d Ed. 2017) § 8:19; B. Gershman, Trial Error and Misconduct § 3–3 (a) (2) (1997).

Most recently, in Harrington v. Richter , 562 U.S. 86, 111–12, 131 S.Ct. 770, 178 L.Ed. 2d 624 (2011), the high court offered more specific and practicable guidance as to how courts are to apply Strickland 's reasonable probability standard. "In assessing prejudice under Strickland ," the court explained, "the question is not whether a court can be certain counsel's performance had no effect on the outcome or whether it is possible a reasonable doubt might have been established if counsel acted differently.... Instead, Strickland asks whether it is reasonably likely the result would have been different.... This does not require a showing that counsel's actions more likely than not altered the outcome, but the difference between Strickland's prejudice standard and a more-probable-than-not standard is slight and matters only in the rarest case .... The likelihood of a different result must be substantial, not just conceivable ." (Citations omitted; emphasis added; internal quotation marks omitted.) Id.

Both the United States Court of Appeals for the Second Circuit and this court have since applied Harrington when evaluating prejudice claims under Strickland . See Santone v. Fischer , 689 F.3d 138, 155 (2d Cir.), cert. denied, 568 U.S. 926, 133 S.Ct. 390, 184 L.Ed. 2d 231 (2012) ; Anderson v. Commissioner of Correction , 313 Conn. 360, 376, 98 A.3d 23 (2014), cert. denied sub nom. Anderson v. Semple , ––– U.S. ––––, 135 S.Ct. 1453, 191 L.Ed. 2d 403 (2015). In Michael T. v. Commissioner of Correction , 307 Conn. 84, 52 A.3d 655 (2012), for example, we treated the Strickland prejudice standard as essentially equivalent to a preponderance of the evidence test. See id., at 102, 52 A.3d 655 (prejudice exists when "new evidence undermines the confidence in the result reached such that it can be said that an injustice was likely done and that it is probable that the new trial would produce a different result" [emphasis added] ).

As I explain more fully hereinafter, in my view, in the majority opinion applies Strickland in a far looser manner than Harrington and Anderson permit.² Those cases teach that, to demonstrate prejudice, the petitioner bears the burden of proving that it is nearly as likely as not that, but for the errors of counsel, a different outcome would have obtained. In the present case, however, the majority simply speculates as to what could have been different had Ossorio testified, instead of requiring the petitioner to demonstrate that he had a reasonable probability of obtaining a different result. The jury might have found Ossorio's testimony more credible than that of the state's witnesses who testified against the alibi story. The jury "could have found that the victim was murdered prior to 11:15 p.m." "[T]he jury could have viewed the testimony of [the state's] witnesses differently if [defense counsel] had presented [a stronger] alibi defense ...." The jury, which initially might have been persuaded by the prosecutor's unsubstantiated references to a Skakel family conspiracy, might then have concluded that there was no such conspiracy after all. Strickland and its progeny caution against just this sort of speculation as to what conceivably might have gone differently, in a different trial, if different witnesses had testified. That is precisely why prejudice is—and is meant to be—so difficult to establish.

To be clear, each of the following propositions would have to be true for the outcome of the trial to have been different: (1) the jury concluded that the time of the murder mattered, that is, that the petitioner's various confessions and other evidence of his guilt was not so compelling that the jury could safely conclude that, regardless of when he did it, he must have killed the victim;³ (2) the jury concluded that the victim was killed during the alibi period, not later, and rejected testimony that the petitioner himself admitted to having seen her alive later that evening; and (3) the jury, although finding all of the petitioner's other family and independent alibi witnesses not to be credible, would have concluded that Ossorio's alibi testimony was more credible than that of the state's witnesses.

The majority repeatedly contends that the petitioner need only demonstrate that the jury could have concluded that the crime was committed during the alibi period or could have viewed the evidence differently in the light of Ossorio's testimony. But that isn't enough. In order for those three propositions collectively to be as likely as not true, or at least nearly so, which is what the petitioner bears the burden of establishing, each proposition individually must at least be highly probable.⁴ A simple gut feeling that things might well have gone differently for the petitioner if Ossorio had testified is not enough, as a matter of law. As I explain in parts II and III of this dissenting opinion, the petitioner has not come close to meeting his burden in this regard.

B

Appellate Review

The standards by which we review the granting or denial of habeas relief with respect to an ineffective assistance of counsel claim are well established. "The habeas judge, as the trier of facts, is the sole arbiter of the credibility of witnesses and the weight to be given to their testimony.... The application of historical facts to questions of law that is necessary to determine whether the petitioner has demonstrated prejudice under Strickland , however, is a mixed question of law and fact subject to our plenary review." (Citation omitted; internal quotation marks omitted.) Small v. Commissioner of Correction , 286 Conn. 707, 717, 946 A.2d 1203, cert. denied sub nom. Small v. Lantz , 555 U.S. 975, 129 S.Ct. 481, 172 L.Ed. 2d 336 (2008) ; see also Strickland v. Washington , supra, 466 U.S. at 698, 104 S.Ct. 2052 ("[i]neffectiveness is not a question of basic, primary, or historical [fact]" [internal quotation marks omitted] ).

Consistent with these principles, and applying plenary review, this state's appellate tribunals have, on multiple recent occasions, reversed a habeas court's determination with respect to the prejudice prong of Strickland . See Horn v. Commissioner of Correction , 321 Conn. 767, 782–83 and n.12, 138 A.3d 908 (2016) (interpreting trial evidence differently than habeas court and emphasizing that "we [are not] required to defer to the trial court's legal determination that there is a reasonable probability that newly discovered evidence would have resulted in a different verdict if credited by the jury, i.e., that it undermines confidence in the verdict"); see also, e.g., Michael T. v. Commissioner of Correction , supra, 307 Conn. at 102–103, 52 A.3d 655 ; Crespo v. Commissioner of Correction , 149 Conn. App. 9, 20, 87 A.3d 608, cert. denied, 311 Conn. 953, 97 A.3d 984 (2014) ; cf. Lapointe v. Commissioner of Correction , 316 Conn. 225, 266, 112 A.3d 1 (2015).

Moreover, although we must defer to the habeas court's factual findings with respect to the credibility of witnesses, such as Ossorio, who testified before that court, we are not bound by the habeas court's ultimate determination as to whether the jury would have credited those witnesses over others who testified at trial. As we explained in Horn , "there is no requirement that we defer to the habeas court's legal determination that new evidence is so compelling that a reasonable juror could not fail to credit it."⁵ Horn v. Commissioner of Correction , supra, 321 Conn. at 783 n.12, 138 A.3d 908. A fortiori, we review de novo a habeas court's legal determinations, formed on the basis of its review of the cold trial record, such as whether a proffered alibi defense would have constituted a full alibi or only a partial alibi. II

I turn now to the question of whether, if we assume for argument's sake that defense counsel's failure to procure and present Ossorio's alibi testimony constituted deficient performance, that deficient performance was prejudicial. In other words, whether there is a reasonable probability that the jury, having heard Ossorio's testimony, would not have convicted the petitioner. The habeas court concluded that defense counsel's errors were prejudicial on the basis of its determinations that (1) there was "weighty evidence that the victim was murdered in the time range of 9:30 p.m. to 10 p.m. on October 30, 1975," and (2) Ossorio's testimony was sufficiently credible that it would have greatly enhanced the persuasiveness of the petitioner's alibi for that time period. In this part of my dissenting opinion, I explain how the fact that the victim reasonably could have been killed outside of the alibi period renders the alibi—and, therefore, defense counsel's allegedly deficient performance—irrelevant. In part III of this dissenting opinion, I explain why a different result would have been unlikely even if the petitioner had offered a full rather than a partial alibi.

A

Governing Law

It frequently has been remarked, and the majority itself concedes, that "a partial alibi is no alibi at all." (Internal quotation marks omitted.) Williams v. State , 185 So.3d 1270, 1271 (Fla. App. 2016), review dismissed, Florida Supreme Court, Docket No. 16–605, 2016 WL 1394643 (April 8, 2016). The reasons behind this maxim are readily apparent.

"The literal significance of the word 'alibi' is 'elsewhere.' ... [T]o make out a defense of alibi, the range of the evidence in respect to time and place must be such as reasonably to exclude the possibility of the defendant's presence at the scene of the offense at the time of the commission of the crime.... In other words, by an alibi the accused attempts to prove that the accused has been at a place so distant that the accused's participation in the crime has been impossible ." (Emphasis added; footnotes omitted.) 21 Am. Jur. 2d 308, Criminal Law § 179 (2016).

"To be successful, a defendant's alibi must cover the entire time when the defendant's presence would have been required for the accomplishment of the crime .... [S]ince an alibi defense derives its potency from the physical impossibility of the accused's guilt, a purported alibi that leaves it possible for the accused to be the guilty person is no alibi at all." (Emphasis added; footnotes omitted.) 21 Am. Jur. 2d, supra, § 181; see also Black's Law Dictionary (10th Ed. 2009) (defining "alibi" as "[a] defense based on the physical impossibility of a defendant's guilt" [emphasis added] ).

Consistent with these principles, cases almost universally hold that defense counsel's failure to investigate, identify, or present an alibi witness either does not constitute deficient performance or is not prejudicial when that alibi would cover only a portion—even a substantial portion—of the time period during which the crime could have been, or was alleged to have been, committed. See, e.g., Johnson v. Secretary, Florida Dept. of Corrections , 680 Fed. Appx. 869, 872 (11th Cir. 2017) (no prejudice when prison records would have shown only that petitioner "was in jail for a substantial portion of the timeframe that the victim testified that the acts occurred" [internal quotation marks omitted] ); Matthews v. Mazzuca , 120 Fed. Appx. 856, 858 (2d Cir. 2005) (no prejudice when alibi would have covered no more than three of five hours during which burglary could have occurred); Fargo v. Phillips , 58 Fed. Appx. 603, 608–609 (6th Cir.) (when uncalled alibi witnesses would have left narrow window of time during which crime and various related actions could have been committed, majority rejected habeas court's conclusion that failure to call partial alibi witness was potentially prejudicial), cert. denied, 539 U.S. 932, 123 S.Ct. 2585, 156 L.Ed. 2d 613 (2003) ; Fawaz v. Wolfenbarger , United States District Court, Docket No. 09–14965 (DML) (E.D. Mich. April 5, 2013) (failure to present alibi that would have covered all but three hours of night when murder occurred was deficient but not prejudicial); Rector v. Wolfe , United States District Court, Docket No. 5:07CV1229 (CAB), 2009 WL 1788569 (N.D. Ohio June 23, 2009) (no prejudice when indictment alleged that rapes occurred on or about Thanksgiving and Christmas holidays and alibi witnesses would have eliminated Thanksgiving Day, Christmas Eve, and Christmas Day), aff'd, 499 Fed. Appx. 532 (6th Cir. 2012) ; Halton v. Hesson , 803 F.Supp. 1272, 1277 (M.D. Tenn. 1992) (evidence that petitioner worked as usher during afternoon of crime would not have foreclosed possibility that he committed assault during remainder of afternoon or evening); Spearman v. Commissioner of Correction , 164 Conn. App. 530, 552, 562–63, 138 A.3d 378 (despite fact that state's case was relatively weak and rested primarily on testimony of one eyewitness of questionable credibility, failure to present alibi witnesses deemed neither deficient nor prejudicial because petitioner could have helped set fire, run home, and pretended to be asleep before arson became apparent), cert. denied, 321 Conn. 923, 138 A.3d 284 (2016) ; Ostolaza v. Warden , 26 Conn. App. 758, 767, 603 A.2d 768 (finding no deficient performance because "the petitioner's work schedule constituted only a partial alibi that demonstrated his opportunity to commit the crimes charged on several of the days alleged in the information"), cert. denied, 222 Conn. 906, 608 A.2d 692 (1992) ; Beasley v. State , 18 So.3d 473, 492 (Fla. 2009) (no prejudice when alibi would have left between seventy five and ninety minutes unaccounted for on day of murder); White v. State , 293 Ga. 825, 827, 750 S.E.2d 165 (2013) (although failure to carefully peruse file and raise alibi defense constituted deficient performance, there was no prejudice because "there was a window of time on the night of the murder for which appellant's whereabouts could not be verified" by alibi); People v. Wynekoop , 359 Ill. 124, 136, 194 N.E. 276 (1934) ("[a]s the crime is shown to have been committed some time during the five and one-half hour period between 3 and 8:30 [p.m.] the failure of the defendant to offer any alibi proof during the three periods of time aggregating nearly two hours is fatal"), cert. denied, 295 U.S. 758, 55 S.Ct. 915, 79 L.Ed. 1700 (1935) ; King v. State , 505 S.W.3d 419, 426 (Mo. App. 2016) (evidence that provides only partial alibi "cannot entitle movant to [postconviction] relief"); State v. Razo , Ohio Court of Appeals, Docket No. 05CA008639, 2005 WL 1763611 (July 27, 2005) (failure to interview potential alibi was not prejudicial even though witness would have placed defendant out of state during large portion of period during which rapes allegedly were committed), appeal denied, 107 Ohio St.3d 1684, 839 N.E.2d 404 (Ohio 2005) ; State v. Jones , Ohio Court of Appeals, Docket No. 97–CA–648 (April 14, 1998) (when circumstantial evidence of guilt was strong, no prejudice even though partial alibi would have covered "a large portion of the time" when crime could have occurred); Carruthers v. State , Tennessee Court of Criminal Appeals, Docket No. W2006–00376–CCA–R3–PD, 2007 WL 4355481 (December 12, 2007) (no prejudice when, even if testimony related to date of crime, it provided partial alibi at best); Johnson v. Virginia , 210 Va. 16, 20, 168 S.E.2d 97 (1969) (trial court properly declined to give requested alibi instruction when defendant's alibi would have covered all but one hour and forty-five minutes of nine hour period during which crime could have been committed); Tinsley v. Commonwealth , Virginia Court of Appeals, Docket No. 1026–11–2, 2012 WL 1499352 (May 1, 2012) (explaining that alibi must cover sufficient time to render defendant's presence impossible or highly improbable, and concluding that evidence that appellant was twenty to thirty minutes away from crime scene fifteen minutes before crime occurred did not justify alibi instruction). In fact, many jurisdictions apply what amounts to a per se rule that a petitioner cannot establish prejudice unless the undiscovered alibi would have covered the entire time period during which the crime might have been committed.⁶ See Johnson v. Virginia , supra, at 19–20, 168 S.E.2d 97.

The majority, while generally acknowledging the principle for which these cases stand, argues that none of them is on point and that the present case does not truly present a partial alibi. The distinction to which the majority appeals—a distinction that apparently has never previously been articulated by either court or commentator—is set forth in two hypothetical scenarios. In the first scenario, the crime could have been committed on any one of ten days and the defendant has an alibi for nine of those days. In the second scenario, the evidence suggests a high chance that the crime was committed on days one through nine, and the defendant has an alibi for those days, but not the tenth day. It is the view of the majority that the first scenario presents a partial alibi but that, for reasons that are not entirely clear, the second scenario presents a complete alibi, regardless of the fact that the crime reasonably could have been committed outside of the alibi period in both instances. I am not certain that the novel distinction that the majority is trying to draw is a meaningful one. In the absence of any additional information, if a crime could have been committed on any one of ten days, then the likelihood that it was committed on the particular day that the hypothetical defendant lacks an alibi is ten percent. Such a case is no different from one in which the evidence suggests a 90 percent chance that a crime was committed on a day where the defendant had an alibi and a 10 percent chance that it was committed on a different day. The probabilities are the same.

But we needn't count the angels on the head of a pin. There are at least two other reasons why the argument of the majority fails. First, contrary to the assertions of the majority, the cases that I have cited as authority for the partial alibi rule are not all of the nine in ten day variety. In a number of them, the evidence indicated that it would have been extremely difficult for the petitioner to have committed the crime outside of the period during which he had a potential alibi. Because the evidence left open some realistic possibility that the crime had been committed outside of the alibi period, however, the court applied the partial alibi rule. Consider Fargo v. Phillips , supra, 58 Fed. Appx. 603. In that case, the overlooked alibi witnesses, whom the trial court found to be truthful, would have testified that they were with the petitioner on the day in question. Some of those witnesses were with the petitioner until just before twilight and others beginning just after dark. See Fargo v. Phillips , 129 F.Supp.2d 1075, 1080–84 (E.D. Mich. 2001), rev'd, 58 Fed. Appx. 603 (6th Cir. 2003). As the dissenting judge explained, once the habeas court's erroneous assumptions regarding the time of sunset were corrected, the testimony of the alibi witnesses would have left only a very brief window of time, potentially as narrow as nineteen minutes, for the petitioner to bid goodbye to his friends, leave his house, pick up the complainant from her house, drive her back to his place, watch television with her, engage in foreplay with her, perpetrate a sexual assault, get dressed, drive the complainant to a friend's house, and then drive to the home of another alibi witness. Fargo v. Phillips , supra, 58 Fed. Appx. at 609 (Moore, J., dissenting). Nevertheless, the majority in that case concluded that the failure of defense counsel to interview and call the alibi witnesses was not prejudicial because there was a sufficient window of time in which the crime could have been committed on that evening. Id., at 608 ; see also Spearman v. Commissioner of Correction , supra, 164 Conn. App. at 559–60, 138 A.3d 378 (failure to call partial alibi witness in arson case deemed nonprejudicial when alibi would have left just one minute for petitioner to set fire, run home, return to upstairs bedroom, and feign sleepiness); Tinsley v. Commonwealth , supra, Virginia Court of Appeals, Docket No. 1026–11–2 (alibi that would have placed defendant estimated twenty to thirty minutes away from scene of crime fifteen minutes before it occurred deemed incomplete because it was not impossible that favorable road conditions and aggressive driving could have allowed him to arrive there sooner). Accordingly, in my view, the majority is simply incorrect when it pronounces, without citation to a single authority, that "when a true partial alibi is at issue, it is invariably the case that the defendant just as likely could have committed the crime during a period of time not covered by the alibi."

The second issue that I have with the majority's characterization of the partial alibi rule is that it is inconsistent with the very concept of an alibi. As the sources that I have cited previously in this dissenting opinion make clear, an alibi defense has long been understood to mean that it is impossible for a person to have committed a crime because he or she was at another location when the crime was being committed. The second hypothetical scenario that the majority discusses, thus, presents no more of an alibi than does the first; in both instances, there is a realistic possibility that the crime was committed outside of the alibi period.

The majority dismisses the idea that an alibi could be defeated, as a matter of law, if there were only a small possibility that the crime was committed at a different time. Of course, that is exactly what the word "impossible" means, and other jurisdictions have adopted the stringent version of the partial alibi rule that the majority criticizes as "clearly [defying] reason and common sense." See State v. Tutson , 278 Conn. 715, 733 n.10, 899 A.2d 598 (2006). For present purposes, however, it is unnecessary to resolve whether, under Connecticut law, a legally cognizable alibi must render the defendant's presence at the crime scene impossible or merely highly improbable. In the present case, as I explain in the remainder of this part of my dissenting opinion, there was a substantial possibility that the victim was killed outside of the alibi period. Indeed, the jury reasonably could have credited testimony indicating that the petitioner himself admitted to having seen the victim alive later that evening. The only evidence that the victim was killed during the alibi period, in fact, are inferences from the barking of dogs, ambiguous testimony regarding the victim's curfew, and pure speculation regarding how an independent, precocious teenaged girl might or might not have spent her evening. There is literally nothing more. By whatever standards we define a partial alibi, this was one.

Finally, it bears noting that the majority consistently states that the petitioner is only required to demonstrate that the jury reasonably could have found that the murder took place during the alibi period. The majority cites no authority in support of this proposition. Indeed, as I have previously explained in this dissenting opinion, such a standard is inconsistent with Strickland and its progeny. See part I A of this dissenting opinion. The argument of the majority, however, also fails on its own terms. If the evidence suggested that there was a 51 percent chance that the crime was committed between 9:30 and 11 p.m., for example, then a jury certainly could reasonably determine that it was, in fact, committed within that time. But I am not aware of any court that has held that a defendant has a legally cognizable alibi defense when the alibi fails to account for nearly half of the time during which the crime could have been committed. That cannot be the correct standard.⁷ B

Analysis

The dispositive question on reconsideration of our decision in Skakel v. Commissioner of Correction , supra, 325 Conn. at 426, 159 A.3d 109, is, therefore, whether the petitioner's alibi story, if believed by the jury, would have constituted a full alibi or only a partial one. If, as the state contends, his purported visit to the Terrien home was merely a partial or incomplete alibi, accounting for less than two hours out of the approximately four to eight hour period during which the murder reasonably could have been committed then, as a matter of law, defense counsel's failure to bolster the alibi by presenting Ossorio's testimony could not have been prejudicial, regardless of the overall strength of the state's case. Accordingly, I turn my attention to the evidence that was presented at trial with respect to the timeframe during which the victim could have been killed and the limited portion of that timeframe encompassed by the purported alibi.

1

Scope of the Alibi

The jury heard the following evidence with respect to the time during which the petitioner purportedly visited the Terrien home. It was undisputed that James Terrien, Rushton Skakel Jr. (Rushton), and John Skakel (John) left the Skakel residence in a family car, a Lincoln, just before 9:30 p.m. on October 30, 1975, shortly before Helen Ix (Helen) and Geoffrey Byrne took their leave of the Skakel residence and Julie Skakel (Julie) drove her friend Andrea Shakespeare home. It also was undisputed that, before returning from the Terrien home, Rushton and John watched a Monty Python television program. Although memories varied somewhat as to the precise broadcast time of the show, a published television schedule from that week indicates that the show aired on channel thirteen from 10 to 10:30 p.m. The accuracy of that document is undisputed.

What is in dispute is the time at which Rushton and John—with or without the petitioner—returned to the Skakel residence. A few weeks after the murder, the petitioner himself informed the police that he and his brothers had arrived back from the Terrien home at about 10:30 or 11 p.m. Although Rushton could not be certain of the exact times, he testified that the boys stayed at the Terrien home for approximately fifteen to twenty minutes after the show ended at 10:30 p.m., and that the drive home took between twenty and twenty-five minutes. This would have placed the boys back at the Skakel residence sometime between 11:05 and 11:15 p.m. Terrien testified that the Skakel boys left his house a little before 11 p.m. Finally, John recalled that they left the Terrien home at about 11 p.m. and made it home around 11:15 p.m. Accordingly, the evidence would have placed the boys back in the Belle Haven neighborhood of Greenwich sometime between 10:30 and 11:15 p.m., with the preponderance of the evidence pointing to the period between 11:00 and 11:15 p.m. I do not understand the majority to disagree with this analysis.

2

Possible Time of Death

The principal question to be resolved, then, is whether it is highly likely that the victim was killed prior to the 10:30 to 11:15 p.m. time period or whether the jury reasonably could have found that she was killed later that night, after the Lincoln and its occupants returned to Belle Haven. If the former, then Ossorio's testimony would have supported a full alibi, and we must determine whether presenting that testimony to the jury would have been reasonably likely to result in a different outcome. If the latter, however, then the trip to the Terrien home represented merely a partial or incomplete alibi and, as a matter of law , defense counsel's failure to present one additional witness in support of that partial alibi was, at worst, a harmless error.

In support of his claim that it is highly likely that the jury concluded that the murder was committed between 9:30 and 10 p.m., the petitioner offers the following five arguments: (1) the state itself believed and indicated at trial that the time of death was before 10 p.m., (2) the forensic evidence supports that timeline, (3) the victim's curfew supports that timeline, (4) the evidence of barking dogs supports that timeline, and (5) the jury's request to have evidence read back relating to the alibi indicates that the jury believed that the alibi was relevant. In addition, the majority offers three additional arguments in support of the petitioner's position: (1) that the state failed to articulate a plausible explanation of where the victim might have been between the time she left the Skakel residence and approximately 11 p.m., when the Lincoln returned from the Terrien home, (2) that the prosecutor argued to the jury that the victim never made it home on the night of the murder, precluding the possibility that she went home and then back out again, and (3) that the victim was killed along the most direct route between the Skakel residence and her home. I consider each of these arguments in turn.

a

The State's Case

The petitioner first argues that the state itself has proceeded on the assumption that the murder had occurred by 10 p.m. and, specifically, that the prosecution's closing argument demonstrated that belief to the jury. I disagree.

In fact, throughout the state's closing argument, the prosecution repeatedly informed the jury that it could convict the petitioner of the murder, even if it credited his alibi story, because the crime reasonably could have been committed after he returned from the Terrien home. The prosecutor began his closing argument by discussing the charges against the petitioner: "What's in the information. First of all, the when and the where. Between 9:30 p.m. and 5:30 a.m., at Walsh Lane, Greenwich, Connecticut—it's no more specific than that ...." He then walked the jury through the state's timeline of the night of the murder. With respect to the trip to the Terrien home, the prosecutor explained: "Exactly who went [to the Terrien home] is one of our controversies in this trial. But, as you will see, it is not one that the state necessarily has to resolve in order for you to convict." Later, when the prosecutor first addressed the petitioner's alibi defense, he stated as follows: "[T]he alibi, that is the cornerstone of the defense here. It is a somewhat unbalanced alibi because due to the defendant's ongoing tales in the 1990s, you can accept the alibi at face value and still convict the defendant but you of course will want to take a careful look at that alibi."

The prosecutor continued to follow this belt and suspenders approach in his rebuttal, after defense counsel had laid out the alibi theory. Early in the rebuttal, the prosecutor argued as follows: "Let's talk about time. I spoke about this in my opening. The concept of exact time for a murder is obviously of great concern for the defense, as it should be. Because from 1975 until 1992 or thereabouts, [the petitioner] had a nice, neat 9:30 [or] 10 [p.m.] type alibi. But, as you will see ... the [petitioner] has dug himself a hole that throws his alibi somewhat to the wind.

"Keep in mind that as regards time, the state has to prove beyond a reasonable doubt only that [the victim] was murdered between 9:30 p.m. and 5:30 a.m. I am sure you have noted that the defense has presented a partial alibi only , the trip to [the Terrien home from] 9:30 to 11 [p.m.] or so ... at night." (Emphasis added.)

Moreover, the prosecutor specifically explained that the jury need not reach a unanimous decision with respect to the alibi and time of death: "As regards time, you must be unanimous that the crime occurred during the time set in the information, 9:30 [p.m.] to 5:30 [a.m.] and that's all.... For that matter, if half of you ... figured the crime happened early and not accept the alibi and the other half of you were to accept the alibi and conclude the [petitioner] ... came by later on at night and did it ... you must convict.

"For that matter, all [twelve] of you ... could each come up with his own personal time. As long as every-body's time came up between 9:30 [p.m.] and 5:30 [a.m.] and you are convinced beyond a reasonable doubt that [the petitioner] murdered [the victim], you must convict."

Finally, the prosecutor addressed the forensic evidence, arguing against the conclusion of the defense expert that the murder probably happened around 10 p.m. The theory pressed by the state, as I will discuss more fully hereinafter, was that that medical evidence was fully consistent with a time of death as late as 4 a.m. The prosecutor also argued that the only evidence favoring the 10 p.m. time of death—barking dogs and the victim's purported curfew—was "open to any number of alternative constructions." He concluded that portion of his argument with the observation that "all of this falls within 9:30 [p.m.] to 5:30 [a.m.] and that's all that is alleged in the information."⁸ Only at that point in his rebuttal, after having argued at length that the Terrien alibi was incomplete and immaterial because the victim could have died well after 10 p.m., did the prosecutor proceed to argue, in the alternative , that the alibi story itself was suspect.⁹ Moreover, although it is true that the prosecution implicitly acknowledged that the state initially had assumed that the victim was killed soon after everyone departed from the Skakel residence at 9:30 p.m., the prosecution made clear that the reason that the state had expanded its working theory of the timeframe during which the murder could have been committed in the 1990s was because it became aware at that time that the petitioner himself had admitted to having seen the victim alive later in the evening.¹⁰ Accordingly, there simply is no reasonable possibility that the jury could have concluded on the basis of the prosecution's closing argument that it continued to be the state's position at the time of trial that the murder must have been committed before 10 p.m. If any doubt remains in this regard, it surely is resolved by the fact that, at trial, defense counsel repeatedly acknowledged in the presence of the jury that the state viewed the trip to the Terrien home as only a partial alibi. Indeed, in his opening remarks, defense counsel recalled that, during jury selection, "[the state] half apologized almost to all of you that they are only going to be able to prove that the crime was committed between 9:30 [p.m.] and 5:30 [a.m.] and that will become very important." Then, in his own closing, defense counsel reflected on the import of a medical examiner's report as follows: "[A]ll we know [is that this] report ... says ... that [the murder] occurred at 9:30 [p.m.] to 4:30 or 5:30 [a.m.] the next morning. And [the state] points out, well, that's all we have to tell you."

8 Throughout its opinion, the majority consistently downplays or ignores evidence and arguments that contradict or fail to support its own theory of the case, notwithstanding the fact that the jury, which was in the best position to assess the evidence and arguments presented at trial, concluded that the petitioner murdered the victim. In this context, for example, the majority ignores all of these statements in which the prosecutor repeatedly made clear that the state was aggressively arguing a partial alibi theory. The majority acknowledges only that State's Attorney Jonathan C. Benedict "observed during closing argument that the state did not have to disprove the petitioner's alibi for the jury to find him guilty," which hardly does justice to the force of the state's partial alibi theory.

9 Accordingly, the majority's statement that "[t]he respondent has identified no case in which a partial alibi was found to exist and in which the state's primary theory of the case, and the only one toward which its evidence was geared, was that the crime most likely occurred during the period of time covered by the defendant's alibi," while quite possibly true, is simply irrelevant. That was not the state's argument in the present case.

10 It is, therefore, disingenuous of the petitioner to insinuate in his brief that the state changed its theory as to the time of death in order to pin the crime on him. The petitioner neglects to mention that it was his own subsequent admissions that he initially misled law enforcement about his activities and whereabouts after returning from the Terrien home, and that he had seen the victim alive later that evening, that led the state to reexamine the time of death and the importance of the Terrien alibi.

Later in his closing, defense counsel returned at several points to the time of death, each time remarking on the state's partial alibi theory. Specifically, defense counsel stated the following: "[t]ime of death, the state would have you believe that this isn't such a big deal, they can go both ways on this"; "you [may] buy the state's alternative theory that [the petitioner] came back after 11 [p.m.] and then went out and did this horrible thing"; and "[t]he trip to the Terrien house ... I am guessing at least that the state is going to say well, he didn't go but maybe he did, we are not sure. Well, I don't believe in offering up a buffet table of excuses and I don't believe that they should be giving you an alternate choice as well, maybe he did and maybe he didn't."¹¹

11 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

Indeed, the trial court itself highlighted the state's partial alibi theory, twice instructing the jury as follows: "You should also bear in mind the state's claim that even if you find that the defendant was where his testimony and that of his witnesses indicates, he could still have reached the scene of the crime in time to have committed it ...." Defense counsel later took exception to the fact that the court's jury instructions placed too much emphasis on the fact that time of death was unimportant.¹² In light of the repeated statements by the prosecutor, defense counsel, and the trial court that the state was alleging only that the murder had been committed between 9:30 p.m. and 5:30 a.m. and that the petitioner could have committed the crime after returning from the Terrien home, there simply is no reasonable possibility that the jury could have concluded that it was the state's view at the time of trial that the crime had been committed by 10 p.m.

12 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

b

The Forensic Evidence

I next address the petitioner's argument that the forensic evidence demonstrated that the time of death was approximately 10 p.m. The following additional facts are relevant to this argument. At trial, the jury heard evidence, either directly or indirectly, regarding the opinions of three different medical experts who, at some point, had opined as to the probable time of the victim's death. The first opinion was that of Elliot M. Gross, a physician who was the state's chief medical examiner when the murder occurred in 1975 and who had performed the official autopsy on the victim. Gross was not available to testify at the petitioner's trial in 2002, and the report that resulted from his autopsy does not state a conclusion as to the time of the victim's death. Nevertheless, Thomas Keegan, a captain in the detective division of the Greenwich Police Department, testified that, during his investigation of the murder, he sought Gross' opinion as to the time of death. Although Keegan did not testify as to Gross' precise opinion as a result of a hearsay objection,¹³ Keegan did indicate that the time window that Gross had given to him on the basis of the forensic evidence was too broad to be helpful in investigating the crime.¹⁴ The only reasonable conclusion the jury could have drawn from that testimony was that Gross' conclusion was consistent with the state's theory that the crime need not have occurred between 9:30 and 10 p.m.

13 As previously noted in this dissenting opinion, during closing arguments, defense counsel stated that "all we know from ... Gross is [that his] report ... says well, my conclusion is that it occurred at 9:30 [p.m.] to 4:30 or 5:30 [a.m.] of the next morning." We must assume that, pursuant to the court's instructions, the jury disregarded counsel's recitation of facts that were not in evidence. State v. McIntyre , 250 Conn. 526, 534, 737 A.2d 392 (1999).
Unfortunately, the majority fails to honor this same, well established principle within its own reasoning. Indeed, the majority's prejudice analysis hinges to a large extent on the dubious assumption that the jury was persuaded not by the abundant, actual evidence of the petitioner's guilt; see part III of this dissenting opinion; but, rather, by the prosecutor's speculative suggestion during closing argument that the Skakel family had engaged in a protracted conspiracy to cover up the petitioner's crime, a conspiracy theory that defense counsel easily rebutted and that found virtually no support in the evidence presented at trial. See Skakel v. State , supra, 295 Conn. at 687–95, 991 A.2d 414 (Palmer, J. , dissenting). The majority fails to explain why we should attribute such irrational decision making to the jury, which it accuses, without any support or evidence, of having decided the case on the basis of a feeling of outrage that a wealthy family was able to " 'trick' " the police.

14 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

The jury also heard the testimony of Harold Wayne Carver, the state's chief medical examiner at the time of trial. Carver explained that three forensic factors—lividity, rigor mortis, and digestion—could be used to help identify the time of death.

Carver first discussed lividity, or livor mortis, the tendency of blood to pool in and discolor the lower portions of a body after death. He explained that lividity begins to occur within a couple of hours and that it becomes fixed "somewhere over four, usually six or more hours ...." In the present case, Carver explained that lividity could not be used to pinpoint the time of death because the autopsy had not been performed until more than one day after the body was discovered and also because the body was turned over five or six hours after it was found. Moreover, even if lividity had been fixed by the time that the victim's body was discovered at 12:30 p.m. on October 31, on the basis of Carver's testimony the jury best could have concluded that she probably had been killed sometime prior to 6:30 a.m. on that day.¹⁵ Second, Carver considered the importance of rigor mortis, the stiffening of muscles that occurs after death. Carver testified that the speed at which the process occurs can vary depending on the ambient temperature and the physical condition of the deceased but that, generally, rigor mortis begins to set in after several hours and persists for between twelve and twenty four hours. In the present case, Keegan observed that the victim was in rigor when he examined her around 1:15 p.m. on October 31. In light of the variable and relatively lengthy time that rigor mortis persists, Carver stated that he could not determine the victim's time of death with precision. He could opine only that "[s]he died several, many hours before she was found" and that the time of death was probably "closer to 9:30 [p.m. on October 30] than [noon on October 31]." In other words, the victim likely died at some time prior to 5 a.m. on October 31, and certainly before 9:30 a.m. or so.

15 Keegan's investigation report indicates that he observed slight lividity when he examined the body during the afternoon of October 31, but there is no indication whether lividity was fixed at that time. Gross did note lividity during the autopsy on November 1.

Finally, Carver discussed the forensic evidence relating to the victim's digestive process. He noted that the victim last ate between 6 and 6:30 p.m. on October 30, which was consistent with the testimony of the victim's mother, Dorothy Moxley, that the victim began her evening meal around 5:45 p.m. and that she ate "just before she went out" at around 6:30 or 6:45 p.m. Carver further testified that, at the time of the autopsy, the stomach contained only some blackish fluid and the small intestine contained some semi-liquid feces. He further explained that food typically passes from the stomach into the small intestine between one and two hours after eating, and then takes approximately twenty-four to forty-eight hours to pass through the small intestine. Carver then testified that digestion more or less terminates upon death. On the basis of this information, Carver was unable to draw any useful conclusions regarding the victim's likely time of death. The meal that she ate at 6 or 6:30 p.m. presumably would have exited her stomach by approximately 8:30 p.m. and, in the normal course of digestion, would have remained in her small intestine—where it remained at the time of death—until at least the next morning.

Considering all of these factors in tandem, Carver concluded that the forensic evidence revealed by Gross' autopsy was consistent with the defense's theory that the victim died between 9:30 and 10 p.m. but also was consistent with the state's theory that she could have been killed any time before midnight or 1 a.m. Accordingly, Carver agreed with Gross' opinion that it was impossible to pinpoint or even meaningfully narrow the time of death on the basis of forensic evidence.

Unfortunately, the majority fails to acknowledge any of Carver's extensive forensic analysis, which makes clear that there was absolutely no forensic support for the conclusion that the victim was likely killed before 10 p.m. Moreover, the majority represents that "Carver ... opined that it was within the realm of scientific possibility that the victim died any time between 9:30 p.m. on October 30, 1975, and 'many hours before she was found' the next afternoon," creating the misleading impression that the expert somehow suggested that a time of death after 10 p.m., while theoretically possible, was extremely unlikely. (Emphasis added.) In fact, Carver said nothing of the sort. He gave absolutely no indication that it was more likely that the victim died between 9:30 and 10 p.m. than at 11:15 p.m. or later that evening. The third witness to testify with regard to the forensic evidence was the defense expert, Joseph A. Jachimczyk, a well respected retired forensic pathologist and medical examiner from Texas, who consulted with the Greenwich Police Department during the initial investigation of the victim's murder. Although Jachimczyk's testimony differed slightly from that of Carver with respect to the forensic evidence, none of his medical opinions could support a determination that the victim likely died between 9:30 and 10 p.m.¹⁶ Moreover, although there is no doubt that Jachimczyk is a highly qualified expert, neither Carver nor Gross' expertise was in any way impugned at trial, and there is no basis for concluding that the jury would have found the testimony of Jachimczyk to be more credible to the extent that it may have differed.

16 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

With regard to lividity, Jachimczyk testified that the process usually begins between two and four hours of death and becomes fixed between eight and twelve hours. Because lividity was first noted—and then only faintly—by Keegan at 1:15 p.m. on October 31 and the victim's body was not moved¹⁷ until later that afternoon, the most that can be said on the basis of Jachimczyk's testimony is that the victim probably died sometime prior to dawn on October 31.

17 The extent to which lividity has been fixed is ascertained by turning a body over and noting the degree to which the color shifts downward. Even if the detectives had noted fixed lividity when they first examined the body at 1:15 p.m., at best that would indicate that the victim had died sometime prior to the early morning hours of October 31.

In terms of rigor mortis, Jachimczyk explained that the process usually begins between four and eight hours and persists for about twenty-four hours. Importantly, he also agreed with Carver, and emphasized on several occasions, that rigor mortis is variable and its onset and duration can depend on factors such as ambient temperature. To the extent that rigor mortis was of any use in estimating the time of death, then, Jachimczyk's testimony would support a conclusion only that the victim died sometime after 1:15 p.m. on October 30 and before the morning of October 31.

Lastly, with respect to digestion, whereas Carver was under the impression that the victim had taken her last meal between 6 and 6:30 p.m., Jachimczyk proceeded under the assumption that she last had eaten at 5:30 p.m. Notably, whereas Carver testified that the stomach empties its contents within one or two hours, Jachimczyk was of the belief that the process takes four hours on average. This means that if the victim last ate around 6, as testimony from her mother suggests, then the earliest she could have been killed, given that her stomach was empty at the time of death, was approximately 10 p.m.¹⁸ This insight will become important when I discuss the nonforensic factors.¹⁹

18 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

19 Jachimczyk acknowledged that the only determination that could be made on the basis of an examination of the victim's digestive tract was that she died sometime after 9:30 or 10 p.m. on October 30. He agreed that it was very difficult to say how long she had lived past 10 p.m.

To summarize the forensic testimony that was presented to the jury, we know for certain that the victim died sometime after 9:30 p.m. on October 30, which is when she was last seen alive at the Skakel residence. The fact that rigor mortis and some degree of lividity were noted when her body was found in the early afternoon on October 31 suggests that she likely died sometime before dawn that day, consistent with the state's charging document. In other words, the forensic evidence itself really tells us almost nothing beyond the undisputed fact that the victim was still alive at 9:30 p.m. on October 30 and that she had died by approximately 5:30 a.m. the following day.

Specifically, nothing in the testimony of any of the medical experts regarding the forensic evidence supports a conclusion that the victim was more likely to have died between 9:30 and 10 p.m. on October 30 than at, say, 11:15 p.m. Although Jachimczyk did make the conclusory statement that he considered the forensic evidence when forming his opinion that the victim likely died around 10 p.m., at no time did he specifically tie any of the forensic factors to that time frame in any manner, nor does any of his specific forensic testimony support such a conclusion. Moreover, when pressed, Jachimczyk acknowledged that forensic factors such as rigor mortis were consistent with a much broader timeframe, including a time of death as late at 4 a.m. on October 31. He also conceded that "this isn't a precision type of thing. It is a range. [There] can be an honest difference of medical opinion." In fact, it became clear from his testimony and from his report that, when he settled on the specific time of 10 p.m. as the probable time of death, he did so largely on the basis of nonmedical factors, such as reports of dogs barking at that time and his belief that the victim had a 10:30 p.m. curfew. Jachimczyk testified that, at best, he could pinpoint the time of death "plus or minus an hour or so." Adding about one hour or so onto his 10 p.m. estimate would, of course, put the time of the victim's death between 11 p.m. and midnight, right when the occupants of the Lincoln were returning from the Terrien home. It is particularly troubling that the majority, which relies so heavily throughout its opinion on the premise that "the substantial weight of the evidence indicated that the murder most likely was committed between 9:30 and 10 p.m.," fails even to acknowledge that Jachimczyk, the only witness who even came close to pinpointing the murder at that time, conceded that his estimate was only accurate to within a few hours. See footnote 8 of this dissenting opinion.

Ultimately, then, the petitioner is unable to point to a single piece of medical evidence or testimony indicating that the victim was killed at 10 p.m., or even anywhere close to that time. In fact, the forensic evidence implied that 10 p.m. was likely the earliest time that the victim could have been killed, because of the four hours required for her 6 p.m. meal to pass into her small intestine, but far from the latest. Certainly none of the specific expert testimony indicated that an 11:15 p.m. time of death was at all improbable, and neither expert offered any medical rationale as to why 10 p.m. would have been more likely than 11:15 p.m. Although Jachimczyk was, of course, free to consider nonmedical evidence such as curfews and barking dogs in forming his opinion as to the likely time of death, there was no suggestion that he had any special expertise in the fields of teenage or canine behavior. One may assume that the jury, which had access to more detailed, accurate, and timely evidence about the barking dogs and the victim's curfew than did Jachimczyk, would have felt free to reach its own conclusions in that regard.

c

The Curfew

I next consider the petitioner's argument that the victim had a 9:30 p.m. curfew on the night of October 30 and that she likely died at approximately 10 p.m., while walking home from the Skakel residence to comply with that curfew. There are many problems with this argument.

First, and most obviously, the petitioner's argument fails on its own terms. The argument is predicated on the assumption that the victim reliably honored her curfew and, therefore, that she would not voluntarily have stayed out past 9:30 p.m. on October 30 participating in the mischief night festivities, fraternizing with boys, or otherwise occupying herself. The problem is that the victim's last known whereabouts were at the Skakel house at approximately 9:30 p.m., the purported curfew time. That was when the other teenagers visiting the Skakel residence that evening left for home, including the victim's friend Helen, who made a point of being home by 9:30 p.m. to comply with her own curfew. The victim lived right across the street, less than one minute's walk from the Skakel residence. If, as Jachimczyk seemed to postulate, the victim was killed en route while walking home to comply with her curfew, then one would expect that the time of death would have been just moments after 9:30 p.m., rather than closer to 10 p.m.

Although the difference between a 9:30 and 10 p.m. time of death may seem insignificant, there are several reasons why, on the petitioner's theory of the case, it is unlikely that the victim died at 9:30 p.m. As I have noted, Jachimczyk's testimony suggested that it was unlikely that she would have fully digested her dinner by that time. Moreover, as I discuss more fully hereinafter, the dog barking on which the petitioner relies did not commence until at least 9:40 or 9:45 p.m.

The second problem with the petitioner's curfew theory is that it is extremely unlikely that the jury concluded that the petitioner had a 9:30 p.m. curfew on the night in question. Although the victim's mother initially may have told the police that she expected the victim home by 9:30 p.m. on school nights, at trial she made clear that any curfew would have been at least one hour later on the night in question, because the victim's school was not in session on Friday, October 31. See footnote 33 of this dissenting opinion. In fact, the victim's mother explained at trial that, within one week of the murder, she clarified to the police that, because Thursday, October 30, had not been a school night, she would have expected the victim home at approximately 10:30 p.m. rather than 9:30 p.m.

More importantly, the victim's mother repeatedly explained at trial that, because the victim was generally responsible and well behaved, she really had no formal curfew at all. Specifically, she testified that there was just a general expectation that the victim would come home at a reasonable hour, which could have been as late as 11 p.m. on the night in question.²⁰ For example, the victim's mother variously testified as follows:

20 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

• "I thought she would probably be home about 10:30 [p.m.]."

• "I thought she would be home [at] about 10:30 [or] 11 [p.m.]."

• "[P]robably 10:30 or 11 [p.m.] would have been all right."

• "You know, we really never had a specific curfew time. I mean, we didn't have a specific time when the kids had to be home because they were always so good, we never had to do that.... So, for me to say that it was exactly 9:30 [p.m.] or, you know, it's difficult

because we just didn't have those.... [I]t wasn't a curfew. I mean, it's school tomorrow, come home at a decent hour."

• "[W]e did not have a set time when the kids had to be home—I mean, it wasn't, you go out and you be home at 9:30 [p.m.]. We didn't do that. We didn't have to.... I am sure I put it some time ... between 9:30 and 10:30 [p.m.] because that's a logical time I would think for kids to be home if they were going to school."

• "[Expecting the victim home at around 9:30 p.m. on school nights and 10:30 p.m. on other nights] wasn't written in stone.... I didn't have to .... [W]e didn't have these ...."

• "[T]he time was not set in stone."

The victim's brother, John Moxley, provided further support for the conclusion that there was no firm curfew at 9:30 p.m. or even 10:30 p.m., and certainly not one that the victim reliably kept. His testimony indicated that he was not concerned when the victim had not come home by 11 p.m., as he assumed that she was out celebrating mischief night, and that his mother was only "a little worried" at that time. If the victim invariably came home each night by 9:30 p.m., as the petitioner's argument posits, then one would expect her mother to be more worried when she had not returned by 11 p.m.

On the basis of the trial evidence, then, it is reasonable to assume that the jury concluded that, to the extent that the victim would have felt compelled to return home by any particular time on the evening of October 30, that time would have been no earlier than 10:30 or 11 p.m. Indeed, that is precisely how this court summarized the trial evidence when we reviewed this case on direct appeal. See State v. Skakel , supra, 276 Conn. at 641, 888 A.2d 985 (noting that victim's mother "expected that the victim would be home that evening by 10:30 or 11 p.m."). Third, despite the testimony of the victim's mother, it is doubtful that the jury concluded that the victim always complied with her parents' informal curfews or expectations. It is true that the victim's mother testified that her children did not require a curfew because they were well behaved and reliably came home at a reasonable hour. Nevertheless, even a quick perusal of the victim's diary—excerpts of which were admitted into evidence—would have left the jury with a rather different impression. A common subject of the victim's diary entries was her evening social activities. In none of those entries did she make any mention whatsoever of a curfew or indicate that she felt compelled to be home by a certain time. More importantly, several entries logged in the weeks before her death suggest that she did not reliably return home by 9:30 p.m. on school nights or 10:30 p.m. on other nights. For example, on Thursday, September 4, she wrote that her mother was "really ticked" at her after she returned from pool hopping on a school night with the petitioner, David Skakel, and Thomas Skakel (Tommy), as well as her friend Jackie Wettenhall, and that as a result she might not be allowed to attend an upcoming concert. Three days later, on Sunday, September 7, the victim stayed at the Skakel house at least as late as 9:40 p.m. Then, on September 21, she wrote that she had walked home from a block party the day before at 11:30 p.m., after which she went over to Wettenhall's house at 12:45 a.m., went to the Skakel house for one-half hour, and then went home, after which Tommy and others came over to visit her from approximately 2:15 to 4:30 a.m.²¹ Fourth, it is entirely possible that the victim did, in fact, return home prior to 10:30 p.m., only to go back out later in the evening. The victim's mother testified that, between 10 and 11 p.m. on October 30, she was occupied in her bedroom, cleaning up painting supplies and showering and, therefore, that it was possible—albeit unlikely, in her opinion—that the victim had returned home during that time and then gone out again without her knowledge. Although the victim's mother indicated that she was not aware that her daughter had done that before, the jury, presumably having read the victim's diary, knew that she had in fact returned home and then gone back out again late at night to fraternize with the Skakel boys on at least one prior occasion during the fall of 1975.²²

21 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).
It is not entirely clear from the diary entry whether the activities that ensued took place at the Moxley home or at Wettenhall's home. Curiously, the entry indicates that both girls' mothers were awakened at 4:30 a.m. when the boys left.

22 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

For similar reasons, the jury may have hesitated to credit the mother's statement that it was unlikely that the victim had gone back out that night without informing her because the victim "was very good at telling [her] everything that was going on." In fact, the victim's mother also testified that, to the best of her knowledge, the victim had never visited the Skakel house, the Skakel boys had never been to her house, and the victim did not even know the Skakel children. In her diary, however, the victim reveals that she spent at least eight evenings with Tommy and or the petitioner between September 4 and October 4 of that year, several of which involved her visiting the Skakel house or their recreational vehicle.

Accordingly, I think it is extremely unlikely that the jury would have concluded on the basis of the evidence presented at trial that the victim would have felt compelled to return home by 9:30 p.m., or even 10:30 p.m., and, therefore, that her death must have occurred by 10 p.m. Rather, the trial testimony of her mother and brother, as well as her own words as memorialized in her diary, give every indication that she could have remained out—or gone back out—until at least 11 p.m. or so, just around the time that the occupants of the Lincoln were returning from the Terrien home.

d

The Barking Dogs

The petitioner next argues that the "unusual behavior and incessant barking" by certain neighborhood dogs at approximately 10 p.m. supports the conclusion that the victim was killed at that time. This argument echoes the conclusions of Jachimczyk and the habeas court that reports of barking dogs in the vicinity of the Moxley home between 9:30 and 10 p.m. on the night in question indicate that the murder likely occurred at that time. I am not persuaded.

Dogs, of course, are wont to bark, and the jury heard undisputed testimony from multiple witnesses that both of the dogs at issue—the Skakel family's German shepherd, Max, and the Ix family's Australian Shepherd, Zock—as well as other neighborhood dogs, were chronic barkers. Although my search of the case law revealed scant authority on the question of canine cacophony—and that mostly of the dog that didn't bark variety—what little there is confirms that a jury may reasonably disregard evidence of barking as having limited probative value. See, e.g., Robinson v. Pezzat , 818 F.3d 1, 11 (D.C. Cir. 2016) (observing that "self-respecting" dogs bark).

As defense counsel readily conceded in his closing argument, there was no expert testimony presented at trial to support the petitioner's theory that the barking or agitation of a few neighborhood dogs provides reliable evidence of the time of a murder. It is true that Jachimczyk, in concluding that the victim likely died around 10 p.m., reasoned that "there were at least two dogs barking and agitated and something was obviously bothering them ... right around that time ...." There is no indication, however, that Jachimczyk possessed any expertise in canine behavior, that he was more qualified than the jury to interpret the meaning of a dog's bark, or that he was aware of, or ever even considered, whether the behavior of those dogs might have been in reaction to the mischief night festivities. Indeed, the fact that both of the medical experts who testified agreed that the crime could have occurred later in the evening indicates that neither was persuaded that the canine disturbance necessarily was related to the crime.

More importantly, it is highly unlikely that the jury would have concluded that the dogs at issue were good barometers of criminal activity or that their behavior could be used to pinpoint the time of the victim's death. With respect to Zock, although it is true that Helen characterized his vocalizations on that evening as angry, violent, and somewhat atypical, at other times she indicated that his barking was unusual more for its duration than its intensity. Perhaps more importantly, Helen also acknowledged what jurors' own common sense must certainly have told them: that Zock might have barked more protractedly, vociferously, and fearfully on October 30 because teenagers and other children were "out and about around the Belle Haven area" for "mischief night."²³ On that night, as the petitioner himself colorfully described, bands of local teens raced through the yards of Belle Haven egging cars, setting off fireworks, and discharging homemade ballistics such as "funnelators," smoke bombs, and projectiles fabricated from shaving cream cans and butane lighters.²⁴ Helen specifically acknowledged that that sort of destructive behavior could have caused Zock to bark more violently than usual. The jury was free to credit that testimony.²⁵ It is true that Helen testified not only that Zock was barking excessively on the evening of October 30th, but also that he would not come in when she called him. It is difficult to know how much stock the jury may have placed in the fact that Zock, an Australian shepherd, declined to comply when a fifteen-year-old tried to call him into the house. We do know, however, that the jury heard evidence that Zock was a "difficult" dog that was loved only by his family. They also learned that neighborhood teens, including the petitioner, had been known to taunt the poor creature and even shoot him with BB guns. In the face of such treatment, it would be little surprise if Zock were to approach mischief night in a spirit of aggressive trepidation and recalcitrance. Certainly, the jury permissibly could have drawn that inference.

23 It is, therefore, quite misleading for the majority to argue that State's Attorney Jonathan C. Benedict made no attempt to proffer "an alternative explanation as to what had caused the agitated barking and other unusual noises in the victim's yard between 9:30 and 10 p.m." Although it is technically true that Benedict himself only alluded to this point, it is also true that another state's attorney specifically questioned Helen as to whether teenagers and other children were out that night and induced testimony indicating that Helen had initially assumed that this activity was what Zock had been barking at.

24 There is no evidence in the record as to whether the Ix' had owned Zock for more than one year or whether Zock had ever before experienced mischief night and, if so, how he had reacted to that sort of mayhem.

25 The majority dismisses this testimony, noting that there was no specific evidence at trial that anyone was engaged in mischief night activities at that particular time and place. However, there was abundant evidence, including statements from the petitioner himself, that mischief night typically involved the setting off of loud and destructive percussive devices. Anyone who has owned a dog or spent time around dogs will know that the explosion of a firecracker or similar pyrotechnic anywhere in the vicinity typically will be enough to send neighborhood dogs racing to their doors or to the edge of their properties, where they will stand barking at the outside world. That is precisely how Zock behaved.
It is important to note in this respect that this is a habeas case. The petitioner bears the burden of establishing that the crime occurred during the alibi period and, therefore, that the alibi was legally relevant. See Wong v. Belmontes , 558 U.S. 15, 27, 130 S.Ct. 383, 175 L.Ed. 2d 328 (2009) ; Lawrence v. Armontrout , 31 F.3d 662, 668 (8th Cir. 1994). The state is not obligated to prove the specific details of every scenario that is incompatible with the petitioner's claim. In any event, there was more than enough evidence regarding mischief night for the jury to have dismissed Zock's behavior as unimportant.

The plausibility of the petitioner's two dog night theory also is seriously undercut by the fact that the pets at issue—Zock and Max²⁶ —apparently became agitated at completely different times and were barking in different directions. The Skakel's tutor, Kenneth Littleton, in a recorded conversation with his ex-wife that was read to the jury, related that, on the evening of October 30, the Skakel's nanny asked him to go outside to check on barking dogs, including Max, who was barking in the vicinity of the recreational vehicle.²⁷ Littleton indicated that this disturbance, during which he heard rustling sounds in the trees, happened at approximately 9 p.m. He specifically noted that it occurred before he sat down to watch The French Connection, which aired beginning at 9 p.m.²⁸

26 Although Jachimczyk may have relied on police reports relating to a third dog that allegedly acted peculiarly on the night in question, that evidence was not presented to the jury and, therefore, the habeas court properly determined that it was not relevant to the prejudice analysis.

27 Littleton explained that the recreational vehicle was parked in front of the Skakel house. It was, therefore, on the opposite side of the house from the Moxley residence, and far removed from where Zock was barking across from the front of that property.

28 The habeas court proceeded on the assumption that these events transpired "at around 10 p.m.," rather than at 9 p.m. Insofar as I am unable to identify any evidence in the trial record to support that alternative timeline, I conclude that the opinion of the habeas court was clearly erroneous in this respect.

Zock, by contrast, did not begin barking until at least 9:40 p.m. Helen testified that she clearly recalled that she returned home at precisely 9:30 p.m.—the time of her curfew—and shortly thereafter she commenced a telephone conversation, during which Zock began barking. She specifically testified that the barking started at approximately 9:40 or 9:45 p.m. and lasted until 10:15 p.m.²⁹ It is clear, therefore, that the testimony that was presented at trial with respect to barking dogs related to two or more distinct events that occurred over the course of mischief night. Accordingly, there was no reasonable basis for the jury, the habeas court, or Jachimczyk to conclude that simultaneous barking somehow indicated that the victim was murdered at 10 p.m.

29 This was consistent with testimony from the victim's mother indicating that she heard voices outside and the barking of dogs between around 9:30 and 10 p.m.

e

The Jury Requests

The petitioner next points to the fact that, during deliberations, the jury expressed some interest in reviewing testimony related to the petitioner's alibi story and, specifically, whether he was in the Lincoln when it left for the Terrien home at 9:30 p.m. The petitioner's argument appears to be that evidence for or against his alibi story would not have been of interest unless the jury had concluded that the victim was killed during the time when Rushton and John were at the Terrien home.

The following additional facts and procedural history are relevant. The jury began its deliberations on June 4, 2002. On June 5, the jury requested that the testimony of the following six witnesses be read back in full: Julie; Shakespeare; Helen; Andy Pugh, a childhood friend of the petitioner; John Higgins, a former resident with the petitioner at the Elan Therapeutic Boarding School (Elan); and Henry Lee, a professor of forensic science. After having heard a portion of Julie's testimony, the jury sent the court a revised request, indicating that, after completing the read back of all of Julie and Shakespeare's testimony, it was interested in rehearing only that portion of Helen's testimony relating to "who was in the driveway and who [was] left in the car," and only the last two pages of Lee's testimony relating to his statement that there was no direct evidence implicating the defendant in the crime. During the reading of Shakespeare's testimony, the jury also indicated that it did not require a replaying of defense exhibits L and N, which were a tape recording and transcript of Shakespeare's 1991 interview with police detectives that defense counsel introduced as a prior inconsistent statement with respect to the alibi question.

The following day, on June 6, the jury again amended its request. It withdrew its request for a play back of Higgins' testimony and it added a request to rehear various jury instructions. The jury also requested a read back of the rebuttal portion of the state's closing argument, a request that the court denied. The jury finished rehearing Pugh's testimony on the afternoon of June 6. The following morning, the jury delivered its verdict.³⁰

30 The majority's argument that the fact that the jury took several days to reach a verdict indicates that the jury thought that this was a close case is belied by the fact that very little of that time was spent in actual deliberations. The jury spent only two full days and part of two others deciding the case, and much of that time was spent rehearing the requested testimony. The present case is thus readily distinguishable from those on which the majority relies.

After reviewing this procedural history, I am not persuaded by the petitioner's argument that the jury's request to rehear the testimony of Julie, Helen, and Shakespeare, each of whom testified, among other things, as to whether the petitioner went to the Terrien home, indicates that at least some members of the jury must have concluded that the murder was committed during the alibi period. Of course, common sense suggests that a jury's request to rehear particular testimony indicates that the testimony at issue was of some interest to the jury in its deliberations. See State v. Santiago , 224 Conn. 325, 334, 618 A.2d 32 (1992). This is especially true when, for example, the jury requests the testimony of only one or two witnesses or only testimony relating to a particular issue. See id. In the present case, however, the jury asked for a read back of the testimony of six different witnesses, only three of whom even touched on the alibi story. Each of the other three had provided highly inculpatory testimony regarding the petitioner. Lee testified that there was indirect evidence that the petitioner committed the crime, and it was that portion of his testimony that the jury specifically asked to rehear. Higgins testified that the petitioner had confessed to running through the woods with a golf club on the night of the murder and, ultimately, to having committed the crime. Pugh provided a possible motive for the crime, testifying that the petitioner had a crush on the victim. Pugh also testified that the petitioner was agitated the day after the crime, that he was disliked by Zock, and, most significantly, that he had admitted that, on the night of the murder, he had gone into the victim's yard and had been masturbating in the tree under which the victim's body was later discovered.³¹ The fact that the jury reached its verdict soon after it reheard Pugh's testimony—testimony indicating that the petitioner had placed himself at the scene of the murder, under highly suspicious circumstances, after the Lincoln had returned from the Terrien home—strongly suggests that the jury ultimately concluded that the question of whether the petitioner's alibi was valid or fabricated was simply irrelevant.³²

31 Pugh's testimony also reasonably can be understood to evidence the petitioner's consciousness of guilt. See part III B 3 of this dissenting opinion. Pugh testified that the petitioner tried to provide him with an exculpatory account of his activities on the night in question and then urged Pugh to speak with an investigative agency that had been hired to clear the petitioner's name.

32 The majority opinion barely mentions Pugh's testimony. See footnote 8 of this dissenting opinion.

With respect to Julie, Helen, and Shakespeare, it is reasonable to assume that the jury's request to have that testimony read back indicates that, early in its deliberations, the jury gave some consideration to the alibi question. As we explained in Gigliotti v. United Illuminating Co. , 151 Conn. 114, 120, 193 A.2d 718 (1963), however, such a request may reveal nothing more than a jury's "conscientious effort ... to cope with perhaps the most important factual question in the case as it had been submitted to them." In the present case, in light of the substantial evidence of the petitioner's guilt; see part III of this dissenting opinion; and what Justice Palmer recognizes to be the obvious flaws in defense counsel's other strategy of painting Littleton as the likely killer; see Skakel v. Commissionerof Correction , supra, 325 Conn. at 589–97, 159 A.3d 109 (Palmer, J. , dissenting); it stands to reason that the jury, before convicting the petitioner, would have given due consideration to his principal defense, namely, the Terrien alibi. Nothing in the jury's various requests, however, provides even the slightest support for the petitioner's bald speculation that the jury was just one witness away from believing his alibi story and acquitting him.

Furthermore, the fact that the jury requested a read back of only those portions of Helen's testimony relating to whether the petitioner accompanied his brothers to the Terrien home, but asked to rehear all of Julie and Shakespeare's testimony, suggests that the alibi issue was not the jury's primary concern with respect to those latter two witnesses. In Julie's testimony, for instance, the jury may have been interested in her efforts to provide an innocent explanation for the petitioner's attempted suicide in 1977 and his inculpatory statement to the family chauffeur, Lawrence Zicarelli, that "he had done something very bad and he either had to kill himself or get out of the country." See part III B 1 c of this dissenting opinion. The jury also may have been interested in Julie's testimony that the petitioner had been dismissed from several high schools after the murder, that he was drinking alcohol daily by the age of fifteen, and that he was consuming various other controlled substances at that age. All of that testimony reinforced the testimony of other witnesses that the petitioner was in a drug and alcohol induced blackout on the night of the murder and was uncertain whether he had committed the crime.

Turning to Shakespeare, it is even clearer that the jury may have been less interested in her testimony related to the alibi than in what light that she could shed on the petitioner's culpability. Shakespeare testified twice, once for the prosecution and once for the defense. In both instances, she testified that she was certain that the petitioner remained at the Skakel residence after the Lincoln departed. Curiously, however, the jury informed the court that it was not interested in reexamining defense exhibits L and N, which pertained to Shakespeare's prior statements to the police suggesting that (1) she had no firsthand knowledge of whether the petitioner had gone to the Terrien home, and (2) her recollection that he had stayed home was based largely on "tales" and hearsay. It was Shakespeare who provided the strongest support for the state's theory that the petitioner had not gone to the Terrien home. If the jury had been interested in assessing her testimony and credibility on that question, then one would expect that jurors would have asked to reexamine her prior inconsistent statements as well.

Why, then, would the jury have requested a play back of Shakespeare's testimony, if not to explore the alibi question? Although we do not know for certain, one clue may be found in the state's closing argument, during which the prosecutor listed nearly one dozen of the petitioner's admissions, confessions, and other inculpatory statements regarding the murder. Notably, the prosecutor began this portion of his argument by reminding the jury: "On October 31, to [Shakespeare], [the petitioner] said 'Martha is dead, Tommy and I were the last to see her.' " In fact, Shakespeare's precise testimony was that officials at the school that she and Julie attended had instructed the two girls to go home to the Skakel house "immediately," before the end of the school day, on that Friday. See footnote 33 of this dissenting opinion. She further testified that, as they pulled up to the Skakel residence, the petitioner came up to their car and informed them "that Martha had been killed and he and Tommy were the last to see her that night."

Although Shakespeare testified as to this conversation on two separate occasions, her brief references to the petitioner's odd statement did not generate significant attention at trial, and it is possible that the jury did not perceive the true importance of the statement until the prosecutor highlighted it during his closing argument. The petitioner's story was that he had departed for the Terrien home in the Lincoln, while the victim remained standing outside the Skakel residence. When he spoke to the police after the murder, he indicated that she had been standing alone with Tommy. To Richard Hoffman, the petitioner's would-be biographer, he indicated that Helen, Byrne, Wettenhall, Marjorie Walker, and, possibly, Robert Ix also remained at the back door of his house with the victim after he departed. He further informed the police that, upon returning home, he went straight to sleep. In Hoffman's tapes, by contrast, the petitioner recounted that he was unable to sleep and, therefore, that he had gone back out to peep at a female neighbor and, later, to try to "get a kiss" from the victim. Still, he never indicated to Hoffman that he ever saw the victim again that night. The question thus becomes on what basis could the petitioner have concluded that he and his brother were the last people to see the victim alive? Clearly Tommy was with her after the Lincoln departed and might have been among the last to see her. But Helen, Byrne, and perhaps a number of other neighborhood children were with the victim at least as long as was the petitioner, if not longer. In addition, Julie and Shakespeare themselves had remained at the Skakel residence after the Lincoln departed, and so the petitioner would have had no way of knowing whether they saw or spoke with the victim after he departed.

More fundamentally, if the petitioner's account of the evening's events were true, then he could not possibly have known what the victim did, or whom she met, after she left the Skakel residence. This was decades before the age of cell phones and social media. When the petitioner made the incriminating statement to Julie and Shakespeare, the school day had not yet ended³³ and friends and family were just beginning to learn and discuss the tragic news. The petitioner would not yet have had any opportunity to canvass other members of the victim's circle of friends to determine whether she had met up with any of them to celebrate mischief night after leaving the Skakel residence.

33 Although that Friday was a teacher conference day for the local public school that the victim attended, those who attended various private schools in the area of Belle Haven, such as the Skakel children, did have school that day.

In short, if he were innocent, and if his alibi story were true, then the petitioner would not have been one of the last two people to see the victim alive. Nor could he possibly have known who, if anyone, had spent time with the victim after he purportedly left for the Terrien home. Rather than informing Julie and Shakespeare that he was the last to see her alive, it would have made far more sense for him to have asked them whether, and under what circumstances, they had seen the victim after he left. Moreover, Julie and Shakespeare, having remained at the house after the petitioner purportedly left, would have known that the petitioner could not have seen the victim any later than they did—unless of course he was involved in her murder. Accordingly, the petitioner's statement, as recounted by Shakespeare, provided compelling evidence of his consciousness of guilt, and it is unsurprising that the jury asked to rehear her testimony after the prosecutor had highlighted its significance.³⁴ Three final points bear emphasizing in this regard. First, in addition to requesting a read back of the testimony of these witnesses, the jury also asked the court to repeat its instruction regarding "the requirements for a conviction or acquittal." When the court invited the jury to elaborate, the jury asked to rehear the court's instructions regarding "reasonable doubt, inferences, weighing testimony and then the intentional murder charge, elements, time of offense , proximate cause, intent, motive, alibi, and ... concluding instructions ...." (Emphasis added.) Thus, although it would appear that the credibility of the alibi witnesses was on the minds of the jurors as they deliberated, it is equally apparent that (1) that was only one issue among many that concerned them, and (2) they also were focused on the court's time of offense instructions, which made clear that the state was required to prove only that the crime occurred sometime prior to 5:30 a.m. on October 31. Taken together, then, the jury's requests provide at least as strong support for the proposition that the jury accepted the state's partial alibi theory and concluded that the petitioner had committed the crime after returning from the Terrien home.

34 As I have discussed previously in this dissenting opinion, during closing argument the petitioner's statement to Shakespeare and Julie headlined the list of confessions and other inculpatory statements that the prosecutor highlighted for the jury. Nevertheless, the majority, in dismissing the importance of that statement, contends that I have parsed it overmuch. Indeed, the majority goes so far as to allege that, by merely discussing the evidence that the state set before the jury, I have demonstrated my "one-sided approach" to the petitioner's motion for reconsideration.
Apparently eager to avoid my missteps, the majority parses this key element of the state's case not at all. In any event, one need not analyze Shakespeare's testimony too deeply to recognize that it dovetailed perfectly with unrefuted testimony in the record indicating that the petitioner and Tommy really were the last ones to see the victim alive, later that evening, when the petitioner stole onto the victim's property to watch her undressing. See part II B 3 of this dissenting opinion.

Second, to the extent that the jury was focused on the alibi issue when it asked to rehear the testimony of Julie, Helen, and Shakespeare, that tends to diminish the likelihood that the outcome would have been different if Ossorio had testified at trial. While instructing the jury on the alibi, the court specifically directed the jury's attention to the testimony of those three witnesses. On the one hand, the court identified Helen as a disinterested alibi witness, one who was not a member of the petitioner's family. Although the majority largely ignores both Helen's testimony and the court's instruction, the trial court properly recognized that Helen was an independent witness whose testimony corroborated the petitioner's alibi. Specifically, Helen testified on multiple occasions that, although she could not be 100 percent certain, her best recollection was that she had seen the petitioner leave for the Terrien home with his brothers.³⁵

35 In light of this testimony and the trial court's clear instruction thereon, it is difficult to understand how the majority can represent that the "alibi defense [was] comprised solely of the testimony of family members." See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder). Also concerning is the majority's persistent exaggeration of the weakness of the petitioner's alibi, suggesting that an alibi defense that included the testimony of multiple family members as well as Helen's independent testimony was so "far weaker" and so " 'poorly investigated' " that proffering the defense actually constituted a " 'disservice' " to the petitioner.

On the other hand, the court reminded the jury of Shakespeare's testimony that the petitioner had remained at the Skakel residence after the Lincoln departed, and also alluded to Julie's initial statements to the police that she thought she saw the petitioner run by at that time. Accordingly, even if the petitioner's interpretation of the jury requests was accurate, the only reasonable conclusion would be that the jury ultimately credited Shakespeare's testimony and Julie's statements to the police, despite knowing that at least one disinterested witness—as well as several family members—had confirmed the petitioner's alibi. To demonstrate that there is a reasonable probability that the outcome would have been different but for defense counsel's deficient performance, then, the petitioner must establish not only that the jury would have credited Ossorio, but also that it would have found him so credible that his testimony would have overshadowed not only the abundant evidence of the petitioner's guilt; see part III of this dissenting opinion; but also both Shakespeare and Julie's statements that the petitioner had not been in the Lincoln when it left for the Terrien home.³⁶

36 Had Ossorio testified at trial that he had seen the petitioner at the Terrien home on the night of the murder, the state undoubtedly would have impeached that testimony on the grounds that (1) the Skakel boys themselves had not recalled watching television with Ossorio on that night, and (2) it seems highly unlikely that, decades after the fact, Ossorio could have recalled with any precision whether it was that particular evening and not some other Thursday in 1975 in which he watched television with the petitioner.

Third, it is noteworthy that the jury did not ask to rehear the testimony of Jachimczyk. He was the only witness who opined that the murder probably occurred around 10 p.m. Nevertheless, he failed to provide any medical support for that conclusion, and he acknowledged that his estimate was only accurate within "an hour or so." If the jury had been focused on the 10 p.m. timeframe, as the petitioner suggests, then one would have thought that jurors would have reexamined Jachimczyk's testimony as well. The fact that they did not strongly suggests that they ultimately were persuaded that the petitioner had committed the crime regardless of when it occurred.

f

The Ninety Missing Minutes

Although the petitioner himself does not make the argument, the majority argues that the jury could not reasonably have concluded that the victim might have been killed after the alibi period because we cannot account for her whereabouts between 9:30 p.m., when she allegedly left the Skakel residence, and approximately 11 p.m., when the Lincoln and its occupants returned to Belle Haven. The argument appears to be that the victim was a young woman who rarely spent time alone outside her home and that, (1) if she had been out socializing, someone else would have reported having seen her between 9:30 and 11 p.m., and (2) if she had been at home during that period her mother would have been aware of her presence.

i

Needless to say, this argument is highly speculative, and there simply is no way to know whether the jury even would have considered it, let alone found it persuasive. In any event, there are countless plausible explanations for where the victim could have been during the alibi period. In an age before cellphone communications, a teenage girl could have been walking around the neighborhood looking for her friends. She could have been engaging in mischief night festivities with her friend Byrne, who died a few years after the murder, or hanging out at the Skakel residence with Tommy, neither of whom testified at trial. She could have been out with some other young man who, presumably, would not have been especially eager to come forward after the murder and inform law enforcement that he had been the last person to see her alive.³⁷ She might have dozed off in the Skakel's recreational vehicle after drinking too heavily.³⁸ Or she could have come home while her mother was busy painting,³⁹ been reading a book in her room, become bored, changed clothes, and gone back out again, consistent with the petitioner's own confession to Michael Meredith⁴⁰ that he had peeped at the victim while she was undressing later that evening. As anyone who has parented a precocious teenager will know, the possibilities are endless.

37 See footnote 3 of the majority opinion (discussing testimony from habeas trial that victim had sexual encounter with Tommy on Skakel property beginning around 9:30 p.m., after other teenagers departed, which encounter Tommy neglected to report to law enforcement).

38 At trial, the victim's mother testified that, when the victim had not come home by the morning of October 31, she believed that the most likely explanation was that the victim, who had developed a fondness for beer, had been drinking in the recreational vehicle with Tommy and had fallen asleep. There is no doubt, then, that the jury would have been cognizant of this possibility.

39 See part II B 2 c of this dissenting opinion (explaining that state specifically asked victim's mother whether victim could have returned home for a while without her mother's knowledge and victim's mother conceded that it was possible, which was fully consistent with victim's description of her late night activities in her diary).

40 See part II B 3 of this dissenting opinion.

Ultimately, the burden falls on the habeas petitioner to establish that the victim could not reasonably have been alive after 9:30 or 10 p.m., and not, as the majority repeatedly implies, on the state to prove her precise whereabouts throughout the evening. See Skakel v. Commissioner of Correction , supra, 325 Conn. at 460, 159 A.3d 109 ; Hampton v. Commissioner of Correction , 174 Conn. App. 867, 886, 167 A.3d 418 (2017) ; cf. State v. Evans , 205 Conn. 528, 536, 534 A.2d 1159 (1987) (state not obliged to pinpoint exact time of offense even though failure to do so may make it difficult for defendant to establish complete alibi), cert. denied, 485 U.S. 988, 108 S.Ct. 1292, 99 L.Ed. 2d 502 (1988). Indeed, the majority fails to cite even a single case in support of its novel and bizarre theory that the state was obliged to account for the victim's whereabouts during the entire time that the state charged and the experts said she could have been killed and that its failure to do so means that we must assume that the jury concluded that the victim was killed at the earliest possible time, between 9:30 and 10 p.m. Neither law nor logic supports such a theory.

ii

Indeed, the only analysis that the majority does offer with respect to the whereabouts of the victim after 9:30 p.m. relies on facts outside the scope of the trial record. The majority contends that "the evidence in the state's possession at the time of trial ... indicates that the police interviewed hundreds of people at the time of the murder, including Byrne and the victim's other friends and neighbors, subjecting many of them to multiple lie detector tests; and yet not one of them professed any knowledge of the victim's whereabouts after 9:30 p.m." This is just one example of a troubling pattern of the majority relying on or citing to facts and evidence that were not part of the trial record and could not have been considered by the jury in its deliberations.

In a number of instances, for example, the majority discusses evidence unrelated to Ossorio's testimony that was presented only at the habeas trial or that is completely outside the record. See, e.g., footnote 3 of the majority opinion (discussing statement from Tommy indicating sexual liaison with victim beginning at 9:30 p.m.); footnote 4 of the majority opinion (discussing behavior of third dog around 9:45 p.m.); footnote 5 of the majority opinion (discussing conclusion by police that victim must have died between 9:30 and 10 p.m. because 400 people were interviewed and none reported seeing her after that time); footnote 22 of the majority opinion (discussing assertions made in Mark Fuhrman's book and passages in victim's diary that were never entered into evidence); footnote 26 of the majority opinion (discussing habeas testimony of former Elan resident John Simpson that contradicted testimony of state's key witness); part I of the majority opinion (discussing publicity that led law enforcement to reopen case and focus attention on the petitioner); part V B 1 of the majority opinion (discussing what state's theory of murder had been prior to trial); part VI A of the majority opinion (discussing what police had believed about case "for the better part of twenty-five years").

Respectfully, such evidence, which relates to key questions such as the time of the victim's death, the believability of the petitioner's confessions, and other possible suspects in the victim's murder, is not properly the subject of this court's consideration in a Strickland analysis. That evidence is, therefore, inappropriate for the majority to consider in reaching the determination that the jury would not have convicted the petitioner had Ossorio's testimony been presented. In some instances, the majority cites such materials without offering any justification whatsoever for considering facts and evidence that the jury itself never saw. In other instances, the majority purports not to have considered those materials but repeatedly draws attention to them.

I have confined my own analysis in the present appeal to the record that was before the jury and the inferences that the jury reasonably could have drawn therefrom, as properly supplemented by the habeas testimony of Ossorio. I merely note, however, that if I were to follow the lead of the majority and freely discuss materials from outside of the trial court record, there is plenty of information in the public domain and the habeas record that, if it had been before the jury, would have inculpated the petitioner. To name just a few: that he tried to kill a police officer, that he bludgeoned a cat to death with a golf club, and that experts concluded that the victim was likely murdered by a serial peeping tom like the petitioner. There also is evidence from the habeas trial, demonstrating that the victim did not immediately go home after leaving the Skakel residence. Specifically, she spent time with Tommy, who subsequently lied to the police about having been with her after 9:30 p.m. In addition, entries from the victim's diary suggest that her social circle was far wider at that time than her parents were likely aware of and her evening social life far more private than they knew, and also that she had misgivings about her relationship with the boy whom the majority characterizes as her "steady" boyfriend. If we are going to take evidence from outside of the trial record into account when speculating about these matters, then we should consider it all.

iii

The majority also dismisses out of hand the quite realistic possibility that the victim could have returned home for a while after leaving the Skakel residence and then gone out again later that evening, without her mother's knowledge. The majority dismisses this possibility as incompatible with the state's "theory of the case ...." As I will discuss more fully hereinafter, the majority relies on similar reasoning to dismiss other key trial evidence, such as the testimony of Meredith, who testified without contradiction that the petitioner admitted to having seen the victim alive, later that evening, after purportedly returning home from the Terrien home. The view of the majority appears to be that any consideration of the testimony of the victim's mother that the victim could have returned home and then gone back out again on the evening of October 30 is somehow off limits because that evidence contradicts the state's arguments at trial. There are a number of problems with this position. First, to the extent that the majority implies that we may consider only that evidence and those inferences from the evidence that defense counsel expressly set forth during closing argument,⁴¹ this court has flatly rejected such a rule. In State v. Robert H. , 273 Conn. 56, 866 A.2d 1255 (2005), for example "we emphasize[d] ... the well established principles ... that when evaluating the evidence in support of a conviction, we generally do not confine our review to only that evidence relied on or referred to by counsel during the trial.... We also assume that the fact finder is free to consider all of the evidence adduced at trial in evaluating the defendant's culpability, and presumably does so, regardless of whether the evidence is relied on by the attorneys." (Citations omitted.) Id., at 81–82, 866 A.2d 1255 ; see also State v. King , 321 Conn. 135, 153, 136 A.3d 1210 (2016) ("a jury may consider all evidence properly before it").

41 The majority argues, for example, that we should not consider the possibility that the victim stopped home to shower and change clothes on the evening of the murder, because the prosecutor did not expressly discuss that scenario during his closing argument.

Indeed, to adopt a contrary rule would directly contradict the time limitations on counsel's closing arguments. This was a trial that played out over the course of an entire month. More than fifty witnesses took the stand. Their testimony fills literally thousands of pages of transcripts. More than 100 exhibits, some of them quite lengthy, were entered into evidence. To suggest that the state may rely on—and that a reviewing court may consider—only the evidence and analysis that the prosecution had time to specifically highlight during its limited closing argument would place an impossible burden on the state and would, in all likelihood, transform closing argument into a pointless exercise in speed reading rather than an opportunity to provide the jury with a useful, thoughtful framework by which to evaluate all of the evidence of record. The majority has not articulated any rationale or authority as to why these principles should apply differently in the Strickland context. Indeed, Strickland expressly requires that "a court hearing an ineffectiveness claim must consider the totality of the evidence before the ... jury." (Emphasis added.) Strickland v. Washington , supra, 466 U.S. at 694, 104 S.Ct. 2052. The cases cited by the majority are not to the contrary. For example, in Weeden v. Johnson , 854 F.3d 1063 (9th Cir. 2017), the United States Court of Appeals for the Ninth Circuit merely declined to engage in pure counterfactual speculation as to what rebuttal evidence the state might have presented had defense counsel properly sought a psychological evaluation of petitioner. Similar reasoning underlay that court's decision in Hardy v. Chappell , 849 F.3d 803, 823 (9th Cir. 2016). In the present case, by contrast, the question is simply whether we can, and should, assume that the jury rationally considered all of the evidence that the state did , in fact, present at trial.⁴² Moreover, I am not aware of a single case that holds that we must assume that the jury agreed with all of the state's comments and arguments during closing argument, and the majority has not cited any. Indeed, the trial court expressly instructed the jury that the arguments and statements of counsel "are not evidence and you may not consider them in deciding what the facts are." (Emphasis added.) The jury was, instead, properly instructed that its own recollection and understanding of the facts must control. Nor does the majority cite any authority for its contention that we are somehow constrained by three brief rhetorical comments that the prosecutor made during argument. First, during his opening statement, the prosecutor argued that the victim "went out that evening and with the next day being a holiday for Greenwich High School was due in about 10:30 [p.m.]. She never made it home ...." (Emphasis added.) Second, during closing argument, the prosecutor stated that the victim "didn't have school the next day so wasn't supposed to be in until about 10:30 [p.m.] or so that night. Of course, she never got there ." (Emphasis added.) Subsequently, he argued as follows: "[The victim's mother] didn't become concerned until after 11:00 [p.m.] or so. Needless to say, [the victim ] never did make it home ." (Emphasis added.) The majority suggests that these statements preclude us from considering the possibility that the victim stopped home, unbeknownst to her mother, sometime after 9:30 p.m. and then went back out and was killed later that evening.

42 I recognize that it might be inappropriate for a reviewing court to rely on inferences from the trial evidence that are so esoteric or obscure that it is unreasonable to assume that a lay jury would have imagined them on its own. We ought not to assume, for example, that a jury would have performed its own statistical analysis of evidence presented in a trial involving intensive amounts of data. In the present case, by contrast, the evidence and inferences at issue, such as the testimony of the victim's mother that it was possible that the victim came home and went back out without being seen, all were transparently before the jury.

Respectfully, I disagree, for at least three reasons. First, as I have explained, the jury was well aware that arguments of counsel are not evidence and that the jurors alone were empowered to determine what happened on the evening in question. There is no reason, then, to think that the jury would have felt itself constrained by the statements of counsel. I also am aware of no authority for the proposition that a reviewing court should confine its Strickland analysis to only those facts and possibilities that comport with the inferences that the state urged the jury to draw.

Second, the state did, in fact, seek to elicit testimony that the victim could have returned home and then gone back out again. The prosecutor specifically asked the victim's mother the following question: "[T]he previous night before you had fallen asleep, before you came down to watch the news, when you were painting and taking a shower and stuff, is it possible that [the victim] could have come home at some point when you were working or in the shower or dozing and left again and you would not have necessarily known that she had been home?" There would have been no reason to elicit that testimony if the state's view was that the victim could not have returned home and then gone back out.

Third, it is clear from the context that the prosecutor was not intending to literally foreclose the possibility that the victim stopped home and then went back out. Rather, his brief, offhand comments are clearly rhetorical statements meant to highlight for the jury the tragedy that was the victim's death. The fact that he prefaces the remarks with phrases such as "of course" and "needless to say" indicates that he is not intending to eliminate any realistic possibilities, but merely to emphasize the indisputable fact that, ultimately, the victim did not end up safe in her home that night.

Indeed, in State v. King , supra, 321 Conn. at 135, 136 A.3d 1210, we specifically rejected the argument that, when evaluating the scope and nature of the state's theory of a case, our review should be constrained by the sort of offhand prosecutorial comments at issue here. We explained that "closing arguments are often ambiguous and imprecisely phrased given that most attorneys do not appear before the jury like an actor on the stage with every word, phrase, and inflection memorized and exhaustively rehearsed in advance.... [Because] closing arguments of counsel ... are seldom carefully constructed in toto before the event ... improvisation frequently results in syntax left imperfect and meaning less than crystal clear ...." (Citations omitted; internal quotation marks omitted.) Id., at 155, 136 A.3d 1210 ; cf., State v. Albino , 312 Conn. 763, 796, 97 A.3d 478 (2014) (Palmer, J. , concurring) (arguments of prosecutor must be afforded "generous latitude" with respect to "occasional use of rhetorical devices" [internal quotation marks omitted] ).

g

The Most Direct Route Home

The majority also contends that the murder probably took place during the alibi period because "the victim was attacked ... along what would have been the most direct route between where she was last seen and her parents' home." This theory not only is highly speculative but was never articulated by the defendant or, indeed, by any of the more than fifty witnesses who testified at trial.⁴³

43 Relatedly, in the very first paragraph of its summary of the facts of the case, the majority states, as if it were an established truth, that the victim "was likely murdered as she made her way home from the Skakel driveway." In fact, the only support for this unproven statement is the fact that the Skakel house was located catty corner from the victim's house and she was killed near the driveway in front of that side of her home.

The facts are simply that the victim initially was assaulted on or near the westerly leg of the horseshoe shaped driveway in front of her house. Because there apparently was no sidewalk, walkway, or other path leading to the Moxley house from Walsh Lane, the victim's choice when coming home, from anywhere really, would have been between walking up the easterly or the westerly leg of the driveway to the house. Although there is no evidence in the record one way or the other, one would assume that if she was walking home along Walsh Lane from the east she would opt for the easterly route, and the westerly route if coming from the west. Because the Skakel house lies to the northwest of the Moxley house, it is fair to think that she would have walked along that westerly leg of the driveway if returning directly from the Skakel's. But half of Belle Haven also lay to the west of the Moxley residence, and so it is equally fair to assume that the victim would have taken that same route when coming home from any number of friends' houses. Nothing in the record points uniquely to the Skakel residence, as the majority opinion seems to imply.

More importantly, even if the victim was assaulted on her way home from the Skakel house, that in no way implies that the attack must have occurred soon after the Lincoln departed at 9:30 p.m. The victim's diary indicates that, in the weeks before the murder, she frequently spent time hanging out in the Skakel's recreational vehicle. It is certainly possible, then, that she spent some time in the recreational vehicle before returning home on the night of the murder. Indeed, her mother testified that, when the victim had not returned home by Friday morning, she believed that the most likely explanation was that the victim had been drinking beer in the recreational vehicle and had fallen asleep. The jury certainly could have come to the same conclusion.

h

Deference to the Habeas Court's Legal Conclusions

Lastly, I note in this regard that the majority repeatedly relies on what the majority characterizes as the conclusion of the habeas court that "the substantial weight of the evidence indicated that the murder most likely was committed between 9:30 and 10 p.m. on October 30." This is problematic because the questions at issue here—whether the alibi was a complete or partial one, whether the jury reasonably could have concluded that the crime was committed after the alibi period—are either pure questions of law or mixed questions of law and fact, over which our review is plenary. See Small v. Commissioner of Correction , supra, 286 Conn. at 717, 946 A.2d 1203. The habeas court did not hear any testimony, take any new evidence, or make any factual findings regarding the evidence as to the likely time of death. Any conclusions in that regard were formed on the basis of the same cold trial record that now sits before us. The majority was, therefore, obliged to conduct its own comprehensive, objective review of the trial evidence to determine whether the substantial weight of the evidence did, in fact, point to a time of death between 9:30 and 10 p.m. And yet, one searches the majority opinion in vain for even a reference to, let alone an analysis of, most of the relevant facts that I have discussed herein. As I believe I have shown, an objective and comprehensive review of the trial record reveals that the evidence pointing to a time of death during the alibi period was anything but substantial.

3

Evidence Suggesting a Later Death

Against the virtually nonexistent evidence pointing specifically to a 10 p.m. time of death—little more than the notable recalcitrance of a notably recalcitrant dog, as seen through the eyes of a fifteen year old girl—the jury weighed the testimony that a human , the petitioner himself no less, actually saw the victim alive later that evening. The likelihood that the jury disregarded that testimony, and yet still concluded that the petitioner was the killer, is extremely low. That possibility, although certainly conceivable, is not sufficiently probable to surmount Strickland 's high bar.

Near the end of the state's case in chief, the jury heard the testimony from Meredith, a former Elan resident who stayed in the petitioner's home during the summer of 1987. According to Meredith, the petitioner admitted to him during the visit that, on the night of the victim's murder, he had climbed a tree on the Moxley property and masturbated while watching the victim through her window, undressing. The petitioner also told Meredith that, while he was in the tree, he had seen his brother Tommy walk across the Moxley property toward the victim's home.⁴⁴ Meredith further testified that the petitioner told him that the spying incident was "the last time he saw [the victim] alive." The following day, Meredith left the Skakel residence and terminated his relationship with the petitioner, having developed a fear of the petitioner and a feeling that the petitioner "had a violence kind of boiling under the skin ...."

44 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

Meredith's testimony not only established a motive for the murder, reinforcing the state's theory that the petitioner murdered the victim out of jealousy over her relationship with his older brother, but also established that the petitioner had been at the crime scene on the night of the murder while the victim was still alive . Importantly, the events that Meredith described could not have taken place prior to 9:30 p.m.⁴⁵ Accordingly, his testimony, if believed by the jury, conclusively demonstrated that, if the petitioner did go to the Terrien home, then the victim must have been murdered after the Lincoln returned to Belle Haven.

45 Testimony of various witnesses placed the petitioner, the victim, and Tommy at the Skakel residence from the time the Skakel family returned from dinner around 9 p.m. until the Lincoln departed for the Terrien home just before 9:30 p.m.

The majority makes three arguments in an attempt to downplay the importance of Meredith's devastating testimony, which simultaneously sidelined the petitioner's alibi defense and established his motive and opportunity to commit the crime. First, the majority argues that we cannot consider Meredith's testimony because it somehow contradicts the state's "theory of the case ...." I already have pointed out the problems with that argument. See part II B 2 f iii of this dissenting opinion.

Second, the majority suggests that Meredith's testimony was undermined by the testimony of the victim's mother and brother. As majority notes, the victim's mother testified that there were no climbable trees next to the house and that one would have to be "like a monkey" or wearing cleats to climb the trees directly adjacent to the victim's bedroom. The victim's brother also implied that the trees directly in front of the house, which the petitioner told Hoffman he had climbed on the night of the murder, were not climbable.

This is all true, as far as it goes. The majority neglects, however, to discuss virtually all of the evidence regarding the trees that was actually relevant to Meredith's account of the petitioner's confession.⁴⁶ First, the victim's mother elaborated that, although the trees that were right next to the house would have been difficult to climb because the branches were kept trimmed, "[t]here were all kinds of trees all over the place."⁴⁷ Indeed, she specifically testified that "[w]e had Norwegian spruce trees where the branches came down to the ground and I could see my grandchildren having a good time scurrying up and down those."

46 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

47 The victim's corner bedroom had unobstructed windows that faced both to the south and to the west. Her room would, therefore, have been in view of the trees in both directions on the Moxley property.

The jury would have had little difficulty confirming that the victim's mother spoke truly in that regard. The trial exhibits included numerous photographs of the Moxley yard that depicted trees of all sorts, trees that had a direct line of sight to the Moxley home. Among those were divers trees with large, low branches that almost certainly would have been climbable by a teenager in reasonably good physical condition, and certainly by a star high school athlete such as the petitioner.

Notably, at no point did Meredith testify that the petitioner had told him that the tree in question was directly adjacent to the Moxley home. That was the Hoffman story. Meredith's testimony was that "[the petitioner] told me on the evening of the murder that ... he had climbed a tree outside of his house and Martha's house where he could see through her window." (Emphasis added.) Beyond that, Meredith simply assented when the prosecutor asked the following: "So, he told you that the night that [the victim] was killed, he climbed a tree outside of her bedroom window?" Moreover, Meredith testified that the petitioner claimed, while in the tree, to have "seen his brother, Tommy, crossing the yard [toward the Moxley] house. And [the petitioner], of course, didn't want to be seen. So after [Tommy] was out of sight, [the petitioner] climbed down the tree ...." That testimony strongly suggests that the tree in question was not immediately adjacent to the Moxley home. If it were, then Tommy could not have passed the tree and passed out of the petitioner's line of sight on his way to the Moxley home, other than by entering the house itself. Accordingly, although the jury reasonably could have concluded that the petitioner's statement to Hoffman that he climbed a tree directly in front of the Moxley house was untrue—just as the state contended that it was—nothing in the testimony of the victim's family would have undermined Meredith's more incriminating account of the petitioner's confession.

Third, the majority argues that the prosecutor disavowed Meredith's story when he argued that the petitioner had not in fact masturbated in a tree on the night of the murder. It is true that the prosecutor argued at trial that the petitioner had fabricated the masturbation component of the story in the event that his DNA was later identified on the victim's body. The state never argued, however, that the central import of Meredith's testimony—that the petitioner spied on the victim and watched her undressing after he returned home from the Terrien home on the night of the murder—was untrue. Indeed, the state emphasized that this new evidence was the very reason that the petitioner's alibi was no longer a valid defense.

Fourth, the majority suggests that Meredith's testimony was "flatly contradicted" by the testimony of the victim's mother and brother. This argument appears to be that if the victim had returned home and showered or changed clothes before going back out, then her family would have seen her. In reality, however, the victim's mother testified that the Moxley's had "a very large house" and she twice acknowledged that it was possible, albeit unlikely, that the victim could have come back and left again while she was painting in her room, showering, or napping. It is simply incorrect, then, to say that the testimony of the victim's mother "flatly contradicted" that of Meredith. The statement of the majority makes even less sense with respect to the victim's brother, who might have returned to the Moxley house as late as 11:30 p.m. and, therefore, could not possibly have known whether the victim had stopped home before then.

4

Conclusion

To summarize the evidence with respect to the time of death, two of the three medical experts who reviewed the case were of the opinion that the victim could have been killed well outside the 9:30 to 10 p.m. timeframe. The third expert, Jachimczyk, concluded that she had been killed around 10 p.m., but he conceded that his estimate was only accurate to within "an hour or so." Moreover, his review of the forensic evidence suggested that 10 p.m. was quite possibly the earliest time that the victim could have been killed, and certainly not the latest.

Nor did the other evidence presented at trial bear out Jachimczyk's assumption that the victim was most likely killed around 10 p.m. because that is when neighborhood dogs began barking and when the victim was expected home. The evidence suggested that neighborhood dogs—all chronic barkers—became agitated at different times throughout mischief night, in some instances long before the victim was last seen alive. Moreover, the testimony of the victim's family regarding her alleged "curfew" was consistent with a time of death well after the Lincoln returned from the Terrien home. Specifically, the victim's family did not expect her home until approximately 10:30 or 11 p.m. on October 30, because it was not a school night, and 11 p.m., or soon thereafter, was just around the time that the Skakel brothers returned from the Terrien home.

Finally, and most devastatingly, the jury heard unrefuted evidence that the petitioner himself had admitted to having watched the victim undressing, in her room, later that same evening. If the jury credited that testimony, and there is no reason to believe that it did not, then the petitioner's alibi story was simply immaterial.

In light of this record, the petitioner's contention that the jury could only reasonably have concluded that the victim died at approximately 10 p.m. amounts to pure speculation. Accordingly, I conclude that the Terrien alibi was an incomplete one and, therefore, that defense counsel's failure to buttress it with Ossorio's testimony could not have been prejudicial.⁴⁸ III

48 The majority, begging the question, contends that if the jury had concluded that the crime was committed at 10 p.m., then Ossorio's alibi would have been a complete one. That is undoubtedly true. I have dedicated no fewer than fifty pages of this dissenting opinion, however, to reviewing in detail the evidence that was before the jury and explaining why we cannot assume that the jury concluded that the murder was committed at that time. The majority's counter analysis amounts to little more than a few paragraphs of sheer speculation: why would a dog have barked and become agitated if not because of a murder, and how could a teenaged girl have passed an hour or so unnoticed? Respectfully, I do not believe that that is the sort of objective review of the entire trial record that Strickland demands.

In part II of this dissenting opinion, I explained how defense counsel's failure to procure and present the testimony of one additional alibi witness could not have been prejudicial, as a matter of law, because the petitioner's alibi was at best a partial one. This is necessarily so because: the state was required to prove only that the petitioner killed the victim sometime between 9:30 p.m. and 5:30 a.m.; the limited evidence suggesting that the murder was committed around 10 p.m., rather than later that night, was contested and highly speculative; and the jury could have credited testimony indicating that the petitioner himself admitted to having sought the victim out and seen her alive, near the crime scene, after he purportedly returned from the Terrien home around 11 p.m. None of this would have been altered in the least by calling Ossorio as a witness, however credible his testimony might have been.

Even if we were to assume for the sake of argument, however, that the jury did conclude that the crime took place at approximately 10 p.m., the petitioner still would face a Herculean task in establishing prejudice. Specifically, the petitioner bears the burden of demonstrating that, even though the jury found the evidence in favor of the petitioner's guilt so much more compelling than the evidence in favor of his alibi that it found him guilty beyond a reasonable doubt, there is a reasonable probability that adding the testimony of just one additional alibi witness, on top of the numerous witnesses who already had testified to his alibi, would have altered the outcome. This he has failed to do.

A

Governing Law

In evaluating whether defense counsel's failure to procure Ossorio's testimony was prejudicial, the habeas court was of the opinion that "the state did not possess overwhelming evidence of the petitioner's guilt." This was so, the habeas court reasoned, because the state's case was largely circumstantial, consisting primarily of (1) consciousness of guilt evidence and, (2) in the words of the habeas court, "testimony from witnesses of assailable credibility who asserted that, at one time or another and in one form or another, the petitioner made inculpatory statements." The opinion of the habeas court differed in this respect from that of the trial court, Karazin, J. In rejecting the petitioner's motion for a new trial, that court characterized the evidence of guilt presented at trial as "strong ...."⁴⁹

49 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

We owe no deference, however, to either court's assessment of the strength of the state's case. Although "[t]he habeas judge, as the trier of facts, is the sole arbiter of the credibility of witnesses and the weight to be given to their testimony ... [t]he application of historical facts to questions of law that is necessary to determine whether the petitioner has demonstrated prejudice under Strickland ... is a mixed question of law and fact subject to our plenary review." (Citation omitted; internal quotation marks omitted.) Small v. Commissioner of Correction , supra, 286 Conn. at 717, 946 A.2d 1203. Moreover, as this court recently explained in Horn v. Commissioner of Correction , supra, 321 Conn. at 783 n.12, 138 A.3d 908, "there is no requirement that we defer to the habeas court's legal determination that new evidence is so compelling that a reasonable juror could not fail to credit it.... Nor are we required to defer to the [habeas] court's legal determination that there is a reasonable probability that newly discovered evidence would have resulted in a different verdict if credited by the jury ...." (Citation omitted.) 1

Types of Inculpatory Evidence

Before I review the evidence presented at trial, the strength of the state's case against the petitioner, and the likely impact that Ossorio's testimony would have had on the jury, it will be instructive to set forth the well established legal principles that guide that analysis. Although highly reliable modern forms of scientific identification such as DNA analysis; see General Statutes § 54–86k (a) ; were not yet available to law enforcement at the time of the victim's death in 1975, this court has frequently recognized that more traditional types of evidence, including confessions, consciousness of guilt, and other forms of circumstantial evidence, may provide equally persuasive proof of a defendant's guilt.

As this court explained in State v. Miguel C. , 305 Conn. 562, 581, 46 A.3d 126 (2012), "confessions have a particularly profound impact on the jury, so much so that we may justifiably doubt [the jury's] ability to put them out of mind even if told to do so." (Internal quotation marks omitted.) "A confession is like no other evidence. Indeed, the defendant's own confession is probably the most probative and damaging evidence that can be admitted against him.... [The ] admissions of a defendant come from the actor himself, the most knowledgeable and unimpeachable source of information about his past conduct .... While some statements by a defendant may concern isolated aspects of the crime or may be incriminating only when linked to other evidence, a full confession in which the defendant discloses the motive for and means of the crime may tempt the jury to rely upon that evidence alone in reaching its decision." (Citations omitted; internal quotation marks omitted.) Arizona v. Fulminante , 499 U.S. 279, 296, 111 S.Ct. 1246, 113 L.Ed. 2d 302 (1991). Indeed, as Justice Katz' dissent explained in State v. Lawrence , 282 Conn. 141, 202–204, 920 A.2d 236 (2007), "[m]ock jury studies have shown that confession evidence has greater impact than eyewitness testimony, character testimony and other forms of evidence.... [T]riers of fact accord confessions such heavy weight in their determinations that the introduction of a confession makes the other aspects of a trial in court superfluous ...." (Citations omitted; internal quotation marks omitted.)

We also have recognized the persuasive force of evidence tending to show that a criminal defendant was possessed of a guilty conscience. "As we have stated, [t]he state of mind which is characterized as guilty consciousness or consciousness of guilt is strong evidence that the person is indeed guilty ...." (Citations omitted; internal quotation marks omitted.) State v. Weinberg , 215 Conn. 231, 255, 575 A.2d 1003, cert. denied, 498 U.S. 967, 111 S.Ct. 430, 112 L.Ed. 2d 413 (1990) ; see also State v. Felix R. , 319 Conn. 1, 22, 124 A.3d 871 (2015) (McDonald, J. , concurring) (defendant's consciousness of guilt was most significant factor in determination that state's case was strong for purposes of assessing whether prosecutorial impropriety deprived defendant of fair trial); State v. Coccomo , 302 Conn. 664, 721, 31 A.3d 1012 (2011) (Eveleigh, J. , dissenting) ("[C]onsciousness of guilt evidence is second only to a confession in terms of probative value.... Indeed, nothing but an hallucination or a most extraordinary mistake will otherwise explain why a person would harbor a guilty conscience without actually being guilty." [Citation omitted; internal quotation marks omitted.] ); IA Wigmore on Evidence (Tillers Rev. 1983) § 173 ("The inference from consciousness of guilt to 'guilty' is always available in evidence. It is a most powerful one ...."); 2 Wigmore on Evidence (Chadbourn Rev. 1979) § 273 ("[n]o one doubts that the state of mind that we call 'guilty consciousness' is perhaps the strongest evidence ... that the person is indeed the guilty doer ...." [Citation omitted.] ).

Courts and commentators have identified a wide range of conduct that may be inconsistent with a claim of innocence and indicative of a guilty conscience. This includes attempted flight, attempts to fabricate an alibi or inculpate an innocent party, and any other statements made subsequent to a criminal act that tend to identify the speaker as the perpetrator. See State v. Reid , 193 Conn. 646, 656, 480 A.2d 463 (1984) ; State v. Coccomo , supra, 302 Conn. at 709, 31 A.3d 1012 (Eveleigh, J. , dissenting); 2 Wigmore on Evidence (Chadbourn Rev. 1979) §§ 273 and 276.

Furthermore, it does not follow from the fact that the state's case rested primarily on circumstantial evidence⁵⁰ that it was not a strong one. State v. Smith , 156 Conn. 378, 382, 242 A.2d 763 (1968). "The law recognizes no distinction between circumstantial evidence and direct evidence so far as probative force is concerned." (Internal quotation marks omitted.) Id. ; see also Spearman v. Commissioner of Correction , supra, 164 Conn. App. at 545, 138 A.3d 378 (noting that circumstantial evidence can be used to disprove alibi defense). Accordingly, the habeas court went astray, as a matter of law, insofar as that court concluded that the state's case against the petitioner was weak simply because it rested primarily on confessions, admissions, consciousness of guilt, and other circumstantial evidence.

50 It bears noting, however, that the state's case was not based entirely on circumstantial and confession evidence. Most notably, it is undisputed that the victim was killed with a golf club that had belonged to the petitioner's mother and that was typically stored in the petitioner's home. Although the use of the club as the murder weapon does not directly implicate the petitioner, he is one of only a few potential suspects in the crime who had regular access to the weapon.

2

Importance of Objective Prejudice Analysis

Also of concern is the apparent willingness of both the habeas court and the majority to substitute their own credibility determinations for those of the jury.⁵¹ When the state's case is predicated primarily on witness testimony, "[t]he guilty verdict necessarily establishes that the jury found the [s]tate's witnesses to be credible and believed the [s]tate's version of events." Hope v. Cartledge , 857 F.3d 518, 525 (4th Cir. 2017), cert. denied, ––– U.S. ––––, 138 S.Ct. 646, 199 L.Ed. 2d 530 (2018) ; see also Michael T. v. Commissioner of Correction , supra, 307 Conn. at 102, 52 A.3d 655 ; Ayala v. Commissioner of Correction , 159 Conn. App. 608, 616–18, 123 A.3d 447, cert. denied, 319 Conn. 933, 125 A.3d 207 (2015) ; Hunnicutt v. State , Docket No. 05-00-01867-CR, 2001 WL 995972, *6 (Tex. App. 2001) ; In re Towne , 195 Vt. 42, 52, 86 A.3d 429 (2013). The argument that the state's case was weak because the reviewing court believes that the state's witnesses lacked credibility is, therefore, generally without merit because the jury necessarily resolved those questions in favor of the state. See Hope v. Cartledge , supra, at 525 ; see also Wyatt Energy, Inc. v. Motiva Enterprises, LLC , 308 Conn. 719, 737, 66 A.3d 848 (2013) ("Credibility must be assessed ... not by reading the cold printed record, but by observing firsthand the witness' conduct, demeanor and attitude.... [A reviewing] court must defer to the trier of fact's assessment of credibility because [i]t is the [fact finder who has] an opportunity to observe the demeanor of the witnesses and the parties; thus [the fact finder] is best able to judge the credibility of the witnesses and to draw necessary inferences therefrom." [Internal quotation marks omitted.] ); State v. Hart , 198 Conn. 424, 427, 503 A.2d 588 (1986) ("[t]he credibility of witnesses is a matter to be resolved solely by the jury" [internal quotation marks omitted] ).

51 It was, of course, proper for the habeas court to assess the credibility of Ossorio and other witnesses at the habeas trial. My concern here is that the habeas court appears to have determined that the state's trial witnesses lacked credibility solely on the basis of its review of the cold trial record, notwithstanding the fact that the jury, which had the opportunity to observe the demeanor of those witnesses firsthand, clearly credited at least some of their testimony.

That is not to say that, in assessing Strickland prejudice, it is inappropriate for a reviewing court to draw its own independent conclusions regarding the strength of the state's evidence with respect to the issues directly impacted by the errors of counsel. That it may do so is well established. As the United States Supreme Court explained in Strickland , however, in assessing whether counsel's deficient performance prejudiced the petitioner, the reviewing court must take as given any findings unaffected by the error. Strickland v. Washington , supra, 466 U.S.at 696, 104 S.Ct. 2052. For example, in Spearman v. Commissioner of Correction , supra, 164 Conn. App. at 530, 138 A.3d 378, the Appellate Court properly held that defense counsel's failure to call potentially credible and noncumulative alibi witnesses was not prejudicial because, among other things, (1) the credibility of the primary eyewitness to the petitioner's culpability already had been placed at issue before and assessed by the jury and (2) the alibi testimony would not have directly contradicted her testimony, much less other evidence of the petitioner's consciousness of guilt that was "wholly unaffected by the proposed alibi testimony." Id., at 571–73, 138 A.3d 378 ; see also Cox v. Horn , 174 Fed. Appx. 84, 87 (3d Cir. 2006) (credibility of petitioner's confession not affected by errors of counsel); United States v. Andrews , 953 F.2d 1312, 1327 (11th Cir.) (no prejudice when partial alibi would not directly have refuted testimony of defendant's involvement in crime), cert. denied, 505 U.S. 1210, 112 S.Ct. 3007, 3008, 120 L.Ed. 2d 882 (1992); In re Towne , supra, 195 Vt. at 51–52, 86 A.3d 429 (circumstantial evidence not affected).

Although Spearman , like all cases, may be distinguished on its facts from the present case, that case does stand for an important proposition. Namely, that, in assessing Strickland prejudice, we must focus our analysis primarily on the impact to the state's evidence that would have been most directly undermined or contradicted had the omitted evidence been presented to the jury. Contrary to the majority's characterization of my position, I recognize that Strickland permits a reviewing court to consider the overall strength of the state's case when assessing prejudice and, therefore, that there is a sense in which all of the state's evidence is subject to appellate scrutiny. Nevertheless, the primary focus of the prejudice analysis must be on the most directly affected evidence, rather than on a speculative relitigation of every aspect of the trial. See Cox v. Horn , supra, 174 Fed. Appx. at 87 ; People v. Foster , 6 Cal. App. 4th 1, 12–13, 7 Cal.Rptr.2d 748 (1992).

What this means is that, in most habeas cases in which it is alleged or determined that defense counsel failed to identify and present potentially credible alibi witnesses, the focus of the prejudice analysis is on whether the omitted alibi would have called into question other evidence that placed the petitioner at the scene of the crime at the time that it was being committed. Because eyewitness testimony is perhaps the most common means of establishing that presence, and because other means, such as forensic evidence, is not as readily refutable by alibi testimony, the focus of analysis in such cases frequently is on a weighing of the omitted alibi evidence relative to the strength of eyewitness testimony. See, e.g., Griffin v. Warden , 970 F.2d 1355, 1359 (4th Cir. 1992) ("[e]yewitness identification evidence ... is precisely the sort of evidence that an alibi defense refutes best"); Spearman v. Commissioner of Correction , supra, 164 Conn. App. at 545, 138 A.3d 378 ("alibi testimony is frequently the best way to counter eyewitness testimony of a defendant's involvement in a crime").

By contrast, in the present case, not only was the petitioner not convicted on the basis of eyewitness evidence, but, as I shall explain more fully hereinafter, none of the confession, consciousness of guilt, or other inculpatory evidence offered by the state was linked to any particular time of death or required that the petitioner be present at the crime scene during the purported alibi period. Accordingly, the relevance of Ossorio's testimony with respect to the prejudice prong of Strickland is substantially less than is typically the case with credible alibi evidence. In short, the jury could well have convicted the petitioner without coming to any particular conclusion about when the crime was committed. Of course, if a petitioner has been convicted on the basis of types of evidence other than eyewitness testimony, then it is appropriate for the reviewing court to consider, as a general matter, the overall strength of that evidence. Still, I am not aware of any other case in which a reviewing court has gone to such lengths to criticize and deconstruct the state's case. The majority examines each of the state's witnesses, explaining—from a cold trial record—why it does not find their testimony to be believable and, therefore, why a jury also conceivably might not credit them. In so doing, the majority cites evidence from outside of the trial court record and relies on speculative arguments, which the petitioner himself has never made, requiring credibility determinations best left to the trier of fact. Such a review is, in my view, simply inappropriate in the context of a Strickland analysis.

Several facets of the majority's analysis are especially troubling in this respect. First, the majority focuses less on specific defects in the testimony of the state's witnesses and more on what it perceives to be the hurdles that any witness for the state must overcome in this unique, high profile case. Specifically, the majority implies that the testimony by any witness that the petitioner had confessed to killing the victim would be highly suspect because (1) they might have been exposed to Fuhrman's book and other publicity regarding the crime, (2) their memory could have faded in the intervening years, (3) they might have been motivated by rewards offered by the victim's family, and (4) the petitioner himself might have been tricked by the staff of Elan into believing that he had committed the crime.

Second, it is true that other courts have, in weighing the strength of the state's case, considered whether the defendant successfully impeached the state's key witnesses at trial. It would not necessarily be inappropriate, for instance, for the majority to note in the present case that the testimony of Gregory Coleman; see part III B 1 a of this dissenting opinion; was readily impeached. Coleman admitted to being a career criminal who was on heroin at the time that he testified and had altered his story in various respects. No reasonable observer could dispute that his credibility was suspect. The majority, however, then continues to find that the credibility of Coleman's former wife also was suspect, simply on the basis of unfounded, speculative theories that were never raised at trial, such as that she might have lied to obtain a reward. In my view, such scrutiny by a reviewing court is improper. If anything, where a jury has convicted a defendant on the basis of witness testimony, courts will presume the credibility of such evidence. See, e.g., Bridges v. Thaler , 419 Fed. Appx. 511, 516 (5th Cir. 2011).

Third, the majority fails to take any account, in assessing the evidence against the petitioner, as to how directly that evidence would have been impacted by Ossorio's testimony. When courts have considered the credibility of the state's witnesses for purposes of a Strickland prejudice analysis, it usually involves a situation where the improperly omitted evidence or testimony would have directly contradicted the state's key witnesses. The cases on which the majority relies are of that ilk. See, e.g., Kyles v. Whitley , 514 U.S. 419, 441–45, 115 S.Ct. 1555, 131 L.Ed. 2d 490 (1995) ; Gaines v. Commissioner of Correction , 306 Conn. 664, 690–92, 51 A.3d 948 (2012).⁵² By contrast, the more attenuated the relationship between the omitted evidence and the evidence of guilt, the less appropriate it is to relitigate the case with respect to the latter. See, e.g., People v. Foster , supra, 6 Cal. App. 4th at 12, 7 Cal.Rptr.2d 748 ("If this case involved a single, crucial credibility conflict we would be inclined to find prejudice.... With each ever-more-complicated and farfetched episode appellant described ... the credibility significance of 'who dropped the cocaine' shrank."). In the present case, as I explain more fully hereinafter, there is no direct relationship between the evidence of the petitioner's guilt—his various confessions, his consciousness of guilt, his motive and opportunity to commit the crime—and Ossorio's testimony that the petitioner was at the Terrien home at a particular time. Finally, in my view, the majority persistently overstates the likely effects of Ossorio's testimony. In the first paragraph of its prejudice analysis, for example, the majority cites a case for the proposition that if counsel had presented an additional, independent alibi witness, then the jury might also have given greater credence to those family alibi witnesses who did testify at trial. Notably, the case on which the majority relies, Montgomery v. Petersen , 846 F.2d 407, 415 (7th Cir. 1988), states this proposition in an appropriately measured way, positing that "the jury might well have viewed the otherwise impeachable testimony of the [family alibi] witnesses ... in a different light ...." (Emphasis added.) The majority, by contrast, declares that "Ossorio's testimony ... necessarily would have bolstered the credibility of those family alibi witnesses substantially ...." (Emphasis added.) A decision of the United States Court of Appeals for the First Circuit, Gonzalez–Soberal v. United States , 244 F.3d 273, 278 (1st Cir. 2001), offers an instructive contrast to the approach followed by the majority. In that case, the alleged deficient performance was counsel's failure to use two pieces of documentary evidence that would have directly impeached the testimony of the state's two key witnesses. Id., at 274. It was, therefore, necessary for the court to consider the strength of their testimony in assessing prejudice. Id. Notably, even then the First Circuit concluded that the prejudice analysis was a "close call." Id., at 279. Rather than independently evaluate the credibility of the state's trial witnesses on the basis of a cold record, the court remanded the case to the District Court that had presided over the habeas trial to make the necessary credibility assessments firsthand. Id.

52 In Kyles , for example, the suppressed evidence directly undermined eyewitness testimony that constituted the essence of the state's case. Kyles v. Whitley , supra, 514 U.S. at 441, 115 S.Ct. 1555. Similarly, in Gaines , we emphasized that (1) the omitted alibi evidence would have called into question the most essential elements of the state's case, and (2) the habeas court was the sole arbiter of witness credibility. Gaines v. Commissioner of Correction , supra, 306 Conn. at 677, 690–92, 51 A.3d 948.
None of the other cases on which the majority relies authorizes the level of independent scrutiny in which the majority engages here. For example, in Lapointe v. Commissioner of Correction , supra, 316 Conn. at 293, 112 A.3d 1, the majority took pains to explain that, "[n]eedless to say, it is not the role of this court to make credibility determinations ...." Moreover, Chief Justice Rogers, whose vote was necessary to the result, emphasized in her concurring opinion that "[t]he majority is not holding, and I would strongly reject any suggestion, that this court may ever second-guess the factual findings of the ultimate finder of fact ...." (Emphasis in original.) Id., at 352, 112 A.3d 1. Whereas, in that case, we simply considered how evidence the jury never had an opportunity to hear might have been received by the fact finder, the majority in the present case appears to relitigate portions of the underlying trial a priori.

B

Analysis

Having set forth the governing legal principles, I next consider whether there is a reasonable probability that, if the jury had heard Ossorio's testimony, the result of the petitioner's trial would have been different. In this part of the opinion, I summarize the state's case against the petitioner and the evidence of his guilt. That evidence consisted of: various confessions, admissions, and other inculpatory statements; evidence that the petitioner had both the motive and opportunity to commit the crime; and diverse consciousness of guilt evidence. Finally, and powerfully, the petitioner wove together all of these threads in his incriminating statements to Hoffman, his would-be biographer. In light of this compelling evidence of the petitioner's guilt, none of which identified the murder as having occurred at any particular time, I conclude that there is no reasonable probability that the jury, having heard Ossorio's alibi testimony covering the period from approximately 9:30 to 11 p.m., would have reached a different result.

1

Confessions & Admissions

As I have explained previously in this dissenting opinion, a defendant's confession that he committed a crime is, to the typical juror, among the most powerful and compelling proofs of guilt. This is especially true when the confession illuminates the defendant's motives or the means by which the criminal act was accomplished. More than one dozen witnesses testified at the underlying trial regarding the petitioner's numerous confessions, admissions, and other statements that tended to implicate him in the victim's murder. The petitioner made these statements over the course of more than two decades, in three different states, to family members, friends, acquaintances,⁵³ employees of the Skakel family, and Elan classmates and staff. Some of the witnesses to these admissions revealed them to law enforcement soon thereafter, others shared them with family or friends long before the petitioner had become a suspect in the case or a source of public interest. Still others kept the information to themselves and testified only reluctantly, after having been approached by law enforcement or encouraged to testify by the victim's family. Individually, as is frequently the case in criminal trials, certain of the state's witnesses were subject to reasonable impeachment.⁵⁴ Others, however, were beyond reproach. Taken together, these witnesses presented the jury with an overwhelming picture of the petitioner's guilt.

53 It adds to the probative value of these highly incriminating statements that many were made by the petitioner, while he was visibly emotional, to close friends and family members. See State v. Quail , 168 Conn. App. 743, 765, 148 A.3d 1092, cert. denied, 323 Conn. 938, 151 A.3d 385 (2016).

54 It often has been noted that the prosecution is not free to pick and choose its witnesses and that, ultimately, it is the offender himself who determines who will bear witness to his crimes. See State v. Fronning , 186 Neb. 463, 465, 183 N.W.2d 920 (1971) ; State v. Niblack , 74 Wash. 2d 200, 207, 443 P.2d 809 (1968). This maxim assumes particular significance in the context of confession and admission evidence, as a wrongdoer may not choose to place his confidence in "nuns, teachers [and] engineers ...." (Internal quotation marks omitted.) Virakitti v. Mills , United States District Court, Docket No. CV–07–306–BR (AJB), 2010 WL 429992 (D. Or. February 4, 2010).

a

Confessions

The state presented six witnesses who testified that, on three separate occasions, the petitioner directly confessed to having murdered the victim. Those confessions are noteworthy in that they not only explain the petitioner's motive for committing the murder—he had unrequited romantic feelings for the victim, who rebuffed his advances and chose instead to become involved with his arch rival, his older brother Tommy—but also relate specific details of the crime that are largely consistent with the facts of the case.

The state's most important confession witness was Coleman. "Coleman, a resident at Elan from 1978 to 1980, testified about an exchange that he had had with the [petitioner] while Coleman stood 'guard' over [him] following the [petitioner's] failed escape attempt from Elan. During this conversation, the [petitioner] confided in Coleman about murdering a girl who had rejected his advances. According to Coleman, the [petitioner] had admitted killing the girl with a golf club in a wooded area, that the force with which he had hit her had caused the golf club to break in half, and that he had returned to the body two days later and masturbated on it." State v. Skakel , supra, 276 Conn. at 648, 888 A.2d 985.

It is true that Coleman was less than a model witness for the state. Defense counsel was able to impeach his credibility on several grounds, including his reputation for truthfulness, his history of providing inconsistent testimony, and his ongoing struggles with substance abuse, which included having testified while under the influence of controlled substances. Even if we were to assume that the jury found Coleman's personal credibility to be suspect, however, two independent witnesses validated his account of events.

First, Coleman's former wife testified at trial that Coleman had related the petitioner's confession to her when they first met in 1986, more than one decade before the petitioner became a suspect in the case. She further testified regarding an incident that transpired in the mid–1990s, when Coleman became visibly outraged while watching a television show that suggested that Tommy, rather than the petitioner, was the killer. At that time, Coleman again referenced the petitioner's confession and indicated that he was going to call the television network in an attempt to set the record straight.

Coleman was deceased at the time of trial, and I perceive nothing in the record that would have given the jury cause to question the veracity of his former wife. Although she did not claim to have personally witnessed the petitioner's confession, her testimony suggested that, at the very least, Coleman had sincerely believed that the petitioner murdered the victim and that Coleman had articulated the confession to her years before he might have had anything to gain by fabricating it. There was absolutely no evidence in the record to support the majority's baseless speculation that Coleman's former wife might have stood to gain financially from the petitioner's conviction, and the petitioner himself has never made such an argument.

A second corroborating witness, Jennifer Pease, also verified that Coleman had recounted the petitioner's confession long before the petitioner became a suspect in the victim's murder. Pease, who was a housemate of Coleman's at Elan, testified that, in 1979, Coleman, whom she trusted, told her that the petitioner had admitted "that he had beat some girl's head in and killed her with a golf club." Although the majority tries to undermine Pease's credibility by suggesting that she decided to testify out of an "intense dislike of another former Elan witness, [Alice] Dunn"; footnote 27 of the majority opinion; the majority fails to articulate any reason why Pease's dislike for Dunn would plausibly have led her to fabricate a corroborating account of Coleman's testimony regarding the petitioner's confession. Perhaps because Coleman's testimony was corroborated by two independent witnesses, whose own credibility was largely unassailed, the habeas court itself ultimately characterized him as a "powerful witness" in support of the state's case whose testimony proved "particularly troublesome" for the defense.⁵⁵

55 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

The state also presented the testimony of Higgins, another former resident of Elan. Higgins "recounted certain emotional admissions that the [petitioner] had made to him while the two were on guard duty one night on the porch of the men's dormitory at Elan. In particular, Higgins testified that the [petitioner] had told him that, on the night of the murder, there was a 'party of some kind or another' at the defendant's home. The defendant also told Higgins that he remembered rummaging through his garage looking for a golf club, running through the woods with the club and seeing pine trees. Higgins further stated that, as the conversation continued, the [petitioner's] acknowledgment of his culpability in the victim's murder progressed from 'he didn't know whether he did it' to 'he may have done it' to 'he must have done it,' and finally to 'I did it.' " State v. Skakel , supra, 276 Conn. at 648, 888 A.2d 985. Higgins testified that he disclosed this confession to another Elan resident the following day, to former Elan resident Charles Seigan sometime in the late 1980s or early 1990s, and, reluctantly, to law enforcement prior to the trial after having been persuaded to testify by the victim's mother.

In his own testimony, Seigan verified that, in 1996, Higgins confided in him that the petitioner was involved in a Connecticut murder. Seigan subsequently shared that information with law enforcement. As with Coleman, then, independent testimony corroborated that Higgins had related the petitioner's confession to other individuals prior to 1998, when the publication of Fuhrman's book led the state to begin focusing on the petitioner as a possible suspect.⁵⁶ See Skakel v. Commissioner of Correction , supra, 325 Conn. at 571, 159 A.3d 109 (Palmer, J. , dissenting).

56 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

Evidence of a third confession was introduced through the testimony of Geranne Ridge. Ridge testified that, during a party at her Boston apartment in 1997, she overheard the petitioner say, seemingly in jest, "ask me why I killed my neighbor."⁵⁷ Pursuant to State v. Whelan , 200 Conn. 743, 513 A.2d 86, cert. denied, 479 U.S. 994, 107 S.Ct. 597, 93 L.Ed. 2d 598 (1986), the jury also heard a conversation that was taped in 2002 in which Ridge revealed to her friend, Matt Attanian, that the petitioner, while attending a party at her apartment, had confessed that, on the night of her death, he had watched the victim changing in her bathroom while he masturbated in a tree. He further confessed that he had killed her with a golf club, while high on mind-altering drugs, upon learning that she had had sex with his brother Tommy. At trial, Ridge repudiated her taped statements, testifying that she was not personally acquainted with the petitioner and that she had invented the confession story in order to impress Attanian and put an end to his persistent inquiries about her knowledge of the case. Ridge's attempts to repudiate her taped statements were undermined, however, by the following facts: (1) it was Ridge who had initiated and perpetuated the conversations with Attanian in which she described the petitioner's confessions; (2) it was implausible to think that inventing a salacious confession story would do more to "get [Attanian] off [her] back" and terminate his interest than would a simple statement that she had only met the petitioner briefly at a party and that she had no knowledge of the murder; (3) Ridge appeared to be unreasonably agitated for someone who claimed to have no useful information about the case; and (4) certain tabloid newspapers from which Ridge claimed to have gleaned the details about the murder that she had attributed to the petitioner's confession did not, in fact, contain all of those details. In light of these various deficiencies in her trial testimony, the jury certainly was within its province to determine that her original, taped statement was more credible.⁵⁸ See Sanchez v. Commissioner of Correction , 314 Conn. 585, 608 n.15, 103 A.3d 954 (2014) (declining to disregard Whelan testimony that witnesses recanted at trial when assessing strength of state's case because "we allow the fact finder to determine whether the [ Whelan ] statement is credible upon consideration of all the relevant circumstances" [internal quotation marks omitted] ).

57 At trial, several witnesses testified that, while at Elan, the petitioner had been required to wear a large sign inviting other residents to confront him about the murder of the victim.

58 It is not clear to me why, when a witness such as Ridge has changed their account of events over time in a manner that favors the petitioner, the majority concludes that the latter statements reflect the truth, but that when other witnesses whose testimony evolved in a manner that inculpates the petitioner—such as Shakespeare and the victim's mother—the majority credits, and thereby concludes that the jury must have credited, the witness' original statements.

b

Other Admissions

In addition to the three witnesses who testified that the petitioner directly confessed to the victim's murder and the three additional witnesses who corroborated their testimony, the state proffered additional testimony indicating that, on numerous occasions, the petitioner made various statements that were consistent with his having committed the crime. Specifically, witnesses testified that the petitioner acknowledged that, on the night of the murder, he had been high on illicit drugs as well as "blackout," "blind," "stumbling" drunk and, therefore, that he was unable to rule out the possibility that he was the killer. Witnesses who testified to admissions of this sort included Seigan; Dorothy Rogers, a childhood acquaintance of the petitioner and also a former Elan resident;⁵⁹ Elizabeth Arnold, another former Elan resident; and Alice Dunn, a former Elan resident who, like the petitioner, returned to Elan as a staff member after having graduated from the program.

59 Rogers indicated that she reported the petitioner's admissions to law enforcement soon after she left Elan in 1980, almost two decades before the publication of Fuhrman's book. Richard Haug, a police detective employed by the town of Greenwich, confirmed that he and Rogers discussed the matter while she was under arrest for arson in 1980.

In addition to testifying that the petitioner conceded that he could have committed the crime, several of these witnesses also stated that the petitioner had revealed additional incriminating information. Rogers, for example, testified that the petitioner had admitted to her that his family placed him at Elan because they were scared that he might have committed the murder and they wanted to hide him from law enforcement.⁶⁰ For her part, Arnold told the jury that the petitioner recalled running around outside on the night of the murder and that "he didn't know if he had done it or his brother had done it." She also testified as to the petitioner's motive for the crime, relating that "[h]e said that his brother [had sex with] his girlfriend.... [H]e elaborated and said well, they didn't really have sex but they were fooling around. And [his brother had] stole his girlfriend."⁶¹

60 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

61 Because Arnold's testimony does not go to the alibi issue, the time of death, or the petitioner's presence at the crime scene, the majority's efforts to undermine her credibility on the basis of its own reading of the cold trial record are both improper and largely irrelevant to the legal question presently before this court.

It is true that many of these admissions were made when the petitioner was a resident at Elan, and the jury heard extensive testimony that the therapeutic techniques employed by that school were so draconian that a captive resident might have acknowledged the possibility that he had committed a crime simply to spare himself from unending verbal and physical abuse. The jury also heard testimony, however, that the petitioner continued to make such admissions to trusted friends and relatives even after he had left Elan and was no longer subject to its harsh discipline and unique behavior modification techniques. Dunn, for example, testified about a subsequent dinner date that she had with the petitioner at a restaurant fifteen miles away from Elan, at a time when both of them were employed as staff members at the school and the petitioner was living off-campus in Auburn, Maine. She related that the petitioner continued to indicate that he could not recall the evening in question other than that he had not been in his normal state and that either he or his brother could have killed the victim.⁶²

62 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

Perhaps most damning was the grand jury testimony of a family friend that the petitioner had confided to his own father that he believed that he could have murdered the victim. Mildred Ix (Mildred), a longtime neighbor and confidant of the petitioner's parents, testified before the grand jury in this case that, during a conversation on some undisclosed date, the petitioner's father told her that the petitioner "had come up to him and ... said, you know, I had a lot ... to drink that night and I would like to see ... if I could have had so much to drink that I would have forgotten something and I could have murdered [the victim] .... So he asked to go under sodium pentothal or whatever it was." At trial, Mildred claimed to have misremembered this conversation when testifying before the grand jury, and her grand jury testimony was admitted into evidence pursuant to State v. Whelan , supra, 200 Conn. at 743, 513 A.2d 86. Notably, at trial, Mildred only repudiated that portion of her grand jury testimony relating to the petitioner's alleged admission that he could have killed the victim. She remained firm in the belief that the petitioner had indicated to his father that he desired to take a sodium pentothal test. The only reasonable conclusion that the jury could have drawn from that testimony is that, at the very least, the petitioner was troubled by his lack of recall of the night in question and believed that he could have either murdered or witnessed the murder of the victim. Of course, that would not have been the case if the petitioner had been miles away, at the Terrien home, at the time the crime was being committed.

The majority dismisses, out of hand, not only the petitioner's Elan confessions, but also all of his subsequent confessions and admissions, which he made while no longer under the control or influence of the Elan staff. The majority's rationale bears close scrutiny: "The fact that all of those statements were ... made ... in the aftermath of his experience [at Elan] places them in a light that a jury would be far less likely to disregard in the face of a credible alibi. More specifically, the treatment that the petitioner received at Elan as an adolescent was so brutal and coercive, and so directly related to his alleged involvement in the victim's murder, that the jury reasonably would question how that treatment affected the way the petitioner thought about the murder and how he responded to questions about it." The majority continues this remarkable analysis in a footnote: "We note in this regard that [defense counsel] never sought to explain to the jury that an innocent person—particularly an emotionally troubled adolescent who had been subjected to appalling physical and psychological coercion—could convince himself that he may have killed someone in a drunken stupor but have no recollection of doing so."⁶³ Footnote 25 of the majority opinion.

63 In the majority's discussion of the petitioner's experience at Elan, the majority repeatedly implies that the petitioner was beaten and tortured many times over the course of his stay at the school. In fact, however, there was testimony at trial that he received such treatment only on a single occasion, after he had broken a cardinal rule of the school and attempted to escape.

The majority has concluded that the state's case was weak, and that the petitioner was, thus, prejudiced by defense counsel's alleged errors, because none of his confessions or admissions actually count. They don't count because the majority believes that, after leaving Elan, the petitioner could have persuaded himself that he murdered the victim. They don't count even though the petitioner already was an adult when the relevant events at Elan transpired, and even though his confession to Coleman was made before he had been made the subject of any of the psychologically abusive general meetings. They don't count because, although no expert psychological testimony was presented at trial, the majority's own analysis persuades it that Elan could have tricked the petitioner into spending the rest of his life admitting—and even thinking—that he might be the killer. They don't count even though, as the majority itself readily concedes, the jury itself never was presented with the majority's own theories about the defendant's warped subconscious mind. Yet still, the majority asks us to conclude that the petitioner suffered prejudice because, having weighed Ossorio's testimony in light of the state's case, the jury probably would not have convicted the petitioner. In my view, this is unpersuasive absent any expert testimony in the record.

c

Other Inculpatory Statements

In addition to these various confessions and admissions regarding the petitioner's involvement in the victim's death, the jury heard testimony that the petitioner on several occasions made statements that, although not expressly related to the victim's murder, strongly suggested that he had committed a serious crime. This included the testimony of Zicarelli, a Skakel family employee, and Matthew Tucharoni, a local Greenwich barber.

Zicarelli began working for the petitioner's family as a driver, handyman, and gardener in 1976, the year following the victim's murder. Part of his job was to chauffeur the Skakel children, and he testified that the petitioner trusted and confided in him. Zicarelli further testified that, on one occasion in the spring of 1977, while driving the petitioner to a doctor's appointment in New York City, "[the petitioner] said to me that he was very sorry, that he had a lot of respect for me and [that] I was the only person that he could talk to but he had done something very bad and he either had to kill himself or get out of the country." On the return trip, while they were stopped in traffic on the Triboro bridge, the petitioner exited the vehicle and twice ran toward the side of the bridge as if to jump. After Zicarelli forcibly returned him to the car, the petitioner lamented that "if [Zicarelli] knew what he had done, [Zicarelli] would never talk to him again." Immediately following this incident, Zicarelli terminated his employment with the Skakel family.

In his closing argument, defense counsel attempted to persuade the jury that the petitioner's admissions to Zicarelli related not to the victim's murder but, rather, to the fact that the petitioner had been experiencing feelings of embarrassment over having taken his deceased mother's dress to bed with him. As I have discussed with respect to Meredith's testimony, the jury heard testimony that the petitioner admitted even to casual acquaintances that he had snuck into neighboring women's yards, climbed into nearby trees to peep at them, and masturbated. Apparently the petitioner, who revealed this conduct to Hoffman, was perfectly comfortable including it in his memoirs.

The notion that a man who so shamelessly admits to, and even publicizes, such behavior would at the same time feel so ashamed of having slept with his deceased mother's dress that he would feel compelled to go to such extreme measures as to flee the country or commit suicide defies all logic. Even assuming—and it is an enormous assumption—that the petitioner truly felt that he could no longer show his face in Greenwich after having slept with a dress, why leave the country rather than moving to, say, Oklahoma or Alaska? Leaving the country is something one does to escape the jurisdiction of the criminal justice system. The chance that the jury was persuaded by defense counsel's explanation is, therefore, extraordinarily slim.

The state also introduced testimony from Tucharoni, who recounted an occasion in the spring of 1976 when the petitioner, Rushton, and Julie had come into his barbershop. He testified that, as he was preparing to cut the petitioner's hair, the petitioner declared, "I am going to get a gun and I am going to kill him." When Julie responded, "you can't do that," the petitioner replied, "[w]hy not? I did it before, I killed before."

2

Motive & Opportunity

The state also introduced testimony from several witnesses indicating that the petitioner had both the motive and the opportunity to kill the victim. With respect to motive, as discussed previously in this dissenting opinion, Pugh, the petitioner's childhood best friend, testified that, in 1975, the petitioner had "liked [the victim] quite a bit and had a crush on her," but that the victim "didn't seem as interested ...." Pugh also characterized the petitioner's brother Tommy as his rival and adversary. Arnold testified that the petitioner had variously complained that his brother had "fool[ed] around" with and "stole," his girlfriend. Ridge connected the dots, relating how the petitioner admitted to having killed the victim upon learning that she had sex with Tommy. Hoffman corroborated these accounts. He testified that, on the basis of his conversations with the petitioner, he formed the impression that, as of October, 1975, the petitioner had a crush on the victim and wanted her to be his girlfriend. Hoffman also came to believe that Tommy had been the petitioner's "nemesis."

The majority contends that the only two witnesses who support the state's theory that the petitioner killed the victim in a jealous rage or because she spurned his advances were Ridge and Arnold, and that both witnesses were, in the eyes of the majority, tainted by having Fuhrman's book. See footnote 22 of the majority opinion. The majority appears to have overlooked both Pugh and Hoffman. See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

Turning to opportunity, it was undisputed that the petitioner had ready access to the murder weapon, a Tony Penna golf club, that had belonged to the petitioner's deceased mother and ordinarily was kept at the Skakel residence. As I have discussed previously; see part II B 3 of this dissenting opinion; the jury also heard the testimony of Meredith that the petitioner admitted both to having peeped at the victim, through her bedroom window, on the night of the murder and to having seen Tommy approaching her house. Meredith's testimony, thus, placed the petitioner at the scene of the crime while also bolstering the state's theory of motive.

3

Consciousness of Guilt

As I have discussed previously in this dissenting opinion, consciousness of guilt evidence, if credited by the jury, tends to be highly persuasive. See, e.g., State v. Quail , 168 Conn. App. 743, 765–66, 148 A.3d 1092, cert. denied, 323 Conn. 938, 151 A.3d 385 (2016). A wide range of conduct and statements can provide evidence of a defendant's consciousness of guilt. These include the following: (1) fleeing from law enforcement or attempting suicide, (2) acting unnaturally or demonstrating agitation soon after the discovery of a crime, (3) making false statements or giving inconsistent accounts of one's whereabouts and activities, and (4) attempting to concoct an alibi, pin blame for the crime on other individuals, or intimidate witnesses. See 2 Wigmore on Evidence (Chadbourn Rev. 1979) §§ 273 through 276. The present case may be unprecedented in the scope and range of evidence that was proffered by the state to suggest that the petitioner had a guilty conscience.

First, the petitioner repeatedly and substantially changed his accounts of both his activities and his whereabouts on the night of the murder.⁶⁴ Soon after the murder, the petitioner told the police that he had returned from the Terrien home between 10:30 and 11 p.m. and gone to sleep for the night shortly thereafter. He specifically denied having left the house again after returning home. For years after that, his story, both at Elan and to his father, was that he had been so drunk and high that he was unable to recall any of the evening's events. Subsequently, however, the petitioner somehow regained the ability to recall in elaborate detail his activities and conversations on the night in question. Hoffman's records of his conversations with the petitioner, for instance, contain many pages of notes detailing the petitioner's specific recollections about that evening. But these accounts differed dramatically from, and were far less innocuous than, his initial statements to the police. Most notably, he admitted to having gone back out and masturbated while attempting to spy on the victim and another female neighbor. On the basis of the petitioner's ever changing—and increasingly inculpatory—stories, the jury reasonably could have concluded not only that all of his accounts and denials lacked credibility but also that he had reason to mislead the police as to his true whereabouts and activities on the night the victim was killed.

64 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

Second, there was evidence that the petitioner not only attempted suicide but also considered fleeing the United States in the years following the murder. In a classic demonstration of a guilty conscience, the petitioner, having twice attempted to jump off the Triboro Bridge, confided to Zicarelli that he saw no choice but to kill himself or get out of the country. As I have explained previously in this dissenting opinion, the most reasonable interpretation of that statement is that the petitioner felt the need to escape the jurisdiction of the criminal justice system. That would be consistent with his confession to Rogers that the petitioner's family sent him to Elan "to hide him from the police so the police couldn't put him in jail."

Third, there was evidence that the petitioner tried to fabricate an innocuous account of his activities on the night of the murder, an account that would explain the presence of his DNA should it later be identified on the victim or the murder weapon. Pugh, for example, testified that, when he and the petitioner became reacquainted in 1991, the petitioner, having had no contact with Pugh for fourteen years, volunteered that, on the night of the murder, he had been masturbating in the tree where the victim's body was found. Soon thereafter, Pugh began receiving calls from an investigative agency, which had been hired to clear the petitioner of the crime. Ultimately, the petitioner himself called to urge Pugh to meet with that agency.⁶⁵

65 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

Fourth, as I have discussed previously in this dissenting opinion, soon after the victim's body was discovered, the petitioner made the highly incriminating statement that he, along with Tommy, had been the last person to see the victim alive. If he had gone to the Terrien home while the victim remained at the Skakel residence with one-half dozen other neighborhood children, and not seen her again that night, then he would have known that he was not among the last two people to see her. He also could not possibly have known with whom, if anyone, the victim might have spent time later that evening, as many neighborhood children were still in school at the time that statement was made. See footnote 33 of this dissenting opinion. This statement to Shakespeare and Julie was completely consistent, however, with Meredith's testimony that the petitioner had spied on Tommy and the victim from a tree later that evening.

It also bears noting in this regard that (1) the petitioner decided to cut school the day the victim's body was discovered, (2) he was especially agitated after the murder, and (3) the record indicates that he may have attempted to intimidate the state's witnesses during the trial, which conduct the jury may well have witnessed.⁶⁶ All of this consciousness of guilt evidence, taken together, would have strongly suggested to the jury that the petitioner was involved in the murder.

66 See footnote 8 of this dissenting opinion (noting majority's consistent failure to acknowledge facts undermining its theory of the murder).

4

Hoffman Tapes

Finally, I come to the Hoffman tapes, perhaps the single most important piece of evidence in the state's case against the petitioner. Although the petitioner exercised his right not to testify at trial, the jury was nevertheless able to hear the petitioner, in his own voice, providing his account of the evening's events. This gave the jury a unique opportunity to assess his truthfulness and the credibility of his story.

As I have explained, the statements that the petitioner made to Hoffman inculpated him in various ways. He laid out a motive for the crime: he was attracted to the victim, and even went to "get a kiss" from her when he was feeling "horny" on the night of the murder, but she had rebuffed his advances and declined to go to the Terrien home with him, stating that she had to comply with an early curfew. He confessed to having engaged in criminal misconduct on mischief night: shooting apples out of homemade "funnelators" at other children and moving vehicles and then running away; and peeping in a neighbor's window "hoping to see her naked." He admitted to having consumed numerous alcoholic drinks and smoked marijuana throughout that evening, at the age of fifteen. He expressed a ready willingness to deceive adults, explaining that he had planned to cut school the next day and lie about his whereabouts. He undercut his own alibi, stating that, upon returning from the Terrien home, "he remember[ed] that [Shakespeare] had gone home ...."⁶⁷

67 Because Shakespeare went home after the petitioner had allegedly left for the Terrien home, he should not have had any memory of her leaving.

Perhaps most importantly, however, the Hoffman tapes are simply replete with evidence of the petitioner's guilty conscience. First, he tried to cast suspicion on various Skakel employees. He insinuated that Franz Wittine, the Skakel's handyman, had mysteriously disappeared and suggested that Littleton was possessed of a weird quietness and "wouldn't hesitate to pummel you." In a revelation worthy of Sigmund Freud, the petitioner went so far as to suggest that he had tried to get Littleton romantically interested in the fifteen year old victim, claiming to have told Littleton the following: " 'Oh, you should meet Martha, Martha's hot, she's a "shmoke," .... "Yeah, she's really cute." ' " He also conveniently mentioned that he had planned to tell Pugh that he had seen someone lurking near the victim's house that night.

Second, contrary to previous statements indicating that he had gone straight to bed upon returning from the Terrien home or could not recall the night's events, the Hoffman tapes demonstrate that the petitioner was able to recount his activities, thoughts, and conversations in great detail. He remembered the various types of cocktails that he had been drinking that night. He recalled the victim's exact words when she rejected his invitation to join him at the Terrien home. He knew who was sitting in which seat of the Lincoln, and where the car pulled over to change drivers. He was able to retrace his path through his house, up Walsh Lane, to the window of a neighboring "lady's house," and, finally, to the victim's house and through the murder scene.

Third, the petitioner admitted to feelings of guilt, shame, and panic regarding the evening's events. He recalled that he had gone to sleep hoping that no one had seen his behavior at the victim's house and woken up feeling the same way. He spoke of waking with a feeling of panic, and alluded to his "worry of what I went to bed with ...."⁶⁸ He specifically expressed the fear that people would think that he had committed the crime.

68 The state argued at trial that this was a reference to the missing shaft of the golf club that had been used to kill the victim.

Fourth, upon waking the next morning and being confronted by the victim's mother, the petitioner immediately left Belle Haven on his bicycle. His statement to Hoffman was to the effect that he headed "uptown" to see if he could locate the victim, but concluded that " '[t]his is crazy' " and "turned around and came back." The clear implication is that he was gone only briefly from his house and never actually conducted a search for the victim. The problem, however, is that the petitioner also told Hoffman that, upon his return, police cars were "everywhere" at the scene and the victim had been found dead. This means that the petitioner, who left Belle Haven at approximately 8:30 a.m., did not return until at least 12:30 p.m. The jury may well have determined that the petitioner's unexplained four-hour disappearance from the neighborhood after having been confronted by the victim's mother represented an initial attempt at flight.

Finally, and most significantly, the Hoffman tapes revealed the petitioner's bizarre account of his conduct at the Moxley house after returning from the Terrien home, an account that seems precisely calculated to fabricate a legitimate explanation in the event that anyone saw him assault the victim or his DNA was later tied to the murder. He narrated how he went to the Moxley property that night to get a kiss from the victim, how he climbed a tree to spy on her, how he threw rocks and sticks at the window to get her attention, how he pulled his pants down and masturbated for thirty seconds in the tree, how he started to walk through the oval where the victim had been killed until "something in [him] said, '[d]on't go in the dark over there,' " and how he ran home while picking up sticks, throwing rocks, and yelling obscenities.

In perhaps the most extraordinary portion of its opinion, the majority turns a blind eye to almost all of the petitioner's statements in the Hoffman tapes, summarizing and dismissing them in a single sentence. This was some of the most compelling evidence of the petitioner's guilty conscience. The state considered these recorded statements to be such powerful evidence of the petitioner's guilt that the prosecutor made them the centerpiece of his summation. The habeas court itself recognized that the tapes were "an emotionally powerful tool ...." Most importantly, in his brief to this court, the petitioner characterizes his own taped statements as "creepy" and "highly prejudicial," and admits that the state's use of those statements in its summation was "extremely damning ...." (Emphasis added.) The petitioner concedes that these statements not only corroborated the testimony of the state's witnesses and documented his feelings of guilt and panic surrounding the night of the murder, but also unequivocally placed him at the scene of the crime, drunk and high, masturbating and wanting a "kiss" from the victim, right around the time that she was killed. Yet still, the majority, while purporting to conduct an objective assessment of the state's evidence, dismisses the petitioner's own statements out of hand, writing them off as merely "odd" or "suspicious ...." Footnote 24 of the majority opinion.

Rather than directly address these statements, the majority simply argues that, in any event, the state's case must have been weak because the murder remained unsolved for more than two decades and the police initially pursued various other suspects before finally turning to the petitioner. What the majority fails to acknowledge was that this was not some game of musical chairs in which law enforcement's attention happened to turn to the petitioner only after other, more likely suspects had been cleared. Rather, it was the petitioner's own statements admitting that, while sexually aroused, he had sought out the victim for a kiss, attempted to spy on her, and then ran away from her house while holding "sticks" and yelling obscenities, that he drew the spotlight of suspicion onto himself.

5

Family Conspiracy Theory

Perhaps more remarkably, although the evidence that I have outlined—hundreds of exhibits and many days of trial testimony from dozens of state's witnesses—clearly was sufficient to convince the jury of the petitioner's guilt beyond a reasonable doubt, the majority, having brushed much of it away as merely "odd"; footnote 24 of the majority opinion; instead posits that the jury must have rendered its verdict on the basis of a theory that had virtually no evidentiary support. Specifically, the majority posits that the jury was persuaded by the prosecutor's suggestion during closing argument that the Skakel family could have engaged in a decades long conspiracy to cover up the petitioner's crime, a conspiracy that purportedly included the fabrication of the Terrien alibi, and that the family would not have engaged in such a conspiracy unless the petitioner was guilty. Having set up this straw man, the majority asserts that a stronger alibi defense not only would have made it impossible for the petitioner to have committed the murder around 10 p.m. but also would have refuted the conspiracy theory on which the conviction was purportedly based. In fact, the only evidence that supported this "conspiracy" theory consisted of the petitioner's own admissions that his family had sent him to Elan because they were afraid that he was a killer. That is evidence of the petitioner's consciousness of guilt, and the jury properly could have taken it into account as such. But that has nothing to do with a fabricated alibi defense, and Ossorio's testimony would have done nothing to neutralize it.

The other evidence that allegedly supported the family conspiracy theory, such as the fact that the petitioner's father drove his minor children to the police station to give statements about their whereabouts on the night of the crime, was so innocuous that no reasonable jury could have found it to be incriminating. See Skakel v. State , supra, 295 Conn. at 687–95, 991 A.2d 414 (Palmer, J. , dissenting). There certainly is nothing unusual about a father driving his teenaged children to the police station if their statements have been requested. Moreover, the family conspiracy theory failed, transparently, on its own terms. Soon after the murder, for example, Julie told the police that she thought that she had seen the petitioner running in the bushes outside the Skakel residence after the Lincoln had departed. It defies logic to think that the family would have fabricated a grand conspiracy theory but forgotten to include Julie in the plan, or that it would have proceeded with the conspiracy if Julie refused to go along with it.

In light of the almost complete lack of evidence in support of the family conspiracy theory, the insistence by the majority that that theory—rather than the abundant, actual evidence of the petitioner's guilt—was the basis for his conviction, offends several well established legal rules. First, as I already have noted, the trial court properly instructed the jury that the arguments and statements of counsel "are not evidence and you may not consider them in deciding what the facts are." (Emphasis added.) We are required to assume that the jury complied with that instruction and convicted the petitioner on the basis of evidence, not argument.⁶⁹ PSE Consulting, Inc. v. Frank Mercede & Sons, Inc. , 267 Conn. 279, 335, 838 A.2d 135 (2004).

69 It is notable in this respect that the trial court, in its lengthy instructions to the jury and recitation of the potentially relevant facts, never so much as mentioned the state's family conspiracy theory.

Second, to assume that the jury convicted the petitioner on the basis of the prosecutor's groundless references to a family conspiracy, rather than on the basis of the abundant confession and consciousness of guilt evidence that the state had presented at trial, would run afoul of Strickland 's admonition that "[i]n making the determination whether the specified errors [of counsel] resulted in the required prejudice, a court should presume ... that the ... jury acted according to law. An assessment of the likelihood of a result more favorable to the defendant must exclude the possibility of arbitrariness, whimsy, caprice ... and the like.... The assessment of prejudice should proceed on the assumption that the [decision maker] is reasonably, conscientiously, and impartially applying the standards that govern the decision." Strickland v. Washington , supra, 466 U.S. at 694–95, 104 S.Ct. 2052 ; see also Gaines v. Commissioner of Correction , supra, 306 Conn. at 690, 51 A.3d 948 (we must assume that trier of fact acted properly and considered all relevant evidence at trial); State v. Osman , 218 Conn. 432, 437, 589 A.2d 1227 (1991) (noting, with respect to conspiracy charge, that jury may not "resort to speculation and conjecture [or draw] unwarranted inferences from the facts presented").

6

Conclusion

The truth is that the petitioner's guilt was for the jury to decide. Having heard the petitioner's own words on the Hoffman tapes, the jury had to conclude either (1) that he had the bizarre misfortune of walking right through the crime scene, just after the murder, without noticing the body, murder weapons, or fresh blood, and while acting extremely oddly, sexually, and aggressively, or (2) that he committed the crime in a manner more or less consistent with the evidence presented by the state and later attempted to fabricate an explanation, through Hoffman, to hedge against the discovery of inculpatory DNA evidence or the possibility that anyone had witnessed him committing the crime. Nothing that Ossorio would have said at trial possibly could have transformed the former conclusion into a rational one.

For all of these reasons, I respectfully dissent.

ESPINOSA, J., dissenting.

The majority concludes that the habeas court properly granted the petition for

a writ of habeas corpus filed by the petitioner, Michael Skakel, because his attorney at his criminal trial, Michael Sherman, provided ineffective assistance of counsel when he failed to follow up on a passing reference to a possible alibi witness by another witness, Georgeann Dowdle, during her grand jury testimony. The majority further concludes that there is a reasonable probability that, if not for this supposed deficient performance, the petitioner would have been acquitted. I agree with and join Justice Eveleigh's dissenting opinion, in which he thoroughly and persuasively explains why this conclusion is simply untenable. I write separately in order to highlight the continued and disturbing practice, as I discussed in my dissenting opinions in State v. Santiago , 318 Conn. 1, 388, 122 A.3d 1 (2015) ( Santiago II ), and Lapointe v. Commissioner of Correction , 316 Conn. 225, 439, 112 A.3d 1 (2015), of certain justices of this court ignoring the law and fabricating facts in order to reach their desired result. On this occasion, the majority goes even further and ignores and distorts the policies and rules governing motions for reconsideration that have previously guided this court. Specifically, the majority has allowed the petitioner to use a motion for reconsideration to judge shop and to obtain an unprecedented second bite at the apple. Finally, I write to emphasize that the petitioner has received more than due process, and, at the very least , the effective assistance of counsel to which he is constitutionally entitled.

I first address the majority's role as an enabler of the petitioner's attempt to judge shop, which requires a review of the circumstances surrounding the petitioner's motion for reconsideration. This court's decision in Skakel v. Commissioner of Correction , 325 Conn. 426, 159 A.3d 109 (2016), was released on December 30, 2016. The court was divided, with four justices concluding that the judgment of the habeas court granting the petition for a writ of habeas corpus should be reversed, and three justices contending that the judgment of the habeas court should be affirmed. Justice Zarella authored the majority opinion, in which Justice Eveleigh, Justice Vertefeuille and I joined. Justice Palmer wrote a dissenting opinion, in which Justice McDonald joined, contending, among other things, that Sherman had provided ineffective assistance of counsel when he disregarded and failed to follow up on Dowdle's grand jury testimony that she and her "beau" were together at her parents' home on the night that Martha Moxley (Martha) was murdered, when she heard the voices of her brother, James Terrien, and her cousins, several of the Skakel brothers. See id., at 542, 587–88, 159 A.3d 109 (Palmer, J. , dissenting). Justice Palmer also concluded that the petitioner was prejudiced by Sherman's deficient performance. Id., at 618, 159 A.3d 109. Justice Robinson authored a concurring and dissenting opinion in which he agreed with Justice Palmer that Sherman's failure to follow up on Dowdle's grand jury testimony constituted deficient performance. Id., at 531, 159 A.3d 109 (Robinson, J. , concurring in part and dissenting in part).

In addition to being the date that our decision in Skakel was released, December 30, 2016, was Justice Zarella's last day as a member of this court. Because Justice Zarella intended to take a position at a private law firm, he was not eligible, under the policy of this court, to continue to participate in cases that he had heard while a member of the court, as he would have been entitled to do if he had simply retired from full-time active service. See General Statutes § 51–207 (b).

On January 6, 2017, the petitioner filed a "[m]otion for reconsideration and/or reargument and request for en banc hearing or summoning of replacement member of panel." The petitioner contended that, because any party whose case was heard en banc by a seven member panel normally would be entitled to have his motion for reconsideration heard en banc, he was entitled to have his motion for reconsideration heard by a seven member panel. To support this contention, he relied on Practice Book § 71–5,¹ which provides in relevant part: "A motion for reconsideration shall be treated as a motion for reconsideration en banc when any member of the court which decided the matter will not be available, within a reasonable time, to act on the motion for reconsideration." In addition, the petitioner relied on General Statutes § 51–209,² which provides that a majority of judges on an appellate panel is required to change a prior judgment of any court. In addition, § 51–209 requires that, when an appellate panel considering a case is evenly divided, "the court shall reconsider the case, with or without oral argument, with an odd number of judges." The petitioner contended that, pursuant to these procedural rules, this court was required to "summon a replacement for Justice Zarella" in the event that the vote of the six remaining justices was evenly divided on his motion for reconsideration.

Substantively, the petitioner contended in his motion for reconsideration that the majority had incorrectly applied the governing legal standard when it concluded that Sherman's failure to follow up on Dowdle's reference to her "beau," later identified as Denis Ossorio, did not constitute deficient performance. In support of this claim, the petitioner simply parroted the arguments that Justice Palmer had made in his dissenting opinion.

On January 17, 2017, the respondent, the Commissioner of Correction, filed an opposition to the petitioner's motion for reconsideration. The respondent urged this court to "reject the petitioner's efforts to engineer the composition of the panel that could overturn the appellate judgment against him by requesting that a new member of the panel be appointed in lieu of Justice Zarella ...." The respondent also contended that reconsideration was not warranted because the petitioner made no claim that this court had overlooked any fact or significant legal authority when it rendered its decision. Rather, the petitioner simply disagreed with the decision.

I strongly agree with the respondent on both counts. Under the specific circumstances of the present case—in which the original opinion of the court was divided four to three, a justice in the majority left the court after the opinion was released and the petitioner subsequently filed a motion for reconsideration in which he made no claim of clear error or omission in the original decision, but simply reasserted arguments that the majority previously had rejected—no policy or rule of procedure weighs in favor of, much less requires, adding a seventh judge to the panel on a motion for reconsideration.

Practice Book § 71–5 dates from a time when the standard practice of this court was to hear appeals in panels of five, and the rule presumes that the appeal was not originally heard by an en banc panel. When an appeal has been heard by an en banc panel, and a member of the panel has subsequently left the Judicial Branch, this court has never required that a subsequent motion for reconsideration must be heard by a panel of seven. For example, no judge was added to the reconsideration panels in Tomick v. United Parcel Service, Inc. , 324 Conn. 470, 153 A.3d 615 (2016), or State v. Drupals , 306 Conn. 149, 49 A.3d 962 (2012). In Tomick , a panel of six justices heard the appeal: Chief Justice Rogers and Justices Palmer, Zarella, Eveleigh, McDonald and Robinson. The motion for reconsideration en banc was ready on January 26, 2017, at which time Justice Zarella was no longer with the Judicial Branch. I was added to the panel to make six. Justice Vertefeuille was not added to the panel, and there is no indication that she was disqualified from acting in the case. In Drupals , the case originally was heard by an en banc panel, including Justice McLachlan. After he left the Judicial Branch, the motion for reconsideration en banc was denied by a panel of six justices: Chief Justice Rogers and Justices Palmer, Zarella, Eveleigh, Harper and Vertefeuille. No one was added to the panel, despite the request for reconsideration en banc. In addition to these cases, in Harris v. Bradley Memorial Hospital & Health Center, Inc., 306 Conn. 304, 50 A.3d 841 (2012), cert. denied, 569 U.S. 918, 133 S.Ct. 1809, 185 L.Ed. 2d 812 (2013), a panel of five heard the original appeal: Justices Palmer, Zarella, McLachlan and Harper and Judge Gruendel, who had been summoned from the Appellate Court because the remaining justices of this court were disqualified. A motion for reconsideration en banc was filed after Justice McLachlan's departure from the court, and was heard and denied by the remaining four panel members.

The court presumably has followed this practice because the primary function of a motion for reconsideration en banc, namely, "to secure or maintain uniformity of decision"; Practice Book § 71–5 ; simply has no relevance when the original appeal was heard by an en banc panel of seven justices. Rather, that function comes into play only when there is a possibility that the same court could reach two divergent results with respect to the same issue because of the different composition of the panels addressing the issue. For example, if the vote of a panel of five justices were divided three to two on an issue, the same issue were to arise in a subsequent case, and two of the three justices in the majority in the first case were to be replaced by the two justices who did not hear that case, the second panel could split in the other direction. When a panel of seven justices hears an appeal, however, the parties are assured that a majority of the entire court, as constituted at the time of the original decision, was in agreement about the result. I would note that it has been the policy of this court since 2009 to hear cases en banc whenever possible.

Thus, when a motion for reconsideration is filed after a case has been heard by a panel of seven justices and, for some reason, an original panel member is no longer available, the primary function of adding a new judge or justice to the panel would not be to secure uniformity of decision, but simply to allow a disappointed party to attempt to persuade the newly added judge or justice to vote differently than the departed justice. What Practice Book § 71–5 most assuredly was not intended to encourage or to allow, however, is judge shopping by a disappointed party. Similarly, the provision of § 51–209, requiring that "[w]henever the Supreme Court is evenly divided as to the result, the court shall reconsider the case ... with an odd number of judges," was not intended to allow a party to seek a different panel after a closely divided en banc panel has decided an appeal and a justice in the majority has departed from the court. Rather, it was intended only to ensure that the parties can obtain an appellate decision affirming or reversing the judgment under review in the first instance.

In support of its contention to the contrary, the majority states that "the decision to add Justice D'Auria to the panel in the present case is consistent with the practice of every other sister state court that has addressed the issue posed by the petitioner's motion, that is, whether to add a judge to a panel when an original panel member was unable to participate in the resolution of a timely filed motion for reconsideration." (Emphasis in original.) Part II of the majority opinion. The majority cites five cases that, in its view, support this claim. In one of those cases, however, there is no evidence that the members of the original panel were evenly split on the motion for reconsideration. See Commonwealth v. Franklin , Docket No. 12-P-569, 2015 WL 4663516 (Mass. App. August 7, 2015). In another case, the decision to grant reconsideration and to overrule the prior decision was not based solely on the fact that a new justice had been added to the reconsideration panel; rather, several justices changed their votes. Compare State v. Eriksen , 172 Wn. 2d 506, 259 P.3d 1079 (2011) (Justices Fairhurst, J. Johnson, Stephens and Wiggins and Chief Justice Madsen in majority and Justices Alexander, Owens, C. Johnson and Chambers dissenting), with State v. Eriksen , 170 Wash.2d 209, 241 P.3d 399 (2010) (Justices Sanders, C. Johnson, Chambers, Owens, J. Johnson and Stephens in majority and Justices Fairhurst and Alexander and Chief Justice Madsen dissenting), superseded, 172 Wash.2d 506, 259 P.3d 1079 (2011). Thus, these cases bear little similarity to the present case. With respect to the remaining three cases, two were issued by the same court , only days apart; see University of Michigan Regents v. Titan Ins. Co. , 484 Mich. 852, 769 N.W.2d 646 (2009) (issued July 31, 2009), and United States Fidelity Ins. & Guaranty Co. v. Michigan Catastrophic Claims Assn. , 484 Mich. 1, 795 N.W.2d 101 (2009) (issued July 21, 2009); and all three cases were split decisions, with the dissenting justices making forceful and persuasive arguments as to why reconsideration was inappropriate under the circumstances that are present here. See University of Michigan Regents v. Titan Ins. Co. , supra, at 854, 769 N.W.2d 646 (Young, J., with whom Corrigan, J., joined, dissenting) ("[The] [p]laintiffs have not raised any new legal arguments in their motion for reconsideration.... [The] [p]laintiffs have cited nothing more than their disagreement with prior courts' application of the plain language of the relevant statutes, plain language which could not mislead either the parties or this [c]ourt. There is one significant change since our November 26, 2008 [decision]: the composition of this [c]ourt. Justice Hathaway unseated former Chief Justice Taylor in the 2008 election and took office on January 1, 2008, thereby shifting the philosophical balance on the [c]ourt. There is no palpable error, but there is a new philosophical majority." [Emphasis in original.] ); United States Fidelity Ins. & Guaranty Co. v. Michigan Catastrophic Claims Assn. , supra, at 27, 795 N.W.2d 101 ("The facts have not changed. The text of the statute at issue has not changed. The parties' arguments have not changed. And the rationale advanced in the opinions of this [c]ourt has not changed. Yet, within a matter of months, a decision of this [c]ourt, thoughtfully briefed, argued, and considered by seven justices, is no longer worth the paper it was written on. Even the casual observer, however, does not really need to ask why. The reason is obvious: On January 1, 2009, the composition of this [c]ourt changed."); id., at 29, 795 N.W.2d 101 (majority overruled "[long-standing] and clear principle" that "a rehearing will not be ordered on the ground merely that a change of members of the bench has either taken place or is about to occur" [internal quotation marks omitted); Johnson v. Administrator, Ohio Bureau of Employment Services , 48 Ohio St. 3d 67, 71, 549 N.E.2d 153 (1990) (Holmes, J., with whom Moyer, C. J., joined, dissenting) (dissenting from decision to reconsider and reverse prior decision because "[t]he only change in circumstances in this case was the composition of the court"). Thus, to put it mildly, the persuasive value of these decisions is less than compelling.

Moreover, the very limited purpose of a motion for reconsideration is to "demonstrate to the court that there is some decision or some principle of law which would have a controlling effect, and which has been overlooked, or that there has been a misapprehension of facts." (Internal quotation marks omitted.) Chapman Lumber, Inc. v. Tager , 288 Conn. 69, 94 n.28, 952 A.2d 1 (2008) ; see also E. Prescott, Connecticut Appellate Practice & Procedure (5th Ed. 2016) § 8–5:9.2, p. 496 ("[T]he purpose of [a motion for] reconsideration is generally to point out errors or omissions in the original decision so that they may be corrected. Good grounds for reconsideration would seem to include: [1] over-looking a controlling statute or case; [2] new authority decisive of the appeal; [3] lack of subject matter jurisdiction; and [4] failure to join an indispensable party." [Footnotes omitted.] ). "[A ] motion to reargue [however ] is not to be used as an opportunity to have a second bite of the apple. " (Emphasis in original; internal quotation marks omitted.) Chapman Lumber, Inc. v. Tager , supra, at 94 n.28, 952 A.2d 1. I recognize that the motion at issue in Chapman Lumber, Inc. , was not a motion for reconsideration of an appellate decision, but a motion to the trial court to open a judgment. I can see no reason, however, why a different principle should apply here. If a motion for reconsideration may not be used to obtain a second bite at the same apple, a fortiori, it should not be used to obtain a second bite at a different apple.

In reaching this conclusion, I am mindful that this case does not implicate the doctrine of stare decisis, which comes into play when a party has requested the court to overrule an earlier decision of the court. It is arguable, however, that the principles underlying that doctrine—saving resources, promoting judicial efficiency and promoting the perception that the decisions of this court are stable and based in the law, and not in the personal inclinations of judges—have even greater force in the present circumstances than when this court is asked to overrule an earlier decision. See State v. Peeler , 321 Conn. 375, 416–17, 140 A.3d 811 (2016) (Robinson, J. , concurring) ("Stare decisis is a formidable obstacle to any court seeking to change its own law.... It is the most important application of a theory of [decision-making] consistency in our legal culture and it is an obvious manifestation of the notion that [decision-making] consistency itself has normative value.... Stare decisis does more than merely push courts in hard cases, where they are not convinced about what justice requires, toward decisions that conform with decisions made by previous courts.... The doctrine is justified because it allows for predictability in the ordering of conduct, it promotes the necessary perception that the law is relatively unchanging, it saves resources and it promotes judicial efficiency." [Internal quotation marks omitted.] ); see also id., at 419, 140 A.3d 811 (Robinson, J. , concurring) ("a change in the personnel of the court affords no ground for reopening a question which has been authoritatively settled" [internal quotation marks omitted] ); id., at 379, 140 A.3d 811 (Rogers, C. J. , concurring) ("When neither the factual underpinnings of the prior decision nor the law has changed, the [c]ourt could not pretend to be reexamining the prior law with any justification beyond a present doctrinal disposition to come out differently from [the prior decision]. To overrule prior law for no other reason than that would run counter to the view repeated in our cases, that a decision to overrule should rest on some special reason over and above the belief that a prior case was wrongly decided." [Internal quotation marks omitted.] ); id., at 381, 140 A.3d 811 (Rogers, C.J. , concurring) ("[a] change in the constituency of this court is not a sufficiently compelling reason to warrant departure from a [recent decision]" [internal quotation marks omitted] ).

Unlike the situation in which stare decisis is implicated, there is no risk in the present case of either depriving the parties of the opportunity to obtain a decision of a majority of the court that was in existence when the appeal was filed or of depriving Justice D'Auria of an opportunity to weigh in on an issue that was raised in an appeal filed during his tenure. The parties have obtained such a decision, and Justice D'Auria cannot have had any reasonable expectation that he should be able to weigh in on an issue that was raised in an appeal that was already decided at the time that he became a justice. Accordingly, it is perfectly clear to me that, under the specific circumstances of this case, where the petitioner has not claimed that there was any clear error or omission in the original en banc decision, there are six justices remaining from the en banc panel that decided the appeal and the justices are evenly divided as to whether the original decision should be affirmed or reversed, the petitioner is not entitled to a second bite at a newly constituted apple and the original decision should stand. See General Statutes § 51–209 ("[n]o ruling, judgment or decree of any court may be reversed, affirmed, sustained, modified or in any other manner affected by the Supreme Court or the Appellate Court unless a majority of the judges on the panel hearing the cause concur in the decision").

It is also clear to me that, when the petitioner's motion for reconsideration was ready to be heard by the court on January 17, 2017, it should have been placed on the conference agenda and acted on expeditiously, allowing a reasonable period of time, in the range of two to three weeks, for the justices to consider the parties' arguments, as is normally done. I would note that, between November, 2015 and April, 2017, not including this case and two other cases for which exceptional circumstances existed, the average period of time between the date that a motion for reconsideration was ready to be heard and the date that it was disposed of was approximately twenty-two days. In fact, the petitioner in Taylor v. Commissioner of Correction , 324 Conn. 631, 153 A.3d 1264 (2017), filed a motion for reconsideration on February 24, 2017, that became ready to be heard when the respondent filed an opposition on February 28, 2017, and the motion was placed on the conference agenda for that very same day , despite the fact that the appeal in that case was heard by a panel of six and the petitioner requested reconsideration en banc.³

Needless to say, that was not done in the present case. Instead, although the motion for reconsideration was placed on the conference agenda shortly after it became ready, it was immediately marked over and consideration of the motion was delayed for months. During that period of delay, on March 8, 2017, Justice D'Auria was sworn in as a justice of this court. Thereafter, a majority of the six members remaining from the original panel in this case voted that Justice D'Auria should be added to the panel, thereby effectively voting that a motion for reconsideration may be used as a vehicle for judge shopping. This vote also ensured that a decision on the merits of the motion would be further delayed while Justice D'Auria familiarized himself with the case. Ultimately, there was no vote on the merits of the motion for reconsideration until May, 2018, more than 460 days after the motion for reconsideration was ready.

To my knowledge, this delay is the longest delay in ruling on a ready motion for reconsideration that has ever existed at this court. Indeed, this court took longer to rule on the motion for reconsideration than it did to issue its original decision in this case. See Skakel v. Commissioner of Correction , supra, 325 Conn. at 426, 429, 159 A.3d 109 (case was argued on February 24, 2016, and decision was released on December 30, 2016). Moreover, with the exception of Santiago II , supra, 318 Conn. at 1, 122 A.3d 1, this court has never, to my knowledge, added a new judge or justice to a reconsideration panel when the original appeal was heard by an en banc court. Santiago II was exceptional, however, because, while the original appeal in that case was pending before an en banc panel consisting of Chief Justice Rogers and Justices Norcott, Zarella, McLachlan, Eveleigh, Harper and Vertefeuille, the defendant moved for permission to file a supplemental brief and for additional oral argument addressing the effect on his appeal of Public Acts 2012, No. 12–5 (P.A. 12–5), which prospectively repealed the death penalty for crimes, effective on the date of passage. See State v. Santiago , 305 Conn. 101, 307 n.167, 49 A.3d 566 (2012) ( Santiago I ), superseded in part by Santiago II , 318 Conn. 1, 122 A.3d 1 (2105). This court denied the motion, concluding that the issues raised by the defendant, which had not been previously raised on appeal, would be more appropriately addressed in the context of postjudgment motions. See id., at 308 n.167, 49 A.3d 566. After the decision in Santiago I was released, the defendant filed a motion for "reconsideration," seeking review of the issues involving P.A. 12–5. By that time, Justice Harper had retired and Justice McLachlan had left the court for private practice. Justice McDonald and I had joined the court. In addition, Justice Palmer, who had been disqualified in Santiago I , was not disqualified from considering the issue raised in Santiago II. Ultimately, Justice Palmer, Justice McDonald and I replaced Justice Harper, Justice McLachlan and Justice Vertefeuille, who had taken senior status before Santiago I was heard. Thus, the defendant's motion for "reconsideration" was heard by a panel consisting of Chief Justice Rogers and Justices Norcott, Palmer, Zarella, Eveleigh, McDonald and me. As I have indicated, however, the issue raised in that motion was entirely new and, therefore, the defendant was actually seeking consideration of the issue for the first time, not "reconsideration." Moreover, the court had expressly directed the defendant to raise the claim in a postjudgment motion, and the defendant had not requested a new panel. Accordingly, unlike in the present case, the defendant in Santiago II was not judge shopping, and resolution of the new issue by a new panel created no risk of inconsistent decisions based solely on the composition of the reviewing panel.

In my view, there can be only one explanation for this calculated strategy to delay the resolution of the petitioner's motion for reconsideration until Justice D'Auria could be added to the panel, contrary to the court's ordinary practice of deciding such motions when they are ready: the majority's belief that the petitioner is entitled to special treatment . There are thousands of convicted criminals languishing in Connecticut's prisons, approximately two-thirds of whom are either African–American or Hispanic, who would undoubtedly be thrilled to receive such special treatment. See Justice Center, Council of State Governments, "In Brief: Examining the Changing Racial Composition of Three States' Prison Populations," (March 2015), p. 4, available at https://csgjusticecenter.org/wpcontent/uploads/2015/03/ExaminingtheChangingRacialCompositionofThreeStatesPrisonPopulations.pdf (last visited April 30, 2018). Unfortunately for them, the vast majority do not share the petitioner's financial resources, social standing, ethnicity or connections to a political dynasty. Nor do their cases share the same "glam" and celebrity factor as this cause célèbre, which over the course of decades has spawned dozens, if not hundreds or even thousands, of books, television programs, and newspaper and magazine articles. Indeed, this court has begun to develop a boutique practice of giving special treatment to convicted criminals in such cases. See, e.g., Lapointe v. Commissioner of Correction , supra, 316 Conn. at 225, 112 A.3d 1. In my view, the primary effect of decisions like the majority decision in the present case and in Lapointe will be to erode public confidence and trust that the courts will treat all parties equally and to undermine the rule of law.

With respect to the merits of the petitioner's motion for reconsideration, as I have indicated, I agree with Justice Eveleigh's dissenting opinion in which he explains in detail the flaws in the majority's analysis and the extent to which the majority has ignored, distorted and misrepresented the evidence in this case in order to reach its desired result. I would offer the following sampling of evidence that the majority downplays or ignores outright in its zeal to reverse the petitioner's conviction:

(1) all of the prosecutor's repeated statements that the state maintained that the petitioner had only a partial alibi;

(2) defense counsel's repeated acknowledgment that the state believed that the petitioner had only a partial alibi;

(3) the testimony of Thomas Keegan, a captain in the Greenwich Police Department, that, according to the chief medical examiner who had performed the autopsy on Martha, the time window for her death was too broad to be helpful in investigating the crime;

(4) the testimony of H. Wayne Carver II, the state's chief medical examiner at the time of trial, that makes it clear that there was no forensic support for the conclusion that Martha was killed before 10 p.m.;

(5) the testimony of Joseph Alexander Jachimczyk, a forensic pathologist and medical examiner who testified as an expert for the petitioner, that would support the conclusion only that Martha was murdered sometime prior to dawn on October 31, 1975, and that 10 p.m. on October 30, 1975, was the earliest time at which she could have died;

(6) Dorothy Moxley's testimony that Martha had no set curfew of 9:30 p.m., even when she had to go to school the next day, and any informal curfew would have been at least one hour later than that on the night of the murder;

(7) Martha's diary, showing that she occasionally returned home late or went out again after returning home; (8) evidence showing that the barking of Helen Ix's dog, Zock, could well have been in response to "mischief night" activities such as the egging of cars, the setting off of fireworks and the discharging of homemade ballistics, and not in response to the attack on Martha;

(9) Andrew Pugh's testimony that the petitioner had a crush on Martha, that the petitioner was agitated the day after the murder and that he had admitted masturbating in the tree under which Martha's body was discovered on the night of the murder ; and

(10) Andrea Shakespeare's testimony that, as she arrived home from school on the day after the murder, the petitioner approached her car and informed her that Martha had been killed and that the petitioner and Thomas Skakel were the last persons to see her alive.

Most significantly, the majority simply dismisses the evidence showing that the petitioner confessed to the murder on multiple occasions over the course of decades and that he repeatedly changed his account of his activities on the night of the murder.

Moreover, as Justice Eveleigh also points out, in addition to downplaying and ignoring any evidence that does not support its desired outcome, the majority actually invents evidence to support its theory that the jury would not have convicted the petitioner if Ossorio had testified. Specifically, the majority contends that there is evidence compelling the conclusion that 9:30 p.m. on October 30, 1975, was "the last time any of the victim's friends reported seeing her"; part I of the majority opinion; when there was no such evidence before the jury.

The majority also cites other evidence that either was not before the jury or that is completely outside the record. Specifically, the majority cites evidence that: (1) Thomas Skakel engaged in sexual activities with the victim that night; see footnote 3 of the majority opinion; (2) a third dog was behaving oddly; see footnote 4 of the majority opinion;

(3) the victim likely died between 9:30 and 10 p.m. because none of the hundreds of people who were interviewed by the state police reported seeing her after that time; see part VI A of the majority opinion; and

(4) the testimony of John Simpson at the hearing on the petitioner's new trial petition contradicted testimony by a key witness for the state. See footnote 26 of the majority opinion.

Although the majority denies relying on this evidence, it is clear that the only reason that the majority refers to it is to bolster its conclusion that, if the case had been tried differently, there is a reasonable probability that the petitioner would not have been convicted.

It is essential to remember, however, that, in determining whether Sherman's failure to locate Ossorio and call him as a witness is sufficient to "undermine our confidence in the outcome"; Strickland v. Washington , 466 U.S. 668, 694, 104 S.Ct. 2052, 80 L.Ed. 2d 674 (1984) ; the role of this court is not to speculate as to what would have happened if the petitioner's case had been tried in the manner that the members of the majority—none of whom has ever tried a murder case or presided over a murder trial—thinks that it should have been tried . Nor should the court evaluate the credibility of evidence presented at the criminal trial that was unaffected by the alleged deficient performance, which the jury is presumed to have found credible. The only question before this court is whether, in light of the evidence that was actually presented to the jury and the arguments that were actually made to the jury, it is reasonably probable that there would have been a different outcome if not for the deficient performance. As Justice Eveleigh has made perfectly clear, there is no such reasonable probability in the present case because, first, the jury reasonably could have found that the murder did not occur while the Skakel brothers were at the Terrien residence and, therefore, that the petitioner had only a partial alibi. Second, even if the jury found that the murder did take place during that time, it is not reasonably probable that Ossorio's testimony would have convinced the jury that the petitioner was at the Terrien home in light of the evidence that the petitioner confessed on at least three separate occasions that he committed the murder —evidence that, as I have indicated, the majority completely dismisses, even though it was unaffected by Ossorio's testimony at the habeas trial and was presumptively credited by the jury.

Indeed, although the majority now concludes that the petitioner's confessions to Gregory Coleman and John Higgins must be disregarded because "the treatment that the petitioner received at [the] Elan [School] as an adolescent was so brutal and coercive, and so directly related to his alleged involvement in [Martha's] murder, that the jury reasonably would question how that treatment affected the way the petitioner thought about the murder and how he responded to questions about it"; part V B 2 of the majority opinion; the author of the majority opinion previously has taken the position that "there was nothing inherently coercive about the particular circumstances surrounding the statements to indicate that they had not been given freely. In fact, in the case of each such statement, the defendant appears to have been confiding, voluntarily, in a fellow Elan resident. On appeal, we cannot assume that the atmosphere at Elan was so coercive that any incriminating statement by the defendant necessarily was the product of that coercive environment." (Emphasis added.) State v. Skakel , 276 Conn. 633, 723, 888 A.2d 985, cert. denied, 549 U.S. 1030, 127 S.Ct. 578, 166 L.Ed. 2d 428 (2006). Yet the majority now makes that very assumption. Apparently, in the majority's view, the petitioner's conviction could be upheld only if there were multiple confessions, plus eyewitnesses to the murder, plus DNA evidence, plus incriminating testimony by priests and other members of the clergy. Perhaps even that evidence would not suffice for the majority.

I would also emphasize that the majority's conclusion that no reasonable attorney would have failed to notice or follow up on the portion of the grand jury transcript in which Dowdle testified that she believed that she had heard her brother at the Terrien residence on the night of the murder, but was not sure that she saw him because she was in her mother's library with her "beau" and did not "venture out," is simply belied by the history of this case. The defendant's criminal trial, at which he was represented by Sherman, took place in 2002. Sherman, who had practiced criminal law for more than thirty years, both as a defense attorney and as a prosecutor, enlisted the help of at least three associate attorneys and consulted with, "among others, F. Lee Bailey, William F. Dow III, Richard Emanuel, David S. Golub, David T. Grudberg, and Barry Scheck." Skakel v. Commissioner of Correction , supra, 325 Conn. at 436 and n.6, 159 A.3d 109. None of these attorneys identified Dowdle's "beau" as someone who should be located and interviewed.

After his conviction, the petitioner filed a direct appeal, which was heard by this court on January 14, 2005. See State v. Skakel , supra, 276 Conn. at 633, 888 A.2d 985. The petitioner was represented on appeal by several highly experienced criminal and appellate attorneys, namely, Hope C. Seeley (now Superior Court Judge Seeley), Hubert J. Santos, Steven D. Ecker (now Supreme Court Justice Ecker), Patrick S. Bristol and Sandra L. Snaden. Those attorneys claimed that the petitioner's conviction should be reversed and that he was entitled to a new trial because: "(1) his case improperly was transferred from the docket for juvenile matters to the regular criminal docket of the Superior Court; (2) his prosecution was time barred by the five year statute of limitations for felonies that was in effect when [Martha] was murdered in 1975; (3) the state failed to disclose certain exculpatory evidence in violation of Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed. 2d 215 (1963), thereby depriving him of his right to a fair trial; (4) the state's attorney engaged in pervasive misconduct during closing argument in violation of the [petitioner's] right to a fair trial; (5) the trial court improperly permitted the state to introduce into evidence the prior sworn testimony of a certain witness in violation of the [petitioner's] constitutionally protected right of confrontation; and (6) the trial court improperly permitted the state to present evidence of several incriminating statements that the [petitioner] made while a resident at a school for troubled adolescents in Maine." State v. Skakel , supra, 639–40, 888 A.2d 985. The petitioner's attorneys also challenged the propriety of several other evidentiary rulings of the trial court. Id., at 640, 888 A.2d 985. Despite what the majority characterizes as the obvious "importance of the petitioner's alibi defense, the significance of [Ossorio's] testimony to that defense, the ease with which Ossorio could have been located, and the gravity of the charges and potential punishment that the petitioner faced"; part V A of the majority opinion; all of which, according to the majority, should have been immediately apparent to any reasonably effective attorney who read the transcript of Dowdle's testimony before the grand jury regarding the petitioner's alibi, the petitioner's appellate attorneys raised no claim that Sherman's performance had been deficient because he failed to properly investigate the petitioner's alibi defense. This court rejected the petitioner's claims and affirmed the judgment of conviction. Id., at 770, 888 A.2d 985.

In 2005, while the petitioner's direct appeal was still pending, the petitioner filed a petition for a new trial. See Skakel v. State , Docket No. CV-05-4006524-S, 2007 WL 3380139 (Conn. Super. October 25, 2007). In that proceeding, he was again represented by the law firm of Santos & Seeley, PC. See id., at *1. The revised petition for a new trial originally consisted of nine counts. Id., at *2. By the time of trial, however, the petitioner raised only four claims, specifically, a claim of newly discovered evidence of third-party culpability involving Gitano Bryant, Adolph Hasbrouck and Burton Tinsley, a claim of newly discovered evidence directly contradicting the testimony of Gregory Coleman at the petitioner's criminal trial, a claim that the state had engaged in a pattern of nondisclosure of exculpatory information involving, among other things, a sketch, profile reports of Thomas Skakel and Kenneth Littleton and certain time lapse date, and a claim concerning a "secret pact and book deal between the state's lead inspector, Frank Garr, and author Leonard Levitt." Id. The petitioner also raised allegations concerning Garr's threatening conduct toward witnesses. Id. Again, despite the fact that the petitioner's attorneys had obviously scoured the entire record for any information that would be helpful to the petitioner in postconviction proceedings, they raised no claim that Sherman's failure to follow up on Dowdle's grand jury testimony and to determine whether the "beau" had seen the petitioner on the night of the murder was ineffective assistance of counsel. The trial court denied the petition for a new trial; id., at *24 ; and the petitioner appealed to this court. See Skakel v. State , 295 Conn. 447, 991 A.2d 414 (2010). A majority of this court affirmed the judgment of the trial court. Id., at 522, 991 A.2d 414. In 2007, the petitioner filed an application for a writ of habeas in the United States District Court for the District of Connecticut, and that court granted his motion to amend the petition to add seven additional claims to the five asserted in the original petition, for a total of twelve separate claims involving alleged constitutional violations and newly discovered evidence. See Skakel v. Murphy , Docket No. 3:07cv1625 (PCD), 2009 WL 2253175, *1 (D. Conn. July 27, 2009). Yet again, the petitioner did not raise any claim pertaining to Sherman's failure to follow up on Dowdle's reference to her "beau." The petitioner's attorneys in that proceeding were Santos, Seeley and Snaden, whose surname was then Kuwaye. The District Court granted the petitioner's motion to stay the habeas proceeding to permit him to exhaust any unexhausted claims in state court, and that proceeding is apparently still pending. Id., at *3.

It was not until 2010, after the petitioner's experienced and highly respected attorneys had been searching high and low for any reason to cast doubt on the validity of the petitioner's conviction for eight years , that the petitioner filed the present petition for a writ of habeas corpus claiming that Sherman's failure to follow up on Dowdle's passing reference to her "beau" in her grand jury testimony constituted deficient performance. Yet, the majority insists that, upon reading Dowdle's testimony, any reasonably competent criminal defense attorney would have immediately recognized the purported critical importance of that testimony, attempted to ascertain the identity of the beau and attempted to interview him. If that were the case, it is inconceivable that the petitioner's attorneys—some of the most well-known and highly respected attorneys in the state, and even the country—would have failed to recognize the obvious and critical importance of the testimony for eight years. The majority also states "it could hardly have been easier for Sherman to have ascertained that Ossorio had critical alibi testimony to offer, such that even the most rudimentary of inquiries would have led Sherman directly and immediately to Ossorio." See part V A of the majority opinion. The majority is apparently relying on the habeas court's finding that, if Sherman had asked Dowdle about the beau, "he would have discovered Ossorio and gleaned that Ossorio was prepared to testify that the petitioner was present at the Terrien home during the evening in question." The majority cites absolutely no evidence, however, that would support this finding. The fact that Ossorio lived in the area during the criminal trial and that he would have been available to testify does not mean that Dowdle was aware of his whereabouts and availability. For all we know, Dowdle might not even have remembered his name. In fact, when Sherman was asked at the habeas trial whether he had asked Dowdle whether anyone else was with her on the night of the murder, he stated that he assumed that he had, and we have no idea how the petitioner's attorneys were ultimately able to discover Ossorio's identity or what efforts they were required to make to track him down.

Finally, I would note that, although the majority now claims that Sherman should have recognized immediately that the testimony of an unbiased alibi witness would be critically important to the petitioner's defense, Justice Palmer has previously characterized the alibi evidence that the petitioner presented at trial as strong and the state's theory that the alibi defense was concocted by the petitioner's family as barely plausible . Specifically, in his dissenting opinion in Skakel v. Commissioner of Correction , supra, 325 Conn. at 611, 159 A.3d 109, Justice Palmer claimed that "Sherman could have argued to the jury that this scenario [regarding Thomas Skakel's guilt] required no more speculation—indeed, I would argue that it required considerably less speculation—than the state's argument with respect to the petitioner, namely, that all of his alibi witnesses were lying ...." Similarly, in his dissenting opinion in Skakel v. State , supra, 295 Conn. at 687, 991 A.2d 414, Justice Palmer argued that:

(1) the new evidence on which the petitioner relied likely would result in an acquittal, "not only because the evidence adduced by the state against the petitioner at his criminal trial was not strong, but also because of the strength of the petitioner's alibi";

(2) "the state presented no credible evidence to support its theory of a cover-up"; id., at 688, 991 A.2d 414;

(3) "[a]lthough the state's attorney's argument concerning the allegedly concocted alibi was not so completely lacking in evidentiary support as to be improper, it is abundantly clear that his explanation concerning the manner in which the Skakel family allegedly manufactured the petitioner's alibi was extremely weak"; id., at 690, 991 A.2d 414 ;

(4) the state's theory that the alibi was concocted by the petitioner's family was "factually attenuated"; id. ;

(5) he was "fully persuaded ... that the evidence of a cover-up was sufficiently weak, and the strength of the petitioner's alibi sufficiently strong, that, if a jury were to reconsider the alibi evidence in the context of credible evidence that one or more other persons had the means, motive and opportunity to commit the murder, that jury likely would find the petitioner not guilty of the victim's murder"; id., at 693–94, 991 A.2d 414 ;

(6) the state's conspiracy theory was weak because "Greenwich police officers testified that they were given unfettered access to the Skakel children, as well as their home and property, in the hours following the murder, and for several months thereafter, and that the family was fully cooperative"; id., at 693, 991 A.2d 414 ; (7) the "[e]vidence to support [the state's theory that the petitioner's alibi was concocted by his family] is essentially nonexistent"; id., at 694, 991 A.2d 414 ; and

(8) the petitioner's alibi defense "was challenged by the state on grounds that find only marginal support in the record." Id., at 695, 991 A.2d 414.

The majority does not explain why it should have been obvious to Sherman that it was critically important to track down every possible lead to support an alibi defense that, according to the author of the majority opinion, was already strong and to rebut a theory supported by evidence that, also according to that author, was so weak and speculative as to be "essentially nonexistent." Id., at 694, 991 A.2d 414.

The majority also fails to recognize that, if the jury rejected what, according to Justice Palmer, was already a strong alibi defense, the jury also might have been skeptical of Ossorio's claim that, more than twenty-five years after Martha's murder, he remembered watching television with the petitioner that night, but he never saw any need during that period to reveal that fact to anyone, even after the petitioner was charged, despite the enormous publicity that the murder, the ensuing investigation and the charges against the petitioner generated. I recognize that Sherman acknowledged at the habeas trial that, in retrospect , testimony from an alibi witness who was unrelated to the petitioner would have been helpful. The question that this court must answer, however, is whether, as viewed at the time , the petitioner received effective assistance of counsel. See Siemon v. Stoughton , 184 Conn. 547, 555, 440 A.2d 210 (1981) ("[t]he issue, therefore, is not what counsel should have done to constitute the proper representation of the defendant considering the case in retrospect, but rather, whether in the circumstances, as viewed at the time, the defendant received effective assistance of counsel" [internal quotation marks omitted] ).

Let there be no mistake as to the answer to that question: the petitioner received a fair trial and the effective assistance of counsel to which he was constitutionally entitled. Because the majority's conclusion that he is entitled to a new trial is not based on the evidence or the law, an objective observer might conclude that it must be based on the majority's belief that wealthy, white, politically connected defendants, like the petitioner, are entitled to special treatment from the courts.⁴ Or is it based on the majority's desire to make a splashy statement in a high profile case? Perhaps the majority feels that, even if the petitioner killed Martha, he has been punished enough. After all, the petitioner served eleven years in prison, and, for a prisoner like him, that should be enough time. I would note, however, that in Connecticut the present maximum sentence for intentionally taking a life is life in prison (sixty years) with a mandatory minimum sentence of twenty-five years. See General Statutes §§ 53a–35a (2) and 53a–35b. Is it sympathy then? Admittedly, I really have no way of knowing, but the appearances speak loud and clear. I do know, however, that Martha, a lively and lovely fifteen year old girl with her entire life ahead of her, received no such sympathy at the hands of her vicious and heartless killer, and no such knight on a white charger intervened on her behalf. And where is the sympathy for Martha's mother, who has waited for more than forty years to see justice done? Indeed, perhaps the petitioner is betting that, after the passage of so much time, and the attendant death of witnesses, the deterioration of memories and the disappearance of evidence, the state will never retry him. Because the petitioner has received the justice to which he is entitled, and because I believe that no one is entitled to receive special treatment from the courts, for any reason, I dissent.