IN THE SUPREME COURT OF CALIFORNIA
In re KIMBERLY LOUISE LONG
on Habeas Corpus.
Fourth Appellate District, Division Two
Riverside County Superior Court
November 30, 2020
Justice Liu authored the opinion of the Court, in which Chief Justice Cantil-Sakauye and Justices Corrigan, Cuéllar, Kruger, Groban and Franson* concurred.
Michelle Rogers, under appointment by the Supreme Court; and Alissa Bjerkhoel for Petitioner Kimberly Louise Long.
Michael A. Hestrin, District Attorney, and Alan D. Tate, Deputy District Attorney, for Respondent the People.
Opinion of the Court by Liu, J.
Petitioner Kimberly Louise Long filed an original habeas corpus petition in this court seeking relief from a 2005 conviction for second degree murder of her boyfriend, Oswaldo Conde, for which Long was sentenced to prison for 15 years to life. In 2015, upon finding that Long’s petition established a prima facie case for relief, we issued an order to show cause before the Riverside County Superior Court as to “why trial counsel was not ineffective in his failure to: consult a time of death expert, investigate DNA evidence, present evidence petitioner did not change her clothes, and present evidence of the victim’s application for a restraining order, and why petitioner is not actually innocent of the crime as petitioner claims in [various] grounds.”
Long was convicted by a jury after an earlier jury trial had ended in a mistrial. In adjudicating Long’s habeas corpus petition, Judge Patrick Magers, who had presided over both of Long’s trials, held an evidentiary hearing over several days, adjudicated factual disputes, and ultimately found that Long’s trial counsel rendered objectively deficient performance that prejudiced Long’s case. The court granted Long’s petition, vacated her conviction, ordered a new trial, and released her on a $50,000 bond. The Court of Appeal reversed and reinstated her conviction. We now reverse the judgment of the Court of Appeal on the ground that trial counsel’s failure to investigate the victim’s time of death was objectively unreasonable and prejudicial to Long’s defense.
At the close of Long’s first trial, nine jurors favored acquittal and three favored guilt, and the court declared a mistrial. Long’s second trial took place in December 2005 and resulted in her conviction. The following facts were shown at the second trial.
On October 5, 2003, Long spent the day with Conde, their friend Jeffrey Dills, and others riding their motorcycles and drinking at various bars. Long had approximately 12 beers and 10 shots of hard liquor that day. The others had also been drinking. At some point, an argument between Long and Conde escalated into a physical altercation when the two of them, along with Dills, returned to the couple’s home that evening in Corona. According to Dills, Long accused Conde of “not paying his share and being a loser and not having a job.”
Long and Dills left sometime between 11:00 p.m. and midnight and went to Dills’s home about 2.5 miles away. There, Dills and Long spent time in a jacuzzi, then moved to the bedroom and had oral sex. At some point, Long told Dills she had to return home because her ex-husband was supposed to drop off her child.
Dills informed police that he dropped Long off at her home around 1:20 or 1:25 a.m. Dills then returned to his own home; he recalled seeing his alarm clock by his bedside at 1:36 a.m. Long disputed this, testifying that she was dropped off by Dills around 2:00 a.m.
At 2:09 a.m., Long called 911 and said: “Oh my god something happened to my husband. . . . I just came home. He’s bloody. I don’t know what’s going on. He’s still breathing. Something’s wrong.” She hung up and then called 911 again. During the second call, Long said she was an emergency room nurse but added, “I can’t give him medical attention. Something’s wrong with him.”
Police officers Jeffrey Glenn and Edward Hurtado were dispatched to the scene at 2:10 a.m. They arrived three to four minutes later and found Long waiting in the middle of the street, distraught. When they entered the home, they found Conde slumped over the couch with his feet on the floor, and they saw blood on the walls. Hurtado checked Conde’s body for a pulse and found none. When Hurtado touched Conde’s neck, a blood bubble burst in Conde’s mouth.
Paramedics arrived at 2:20 a.m. They confirmed Conde had no pulse and observed evidence of trauma to the right and back of Conde’s head. The paramedics noted that Conde’s wound was not actively bleeding and that his blood had already coagulated. They described Conde’s skin as “pale or ashen” and “cold” to the touch. Lividity, or skin discoloration resulting from internal pooling of blood, was present on the back of Conde’s arms and the left side of his face. They also observed rigidity, or rigor mortis, in Conde’s arms.
At 5:03 a.m., Deputy Coroner Richard Gomes arrived. Gomes inspected Conde’s body and noted in his report: “Rigor had not started. Lividity was almost fixed, with medium discoloration, and consistent with his position.” On the morning of October 7, Dr. Joseph Pestaner performed an autopsy and noted in his report that “[r]igor mortis is mild to moderate and symmetric. Livor mortis is fixed on the posterior aspect of the body.” The coroner ultimately determined the cause of death was blunt force injuries to the head. Based on the autopsy report and photographs, a pathologist testified that Conde had been hit with a blunt weapon three to eight times and that an injury of that nature would have resulted in Conde’s death within 10 to 15 minutes.
A forensic technician observed blood evidence 360 degrees around Conde’s body, including castoff on all four walls. Police noticed broken glass in the kitchen and saw that the sliding glass door from the kitchen to the back yard was open. A shotgun and shotgun shells were missing from a closet, as were a bowl of change and a stereo from the living room. Investigators found no evidence of an attempt to clean up the house, no sign of the murder weapon, no blood anywhere else in the house, and nothing to suggest the sinks or showers had been recently used. Officers searched the area, but they did not recover the murder weapon, any bloody clothing, or other evidence linked to the crime.
Corona police officers interviewed Long at the station. According to Long, she stayed at Dills’s house for “an hour and a half, two hours” but could not recall exactly when she arrived home. When she arrived home, she noticed that the front door was unlocked. She walked into the home and saw Conde was on the couch in the living room. Given the blood, she initially thought he had gotten into a fight. Long said that Conde was “gurgling,” so she believed he was still alive and breathing. Only after he remained inert did Long realize something was wrong. She turned on the lights and saw he was gravely injured. She told the police that she called 911 less than 10 minutes after arriving home, ran around the house screaming, hung up on 911, ran outside, and called 911 again. In a second interview on October 9, Long told officers that she suspected Conde’s ex-girlfriend Shiana Lovejoy murdered Conde because Lovejoy had repeatedly threatened Long and Conde.
The victim’s brother told law enforcement that on the Monday before Conde was killed, Lovejoy had threatened to “slice their throats” (referring to Long and Conde) and that Lovejoy’s behavior had gotten “pretty bad” in the weeks before Conde’s murder. Two weeks before Conde’s death, Lovejoy sent a letter to Long claiming that she (Lovejoy) was still romantically involved with Conde and calling Long a “whore” and Conde a “broke mother fucker.” Lovejoy repeatedly called Long and Conde and left harassing voicemails. Long and Conde reported these calls to the police and changed their phone number in response. Lovejoy also defaced Conde’s white truck with the phrase “asshole & deadbeat” while it was parked at Long’s house and put glue in the keyhole of his truck so he could not open the door.
A week before his death, Conde sought a restraining order against Lovejoy and orders regarding custody and visitation of their son. In the application, Conde wrote that Lovejoy “called and said that my girlfriend [Long] is going to get it and for me to watch my back,” and that Lovejoy “hates my girlfriend, and she is going to ruin our lives.” His request for a temporary restraining order was denied, but the court set a hearing on Conde’s permanent restraining order application for October 20. On the same day Conde sought a restraining order against Lovejoy, Lovejoy filed a paternity suit against Conde. A few days before Conde’s murder, Lovejoy filed a competing child custody application. After Conde’s death, Lovejoy received Social Security payments on behalf of their son.
Lovejoy told police that on the night of Conde’s murder, she had dinner at T.G.I. Friday’s with Oscar Castaneda, whom she was dating at the time, and then went to the Days Inn in Whittier where they paid cash for a room and stayed until around 1:30 a.m. Police confirmed Castaneda’s credit card was used at the restaurant at 10:24 p.m. The Days Inn manager on duty that night recognized a photo of Castaneda but had no recollection of Castaneda at the motel that evening, and there was no computer record of their stay. Lovejoy’s mother was babysitting that night and told officers that Lovejoy arrived home in Anaheim around 1:30 a.m.
In a police interview, Lovejoy said she had a great relationship with Conde until the last six months, when he moved in with Long. Lovejoy admitted that she lied to the police when they visited her house to investigate the vandalism of Conde’s truck.
Long’s ex-husband, Joe Bugarski, was also investigated as a potential suspect. Long and Bugarski married in 1999 or 2000 and had a son. Bugarski described two incidents in which he said Long was physically violent with him. According to a police report, the second of those incidents, in June 2000, resulted in Bugarski’s arrest for domestic violence. During that incident, Bugarski held a butter knife to Long’s throat and threatened to kill her. In March 2003, Long kicked Bugarski out of the house and ended their marriage. On occasion, Long allowed Bugarski to stay at her home to watch their son when Long worked the graveyard shift. Bugarski installed a spy camera in Long’s bedroom and a voice-activated tape recorder under her bed because he suspected Long was intimately involved with Conde. Bugarski threatened to “kick [Conde’s] ass” and told Long’s mother he wanted to kill Conde. He occasionally stalked Long “around town,” and about a month before Conde’s murder, he spied on Long and Conde at their home, which prompted a neighbor to call 911.
On the night of October 5, Bugarski was supposed to visit Long’s home around 9:00 p.m. to drop off their son, but he left a message on Long’s phone at 7:30 p.m. saying their son did not want to go and would stay with Bugarski that night. Long called back between 8:00 to 9:00 p.m. and told Bugarski that was fine because she could not be home that night. Bugarski said he and his son stayed at the home of Bugarski’s girlfriend, Chelsea Murray, that night and fell asleep on Murray’s bunkbed between 8:00 and 9:00 p.m. Murray told police she fell asleep around 10:00 p.m. and awoke at 5:00 a.m. to Bugarski’s cell phone vibrating.
On October 22, police officers interviewed Alejandro and Juanita Sandoval, who lived next door to Long and Conde, and Phillip Virga, who lived across the street. Alejandro remembered waking up around 12:00 a.m. to the sound of a motorcycle engine revving multiple times from inside his neighbor’s garage. He told police he did not hear any voices at that time. Alejandro went back to sleep and woke up again briefly when he heard Long’s car alarm around 2:00 a.m.
Juanita believed she heard Conde attempting to start his motorcycle inside his garage at 1:30 a.m. The motorcycle stalled multiple times, and she recalled hearing someone yelling “fuck” and “shit” loudly at the motorcycle. Based on the engine’s noises, she believed Conde left home on his bike and returned around 1:45 a.m. Around 2:00 a.m., Juanita also heard Long’s car alarm for “a little while.”
Virga told officers that he went out for dinner with his family and spotted two motorcycles parked in front of Long and Conde’s house around 8:00 p.m. on October 5. By the time Virga returned at 9:30 or 9:45 p.m., the motorcycles were gone. Virga was awakened around 11:50 p.m. to the sound of a motorcycle’s roaring engine from inside the couple’s garage. He thought he heard a muffled female’s voice inside the garage. Virga fell asleep and woke up again around 12:10 a.m. to the sound of a motorcycle engine and a female voice. Around 1:20 a.m., he heard the sound of a motorcycle driving past. Virga woke up again around 2:00 a.m. when he heard Long’s car alarm go off.
Dills died unexpectedly before trial, so his preliminary hearing testimony was admitted. Dills testified that he dropped off Long at her home at 1:20 to 1:25 a.m. and drove back to his own home by 1:36 a.m. on the night of Conde’s murder. Because Dills was the only person who saw Long after midnight, the prosecution at trial used his preliminary hearing testimony to establish a timeline for the crime, with Long arriving home around 1:20 a.m. and mortally wounding Conde some time before her call to 911 at 2:09 a.m.
During closing argument, the prosecutor argued that Long arrived home intoxicated around 1:20 a.m., killed Conde, disposed of the murder weapon, and cleaned herself and the scene before calling 911 at 2:09 a.m. No physical or other direct evidence connected Long to Conde’s killing. The jury found Long guilty of second degree murder. On February 24, 2006, Long was sentenced to a prison term of 15 years to life.
Long moved for a new trial. (Pen. Code, § 1181.) The trial court “reluctantly” denied Long’s motion, saying: “To make a perfectly clear record in this matter, if this was a court trial, if the Court would have heard the evidence in this case, I would have found the defendant not guilty. I would have found that the evidence was insufficient to prove beyond a reasonable doubt.” In particular, the court found “troubling” that Dills’s preliminary hearing testimony featured so prominently in the prosecution’s case, despite the fact that the cross-examination of Dills “during the preliminary hearing was, as in most preliminary hearings, not extensive.” In particular, the court noted that “his level of intoxication was not fully addressed in the examination.” The Court of Appeal upheld Long’s conviction and denied her petition for rehearing. This court denied review.
In February 2010, Long filed a petition for a writ of habeas corpus in federal court. Reviewing Long’s conviction under the standards set forth in the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA; 28 U.S.C. § 2254), the district court denied her petition but noted it was “unfortunate that [Long’s] conviction largely hinged on [Dills’s] preliminary hearing testimony.”
Long appealed, claiming insufficient evidence supported her murder conviction. The United States Court of Appeals for the Ninth Circuit affirmed, explaining that “[w]ere we the jury, we might have entertained a reasonable doubt. Were we sitting as the reviewing court on direct appeal, we might have found the evidence to be insufficient. But under AEDPA, which demands double deference, we are limited to deciding whether the California courts unreasonably applied Jackson [v. Virginia (1979) 443 U.S. 307]. They did not.” (Long v. Johnson (9th Cir. 2013) 736 F.3d 891, 897.) One judge said, “I have grave doubts about whether the State has convicted the right person in this case.” (Ibid. (conc. opn. of Watford, J.).)
Long then sought habeas corpus relief in state court on the grounds of ineffective assistance of counsel and actual innocence. (Cal. Const., art. VI, § 10.) After the trial court and Court of Appeal summarily denied Long’s petitions, this court found that Long had stated a prima facie case for relief and issued an order to show cause why relief should not be granted.
The trial court took judicial notice of the prior proceedings in Long’s case and conducted an evidentiary hearing over several days. The court examined newly presented evidence and heard witness testimony proffered by both parties. Following the hearing, the court found by a preponderance of the evidence that Long’s trial counsel, Deputy Public Defender Eric Keen, had rendered objectively deficient performance by failing to consult and present testimony from a time of death expert and by failing to present supporting evidence about Long’s clothing. The court granted Long’s petition, vacated her conviction, set bail, and ordered a retrial.
The Court of Appeal reversed, finding that “Keen’s representation of defendant did not fall below an objective standard of reasonableness.” (People v. Long (May 3, 2018, E066388) [nonpub. opn.].) Because it found no deficient performance, the court did not examine prejudice. We granted review.
The Sixth Amendment to the United States Constitution and article I, section 15 of the California Constitution guarantee a criminal defendant the “ ‘right to the effective assistance of counsel at trial.’ ” (In re Lucas (2004) 33 Cal.4th 682, 721; see ibid. [a defendant is “ ‘ “entitled to the reasonably competent assistance of an attorney acting as his diligent and conscientious advocate” ’ ”]; Strickland v. Washington (1984) 466 U.S. 668, 687 (Strickland).) “The ultimate purpose of this right is to protect the defendant’s fundamental right to a trial that is both fair in its conduct and reliable in its result.” (People v. Ledesma (1987) 43 Cal.3d 171, 215 (Ledesma).)
We focus here on Long’s allegation that her trial counsel rendered ineffective assistance by failing to consult a time of death expert. To succeed, Long must show that this omission “fell below an objective standard of reasonableness” (Strickland, supra, 466 U.S. at p. 688; see In re Thomas (2006) 37 Cal.4th 1249, 1257) in light of “the professional norms prevailing when the representation took place” (Bobby v. Van Hook (2009) 558 U.S. 4, 7). Long must also show “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” (Strickland, at p. 694.) Long “need not show that counsel’s deficient conduct more likely than not altered the outcome in the case.” (Id. at p. 693.) It is enough to show “a probability sufficient to undermine confidence in the outcome.” (Id. at p. 694.)
“[T]he standard for judging counsel’s representation is a most deferential one.” (Harrington v. Richter (2011) 562 U.S. 86, 105 (Harrington).) We “must indulge a ‘strong presumption’ that counsel’s conduct falls within the wide range of reasonable professional assistance because it is all too easy to conclude that a particular act or omission of counsel was unreasonable in the harsh light of hindsight.” (Bell v. Cone (2002) 535 U.S. 685, 702.) “Unlike a later reviewing court, the attorney observed the relevant proceedings, knew of materials outside the record, and interacted with the client, with opposing counsel, and with the judge.” (Harrington, at p. 105.) Accordingly, we must “reconstruct the circumstances of counsel’s challenged conduct, and . . . evaluate the conduct from counsel’s perspective at the time.” (Strickland, supra, 466 U.S. at p. 689.)
Long prevailed on her habeas petition following an evidentiary hearing with several findings in her favor. In reviewing the trial court’s order granting relief, we apply “well settled” standards of review: “[W]e give great weight to those of the [trial court]’s findings that are supported by substantial evidence. This is especially true for findings involving credibility determinations. The central reason for referring a habeas corpus claim for an evidentiary hearing is to obtain credibility determinations; consequently, we give special deference to the [trial court] on factual questions ‘requiring resolution of testimonial conflicts and assessment of witnesses’ credibility, because the [trial court] has the opportunity to observe the witnesses’ demeanor and manner of testifying.’ [¶] . . . [I]n other areas we give no deference to the [trial court]’s findings. We independently review prior testimony, as well as all mixed questions of fact and law. Whether counsel’s performance was deficient, and whether any deficiency prejudiced the petitioner, are both mixed questions subject to independent review. Ultimately, the [trial court]’s findings are not binding on us; it is for this court to make the findings on which the resolution of [petitioner’s] habeas corpus claim will turn.” (In re Thomas, supra, 37 Cal.4th at pp. 1256–1257, citations omitted.)
The crux of Long’s claim is that Keen, her trial counsel, failed to sufficiently investigate Conde’s time of death. At the evidentiary hearing, Keen testified that the only expert he consulted on time of death was Daniel Vomhof, an accident reconstruction expert with no medical degree or experience in pathology. Keen had contacted Vomhof in an effort to establish that Long was not physically capable of causing Conde’s injuries. During that consultation, Keen learned that Vomhof “was also a time-of-death expert or purported to be.” Although Keen said he “spoke to [Vomhof] about that issue,” he could not recall if Vomhof had seen the paramedics’ report or coroner’s report, nor could he “remember specifically what each of us said.” Keen confirmed that he “didn’t contact an actual pathologist” or “anyone else besides Dr. Vomhof” to discuss time of death. Defense investigator William Sylvester confirmed that Keen never instructed him to seek out a time of death expert in Long’s case.
Keen further testified that his view at the time of Long’s trials was that no medical expert could offer a time of death that excluded the prosecution’s theory of Long’s window of opportunity for committing the crime. Keen based this opinion on “what Dr. Vomhof told [him]” and on his “own experience,” which “at the time” consisted of “nothing more than [mandatory continuing legal education] for attorneys, homicide seminars, stuff of that nature.” Keen said, “It’s my belief that anybody who testified regarding time of death would have to give a range, and that the times we were dealing with . . . were too small, and then necessarily that range would encompass both theories, the prosecution’s theory and our theory.” Keen testified that if he were trying the case today, he would “seek the opinion of a pathologist, somebody who had performed on the order to thousands of autopsies over their career to render testimony.”
The defense’s Strickland expert, Gary Gibson, provided additional context during the evidentiary hearing. Gibson was a law professor at California Western School of Law, teaching California evidence, forensic evidence, advanced criminal litigation, and California sentencing. He had also been a public defender for over 20 years in San Diego County and worked on 200 to 300 homicide cases. He testified that Vomhof’s résumé indicated he took a continuing education course on time of death in 1979 or 1980 but had no other training or expertise on time of death. Gibson said he would never seek a time of death opinion from “somebody who is not only not an M.D., but is [also] not a medical examiner.” He considered Keen’s reliance on Vomhof unreasonable because Vomhof is a biochemical engineer and not a medical doctor with expert knowledge of how to estimate time of death.
The trial court agreed, finding that Vomhof was “not a qualified expert to render an opinion regarding time of death” and that Keen made “no effort to contact or secure the testimony of such” an expert. The trial court reviewed Vomhof’s résumé and noted that he “holds a Ph.D. in biochemistry” but “is not a forensic pathologist or medical examiner” and “does not have a medical degree.” The court found that although Vomhof is an expert in biomechanics and accident reconstruction, “nothing” in Vomhof’s résumé showed he was qualified to provide an expert opinion on time of death.
In general, estimating time of death requires expert knowledge on how to measure and evaluate relevant postmortem bodily processes and indicators such as lividity, rigor mortis, body temperature, and decomposition. (See Henßge et al., Death Time Estimation in Case Work. II. Integration of Different Methods (1988) 39 Forensic Sci. Intl. 77, 77–78 (hereafter Henßge); In re Richards (2016) 63 Cal.4th 291, 303 [citing expert testimony that “time of death is a complex determination”].) In addition to Vomhof’s lack of medical training or expertise to evaluate such indicators, Keen could not recall — and there is no indication in the record — that Vomhof was provided with the paramedics’ report, the coroner’s report, or other information necessary to reach an informed opinion as to Conde’s time of death. The trial court’s finding that Vomhof was not qualified to provide an expert opinion on time of death is supported by substantial evidence, and we adopt it.
The further question, to which we apply our independent judgment, is whether Keen’s failure to consult a time of death expert fell below an objective standard of reasonableness. In addressing this question, we focus not on whether Keen “should have presented” expert testimony on Conde’s time of death, but on “whether the investigation supporting [counsel’s] decision not to [consult a time of death expert] was itself reasonable.” (Wiggins v. Smith (2003) 539 U.S. 510, 523 (Wiggins).)
Not every homicide case presents a significant issue regarding the time of death. Gibson testified that he recalled only two instances in which time of death was at issue among the hundreds of murder cases he worked on as a public defender. But Conde’s time of death was a particularly important issue in this case and therefore gave rise to a duty of defense counsel to investigate. The core of the prosecution’s theory regarding Long’s opportunity to commit the crime rested on Dills’s preliminary hearing testimony that he dropped off Long at her home between 1:20 and 1:25 a.m. Long repeatedly insisted she arrived home around 2:00 a.m., placing her outside the window of opportunity to murder Conde, clean herself up, and dispose of the evidence. A reasonably competent defense attorney would have recognized the importance of investigating Conde’s time of death to examine whether it was inconsistent with the prosecution’s timeline and could cause the jury to have a reasonable doubt about Long’s guilt.
We acknowledge that time of death often cannot be estimated with precision and that reliable estimates are commonly stated as a time range. Even so, Keen’s inquiry on this issue “did not reflect reasonable professional judgment.” (Wiggins, supra, 539 U.S. at p. 534.) Although Keen discussed Conde’s time of death with Vomhof, the discussion was serendipitous. Keen had contacted Vomhof to explore the level of force needed to fatally injure Conde; he did not contact Vomhof to discuss time of death, nor was he aware that Vomhof purported to have time of death expertise, nor is there any indication that he gave Vomhof the paramedics’ report, coroner’s report, or other information necessary to form a reliable opinion as to time of death. Under these circumstances, a competent attorney would have sought the opinion of a time of death expert on what the observations of lividity and rigor mortis in the paramedics’ and coroner’s reports implied for Conde’s time of death. At the time of Long’s trials, such observations had long been recognized in forensics literature and case law as indicators capable of informing time of death estimates. (See, e.g., Henßge, supra, 39 Forensic Sci. Intl. at pp. 78–81; People v. Noguera (1992) 4 Cal.4th 599, 614 [“routine examinations for lividity and rigor mortis” are “two crude measures used to approximate time of death”]; People v. Watson (1956) 46 Cal.2d 818, 825 [medical experts estimated time of death based on rigor mortis]; People v. Adkins (2002) 103 Cal.App.4th 942, 948 [forensic pathologist relied on observations of rigor mortis and lividity in estimating time of death]; People v. Culuko (2000) 78 Cal.App.4th 307, 320 [pathologist estimated time of death based on rigor mortis].)
Keen testified that he did not seek an expert opinion on time of death because he believed any estimate would encompass a range too broad to exclude the prosecution’s timeframe for the crime. But there is scant evidence to indicate he sought expert advice to confirm that belief. Even accepting that time of death estimates are inexact and are stated as a range, it is not clear what led Keen to believe that any range derived from the available evidence would necessarily encompass the prosecution’s timeframe — a timeframe that tended to rule out the possibility that someone other than Long committed the crime. When asked what “specifically” in his training led him not to inquire further into Conde’s time of death, Keen identified nothing specific. Although Keen said he had knowledge about lividity and rigidity, his testimony provides no indication as to whether or how that knowledge informed his judgment. Without further inquiry, Keen had no basis to write off the possibility that a time of death estimate would help exculpate Long.
The Court of Appeal believed it was reasonable that Keen did not consult a time of death expert and instead focused Long’s defense on third party culpability. But the two lines of defense were potentially complementary, not mutually exclusive; an expert opinion that Conde likely died before 1:20 a.m. would have tended to bolster Long’s defense that another party committed the crime. In any event, we see no reason why competent counsel would not have at least inquired. (See Strickland, supra, 466 U.S. at p. 691 [“counsel has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary”]; Ledesma, supra, 43 Cal.3d at p. 215 [counsel must “make a rational and informed decision on strategy and tactics founded on adequate investigation and preparation”].) Without such investigation, Keen was not in a position to make a reasonable decision as to which defense or defenses to focus on at trial.
A reviewing court must apply “ ‘a heavy measure of deference to counsel’s judgments’ ” in assessing the reasonableness of a particular decision not to investigate. (Wiggins, supra, 539 U.S. at p. 522.) But “[i]n assessing the reasonableness of an attorney’s investigation, [we] 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.” (Id. at p. 527.) The prosecution’s theory of the case put Long at the crime scene no earlier than 1:20 a.m., roughly 45 minutes before the police arrived. Time of death was not only a central issue to investigate; it was also a viable issue for investigation in light of the lividity and rigidity observations by the paramedics and coroner. A reasonably competent attorney would have investigated further, and Keen had no reasonable basis for failing to do so. (See id. at p. 534.) We conclude that counsel’s failure to consult a time of death expert “was the result of inattention, not reasoned strategic judgment” (ibid.), and that counsel’s performance in this regard fell below an objective standard of reasonableness.
We now consider whether counsel’s failure prejudiced Long’s defense. We examine what “evidence counsel failed to discover and present in this case” and whether there is “a reasonable probability that a competent attorney . . . would have introduced it” (Wiggins, supra, 539 U.S. at pp. 534–535), and then we address whether “there is a reasonable probability that [the jury] would have returned” a different verdict if it “had . . . been confronted with [that] evidence” (id. at p. 536).
The trial court found that “qualified medical opinions” concluding that Conde’s death occurred significantly earlier than 1:20 a.m. “were available at the time of trial” in 2005. This finding is supported by substantial evidence.
At the evidentiary hearing, Long presented the testimony of two pathology experts. Dr. Zhongxue Hua was a professor of pathology at the Albert Einstein College of Medicine and previously served as chief medical examiner for Union County in New Jersey. He testified that he has performed around 3,000 autopsies and has reviewed “at least the same amount” of autopsy reports. He explained that the science undergirding time of death estimation is not precise down to minutes or seconds, but it can provide a meaningful range of possible times.
Upon reviewing the circumstances in this case, Dr. Hua opined that it was “remotely possible” but “extremely unlikely” that Conde died after 1:20 a.m. Based on “the known set of evidence,” Dr. Hua concluded that Conde’s “time of death occurred long before 1:20 a.m.” In providing this estimate, Dr. Hua assigned particular importance to the paramedics’ observation of rigor in Conde’s arm around 2:20 a.m. He explained that although rigor may present as early as 30 minutes to an hour after death in small muscles such as those in a person’s fingers or jaw, it would take “much, much later than 30 minutes” after a person’s death for rigor to present in the medium-sized muscles of the arms. Dr. Hua also considered the paramedics’ observations regarding the body’s lividity. He noted that some lividity could present as early as 30 minutes to one hour following a person’s death, suggesting that Conde must have died at least 30 minutes before the paramedics’ observation.
Dr. Hua placed less emphasis on the coroner’s report. As noted, the coroner arrived almost three hours after the paramedics; Dr. Hua said he generally finds earlier observations to be more helpful than subsequent observations. He testified that the coroner’s observation of “almost fixed” lividity was inconsistent with the coroner’s notation that rigor “had not started,” since lividity and rigor tend to present together. According to Dr. Hua, the coroner’s notation that rigor had not started was “impossible” as Conde had been dead for over two hours. Dr. Hua testified that any finding that rigor “had not started” by 5:03 a.m., when the deputy coroner arrived, was “extremely unlikely.” Further, Dr. Hua testified that there is no precise definition of “almost fixed” lividity, so he discounted fully fixed lividity by 50 percent in order to derive his “most conservative” time of death estimate. He opined that the earliest time when “almost fixed” lividity (so understood) could present is “minimum, minimum four hour[s]” after death, and he rejected the possibility that “almost fixed” lividity could occur within four hours after death. On cross-examination, Dr. Hua testified that even if the coroner had examined the body closer to 7:00 a.m., his time of death estimate would be “still before” 1:20 a.m. given the paramedics’ observation of rigor in Conde’s arm at 2:20 a.m.
Long’s other expert was Dr. James Bonnell, who had previously served as the chief deputy medical examiner for San Diego County and had performed over 7,000 autopsies and had provided sworn testimony over 585 times at the time of the hearing. Dr. Bonnell estimated that Conde died “much closer to 11:00 [p.m.] than 1 [a.m.].” He placed “great reliability” on the paramedics’ observations, as they “were the first people to actually examine the body.” Dr. Bonnell considered but did not credit the coroner’s findings. In his view, the coroner’s observation that rigor “had not started” was “inaccurate.” Dr. Bonnell highlighted the paramedics’ contrary observation that rigor was present in Conde’s arms at 2:20 a.m. According to Dr. Bonnell, if “rigidity was not present” during the coroner’s examination between 5:03 and 7:13 a.m., it “can equally mean that [rigor] has passed,” not that rigor had not yet started. If Conde had died before midnight, Dr. Bonnell noted, “it wouldn’t surprise [him] that [rigor] has passed at 7 o’clock.” Like Dr. Hua, Dr. Bonnell testified that the presence of “almost fixed” lividity at the time of the coroner’s examination “totally contradicts” the coroner’s finding of absence of rigor. Especially in light of the paramedics’ observations, Dr. Bonnell characterized the prosecutor’s theory that Conde died at or after 1:20 a.m. as “medically impossible.”
Dr. Bonnell also considered the testimony of a neighbor who thought she heard Conde in his garage at 12:30 a.m. He testified that given the strength of the evidence suggesting an earlier time of death, “I don’t place a whole lot of reliability on” the indicia that Conde was alive as late as 12:30 a.m. On cross-examination, Dr. Bonnell said he “wouldn’t rely . . . at all” on Long’s background as an emergency room nurse or her 911 call transcript in which she said Conde was “still breathing” because, “number one, she may have been in shock, and number two, I know she was intoxicated.”
The district attorney called Dr. Joseph Cohen, who was the chief forensic pathologist of the Riverside County Sheriff-Coroner’s Office and had performed over 7,000 autopsies. Dr. Cohen disagreed with Dr. Hua’s assessment of the paramedics’ observation of rigor in Conde’s arms. He testified that rigor may be observable “within minutes” of a person’s death, although he acknowledged that it generally takes “30 to 60 minutes” for rigor to appear. Dr. Cohen said he placed little weight on the paramedics’ report, as “the deputy coroner’s examination generally will trump the first responder’s impression — not always.” Even so, Dr. Cohen largely dismissed the observations of rigidity and lividity in the coroner’s report. Dr. Cohen doubted that rigor could be present at 2:20 a.m. but not at 6:00 a.m., saying that such a finding “doesn’t make sense.” He also “disagree[d] that [lividity] would have been fixed or nearly fixed at the time of the coroner investigator’s examination,” describing the coroner’s finding of “almost fixed” lividity as “hard to believe” and an “impossibility.”
In estimating Conde’s time of death, Dr. Cohen said he would give “appropriate weight” to the testimony of Juanita Sandoval, who said she heard Conde attempting to start his motorcycle in his garage at 1:30 a.m., and to the 911 call in which Long said Conde was still breathing. Dr. Cohen also disagreed with Dr. Hua’s and Dr. Bonnell’s time of death estimates, describing them as “ludicrous,” “skewed,” “too inflexible,” and “too narrow of a range.” Ultimately, Dr. Cohen opined that Conde’s time of death was “as likely” before 1:20 a.m. as after.
After hearing the testimony of all three experts, the trial court found that Dr. Hua and Dr. Bonnell “concluded to a reasonable degree of medical certainty that the postmortem changes observed in the victim’s body could not have occurred in less than an hour. Neither forensic pathologist could give an exact timing of the victim’s death. However, both forensic pathologists testified in this court that the victim’s death occurred significantly earlier than 1:20 a.m., the earliest time the prosecution could place the petitioner at the scene. Their observations were based upon postmortem changes by the first responders as well as the deputy coroner’s report . . . . [¶] Hearing their testimony, this Court finds both opinions to be credible, convincing, and compelling. Their testimony indicates such qualified medical opinions were available at the time of trial and defense counsel failed to seek out medical experts to address the issue.” These findings, reflecting the trial court’s determination of the witnesses’ credibility, are supported by substantial evidence.
The trial court further ruled that the testimony of a qualified time-of-death expert was evidence that a competent attorney would have presented at trial: “In making this ruling, I’m not saying that he should have contacted these two particular experts, but it’s apparent to the Court that these qualified opinions did exist in the medical field . . . . [¶] If such expert would have testified, it would have put the victim’s time of death at a time when petitioner could not have committed the crime, if believed by the jury. Obviously, it’s always a question of fact for the jury to either accept or reject the testimony of any witness that testifies, including an expert.”
The Court of Appeal “disagree[d] with the trial court’s conclusion that no reasonable attorney would have failed to present expert time of death evidence to the jury.” The Court of Appeal observed that “Hua estimated the victim died before 1:20 a.m., but conceded that estimate could be effected [sic] by the timing of when Gomes [the coroner] observed the victim’s body; the record fails to show at precisely what time Gomes examined the victim’s body.” The court also said Dr. Bonnell’s estimate that Conde died between 11:00 a.m. and 12:30 p.m. “was based upon not considering, or disregarding, all evidence of a later time period, such as [a neighbor’s] statement that she heard the victim at 1:30 a.m., defendant’s 2:09 a.m. observation that the victim was breathing, [the coroner’s] conclusion that rigor had not started by 5:03 a.m. on October 6, and Dr. Pestaner’s observation that the victim’s body was in a state of rigor on October 7.” Further, the court explained that Dr. Cohen’s inconclusive testimony “supports Keen’s reasoning” that a time of death estimate would be too broad to exclude the prosecution’s theory. In sum, the court said, “an attorney could reasonably conclude that reliance upon expert testimony, such as Bonnell’s, would have been a risky defense strategy because a jury could reasonably view such testimony with skepticism and therefore it would be unlikely to raise a reasonable doubt.”
In evaluating these competing views, we begin by noting that the Court of Appeal did not disagree with the trial court’s finding that “qualified medical opinions [that Conde died before 1:20 a.m.] were available at the time of trial.” The Court of Appeal’s concern was that Dr. Hua’s and Dr. Bonnell’s opinions were not clear or not persuasive. However, the Court of Appeal appeared to ignore Dr. Hua’s testimony that he would place Conde’s time of death earlier than 1:20 a.m. even if the deputy coroner had not inspected the body until 7:00 a.m. And the court did not mention Dr. Bonnell’s testimony that he did not give weight to the coroner’s observation that rigor had not started because it “totally contradicts” the other time of death indicators, or to Long’s 2:09 a.m. observation that Conde was breathing in light of her shock and level of intoxication.
More fundamentally, all the experts who testified were aware that the various pieces of evidence bearing on Conde’s time of death were potentially inconsistent, and they took those inconsistencies into account. In their responses to extensive questioning, Dr. Hua and Dr. Bonnell explained why they placed greater reliance on some indicators and not others, and both experts placed significant weight on the observations of the paramedics, the first responders on the scene. (Cf. In re Richards, supra, 63 Cal.4th at p. 303 [noting forensic pathologist’s testimony that “the closer to the time of death observations are made, the more accurate the findings will be”].) Notably, all three experts were critical of key findings in the coroner’s report. Further, the trial court acknowledged that Dr. Cohen “is a highly qualified medical examiner” and that Dr. Cohen “disagreed with the expert testimony of Dr. Bonnell and Dr. Hua and opined the time of death was just as likely . . . before 1:20 a.m. as after.” Nevertheless, the trial court, after hearing all three experts’ lengthy testimony over several days, found Dr. Hua’s and Dr. Bonnell’s opinions “to be credible, convincing, and compelling.” This is not to say that contrary time-of-death opinions were not also available, as evidenced by Dr. Cohen’s testimony. But our assessment of prejudice focuses on the availability of credible opinions favorable to Long because, in light of our determination that counsel did not properly investigate the victim’s time of death, we cannot rule out a reasonable likelihood that had he done so, he would have discovered an opinion helpful to Long’s defense.
The Court of Appeal was correct to focus its inquiry not simply on whether expert testimony that Conde died before 1:20 a.m. was available, but on whether a reasonable attorney could have chosen not to present such testimony. It is true that such testimony could have been challenged on various grounds, and the trial court recognized that “[a] jury hearing this case could accept [Dr. Cohen’s] testimony and disregard petitioner’s experts, whoever those qualified experts will be.” But considering the totality of the experts’ testimony as well as the trial court’s findings, we conclude there is a reasonable probability that a competent attorney would have presented such evidence at trial.
As noted, a time of death defense would have potentially complemented defense counsel’s theory of third party culpability; evidence that Conde died before Long got home would have suggested that someone else committed the crime. Although such testimony would have been at odds with Long’s statements during her 911 call at 2:09 a.m. and her statements to investigators, a competent attorney could have argued she was mistaken due to her intoxication and the shock of discovering Conde. Indeed, Keen argued at trial that Long was “intoxicated” and “hysterical” when she came home “and saw [Conde] dead on the couch,” and Keen emphasized various aspects of the paramedics’ report to suggest Conde was dead before Long arrived. In other words, discounting Long’s 2:09 a.m. statement that Conde was still breathing was in fact part of, not in tension with, the defense strategy. At the evidentiary hearing, Keen testified that expert testimony placing the time of death before Long’s arrival “would be consistent with our defense” and that he “would have fought to introduce that at trial” if he had known such testimony was available.
In sum, the record before us does not support the Court of Appeal’s conclusion that presenting a time of death expert “would have been a risky defense strategy.” As the trial court explained, such expert evidence, even if imperfect, would have provided a potentially meritorious defense ruling out Long as the perpetrator. And even though a jury could have ultimately decided to reject it, such evidence would have had enough substance as to pose no serious risk of compromising the overall credibility of the defense. We conclude that there is “a reasonable probability that a competent attorney . . . would have introduced” expert testimony on the victim’s time of death in this case. (Wiggins, supra, 539 U.S. at p. 535.) Keen was “not in a position to make a reasonable strategic choice” as to whether to present such testimony because his investigation into Conde’s time of death was unreasonable. (Ibid.)
We now consider whether “there is a reasonable probability that [the jury] would have returned with a different” verdict (Wiggins, supra, 539 U.S. at p. 536) if counsel had presented the testimony of a time of death expert. The trial court observed that “it’s always a question of fact for the jury to either accept or reject the testimony of any witness that testifies, including an expert.” And the court acknowledged that Dr. Cohen disagreed with the testimony of Dr. Bonnell and Dr. Hua, and “opined the time of death was just as likely . . . before 1:20 a.m. as after.” Nevertheless, the court ruled that expert testimony estimating time of death before Long arrived home “could reasonably raise a reasonable doubt in the minds of the jurors” and “could be fatal to the People’s case.” Applying our independent judgment, we agree.
The prosecution’s evidence against Long was not particularly strong. The case rested on circumstantial evidence of Long’s motive and opportunity to commit the crime. As noted, in the hours leading up to Conde’s murder, Long and Conde had been involved in a verbal and physical altercation. Long was cheating on Conde with Dills, and she was heavily intoxicated. Although no neighbors saw Long or Conde, they heard several noises from the direction of the couple’s house late at night. When Long called 911, she said Conde was still breathing, yet she did not render assistance despite being a vocational nurse. Several items were missing from Long’s house, though it did not appear to police that the house had been ransacked. And the prosecution had called into doubt Long’s uncorroborated testimony that she had not changed her clothes that night.
At the same time, the prosecutor offered no confession, no eyewitnesses, no murder weapon identified or recovered, and no bloody clothes or other physical or forensic evidence linking Long to the crime. Furthermore, the alibis of the other suspects, Bugarski and Lovejoy, were only weakly corroborated, if at all. In the months before Conde’s death, both Bugarski and Lovejoy had threatened the couple and had engaged in suspicious activity around Long’s house. Ultimately, the sole testimony placing Long at the scene of the crime came from Dills. As the trial court observed, “[O]bviously the People’s cornerstone to their prosecution was the reliability of Mr. Dills.” But Dills had also been drinking throughout that day; he was intimately involved with the victim’s girlfriend; he was a suspect in the crime when interviewed by the police; and because he died before trial, he never testified before a jury and was never cross-examined after full discovery.
The centerpiece of the prosecution’s case was its theory that Long arrived at the murder scene at 1:20 a.m. and called 911 at 2:09 a.m. Under that scenario, she would have had approximately 50 minutes to arrive, secure a weapon, bludgeon the victim to death, and then clean up the scene as well as herself and dispose of the weapon. Under the timeframe offered by the defense, Long would have had only 10 minutes to accomplish all of those tasks.
Several facts stand against the prosecution’s theory. There were two other people with a potential motive to kill the victim, and their alibis were not firm. In addition, the front door was unlocked, the kitchen slider was open, and items were missing from the home. It was a bloody crime scene with a 360-degree blood splatter pattern. Yet there was no evidence that a sink or shower had been used to clean up. A search of the home, the car, and the neighborhood, including storm drains, uncovered no weapon, no bloody clothes, and no other evidence.
In addition, the prosecution expert opined at trial that the victim’s head injuries would have caused death within 10 to 15 minutes. The prosecution’s own evidence showed Long was over two miles away between 11:00 p.m. and 1:20 a.m. Defense counsel did not present available evidence from which the jury could have concluded that the victim could have died much earlier, when Long was nowhere near the crime scene. That evidence would have effectively negated the linchpin of the prosecution’s case and, together with the third party culpability evidence and the evidence of a possible break-in and theft, would have significantly enhanced the probability of a jury having a reasonable doubt as to Long’s guilt.
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.” (Strickland, supra, 466 U.S. at p. 696.) Given the totality of the trial evidence, there is a “substantial, not just conceivable,” likelihood that expert testimony that Conde died before 1:20 a.m. would have led to a more favorable result for Long. (Harrington, supra, 562 U.S. at p. 112.) We conclude it is reasonably probable that the presentation of expert testimony as to Conde’s time of death, along the lines of Dr. Hua’s and Dr. Bonnell’s testimony, would have led one or more jurors to harbor reasonable doubt about Long’s guilt. Counsel’s failure to investigate the time of death, in a case where the timeline was crucial, is an error “sufficient to undermine confidence in the outcome.” (Strickland, at p. 694.)
We reverse the judgment of the Court of Appeal.
CANTIL-SAKAUYE, C. J.
- * Associate Justice of the Court of Appeal, Fifth Appellate District, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution. ?
- * Associate Justice of the Court of Appeal, Fifth Appellate District, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution. ?