Johnson v. Premo

499 P.3d 814, 315 Or. App. 1

• Court: Court of Appeals of Oregon
• Decided: October 6, 2021
• Docket: A159635
• Status: Published

Opinion

Argued and submitted August 2, 2019, reversed and remanded October 6, 2021

JESSE LEE JOHNSON, Petitioner-Appellant, v. Jeff PREMO, Superintendent, Oregon State Penitentiary, Defendant-Respondent. Marion County Circuit Court 08C11553; A159635 499 P3d 814

Petitioner was convicted of aggravated murder and sentenced to death. He petitioned for post-conviction relief on a number of grounds, and the post- conviction court denied relief on all of them. He appeals. In petitioner’s seventh assignment of error petitioner contends that the post-conviction court, having determined that counsel performed deficiently by failing to interview a witness, should also have determined that the deficient performance prejudiced him, and should have granted petitioner relief on that basis. The superintendent responds that the post-conviction court correctly determined that petitioner was not prej- udiced by counsel’s performance and the superintendent also cross-assigns error to the post-conviction court’s determination that counsel performed deficiently. Held: The post-conviction court erred by not granting relief on that claim. The post-conviction court correctly determined that counsel performed deficiently, but the court erred in determining that the deficient performance did not preju- dice petitioner. Because the deficient performance could have had a tendency to affect the result, petitioner was entitled to post-conviction relief on that claim. Reversed and remanded.

Garry L. Reynolds, Senior Judge. Ryan T. O’Connor argued the cause for appellant. Also on the briefs were Jed Peterson and O’Connor Weber LLC. Erin K. Galli, Assistant Attorney General, argued the cause for respondent. Also on the answering brief were Ellen F. Rosenblum, Attorney General, Benjamin Gutman, Solicitor General, Shannon T. Reel, Assistant Attorney General, and Jordan R. Silk, Assistant Attorney General. Also on the reply brief were Ellen F. Rosenblum, Attorney General, and Benjamin Gutman, Solicitor General. 2 Johnson v. Premo

Before Armstrong, Presiding Judge, and Shorr, Judge, and Aoyagi, Judge. ARMSTRONG, P. J. Reversed and remanded. Cite as 315 Or App 1 (2021) 3

ARMSTRONG, P. J. Petitioner was convicted of aggravated murder and sentenced to death. He petitioned for post-conviction relief on a number of grounds, all of which were ulti- mately rejected by the post-conviction court. He appeals. We write to address only petitioner’s seventh assignment of error. In that assignment, petitioner contends that the post-conviction court, having determined that counsel per- formed deficiently by failing to interview a witness, should also have determined that their deficient performance prej- udiced him, and should have granted relief on that basis. The superintendent responds that the post-conviction court correctly determined that petitioner was not prejudiced by counsel’s performance and also has cross-assigned error to the post-conviction court’s determination that counsel per- formed deficiently. We agree with petitioner and therefore reverse and remand for the post-conviction court to grant relief on that claim. Because petitioner’s seventh assign- ment of error will result in a new trial in the criminal case, we need not address petitioner’s remaining assignments of error. We briefly set out background facts here and later add more factual detail in relation to petitioner’s claim concerning the failure to interview a witness—Patricia Hubbard—as pertinent to our discussion. The Supreme Court’s opinion in his direct appeal, State v. Johnson, 342 Or 596, 157 P3d 198 (2007), cert den, 552 US 1113 (2008), contains some additional detail. Harriet Thompson was stabbed to death in her home in Salem in the early morning hours of March 20, 1998. Johnson, 342 Or at 598. Thompson’s upstairs neighbor heard screaming at around 4:30 a.m. A witness told police that petitioner had been in the victim’s home on the night of the murder. Id. Another witness described someone whose appearance was similar to petitioner’s walking away from Thompson’s home at about 6:15 a.m. Id. Police arrested petitioner on a probation violation a week after the murder. Petitioner said that he knew the victim but denied that he had ever been inside her home. Police found fingerprints in the victim’s home that matched 4 Johnson v. Premo

petitioner’s, and a cigarette butt found inside the home was later matched to petitioner’s DNA. Other evidence that was more closely associated with the murder, such as the murder weapon and blood, was collected inside the victim’s home. Of the evidence that was DNA tested, none was matched to petitioner. Petitioner had boots that had similar soles to ones that had left prints in blood at the murder scene, but petitioner’s boots did not test positive for blood.1 An informant told police that petitioner had shown him some of the victim’s jewelry and told him that he “offed the bitch to rob her.” An officer said that, when he asked petitioner whether the informant was lying, petitioner had replied, “No, he’s not.” The informant later recanted to one of trial counsel’s investigators, but then at trial testified against petitioner consistently with his statements to police. Petitioner possessed jewelry that was identified by a witness as matching Thompson’s. Petitioner was charged with aggravated murder. Before his criminal trial, he rejected an offer to plead guilty to manslaughter and robbery and receive a 15-year sentence. Petitioner was represented by two lawyers, to whom we refer collectively as trial counsel, or counsel. Lead trial counsel had previously worked on one or two capital cases. Co-counsel had previously been involved in capital cases but had never represented a defendant in the penalty phase of a capital case. He had also worked on some capital cases in the post-conviction context. Counsel had also attended con- tinuing legal education seminars on the death penalty. Petitioner was convicted of aggravated murder and sentenced to death. After his unsuccessful direct appeal, Johnson, 342 Or 596, petitioner filed a petition for post-conviction relief. In his Fourth Amended Petition for post-conviction relief, petitioner alleged in his first claim that he was denied the 1 Petitioner has raised a claim of error concerning his claim that trial counsel performed inadequately and ineffectively in their handling of the boot evidence. We need not resolve that claim, but suffice to say that, at trial, there was conflict- ing evidence available about petitioner’s boots and whether they connected peti- tioner to the crime scene, and petitioner’s trial counsel took conflicting positions during the criminal trial concerning the boot evidence. Cite as 315 Or App 1 (2021) 5

right to adequate and effective assistance of counsel under Article I, section 11, of the Oregon Constitution and under the Sixth Amendment of the United States Constitution due to trial counsel’s failure to investigate. Within that claim, petitioner alleged more specifically that, among other fail- ures, trial counsel failed to interview Patricia Hubbard, Thompson’s neighbor. The parties each addressed that claim in their trial memoranda. The post-conviction court concluded that trial coun- sel’s failure to interview Hubbard had amounted to deficient performance. The court concluded, however, that petitioner had not been prejudiced by that failure: “The court finds that trial counsel failed to use reason- able skill in not interviewing Ms. Hubbard who live[d] right across the street from the murder scene. However, taking into consideration that the trial produced other evidence that coincided with what she would have testified to and the long period of time between the killing and when she was asked to try [to] identify Petitioner[,] the court is not persuaded that the absence of her testimony would have prejudiced Petitioner.” As noted, petitioner’s seventh assignment of error concerns his claim that trial counsel’s failure to interview Hubbard constituted inadequate and ineffective assistance of counsel. The post-conviction court determined that coun- sel performed deficiently by not interviewing Hubbard, “who live[d] right across the street from the murder scene.” But it did not grant relief on that claim, because it concluded that the deficient performance did not prejudice petitioner. Petitioner assigns error to that ruling. The superintendent cross-assigns error, arguing that the post-conviction court erred in concluding that counsel’s performance was defi- cient. We reverse on petitioner’s assignment of error and affirm on the cross-assignment. We begin by expanding on the facts pertinent to this claim. In her deposition, taken in 2013 in the context of this post-conviction case, Hubbard explained that she lived across the street and two houses down from the vic- tim. She had a view of part of the victim’s property. She worked long hours, and regularly was up late at night or in 6 Johnson v. Premo

the early morning. On the night that the victim was mur- dered, Hubbard was awake and was sitting on her porch at about 3:45 a.m. She saw a white man drive up and park his van in the victim’s driveway. She recognized the white man as someone she had noticed at the victim’s home before, “[m]any, many times.” After the man went inside, within seconds, Hubbard heard shouting and screaming coming from the house. There was a male and a female voice. Hubbard recognized the male voice as belonging to the white man she had noticed numer- ous times before at the property. She heard sounds like pots and pans crashing and loud voices and screaming. The screaming “got higher pitched and louder and more intense on the volume the longer it went on.” She heard screaming, then a thud, and then total silence. The white man was in the victim’s house “[u]ntil after the screaming stopped, and he [came] out the back door and—didn’t even hit the steps. Just flew off the steps and took off running at a northwest—flying northwest * * *.” He had been in the house for about 30 minutes. About 10 to 15 minutes after that, Hubbard testi- fied, she saw a Black man walking down the driveway. She could not say whether he had come from inside the house. Hubbard thought it might be a man to whom she previously had been introduced. Petitioner is Black, and the victim had introduced Hubbard to him. When petitioner’s post- conviction team showed her petitioner’s photograph approx- imately 12 years after the murder, Hubbard said that he did not look like the person whom she had seen that night. The Black man was not running. “He was just * * * kind of strolling out.” He rubbed his forehead “like he had a headache or, you know, disbelief of something or, you know, just kind of like—something wasn’t quite right.” Hubbard received a call that she needed to go back in to work, so she went inside. Later, at 11:00 a.m., she received a call at work telling her that she had “better come home.” There was a “commotion” in the neighborhood— “[p]olice and the caution tape and the gawkers walking around the neighborhood.” Cite as 315 Or App 1 (2021) 7

Hubbard approached a uniformed police officer that day. Hubbard told him that she had “some information that might help with the problem you’ve got.” The officer told her that he didn’t need her help and told her to go back home. On another occasion, Hubbard testified, Shalonda Washington, a neighbor who lived in a house only yards away from the victim’s home, brought a Salem police detec- tive to Hubbard’s house. According to Hubbard, “I started telling him what I saw, and he stopped me, and he said, ‘that won’t be necessary.’ ” Moreover, Hubbard recounted, the detective said, “A nigger got murdered, and a nigger’s going to pay for it.” Hubbard was never interviewed by trial counsel’s investigators. Trial counsel’s investigators spent a total of roughly six hours canvassing the neighborhood and speak- ing to witnesses. They did not speak to Hubbard. She was later contacted and interviewed by investigators in this post-conviction case. The trial court found the following facts: Had she been located, Hubbard would have testified that she had heard screaming and arguing at the victim’s house and saw two people leave—a Black man and a white man. She could identify the white man because she had seen him at the victim’s house before. “She did not recognize the Black man, but when she was shown Petitioner’s photograph by Petitioner’s post-conviction team approximately 12 years after the incident she said he did not look like the [man] she saw that night.” The post-conviction court determined that trial counsel failed to use reasonable skill by not interviewing Hubbard, considering that she lived “right across the street” from the murder scene. The court also concluded, however, that petitioner was not prejudiced by that deficient perfor- mance: “[T]aking into account that the trial produced other evidence that coincided with what she would have testified to and the long period of time between the killing and when she was asked to try [to] identify Petitioner the court is not persuaded that the absence of her testimony would have prejudiced Petitioner.” 8 Johnson v. Premo

As pertinent here, we review the denial of a claim for post-conviction relief for legal error.2 For petitioner to prevail on a claim of inadequate assistance of counsel based on Article I, section 11, he “must demonstrate two things: that * * * counsel failed to exercise reasonable professional skill and judgment and that he suf- fered prejudice as a result.” Gable v. State of Oregon, 353 Or 750, 758, 305 P3d 85, cert den, 571 US 1030 (2013). To prevail on a claim based on the federal constitutional right to counsel, petitioner likewise must establish that “counsel’s performance was deficient” and that “the deficient perfor- mance prejudiced the defense.” Strickland v. Washington, 466 US 668, 687, 104 S Ct 2052, 80 L Ed 2d 674 (1984). “[T]he standards for determining the adequacy of legal counsel under the state constitution are functionally equiv- alent to those for determining the effectiveness of counsel under the federal constitution.” Montez v. Czerniak, 355 Or 1, 6-7, 322 P3d 487, adh’d to as modified on recons, 355 Or 598, 330 P3d 595 (2014). When a court considers whether a lawyer’s conduct failed to meet constitutional standards, it must “make every effort to evaluate [the] lawyer’s conduct from the lawyer’s perspective at the time, without the dis- torting effects of hindsight.” Lichau v. Baldwin, 333 Or 350, 360, 39 P3d 851 (2002). We first address the superintendent’s cross- assignment of error challenging the post-conviction court’s determination that trial counsel performed deficiently by not interviewing Hubbard. The superintendent argues that the approximately six hours that defense investigators spent canvassing the victim’s neighborhood was sufficient, that police reports did not indicate that anyone at Hubbard’s address had useful information, and that Hubbard was not mentioned in any police report. In the superintendent’s view, trial counsel’s investigation was reasonable, because police officers had canvassed the neighborhood. According to their reports, police had attempted to interview residents but

2 Petitioner also raises assignments of error in relation to claims on which the post-conviction court granted summary judgment in favor of the superinten- dent. As noted above, in light of our disposition on petitioner’s seventh assign- ment of error, we do not reach petitioner’s other assignments of error. Cite as 315 Or App 1 (2021) 9

“no one was home” at four nearby residences. In addition, according to the reports, “[t]he residents”—none of whom were identified—of four other homes “said they did not know anything about the deceased or her residence.” Petitioner argues to the contrary that the post- conviction court correctly determined that any reasonable defense attorney would have recognized the importance of interviewing the residents of the homes immediately sur- rounding the victim’s home, and that it was not reasonable to stop attempting to contact those residents after only six hours of canvassing the area and speaking to residents. We agree with petitioner and the post-conviction court. Adequate trial counsel would have recognized the importance in a capital murder case of contacting the peo- ple nearby who were likely to have information about the victim, people associated with the residence, and the events of the night in question, when a violent murder occurred in a nearby home in the early morning hours when many res- idents would likely have been at home. The post-conviction court found that trial counsel’s investigators spent approx- imately six total hours of combined time canvassing and interviewing witnesses in the neighborhood. According to Hubbard, in addition to trial counsel’s investigators failing to contact her directly, had they contacted another neighbor, Shalonda Washington, Washington would have brought the investigators to Hubbard or otherwise helped to put them in touch with each other. That is a reasonable inference, con- sidering that Washington brought a police detective to talk to Hubbard just after the murder, and over a decade later Washington also helped to put petitioner’s post-conviction investigator in touch with Hubbard, telling the investigator that Hubbard had information about what happened. Further, the post-conviction investigator explained that Washington’s house was the first one he visited in his investigation, because it is so close to the victim’s house. Although he spent many hours attempting to contact nearby people in nearly 70 residences, the investigator explained that Hubbard’s home was the second one he contacted. Washington was a witness named in police reports, and she pointed the investigator to Hubbard, whom the investigator 10 Johnson v. Premo

was able to contact and interview “with minimal persistence and effort,” despite the fact that she had moved by the time he contacted her. It was not a reasonable exercise of profes- sional judgment to stop investigating after only six hours of canvassing and interviews under these circumstances. Having concluded that the post-conviction court did not err in determining that trial counsel performed defi- ciently, we turn to petitioner’s argument that the court erred when it concluded that petitioner suffered no prejudice as a result of that deficient performance. Petitioner first argues that the post-conviction court erred by applying the wrong legal standard for determining whether trial counsel’s defi- cient performance prejudiced him. He next argues that, if the court did apply the correct standard, it misapplied that standard to the facts and should have come to the opposite conclusion. Specifically, he argues that, had counsel had the information Hubbard could have provided, that could have tended to affect the result, because it could have affected counsel’s strategy and could have raised doubts that would have had a tendency to affect the jury’s verdict. We begin with petitioner’s first argument, that the post-conviction court applied the wrong standard for deter- mining prejudice. Petitioner points to the post-conviction court’s statement that it was “not persuaded that the absence of [Hubbard’s] testimony would have prejudiced Petitioner.” He argues that the court’s use of “would have” rather than “could have tended to” implies that the court applied the incorrect standard for determining prejudice. Putting its statement in context, however, we are not persuaded that the post-conviction court applied an incorrect standard. The post-conviction court wrote a thorough memorandum of its opinion. It correctly laid out the applicable legal standards for claims under the Oregon and federal constitutions. That it later used shorthand for the applicable standard does not establish that the court used an incorrect standard, after having correctly stated the standard. Specifically, the post- conviction court’s statement was not that it was not per- suaded that Hubbard’s testimony would have affected the outcome. Rather, it stated that it was not persuaded that the testimony would have prejudiced petitioner. The reference to prejudice implies that the court was applying the standard Cite as 315 Or App 1 (2021) 11

for prejudice—which it had earlier correctly set out—and not that it was stating the standard itself. Accordingly, we reject petitioner’s argument that the post-conviction court applied the wrong standard. We agree, however, that, applying the correct stan- dard, the post-conviction court reached an erroneous con- clusion. As we will explain, we conclude that the deficient performance by petitioner’s trial counsel did result in preju- dice. We therefore reverse the post-conviction court’s denial of that claim and remand for the court to grant relief. Concerning prejudice, the post-conviction court con- cluded that “the trial produced other evidence that coincided with what [Hubbard] would have testified to,” and it deter- mined that, considering the long period of time between the murder and her viewing of petitioner’s photograph, her response to the question of whether petitioner was the Black man that she saw that morning did not persuade the court that there was prejudice. That is, the court determined that it could not say that Hubbard’s testimony could have tended to affect the outcome of the trial. Petitioner argues that Hubbard’s testimony would not have merely “coincided” with other testimony. He argues that, instead, Hubbard’s testimony would have tended to influence the jury, and also could have influenced coun- sel’s trial strategy in relation to other evidence and trial decisions. The superintendent responds that the trial court correctly concluded that the evidence would not have been helpful to petitioner at trial, because “nothing about Hubbard’s testimony would have rebutted the state’s the- ory of the case.” The superintendent summarized the state’s theory of the case at the criminal trial: At “around 4:30 a.m. on March 20, 1998,” petitioner stabbed the victim to death, “and left the area at around 6:15 a.m., when John Shaw saw him walking in the area.” The superintendent argues that Hubbard’s testimony would have corroborated the state’s evidence. Petitioner had a cellmate after his arrest, Schellong, who testified in the guilt phase of petitioner’s trial that 12 Johnson v. Premo

petitioner had told him that petitioner had been at the vic- tim’s home when a white man and a Black man, Sampson, had also been there. Schellong testified that petitioner had said that the victim was supposed to introduce him to her drug dealer. When petitioner came out of a room, acting like he was drunk, the white man left, but Sampson stayed. Schellong said that petitioner told him that he and Sampson talked, and, before leaving, Sampson told petitioner to go through the victim to get his drugs. In the superintendent’s view, “Schellong’s testimony is consistent with Hubbard’s testimony that she saw a ‘white guy’ run from the victim’s house, and then, five to fifteen minutes later, saw an African American man walking down the alley.” The superintendent also argues that Hubbard’s tes- timony would have corroborated the testimony of another witness, Swafford. Swafford said that he saw petitioner at the victim’s home the night of the murder. The superinten- dent asserts that Hubbard’s observation of a Black man walking down the alley was consistent with Swafford’s obser- vations. Finally, the superintendent argues that Hubbard’s statement would not have been inconsistent with another witness’s testimony. John Shaw testified that, sometime between 6:10 and 6:30 a.m., he saw a man walking near the victim’s home. Shaw did not identify petitioner as the man he saw, but Schellong testified that petitioner had acknowl- edged that he was the person Shaw had seen. We agree with petitioner that, had trial counsel interviewed Hubbard, her testimony at trial could have had a tendency to affect the outcome. Although there was other evidence that “coincided with” Hubbard’s evidence, that could have made her testimony more persuasive, not less important. That is, corroboration works both ways. Hubbard’s narrative—a white man arrived, after which there was loud shouting, sounds like pots and pans crash- ing, and screaming, then the screaming stopped, and the white man fled the house at great speed—was not duplica- tive of the state’s narrative, even if parts of it matched or did not conflict with testimony by some of the state’s witnesses. The state’s evidence was that a petitioner’s former cellmate had testified that petitioner told him that a white man and a Black man named Sampson had been at the house before Cite as 315 Or App 1 (2021) 13

the murder, and that the white man had left first. That nar- rative is different in important respects from Hubbard’s. Hubbard described sounds that could be inferred to have been the murder, after which she saw the white man, whom she recognized as a frequent visitor, “flying” from the vic- tim’s home. In addition, Hubbard’s testimony included evi- dence of racial bias in the police investigation of the murder, and a failure to properly investigate. Finally, trial counsel had strategic choices to make about the defense theory of the case. Counsel ultimately made choices that allowed evidence that had previously been suppressed to be intro- duced at trial in support of a defense theory that petitioner was present for the murder, but less culpable than the per- son who had left a distinct set of shoeprints. See Johnson, 342 Or at 617-23. Hubbard’s information could have helped counsel to settle on a defense theory that was not partially inculpatory. The post-conviction court erred in concluding that the failure to reasonably investigate did not prejudice peti- tioner’s case. A reasonable investigation would likely have led to finding and interviewing Hubbard, which in turn would have led to evidence and testimony that could have tended to affect the outcome of the trial. Accordingly, we reverse and remand for the trial court to grant relief on peti- tioner’s first claim for relief. Reversed and remanded.