Washington v. Ryan

922 F.3d 419

• Court: Court of Appeals for the Ninth Circuit
• Decided: April 17, 2019
• Docket: 05-99009
• Status: Published

Opinion

FOR PUBLICATION

UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT

THEODORE WASHINGTON, No. 05-99009 Petitioner-Appellant, D.C. No. v. CV-95-02460-JAT

CHARLES L. RYAN, Warden, Respondent-Appellee. OPINION

Appeal from the United States District Court for the District of Arizona James A. Teilborg, District Judge, Presiding

Argued and Submitted September 26, 2018 Pasadena, California

Filed April 17, 2019

Before: Ronald M. Gould, Consuelo M. Callahan, and N. Randy Smith, Circuit Judges.

Opinion by Judge Gould; Dissent by Judge Callahan 2 WASHINGTON V. RYAN

SUMMARY *

Habeas Corpus / Death Penalty

The panel reversed the district court’s denial of habeas relief as to the penalty phase, and remanded, in a case in which Arizona state prisoner Theodore Washington, who was sentenced to death for first-degree murder, asserted that his trial counsel rendered ineffective assistance by not investigating and presenting mitigating evidence at the penalty phase.

The panel reviewed the district court’s decision de novo in this pre-AEDPA case and applied the standard articulated in Strickland v. Washington, 466 U.S. 668 (1984) – assessing the state court record to determine whether Washington’s counsel was constitutionally deficient and whether the deficient performance resulted in prejudice.

The panel held that counsel performed ineffectively by not properly investigating Washington’s background, and as a result, the trial court was not presented at the penalty phase with substantial mitigation evidence regarding Washington’s education and incarceration, his diffuse brain damage, and his history of substance abuse. The panel held that this raises a probability that, had the court been presented with the mitigation evidence in the first instance, the outcome would have been different, as the sentencing judge might have decided that Washington should be spared death and be imprisoned for life.

* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader. WASHINGTON V. RYAN 3

Dissenting, Judge Callahan wrote that in second- guessing the performance of Washington’s trial counsel, the majority uses a standard for gross incompetence that doesn’t square with precedent, and doesn’t hold Washington to his heavy burden of prejudice.

The panel addressed other issues in a concurrently filed memorandum disposition.

COUNSEL

Nathaniel C. Love (argued), Sidley Austin LLP, Chicago, Illinois; Gilbert H. Levy, The Law Offices of Gilbert H. Levy, Seattle, Washington; Mark E. Haddad, Sidley Austin LLP, Los Angeles, California; for Petitioner-Appellant.

Laura P. Chiasson (argued), Assistant Attorney General; Lacey Stover Gard, Chief Counsel; Mark Brnovich, Attorney General, Office of the Attorney General, Tucson, Arizona; for Respondent-Appellee.

OPINION

GOULD, Circuit Judge:

Arizona state prisoner Theodore Washington appeals the district court’s denial of his petition for a writ of habeas corpus under 28 U.S.C. § 2254. In 1987, a jury found Washington guilty of six crimes involving the robbery and murder of Sterleen Hill in her Arizona home. The court sentenced Washington to death.

In his habeas corpus petition, Washington challenges his conviction and sentence on the first-degree murder charge. 4 WASHINGTON V. RYAN

He asserts that he is entitled to habeas relief on several grounds, the majority of which are addressed in a separate memorandum disposition filed concurrently with this opinion. This opinion solely addresses Washington’s certified claim for ineffective assistance of trial counsel. Washington contends that his counsel did not investigate and present mitigating evidence at the penalty phase, including evidence of diffuse brain damage, childhood abuse, and substance abuse. The Arizona court previously considered and rejected this claim on post-conviction review.

Because review under the Antiterrorism and Effective Death Penalty Act of 1996, Pub. L. No. 104-122, 100 Stat. 1214 (“AEDPA”), does not apply in this case, we are not bound by the highly deferential “double deference” in considering Washington’s claim of ineffective assistance of counsel and its proper analysis. See Hardy v. Chappell, 849 F.3d 803, 824–26 (9th Cir. 2016) (explaining the interaction of 28 U.S.C. § 2254(d) and the standard for deficiency under Strickland v. Washington, 466 U.S. 668 (1984)). Instead, we apply the familiar standard articulated in Strickland, and assess the state court record to determine whether Washington’s counsel was constitutionally deficient and whether the deficient performance resulted in prejudice. See Bobby v. Van Hook, 558 U.S. 4, 7 (2009) (applying the Strickland analysis in a pre-AEDPA case). Because Washington’s counsel did not properly investigate Washington’s background, the trial court at the penalty phase was not presented with substantial mitigation evidence regarding Washington’s education and incarceration, his diffuse brain damage, and his history of substance abuse. This raises a reasonable probability that, had the court been presented with the mitigation evidence in the first instance, the outcome would have been different. The sentencing judge might have decided that Washington should be spared WASHINGTON V. RYAN 5

death and be imprisoned for life. 1 We reverse the district court’s denial of habeas relief and remand with instructions to grant habeas relief against the death penalty, unless within a reasonable time the state retries the penalty phase or decides to modify the sentence to life in prison.

I

At around 11:45 p.m. on the night of June 8, 1987, two men forced their way into Ralph and Sterleen Hill’s Yuma, Arizona home in what turned out to be a disastrously violent home invasion. The men forced the Hills to lie face down on the floor of the master bedroom and bound their hands behind their backs. One of the men intermittently “screwed” a pistol in Ralph’s ear while both men yelled at the couple demanding that the Hills give them drugs or money. Ralph glimpsed one of the assailants as he ransacked the drawers and closets in the room. The Hills were discovered lying face down in their bedroom. Both had been shot in the back of the head. Ralph survived the horrendous shot to his head, but was seriously injured. Sterleen did not survive the shooting.

Police arrested Fred Robinson shortly after the incident. Robinson was the common law husband of Susan Hill, Ralph

1 In this case, because it predated the rule in Ring v. Arizona, 536 U.S. 584 (2002), the judge had the power the power to select life imprisonment rather than death. For cases after the effective date of Ring, a jury has this power. In either case, the decision maker at the penalty phase need not account for its decision. See generally Ring, 536 U.S. 584; Gregg v. Georgia, 428 U.S. 153 (1976). Accordingly, the decision maker after hearing all mitigation and aggravation evidence, will be left, to borrow a phrase from Milton, with the power “to temper justice with mercy.” John Milton, Paradise Lost, book x, lines 77–78 (1674). 6 WASHINGTON V. RYAN

Hill’s daughter from a prior marriage. Police also arrested Jimmy Mathers and Theodore Washington in connection with the crimes. The state charged the three men with (1) first degree murder for the death of Sterleen Hill, (2) attempted first degree murder, (3) aggravated assault causing serious physical injury, (4) aggravated assault using a deadly weapon, (5) burglary in the first degree, and (6) armed robbery. The three men were tried together, and the jury convicted on all counts.

A

The penalty phase of the trial commenced on January 8, 1988. In this appeal, we are concerned with the penalty phase of Washington’s trial and the death penalty sentence he received.

Washington’s trial counsel, Robert Clarke, called three witnesses to testify on Washington’s behalf: Washington’s friend, Steve Thomas; Washington’s mother, Willa Mae Skinner; and Washington’s half-brother, John Mondy.

Steve Thomas testified that he knew Washington for two years. He testified that Washington was easily influenced but not violent. He also testified that Washington was a dedicated father. When asked if Washington had a drug problem, Thomas testified that he had not noticed one. Willa Mae Skinner testified that Washington was a good child and that he dropped out of school when he was in high school. She also testified that Washington was a good father, and that he was gentle and “liked to party.” Finally, John Mondy reiterated that Washington was affable but easily led. He also confirmed that Washington had trouble in school as a child. WASHINGTON V. RYAN 7

During closing argument, Clarke focused primarily on attacking the sufficiency of the court’s findings under Enmund v. Florida, 458 U.S. 782 (1982), and Tison v. Arizona, 481 U.S. 137 (1987). Regarding mitigation, Clarke did not entirely ignore all mitigation, but rather urged the court to consider Washington’s age, his relatively minor criminal record, his good relationship with his son, and his general demeanor as a caring individual. This appeal is concerned primarily with the mitigation evidence and argument that Clarke did not present.

The trial court found that the state had established two aggravating factors beyond a reasonable doubt: (1) that the murder was committed in an especially cruel, heinous, or depraved manner, and (2) that the murder was committed for, or motivated by, pecuniary gain. With respect to mitigation, the court found that Washington’s age was not a mitigating factor and that the remaining mitigating factors did not outweigh the aggravating factors. The court sentenced all three defendants to death on the first-degree murder charges.

B

Washington, Robinson, and Mathers each appealed their convictions and sentences to the Arizona Supreme Court. The state high court affirmed Washington and Robinson’s convictions and sentences, State v. Robinson, 796 P.2d 853, 865 (Ariz. 1990), but found insufficient evidence to convict James Mathers and vacated his conviction, State v. Mathers, 796 P.2d 866 (Ariz. 1990).

Following the direct appeal process, Washington and Robinson challenged their convictions and sentences on post-conviction review (“PCR”). The court held a joint PCR hearing on September 8, 1993. The Honorable Stewart 8 WASHINGTON V. RYAN

Bradshaw, the same judge who presided over the trial, also presided over the post-conviction review proceeding. Washington, through his appellate counsel, argued that Clarke was ineffective at the penalty phase due to his failure to present mitigating evidence. Specifically, Washington argued that Clarke erred by failing to conduct a more thorough review of his school, medical, and incarceration records. He also argued that Clarke should have obtained a psychological evaluation and presented the results to the court.

The bulk of the new evidence presented at the PCR hearing was elicited through the testimony of Dr. Roy, the defense counsel’s retained psychologist. Dr. Roy evaluated Washington in 1992. He conducted clinical interviews and several psychological tests. Dr. Roy’s interviews with Washington revealed that he suffered abuse as a child in the form of daily whippings with straps and belts and that adults in the home used alcohol as a means to sedate him as a child. His review of Washington’s school and Department of Corrections (“DOC”) records revealed that he was placed in classes for the “educable mentally retarded” when he was five years old and that he had been marked as low-IQ while incarcerated. However, Dr. Roy testified that these records conflicted with his own clinical findings because Washington tested at a low-average IQ of 96.

Dr. Roy’s interviews with Washington also revealed that Washington had substance abuse problems relating to alcohol and cocaine use. Washington told Dr. Roy that he began drinking recreationally at age eight and was a functional alcoholic by age fourteen. He also told Dr. Roy that he was heavily intoxicated on the night of the murder. Washington also said that he was a heavy cocaine user and WASHINGTON V. RYAN 9

that he consumed about $175 in cocaine per day at the time of the crime.

Finally, Dr. Roy testified that he believed that Washington suffered from diffuse brain damage resulting from early and prolonged drug and alcohol use and numerous traumatic head injuries. Dr. Roy testified that diffuse brain damage can result in disinhibition and poor social judgment as well as poor impulse control and an inability to appreciate the long-term consequences of one’s actions. Dr. Roy testified that, in his opinion, Washington’s cocaine addiction and his impaired impulse control likely contributed to his ability to be manipulated by others into making poor decisions.

The state called Dr. Eva McCullars, a psychiatrist who also evaluated Washington. Dr. McCullars reviewed Dr. Roy’s report and conducted clinical interviews with Washington in June 1993. Dr. McCullars testified that she did not review Washington’s DOC records, school records, or adult incarceration records. Dr. McCullars agreed that Washington suffered from diffuse brain damage, but concluded that Washington also suffered from antisocial personality disorder. On direct examination, the state asked Dr. McCullars whether diffuse brain damage could cause hyperkenesis (hyperactive behavior or attention deficit disorder). Dr. McCullars explained that “[hyperkenesis] is one example of diffuse brain damage.” She went on to explain that several prominent individuals including Walt Disney and Thomas Edison exhibited hyperkinetic behavior as children. When questioned on cross examination, Dr. McCullars acknowledged that Washington came from a “significantly dysfunctional family.” She also admitted that several of the markers for antisocial personality disorder, such as early truancy and an inability to maintain 10 WASHINGTON V. RYAN

employment, were more frequently associated with lower socio-economic status black adolescents when compared to the general population.

Robert Clarke, Washington’s trial counsel, also testified at the PCR hearing. Clarke testified that he did not request Washington’s education or corrections records because he believed his interviews with Washington, Skinner, Mondy and Bryant were sufficient. Clarke testified that he had “very extensive discussions” with Washington about what his life was like and any possible substance abuse issues. He also testified that he had “relatively extensive” discussions with Washington’s mother, half-brother, and girlfriend. Clarke testified that, based on these interviews, “there wasn’t anything that clued me in that there was a special problem that would suggest I should obtain those types of records.” With respect to Washington’s drug use, Clarke testified that Washington never told him that he was addicted to cocaine or that he was using cocaine on the night of the murder. When questioned on the matter, Clarke acknowledged that Bryant had told Clarke that Washington had a “cocaine problem,” but that he did not investigate further.

In a written order, Judge Bradshaw held that Washington was not entitled to relief for ineffective assistance of counsel at the penalty phase. Judge Bradshaw credited Dr. McCullars’s findings that Washington had antisocial personality disorder and was poorly adjusted to living in society. However, Judge Bradshaw concluded that “there is nothing . . . which lessened his ability to differentiate right from wrong or conform his actions with the law.” Judge Bradshaw also explained that he had been aware at the time of sentencing that Washington had been doing well while incarcerated. Judge Bradshaw further reasoned that any drug and alcohol dependency “taken separately or with any WASHINGTON V. RYAN 11

other mitigating circumstance or circumstances would [not] have mitigated against the sentence [Washington] has received.”

C

On April 25, 1995, the Arizona Supreme Court summarily denied Washington’s petition for review of the PCR court’s decision. Washington then commenced his habeas action in the federal district court, culminating in this appeal.

In his amended federal habeas corpus petition, Washington raised 17 claims. The district court determined that claims 1, 2, 3 (in part), 6, 7, 8 (in part), 9, 11 (in part), 12, 14 (in part), 16, and 17 were procedurally barred. On April 22, 2005, the district court rejected the remaining claims on their merits and dismissed the petition. Washington filed a motion to alter the judgment on May 5, 2005, which the district court denied on June 8, 2005.

On July 11, 2005, Washington filed an untimely notice of appeal from the district court’s denial of habeas relief. A three-judge panel of this court held that it lacked jurisdiction and affirmed the district court’s denial of Rule 60(b) relief. Washington v. Ryan, 789 F.3d 1041 (9th Cir. 2015). The court then granted Washington’s motion for en banc rehearing. Washington v. Ryan, 811 F.3d 299 (9th Cir. 2015). In a 6–5 decision, the en banc panel held that Washington was entitled to relief under Rule 60(b)(1) and (6) from his untimely notice of appeal and ordered the district court to “vacate and reenter its judgment denying Washington’s petition for writ of habeas corpus, nunc pro tunc, June 9, 2005,” to render the notice of appeal timely. Washington v. Ryan, 833 F.3d 1087, 1102 (9th Cir. 2016). The United States Supreme Court denied the state’s petition 12 WASHINGTON V. RYAN

for writ of certiorari. Ryan v. Washington, 137 S. Ct. 1581 (2017) (mem.).

In his opening brief, Washington raised three certified issues and four uncertified issues. In this opinion, we address Washington’s certified claim for ineffective assistance of counsel at the penalty phase. The remaining issues are addressed in a memorandum disposition filed concurrently with this opinion.

II

We review de novo a district court’s decision to grant or deny a habeas petition under 28 U.S.C. § 2254. See Bean v. Calderon, 163 F.3d 1073, 1077 (9th Cir. 1998). Because Washington filed his habeas petition before the enactment of AEDPA, the provisions of AEDPA do not apply to this case. Id. (citing Jeffries v. Wood, 114 F.3d 1484, 1495–96 (9th Cir. 1997) (en banc)). Instead, we review the claim under the familiar standard set out in Strickland and its progeny without the added deference required under AEDPA.

III

To prevail on his claim for ineffective assistance of counsel, Washington must establish that Clarke’s performance was deficient and that he suffered prejudice as a result. See Strickland, 466 U.S. at 687–89. To establish deficient performance, Washington must show that “counsel’s representation fell below an objective standard of reasonableness.” Id. at 688. To establish prejudice, Washington must show that there is “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Id. at 694. These are formidable barriers to habeas corpus WASHINGTON V. RYAN 13

petition relief in federal court to state prisoners even absent application of AEDPA.

To prevail under Strickland’s performance prong, Washington must show that his “counsel’s representation fell below an objective standard of reasonableness.” See Strickland, 466 U.S. at 688. Even without the added layer of deference to the state court decision under AEDPA, “[s]urmounting Strickland’s high bar is never an easy task.” Padilla v. Kentucky, 559 U.S. 356, 371 (2010). In articulating the standard against which counsel’s performance should be judged, the Strickland Court emphasized the deference due to a lawyer’s decisions both as to scope of investigation and decisions made after investigation: “[S]trategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable. . . .” Strickland, 466 U.S. at 690. We have likewise recognized the wide latitude to be given to counsel’s tactical choices. See, e.g., United States v. Ferreira-Alameda, 815 F.2d 1251, 1253 (9th Cir. 1993) (“Review of counsel’s performance is highly deferential and there is a strong presumption that counsel’s conduct fell within the wide range of reasonable representation.”).

But our deference to counsel’s performance is not unlimited. As the Court explained in Strickland, counsel’s strategic choices made after less than complete investigation are reasonable only to the extent that “reasonable professional judgments support the limitations on investigation.” 466 U.S. at 690–91.

The mitigation obligation applies even when a person is clearly guilty. See Penry v. Lynaugh, 492 U.S. 302, 318–20 (1989), abrogated on other grounds by Atkins v. Virginia, 14 WASHINGTON V. RYAN

536 U.S. 304 (2002) (“[R]ather than creating a risk of an unguided emotional response, full consideration of evidence that mitigates against the death penalty is essential if the jury is to give a ‘reasoned moral response to the defendant’s background, character, and crime.’” (quoting Franklin v. Lynaugh, 487 U.S. 164, 184 (1988), abrogated on other grounds by Atkins, 536 U.S. 304)). Indeed, that is a key point of the penalty phase of a capital case, which proceeds only after a determination of guilt in the earlier phase. See Gregg, 428 U.S. at 191–92 (endorsing the use of a bifurcated trial to determine guilt and penalty). In the penalty phase, the focus shifts from guilt to culpability, and evidence on both aggravating and mitigating factors is properly considered. See Ariz. Rev. Stat. Ann. § 13-751 (2018) (identifying statutory aggravating and mitigating circumstances in capital-eligible cases); Buchanan v. Angelone, 522 U.S. 269, 275–76 (1998) (observing that the penalty phase of capital sentencing involves a determination of eligibility, where “the jury narrows the class of defendants eligible for the death penalty, often through consideration of aggravating circumstances,” and selection, where “the jury determines whether to impose a death sentence on an eligible defendant” and must consider “any constitutionally relevant mitigating evidence”).

At the penalty phase, “[a] decision not to . . . offer particular mitigating evidence is unreasonable unless counsel has explored the issue sufficiently to discover the facts that might be relevant to his making an informed decision.” Lambright v. Schriro, 490 F.3d 1103, 1116 (9th Cir. 2007); see also Ainsworth v. Woodford, 268 F.3d 868, 874 (9th Cir. 2001) (holding counsel’s performance was ineffective where counsel “failed to adequately investigate, develop, and present mitigating evidence to the jury even WASHINGTON V. RYAN 15

though the issue before the jury was whether [the defendant] would live or die”).

Washington asserts that Clarke erred by not reviewing his education records and incarceration records. “In preparing for the penalty phase of a capital trial, defense counsel has a duty to ‘conduct a thorough investigation of the defendant’s background’ in order to discover all relevant mitigating evidence.” Robinson v. Schriro, 595 F.3d 1086, 1108 (9th Cir. 2010) (quoting Correll v. Ryan, 539 F.3d 938, 942 (9th Cir. 2008)). We have recognized that, “[a]t the very least, counsel should obtain readily available documentary evidence such as school, employment, and medical records.” Id. at 1109. Clarke did not request or review any such records and provided no tactical reason for his failure to do so. Instead, Clarke relied entirely on his interviews with Washington, Skinner, Mondy, and Bryant and testified that he did not believe other sources of information would have been fruitful. Clarke’s failure to review these basic and readily-available sources of mitigating information fell below prevailing professional norms at the time. See id.

Washington also alleges that Clarke erred by not retaining an expert to conduct a psychological evaluation. A psychological evaluation at the penalty phase is not always required. Gentry v. Sinclair, 705 F.3d 884, 900 (9th Cir. 2013). Instead, “[t]rial counsel has a duty to investigate a defendant’s mental state if there is evidence to suggest that the defendant is impaired.” Douglas v. Woodford, 316 F.3d 1079, 1085 (9th Cir. 2003).

In assessing counsel’s performance, we must keep in mind that “[i]t is imperative that all relevant mitigating information be unearthed for consideration at the capital sentencing phase.” Caro v. Calderon, 165 F.3d 1223, 1227 (9th Cir. 1999). Whereas evidence of mental impairment is 16 WASHINGTON V. RYAN

relevant at the guilt phase of a capital trial if it tends to negate the mens rea and criminal liability, evidence of mental impairment is relevant at the penalty phase for broader purpose; namely, where it “might provide a basis for refusing to impose the death penalty on a particular individual, notwithstanding the severity of his crime or his potential to commit similar offenses in the future.” Abdul- Kabir v. Quarterman, 550 U.S. 233, 246 (2007); see also McKoy v. North Carolina, 494 U.S. 433, 440 (U.S. 1990) (“Relevant mitigating evidence is evidence which tends logically to prove or disprove some fact or circumstance which a fact-finder could reasonably deem to have mitigating value.” (quotation omitted)). In light of the broad scope of evidence relevant to mitigation at the penalty phase of a capital case, we have recognized that counsel’s duty to investigate possible psychological mitigation evidence is higher at the penalty phase than it might be during the guilt phase of trial: “At the penalty phase, counsel’s duty to follow up on indicia of mental impairment is quite different from—and much broader and less contingent than—the more confined guilt-phase responsibility.” Bemore v. Chappell, 788 F.3d 1151, 1171 (9th Cir. 2015).

We may conclude, arguendo, that based solely on the limited facts known to Clarke during his investigation— which included Clarke’s discussions with Washington and his friends and family, but not Washington’s education and incarceration records—Clarke’s decision not to seek a psychological evaluation was not objectively unreasonable. But our analysis does not end there. In a case such as this, the question under Strickland is “whether the investigation supporting [counsel’s] decision not to introduce mitigating evidence . . . was itself reasonable.” Id. (alterations in original) (quoting Wiggins v. Smith, 539 U.S. 510, 523 (2003)). WASHINGTON V. RYAN 17

As discussed above, Clarke performed unreasonably by not reviewing Washington’s education and incarceration records. We must therefore determine whether, if Clarke had performed reasonably in reviewing those records, he would have uncovered information that would have prompted him to obtain a psychological evaluation for Washington. See Rompilla v. Beard, 545 U.S. 374, 390–93 (2005). We conclude that an objectively reasonable lawyer would have done so. If Clarke had reviewed Washington’s education and incarceration records, he would have seen that Washington’s elementary school had placed Washington in classes for the educable mentally retarded and that the DOC had indicated that Washington had a low IQ. Although the Supreme Court had not yet recognized that the Eighth Amendment bars the execution of mentally retarded individuals at the time of Washington’s penalty trial, Atkins, 536 U.S. at 321, the Court had made clear that evidence of mental retardation was important mitigation evidence that should be presented at the penalty phase, Penry, 492 U.S. at 322–24 (noting that evidence of a defendant’s mental retardation may render him less culpable for his crime). A reasonable attorney would have investigated the potential that Washington may have been mentally retarded after reviewing Washington’s education and incarceration records.

In sum, although Clarke did not perform unreasonably by not requesting a psychological evaluation based on the evidence known to him at the time, his unreasonable failure to review Washington’s education and incarceration records prevented Clarke from gaining information that would have led him to request a psychological evaluation. See Rompilla, 545 U.S. at 390–93 (concluding that counsel erred by failing to review defendant’s prison file, which would have prompted counsel to further investigate potential 18 WASHINGTON V. RYAN

psychological mitigation evidence). As discussed below, these errors prevented Washington from presenting substantial mitigation evidence at the penalty phase.

Washington also argues that Clarke erred by not investigating and presenting evidence of his childhood abuse. Through his conversations with Dr. Roy, Washington revealed that he suffered physical abuse as a child in the form of daily whippings and beatings. Roy was also told that Washington was given alcohol as a child to control his behavior. None of this information later given to Dr. Roy had come to Clarke’s attention during the trial. By contrast, both psychological experts who testified at the PCR hearing agreed that Washington’s childhood was significantly dysfunctional. Unlike other categories of information that are easily verified through documentary evidence, Clarke had to rely entirely on the word of Washington and his family members in determining whether Washington suffered childhood abuse. Because neither Washington nor his family members had then indicated that Washington suffered abuse as a child, Clarke did not err by not further investigating Washington’s childhood abuse, to the extent that he could have, or by not presenting the information he did not have regarding abuse at the sentencing hearing. See Strickland, 466 U.S. at 691 (“[W]hen a defendant has given counsel reason to believe that pursuing certain investigations would be fruitless or even harmful, counsel’s failure to pursue those investigations may not later be challenged as unreasonable.”).

Finally, Washington alleges that Clarke erred by not investigating and presenting evidence of his substance abuse. Again, Clarke did not err by failing to investigate Washington’s substance abuse, because Clarke reasonably relied on his conversations with Washington and his friends WASHINGTON V. RYAN 19

and family, wherein substance abuse was not indicated. Washington had told Clarke that he was heavily intoxicated on the night of the crimes, but he denied that he had any ongoing problems with drugs or with alcohol. Similarly, Washington’s mother described him as someone who “liked to party,” but also did not say that Washington had problems with addiction. Perhaps the single clue Clarke had that might have raised his suspicions about substance abuse was the statement of Washington’s girlfriend that Washington had a “cocaine problem.” However, when set against Washington’s own statements and those of his family members, Clarke’s decision not to further investigate Washington’s drug abuse was not objectively unreasonable.

In summary, Clarke performed ineffectively by not reviewing Washington’s education and incarceration records. If Clarke had performed effectively, he would have known about and acted on information regarding Washington’s potentially impaired cognitive function. While we recognize that, as a general rule, deficient lawyer performance should be found only in exceptional cases presenting extraordinarily poor performance, we nonetheless conclude that the record here amply demonstrates that Clarke’s representation of Washington at the penalty phase was objectively unreasonable and deficient for the reasons articulated above.

Relief for ineffective assistance of counsel under Strickland requires both deficient performance in representation and prejudice. Even in light of Clarke’s performance, Washington can succeed on his ineffective assistance of counsel claim only if Clarke’s performance resulted in prejudice. To establish prejudice, Washington must show “a reasonable probability that, but for counsel’s 20 WASHINGTON V. RYAN

unprofessional errors, the result of the proceeding would have been different.” Strickland, 466 U.S. at 694. “A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Id. To determine whether a reasonable probability exists that the sentencing judge would not have imposed the death sentence in light of the mitigation evidence, we “reweigh the evidence in aggravation against the totality of the available mitigating evidence.” Wiggins, 539 U.S. at 534.

Here, as a result of Clarke’s performance, the sentencing court did not hear any evidence concerning Washington’s potentially impaired cognitive functions or his possible mental retardation. However, to establish that those omissions prejudiced him, Washington must show that there is a reasonable probability that he would have received a different sentence if those materials and evidence had been introduced at sentencing. Strickland, 466 U.S. at 694. Washington faces a significant obstacle in doing so, because we need not guess at whether the outcome would have been different if that evidence had been available to the court at sentencing. The sentencing judge, Judge Bradshaw, considered that evidence during the PCR proceeding and concluded in a written order that those materials would not have made a difference. Judge Bradshaw’s unequivocal ruling might ordinarily persuade us that Washington cannot show prejudice. The materials that Clarke missed, though they undoubtedly had mitigating value, were not so overwhelming that they influenced Judge Bradshaw’s no- prejudice finding.

However, the case of Washington’s co-defendant, Fred Robinson, casts a long shadow on our prejudice analysis WASHINGTON V. RYAN 21

here. 2 In Robinson v. Schriro, we considered Fred Robinson’s habeas corpus petition. 595 F.3d at 1098–99. That case is important here because Washington and Robinson were tried and sentenced together, and their convictions and sentences were affirmed in state court following joint PCR proceedings, in nearly identical written orders. Like Washington, Robinson alleged that he received ineffective assistance of counsel based on his trial counsel’s failure to present mitigation evidence at the penalty phase. Id. at 1108–10. Judge Bradshaw (like here) concluded that the mitigation evidence Robinson produced in the state PCR proceeding would not have made a difference. Yet, we determined that Judge Bradshaw, who presided at both Robinson’s and Washington’s trial and PCR proceedings, applied an unconstitutional “nexus test” in considering Robinson’s newly-presented mitigation evidence. 3

2 We recognize that the Supreme Court has said that each case assessing deficiency or prejudice must be judged on its own facts, in a separate determination, and that a broad general rule could not be used. See Strickland, 466 U.S. at 696 (“[I]n adjudicating a claim of actual ineffectiveness of counsel, a court should keep in mind that the principles we have stated do not establish mechanical rules.”). However, we also recognize that longstanding principles of justice demand that like cases be treated alike, see, e.g., H.L.A. Hart, The Concept of Law 159–166 (3d ed. 2012), and the relevant aspects of Robinson’s and Washington’s cases are substantially congruent for the purposes of our prejudice analysis. 3 Arizona’s nexus test required that mitigation evidence must have a direct or causal relationship with the crime itself. See State v. Djerf, 959 P.2d 1274, 1289 (Ariz. 1998). However, as we explained in Robinson, the nexus test is unconstitutional, and federal law is clear that a sentencing court must consider all mitigating evidence. See, e.g., Smith v. Texas, 543 U.S. 37, 45 (2004) (rejecting a Texas court’s refusal to consider mitigating evidence unless there was a “nexus” between the mitigating circumstance and the murder); Tennard v. Dretke, 542 U.S. 274, 283 (2004) (rejecting a Fifth Circuit test that barred the 22 WASHINGTON V. RYAN

Therefore, we concluded that Judge Bradshaw had not “properly evaluated the mitigating evidence offered in the evidentiary hearing,” 595 F.3d at 1113, and that the possibility that Judge Bradshaw would have imposed a sentence other than death if he had applied the correct standard when considering that evidence was enough to establish prejudice. Id.

Turning now to this case, Judge Bradshaw committed precisely the same error by imposing the nexus test when evaluating Washington’s mitigation evidence. Evidence of brain damage and mental disorders is important to the mitigation analysis. See, e.g., Daniels v. Woodford, 428 F.3d 1181, 1209 (9th Cir. 2005); Caro v. Woodford, 280 F.3d 1247, 1258 (9th Cir. 2002) (“More than any other singular factor, mental defects have been respected as a reason for leniency in our criminal justice system.”); Mitchell v. United States, 790 F.3d 881, 903–04 (9th Cir. 2015) (“Evidence that [Petitioner] was a chronic user of alcohol and drugs from a young age is the kind of ‘classic mitigating evidence’ that counsel must pursue at the penalty phase . . . .”). However, when evaluating Washington’s newly-presented evidence of diffuse brain damage, Judge Bradshaw discounted the value of the evidence because “[there was nothing] at the time of the offenses, which lessened his ability to differentiate right from wrong or to conform his actions with the law.” As we concluded in Robinson, Judge Bradshaw’s analysis erroneously demanded that the newly-presented evidence relate to Washington’s guilt for the charged offense, a standard that has been squarely rejected by the United States

consideration of mitigating evidence unless “the criminal act was attributable to [a] severe permanent condition”). WASHINGTON V. RYAN 23

Supreme Court. See, e.g., Smith, 543 U.S. at 45; Tennard, 542 U.S. at 283.

Thus, we must conclude here that there is a reasonable probability that the outcome of sentencing would have been different if the trial court had been presented with evidence of Washington’s cognitive defects and had properly evaluated that evidence consistently with the Supreme Court decision in Smith. As in Robinson, our confidence in the Arizona court’s imposition of the death sentence has been undermined, and we remand.

IV

Washington received ineffective assistance of counsel at the penalty phase and was prejudiced thereby. He is thus entitled to relief in the form of a new penalty phase trial. We reverse the district court’s denial of a writ of habeas corpus as to the penalty phase and remand with instructions to grant the writ of habeas corpus unless the state court undertakes resentencing proceedings within a reasonable time to be determined by the district court. 4

4 The majority’s reasoning throughout its opinion responds in substance to the contentions asserted by the dissent, and we do not address all of the dissent’s errors of reasoning and disregard of precedent here. However, we include the following points in brief response to the contentions of the dissent: First, the dissent without any proper basis asserts that the majority has stated the Strickland standard but then “might have applied” a different standard. Our majority opinion discusses and applies the Strickland standard. Second, the dissent relies on Judge Bradshaw’s conclusion that the new evidence would not have made a difference. But this ignores Supreme Court precedent that made clear the Arizona sentencing judge could not have constitutionally required a causal nexus between the mitigation evidence and the crimes. As explained supra at 20–22, that erroneous premise almost certainly 24 WASHINGTON V. RYAN

REVERSED and REMANDED for issuance of a writ of habeas corpus. 5

CALLAHAN, Circuit Judge, dissenting:

I dissent. A jury properly convicted Theodore Washington for the murder of Sterleen Hill and the attempted murder of Ralph Hill, and the trial court judge properly sentenced him to death. Washington’s claim of ineffective assistance of counsel presents no basis for vacating his sentence.

The majority errs in its application of both prongs of Strickland v. Washington, 466 U.S. 668 (1984). First, in second-guessing the performance of Theodore Washington’s trial counsel, the majority uses a standard for gross incompetence that doesn’t square with precedent. A post-conviction petition can always point to something that trial counsel should have done differently. Here, more than 30 years after Washington was sentenced to death, the majority grants Washington relief because his trial counsel failed to investigate Washington’s education and incarceration records. But trial counsel’s performance was

affected Judge Bradshaw’s assessment. Third, the dissent does not correctly apply the Supreme Court’s precedent in Eddings v. Oklahoma, 455 U.S. 104 (1982) (suggesting that the sentencer in a capital case cannot “refuse to consider, as a matter of law, any relevant mitigating evidence” offered by the defendant), and Smith v. Texas, 543 U.S. 37 (2004) (holding that there may not be a requirement of a causal nexus between mitigation evidence and the crime). 5 In a separate memorandum disposition filed simultaneously with this opinion, we affirm against all other issues raised by Washington. WASHINGTON V. RYAN 25

reasonable because his extensive discussions with Washington and Washington’s family and friends gave him no reason to suspect that those records contained helpful information.

Second, the majority doesn’t hold Washington to his heavy burden of showing prejudice. Instead, the majority erroneously presumes prejudice. The evidence that the majority concludes Washington’s trial counsel should have unearthed would not have fundamentally altered the narrative counsel competently (even if not perfectly) presented at sentencing. Indeed, we need not guess whether the new evidence would have made a difference at sentencing because the same judge who sentenced Washington to death (Judge Bradshaw) later presided over the post-conviction review proceedings. Judge Bradshaw unequivocally concluded the new evidence would not have made a difference. His “unique knowledge of the trial proceedings”—including his front-row seat to the presentation of evidence showing brutality of the execution- style murder of Sterleen Hill—“render[ed] him ‘ideally situated’” to evaluate Washington’s claim at post-conviction review. Murray v. Schriro, 882 F.3d 778, 818, 821 (9th Cir. 2018) (quoting Schriro v. Landrigan, 550 U.S. 465, 476 (2007)). There is no good reason for us to dismiss Judge Bradshaw’s conclusion from our lofty perch 25 years later.

I.

Before describing the lengthy procedural history of the case, the majority opinion briefly describes the home invasion turned execution-style murder. As this case wound its way through the courts, what has been lost is the cruelty and senselessness of the defendants’ acts. 26 WASHINGTON V. RYAN

The victims, Sterleen and Ralph Hill, were bound, and forced to lie face down on their bedroom floor in preparation to be shot, execution-style, with a shotgun. Sterleen was forced to listen helplessly as her husband was shot first and then wait as the shotgun was reloaded, knowing that she would be next. Had the Hills’ teenage son, LeSean, not run off, it is evident that he would have suffered the same fate. (Ralph testified he heard a voice in the background say, “We better get the kid.”). The murder and attempted murder appear to have been completely unnecessary to the completion of the robbery. It does not appear that the victims could have identified the defendants, and there was no sign of struggle. There simply was no need to kill.

The panel unanimously agrees that substantial evidence supports the jury’s finding that Washington was one of the men who carried out the execution-style murder and attempted murder of the Hills. The heinous nature of the crimes led the sentencing judge to impose the death penalty. The panel also agrees the state court’s application of the aggravating factors warranting the death sentence was proper.

II.

The principles underlying and governing a claim of ineffective assistance of counsel are familiar. But they bear repeating here because the majority strays from them. “The right to counsel is a fundamental right of criminal defendants; it assures the fairness, and thus the legitimacy, of our adversary process.” Kimmelman v. Morrison, 477 U.S. 365, 374 (1986). “[T]he right to counsel is the right to the effective assistance of counsel.” Strickland, 466 U.S. at 686 (quoting McMann v. Richardson, 397 U.S. 759, 771 n.14 (1970)). Under Strickland’s two-part test for claims of ineffective assistance of counsel, a convicted defendant must WASHINGTON V. RYAN 27

show (1) constitutionally deficient performance by counsel (2) that prejudiced the defense. Id. at 687.

“The essence of an ineffective-assistance claim is that counsel’s unprofessional errors so upset the adversarial balance between defense and prosecution that the trial was rendered unfair and the verdict rendered suspect.” Kimmelman, 477 U.S. at 374. “As is obvious, Strickland’s standard, although by no means insurmountable, is highly demanding.” Id. at 382; see also Padilla v. Kentucky, 559 U.S. 356, 371 (2010) (“Surmounting Strickland’s high bar is never an easy task.”). “Only those habeas petitioners who can prove under Strickland that they have been denied a fair trial by the gross incompetence of their attorneys will be granted the writ . . . .” Kimmelman, 477 U.S. at 382.

“When counsel focuses on some issues to the exclusion of others, there is a strong presumption that he did so for tactical reasons rather than through sheer neglect.” Yarborough v. Gentry, 540 U.S. 1, 8 (2003) (citing Strickland, 466 U.S. at 690). Even if inadvertence (not tactical reasoning) results in non-pursuit of a particular issue, “relief is not automatic. The Sixth Amendment guarantees reasonable competence, not perfect advocacy judged with the benefit of hindsight.” Id.

III.

The majority pays lip service to the rules placing an exacting burden on an inmate claiming ineffective assistance of counsel, but it doesn’t actually hold Washington to 28 WASHINGTON V. RYAN

Strickland’s “highly demanding” standard. See Kimmelman, 477 U.S. at 382. 1

A.

Washington failed to meet his burden on the first Strickland prong. I agree with the majority’s conclusion that much of the alleged deficiencies of counsel do not amount to gross incompetence. The majority is wrong, however, in concluding that Washington’s trial counsel, Robert Clarke, provided constitutionally deficient performance by failing to obtain and review Washington’s education and incarceration records.

First, the record does not support the majority’s implicit suggestion that the education records themselves contain meaningful mitigation evidence. The majority states that the sentencing court did not hear evidence concerning Washington’s “possible mental retardation.” Maj. Op. at 20. The only conceivable evidence for such a possibility is a notation in a 1965 education record (from when Washington was five years old) of the need for placement in special classes for the “educable mentally retarded.” But that single, decades-old notation is inconsequential when compared with more than ten additional years of schooling in the general population. And any suggestion that the school records showed a low IQ is contradicted by later IQ testing by Washington’s own expert, Dr. Roy. Washington has never even suggested the possibility of mental retardation. 1 The Supreme Court has on occasion corrected our court for reciting yet straying from Strickland. See, e.g., Wong v. Belmontes, 558 U.S. 15, 27–28 (2009) (reversing where the Ninth Circuit recited the correct standard for prejudice but its analysis “suggested it might have applied” a different standard). WASHINGTON V. RYAN 29

In short, the district court was correct in observing that Washington “presented no evidence that either his school records or his California incarceration records would have revealed potential mitigation.” 2

Second, even assuming it were proper to conclude so reflexively that the failure to obtain the education records was categorically incompetent, that conclusion alone wouldn’t justify the majority’s finding of incompetence. Instead, the majority’s holding necessarily requires a second layer of attorney deficiency. To the majority, if Clarke had reviewed Washington’s education records, he would have seen the 1965 notation about the need for placement in special classes. Upon seeing that notation, the majority reasons, all but the grossly incompetent lawyer would obtain a psychological evaluation—even absent any other indications from Washington’s life since kindergarten of subnormal mental capacity.

2 As for the incarceration records, Washington, through his able habeas counsel, points out that Clarke never obtained them, but he does not even assert that, let alone explain why, such a failure was constitutionally deficient. Nor does Washington indicate in his briefs what the incarceration records would have revealed. The majority doesn’t supply the missing reasoning or the missing description of mitigation evidence in the incarceration records. We have held that Strickland’s prejudice prong cannot be satisfied without “specification of the mitigating evidence that counsel failed to unearth.” Cox v. Del Papa, 542 F.3d 669, 681 (9th Cir. 2008). In the post-conviction review proceedings, there was a suggestion that the incarceration records might show good behavior. But Judge Bradshaw (who both sentenced Washington and presided over the post-conviction review proceedings) stated that he was already aware of Washington’s good behavior at the time of sentencing. 30 WASHINGTON V. RYAN

The majority is too quick to find gross incompetence. If Clarke had obtained the education records and seen the single notation from over twenty years earlier indicating a need for placement in special education classes, that notation would have been considered in context with everything else Clarke learned about Washington’s background. Clarke’s investigation included “extensive discussions” with Washington and Washington’s family and friends. Clarke asked Washington and his family members about whether Washington “had any propensity to violence,” “about his drug use,” “about his alcohol intake,” “about whether or not he was abused, growing up,” about “what discipline was like,” and “things of that nature.” For example, from his interviews, Clarke knew that Washington went to school in the general population and that he struggled in high school, dropping out in tenth or eleventh grade. 3 At the post- conviction review hearing, Clarke testified that, in all the interviews with Washington and his family, nothing triggered any red flags signaling that further investigation would have been fruitful. Clarke stated that he considered that Washington’s family members would make better witnesses than a psychologist who might examine Washington for a relatively brief period. The majority’s conclusion that only the grossly incompetent lawyer would not have obtained a psychological evaluation under these circumstances cannot be squared with Strickland’s deferential standard for determining the competence of counsel. See Yarborough, 540 U.S. at 8 (“The Sixth Amendment guarantees reasonable competence, not perfect advocacy judged with the benefit of hindsight.”).

3 Among the evidence Clarke presented at trial was testimony about Washington struggling in school and dropping out in tenth or eleventh grade. WASHINGTON V. RYAN 31

Under the deferential standard required by Strickland and its progeny, Clarke’s investigation was thorough and his performance was reasonable. If Clarke’s performance amounts to the “gross incompetence” habeas relief is reserved for, it’s doubtful many attorneys could withstand the second-guessing scrutiny of the majority’s approach for determining constitutional competence.

B.

“The 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, 466 U.S. at 686. Strickland “requires showing that counsel’s errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable.” Id. at 687. To prove prejudice, a defendant must show “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.” Id. at 694.

“It is not enough ‘to show that the errors had some conceivable effect on the outcome of the proceeding.’ Counsel’s errors must be ‘so serious as to deprive the defendant of a fair trial, a trial whose result is reliable.’” Harrington v. Richter, 562 U.S. 86, 104 (2011) (citations omitted) (quoting Strickland, 466 U.S. at 693, 687). Although the reasonable probability standard “does not require a showing that counsel’s actions ‘more likely than not altered the outcome,’ . . . the difference between Strickland’s prejudice standard and a more-probable-than- not standard is slight and matters ‘only in the rarest case.’” Id. at 111–12 (quoting Strickland, 466 U.S. at 693, 697); see 32 WASHINGTON V. RYAN

id. at 112 (“The likelihood of a different result must be substantial, not just conceivable.”).

To determine whether Washington has met his burden of showing prejudice, we must “reweigh the evidence in aggravation against the totality of available mitigating evidence.” Wiggins v. Smith, 539 U.S. 510, 534 (2003). This comparison cannot be made without first clearly identifying the evidence in mitigation that would have been offered at the penalty phase of trial but for counsel’s grossly incompetent performance. The majority concludes that Clarke’s incompetence resulted in the omission of “evidence concerning Washington’s potentially impaired cognitive functions.” 4 Maj. Op. at 20. By this, the majority presumably refers to Dr. Roy’s conclusion that Washington had symptoms of diffuse brain damage, likely caused by multiple head injuries while Washington was young. Dr. Roy further concluded that diffuse brain damage contributes to a “lack of judgment” and an “inability to establish stability in life.”

In reweighing this evidence, we must take as our baseline the evidence of aggravation and mitigation offered at trial and the resulting sentence. 5 After considering the details of

4 The majority also suggests the omitted evidence includes Washington’s “possible mental retardation.” Maj. Op. at 20. As discussed above, that possibility has been ruled out and Washington himself has never claimed (even now) a possibility of mental retardation.

5 The majority quotes Milton in suggesting that the sentencing judge has “the power ‘to temper justice with mercy.’” Maj. Op. at 5 n.1. Although that is valid as an abstract proposition, it tells us nothing about—and even obfuscates—the proper analysis under Strickland. Because a sentencing judge (or jury) has the power of leniency in every capital case, it’s always possible that, as the majority states, “[t]he WASHINGTON V. RYAN 33

the brutal, execution-style murder and attempted murder, and weighing it against the mitigation evidence Washington’s counsel presented, Judge Bradshaw sentenced Washington to death. With that starting point in mind, we must undertake the theoretical inquiry of determining whether it is reasonably likely that Washington would have received a different sentence if the new mitigation evidence were to be added to the mix of mitigation evidence that was presented at trial.

Of course, no guesswork is needed here. We know that Washington’s new evidence would not have made a difference because the sentencing judge said so. See Cook v. Ryan, 688 F.3d 598, 612 (9th Cir. 2012) (finding no prejudice where “the same trial judge who sentenced” the petitioner to death stated that the new evidence “would not have made any difference”). Judge Bradshaw “considered all of [the new] information in the post-conviction hearing and” definitively “held that none of it would have altered his judgment as to the proper penalty for” Washington. Gerlaugh v. Stewart, 129 F.3d 1027, 1036 (9th Cir. 1997).

A fair evaluation of the evidence in light of Supreme Court precedent confirms the soundness of Judge Bradshaw’s finding of no prejudice. Because of Strickland’s

sentencing judge might have decided that [the defendant] should be spared death and imprisoned for life.” Id. at 4–5. That is true even if the hypothetical second trial were a redo of the first without any new evidence. Of course, in engaging in the Strickland prejudice analysis, we have to control for that. In other words, because we are rejecting Washington’s other challenges to his sentence (i.e., he was properly convicted and sentenced to death), we must presume that he would have received the same sentence upon the same evidence. That is easier said than done; but we must do so to analyze properly whether Washington has met his burden of showing prejudice. 34 WASHINGTON V. RYAN

“highly demanding” standard, Kimmelman, 477 U.S. at 382, it’s no surprise that petitioners have historically found little success bringing ineffective assistance of counsel claims. However, beginning in 2000, the Supreme Court found Strickland’s “high bar” satisfied in four cases involving claims of ineffective assistance of counsel at the penalty phase of a capital trial. Williams v. Taylor, 529 U.S. 362 (2000); Wiggins, 539 U.S. 510; Rompilla v. Beard, 545 U.S. 374 (2005); Porter v. McCollum, 558 U.S. 30 (2009). These decisions serve as guideposts for determining when relief is warranted in such cases.

In Williams, the jury fixed the punishment at death after hearing evidence of a long history of criminal conduct including armed robbery, burglary and grand larceny, auto thefts, violent assaults on elderly victims, and arson. 529 U.S. at 368. At sentencing, defense counsel offered very little evidence. Id. at 369. 6 In addressing Williams’ Strickland claim, the Supreme Court cited “graphic” details “of Williams’ childhood, filled with abuse and privation,” evidence that Williams was “borderline mentally retarded,” and other significant mitigation evidence that was not unearthed only because of counsel’s deficient performance:

[C]ounsel did not begin to prepare for that phase of the proceeding until a week before the trial. They failed to conduct an investigation that would have uncovered

6 Counsel presented testimony from Williams’ mother and two neighbors who briefly described Williams as a “nice boy” and non- violent. Williams, 529 U.S. at 369. They also played a taped excerpt from a statement by a psychiatrist that merely related “Williams’ statement during an examination that in the course of one of his earlier robberies, he had removed the bullets from a gun so as not to injure anyone.” Id. WASHINGTON V. RYAN 35

extensive records graphically describing Williams’ nightmarish childhood, not because of any strategic calculation but because they incorrectly thought that state law barred access to such records. Had they done so, the jury would have learned that Williams’ parents had been imprisoned for the criminal neglect of Williams and his siblings, that Williams had been severely and repeatedly beaten by his father, that he had been committed to the custody of the social services bureau for two years during his parents’ incarceration (including one stint in an abusive foster home), and then, after his parents were released from prison, had been returned to his parents’ custody.

Id. at 395, 398 (citation and footnote omitted). In concluding Williams had shown prejudice, the Court noted that the same judge who presided over the criminal trial heard Williams’ post-conviction review claims. That trial judge, who initially “determined that the death penalty was ‘just’ and ‘appropriate,’ concluded that there existed ‘a reasonable probability that the result of the sentencing phase would have been different’” if evidence developed in the post- conviction proceedings had been offered at sentencing. Id. 396–97.

In Wiggins, trial counsel focused their strategy at sentencing on arguing that the defendant was not directly responsible for the murder, and they did not present any other mitigation evidence, despite knowledge of at least some of the defendant’s troubled background. 539 U.S. at 523–24. The Court cited “powerful” mitigation evidence that counsel either had, or should have, discovered. Id. 36 WASHINGTON V. RYAN

at 534–35. When Wiggins was a young child, his alcoholic mother frequently left him and his siblings home alone for days without food, “forcing them to beg for food and to eat paint chips and garbage.” Id. at 516–17. The mother beat Wiggins and his siblings and had sex with men while her children slept in the same bed. Id. at 517. On one occasion, the mother forced Wiggins’ hand against a hot stove burner, resulting in his hospitalization. Id. After being removed from his mother’s custody and placed in foster care, Wiggins was physically abused and “repeatedly molested and raped” by one foster father, and gang-raped on multiple occasions by a foster mother’s sons. Id. He ran away from one foster home and began living on the streets. Id. The Court held that had the jury been presented with Wiggins’ “excruciating life history,” rather than virtually no mitigation evidence, “there is a reasonable probability that at least one juror would have struck a different balance.” Id. at 537.

In Rompilla, trial counsel undertook a number of efforts to investigate possible mitigating evidence, “including interviews with Rompilla and some members of his family, and examinations of reports by three mental health experts who gave opinions at the guilt phase,” but none of these sources was helpful. 545 U.S. at 381. Notwithstanding these efforts, the Court found one “clear and dispositive” error by counsel. Id. at 383. Defense counsel knew the prosecution intended to seek the death penalty and would hinge its penalty case on Rompilla’s prior conviction for rape and assault. Id. Counsel nevertheless failed to even look at the court file for the prior conviction; had they done so “they would have found a range of mitigation leads that no other source had opened up.” Id. at 384, 390. The mitigation evidence that would have been available from simply looking at the files included, among other things: WASHINGTON V. RYAN 37

Rompilla’s parents were both severe alcoholics who drank constantly. His mother drank during her pregnancy with Rompilla, and he and his brothers eventually developed serious drinking problems. His father, who had a vicious temper, frequently beat Rompilla’s mother, leaving her bruised and black-eyed, and bragged about his cheating on her. His parents fought violently, and on at least one occasion his mother stabbed his father. He was abused by his father who beat him when he was young with his hands, fists, leather straps, belts and sticks. All of the children lived in terror. There were no expressions of parental love, affection or approval. Instead, he was subjected to yelling and verbal abuse. His father locked Rompilla and his brother Richard in a small wire mesh dog pen that was filthy and excrement filled. He had an isolated background, and was not allowed to visit other children or to speak to anyone on the phone. They had no indoor plumbing in the house, he slept in the attic with no heat, and the children were not given clothes and attended school in rags.

Id. at 391–92. All the evidence counsel failed to discover simply by failing to look at the court file of the prior conviction “add[ed] up to a mitigation case that bears no relation to the few naked pleas for mercy actually put before the jury.” Id. at 393. The Court thus concluded there was a reasonable probability of a different result had counsel performed adequately. Id. 38 WASHINGTON V. RYAN

In Porter, penalty phase counsel offered scant evidence on behalf of Porter. “The sum total of the mitigating evidence was inconsistent testimony about Porter’s behavior when intoxicated and testimony that Porter had a good relationship with his son.” Porter, 558 U.S. at 32. Post- conviction review proceedings revealed several facts about Porter’s “abusive childhood, his heroic military service and the trauma he suffered because of it, his long-term substance abuse, and his impaired mental health and mental capacity.” Id. at 33.

Porter routinely witnessed his father beat his mother, one time so severely that she had to go to the hospital and lost a child. Porter’s father was violent every weekend, and by his siblings’ account, Porter was his father’s favorite target, particularly when Porter tried to protect his mother. On one occasion, Porter’s father shot at him for coming home late, but missed and just beat Porter instead.

Id. Porter’s company commander in the Army also offered a “moving” account of Porter’s heroic efforts “in two of the most critical—and horrific—battles of the Korean War,” for which Porter “received two Purple Hearts and the Combat Infantryman Badge, along with other decorations.” Id. at 30, 34–35, 41. A neuropsychologist “concluded that Porter suffered from brain damage that could manifest in impulsive, violent behavior.” Id. at 36. The expert also testified that “[a]t the time of the crime . . . Porter was substantially impaired in his ability to conform his conduct to the law and suffered from an extreme mental or emotional disturbance,” which would have provided a basis for two statutory mitigating circumstances. Id. WASHINGTON V. RYAN 39

In concluding Porter established prejudice, the Court reasoned that “[t]he judge and jury at Porter’s original sentencing heard almost nothing that would humanize Porter or allow them to accurately gauge his moral culpability. They learned about Porter’s turbulent relationship with [the victim], his crimes, and almost nothing else.” Id. at 41. The Court emphasized the significance of Porter’s military service, both because “he served honorably under extreme hardship and gruesome conditions” and because “the jury might find mitigating the intense stress and mental and emotional toll that combat took on Porter.” Id. at 43–44.

Washington’s case has little in common with Williams, Wiggins, Rompilla and Porter. First, Porter is distinguishable because of the Court’s emphasis on the unique significance of military service in potentially mitigating against aggravating factors. See Porter, 558 U.S. at 43 (“Our Nation has a long tradition of according leniency to veterans in recognition of their service, especially for those who fought on the front lines as Porter did.”). Likewise, Rompilla is distinguishable because there is no analog here to the “dispositive” failure of trial counsel in Rompilla to look at the records that prosecution had indicated would serve as the basis for its case for the death penalty.

Second, although the evidence of Washington’s head injuries suggests a difficult childhood and perhaps provides a more complete picture of his background than was presented at trial, that evidence is not nearly as extreme as the mitigating evidence in the Supreme Court decisions. The head injuries and the suggested harsh discipline of Washington’s mother are not comparable to the outright beatings and criminal neglect of Williams’ parents, the starvation, neglect, physical abuse, molestation and rape, 40 WASHINGTON V. RYAN

and gang-rape Wiggins suffered at the hands of his mother and foster families, Rompilla being locked up with his brother “in a small wire mesh dog pen that was filthy and excrement filled,” deprived of clothing, and beaten by his alcoholic father, or the other harrowing facts in those cases. See Rhoades v. Henry, 638 F.3d 1027, 1051 (9th Cir. 2011) (“Even the more complete picture portrayed in the proffer of Rhoades’s dysfunctional family with its alcoholism, abuse, aberrant sexual behavior, and criminal conduct does not depict a life history of Rhoades himself that is nightmarish as it was for the petitioners in cases such as Rompilla, Wiggins, and Williams. . . .”).

Even if we didn’t have the benefit of Judge Bradshaw’s finding of no prejudice, in considering how Washington’s sentencing might have gone had his counsel presented the evidence that the majority concludes was unfairly omitted, I seriously doubt the sentencing judge would have elected not to impose the death penalty.

Although the majority acknowledges Washington’s heavy burden of showing prejudice, it doesn’t actually hold him to making such a showing. Nor does the majority even engage in the necessary reweighing of evidence. Instead, the majority effectively presumes prejudice, concluding that our decision in Robinson v. Schriro, 595 F.3d 1086 (9th Cir. 2010), casts such “a long shadow” that the question of whether Washington can show prejudice has already been decided. See Maj. Op. at 20–22. The majority perceives itself within the “shadow” of Robinson because Washington and Robinson were co-defendants, tried and sentenced together by the same judge, and the same judge denied their petitions for post-conviction review. But the sharing of a procedural history does not make two cases analogous. Only similarities on the issue in question—here, whether WASHINGTON V. RYAN 41

counsel’s performance was constitutionally deficient and whether any deficiency resulted in prejudice—can render cases analogous. On the issues of attorney competence and prejudice, the facts of Robinson starkly differ from the facts here.

Robinson’s trial counsel “engaged in virtually no investigation” and “did not call a single witness or introduce any evidence” at the sentencing hearing. Robinson, 595 F.3d at 1109. In contrast, here, Clarke investigated potential mitigation evidence by having “very extensive” discussions with Washington about his background and by interviewing—both before trial and after the verdict— Washington’s mother, brother, and common-law wife. Clarke also called three witnesses, each of whom offered testimony supporting a cogent narrative that Washington was friendly yet gullible, non-violent, and a loving father (and son) and that he desired to make something of his life.

In Robinson, the significance we placed on the utter failure of Robinson’s counsel cannot be overstated. For starters, we based our finding of prejudice on counsel’s non- performance because, under Arizona’s death penalty statute at the time of sentencing, the “failure to present a mitigation defense all but assured the imposition of a death sentence.” Robinson, 595 F.3d at 1111 (quoting Summerlin v. Schriro, 427 F.3d 623, 640 (9th Cir. 2005)). We also distinguished two Supreme Court cases—Bobby v. Van Hook, 558 U.S. 4 (2009) and Wong, 558 U.S. 15—on the basis that Robinson’s counsel failed to put on any mitigation evidence. Robinson, 595 F.3d at 1111 n.21 (stating that in both Bobby and Wong “defense counsel presented a significant amount of mitigating evidence”).

Because the utter failure of Robinson’s counsel all but compelled a conclusion of prejudice, the state’s best 42 WASHINGTON V. RYAN

argument perhaps was that the new evidence should be disregarded altogether because it lacked a “causal connection” to the crime. See id. at 1111–12. We rejected that argument based on Supreme Court precedent holding that evidence of a defendant’s background and mental capacity is relevant to mitigation and cannot be ruled inadmissible simply because the defendant fails to show a causal connection between the evidence and the crime. Id. at 1112; see Smith v. Texas, 543 U.S. 37, 45 (2004) (reaffirming the holdings of Eddings v. Oklahoma, 455 U.S. 104 (1982), and Tennard v. Dretke, 542 U.S. 274 (2004)).

There is a difference, of course, between the admissibility of evidence and the weight given to that evidence. Thus, although a court must allow a defendant to present any mitigation evidence (think Smith/Eddings/Tennard), “the failure to establish . . . a causal connection may be considered in assessing the quality and strength of the mitigation evidence,” State v. Newell, 132 P.3d 833, 849 (Ariz. 2006). See McKinney v. Ryan, 813 F.3d 798, 817–18 (9th Cir. 2015) (en banc) (referring to Newell’s rule as “proper[]”). The district court here recognized this difference. It “neither [mis]understood state law to preclude consideration of relevant proffered mitigation, nor to impose a minimum threshold before such mitigation could be considered.” Consistent with that (correct) view of the law, the district court understood Judge Bradshaw to have “considered the mitigation [evidence] proffered to show prejudice, but [Judge Bradshaw] determined that it carried insufficient weight to alter the sentence.” In my view, the district court correctly interpreted Judge Bradshaw’s decision and the majority unfairly assumes that Judge Bradshaw didn’t follow the law when nothing from his order compels such a conclusion. Judge Bradshaw stated that the information revealed from Washington’s psychological evaluation for WASHINGTON V. RYAN 43

the post-conviction review proceedings would not “have altered the sentence imposed.” 7 Notably, Washington’s able and zealous habeas counsel does not even contend Judge Bradshaw committed an Eddings error as to the psychological evidence. 8

Washington has not met his burden of showing prejudice under Strickland. That is, the omission of the new mitigation

7 In discussing Washington’s evidence of substance abuse, Judge Bradshaw concluded that the asserted drug and alcohol dependence did not affect Washington’s “ability to conform his actions to the demands of society.” The quoted language “echoes the causal nexus test of” Arizona’s (former) statutory mitigating factor for diminished capacity. See McKinney, 813 F.3d at 810; Ariz. Rev. Stat. § 13-703(G)(1) (2008). We held in McKinney that the causal nexus test applied in the context of this mitigating factor “does not violate Eddings.” McKinney, 813 F.3d at 810. Had Judge Bradshaw said nothing more, it could be inferred that he failed to consider Washington’s evidence for purposes of non- statutory mitigation. But Judge Bradshaw didn’t stop there; the very next sentence in his order shows that he in fact considered the evidence. He concluded that the evidence of substance abuse, considered alone or together with other mitigation evidence, would not “have mitigated against the sentence [Washington] has received.” The conclusion that the evidence of substance abuse lacked a causal nexus to the crime was thus appropriate because “a court is free to assign less weight to mitigating factors that did not influence a defendant’s conduct at the time of the crime.” Hedlund v. Ryan, 854 F.3d 557, 587 n.23 (9th Cir. 2017).

8 Washington’s only assertion of an Eddings error is his claim that Judge Bradshaw was wrong to conclude that he could not consider the reversal of James Mathers’ conviction as a mitigating factor. That contention is part of a claim we unanimously reject in a memorandum disposition filed concurrently with this opinion. Even if Judge Bradshaw failed to follow the law, it wouldn’t justify the majority’s de facto presumption of prejudice. An Eddings error by the post-conviction review court would mean, at most, that we give little or no weight to the state court’s conclusion on prejudice. Such an error would not eliminate the need to assess prejudice under Strickland. 44 WASHINGTON V. RYAN

evidence did not deprive Washington of “a fair trial,” see Strickland, 466 U.S. at 687, nor does the omission undermine my confidence that the trial “produced a just result,” see id. at 686.

* * *

Washington and his two co-defendants were convicted and sentenced to death for the murder of Sterleen Hill and the attempted murder of Ralph Hill. Anyone following this case is aware that one of Washington’s co-defendants had his conviction overturned and the other had his sentenced vacated. Under these circumstances, there is a temptation to bend the governing legal standards to equalize the outcomes for the three defendants in an effort “to achieve what appears a just result.” Holland v. Florida, 560 U.S. 631, 673 (2010) (Scalia, J., dissenting). However enticing the impulse, that is not our role. Ours is the duty to apply the law to determine whether Washington has met his high burden of showing a constitutional violation that deprived him of a fair trial. He has not.

Today’s decision is neither just nor faithful to Strickland’s standard. The majority “succumbs to the very temptation that Strickland warned against”—cherry-picking the record “to second-guess counsel’s assistance” through the “distorting” lens of “hindsight.” Rompilla, 545 U.S. at 408 (Kennedy, J., dissenting). The majority also discredits the sentencing judge’s own conclusion—based on his front-row seat at trial—that the evidence presented on post-conviction review would not have made a difference. Washington received a fair trial. The only injustice here is our unwarranted interference with Arizona carrying out the penalty lawfully chosen by the sentencing judge. We cannot bend the legal standards to “correct” Judge Bradshaw’s choice against leniency. Although Judge Bradshaw had the WASHINGTON V. RYAN 45

power to temper justice with mercy, in our role as federal appellate judges on habeas review, we do not.

I respectfully dissent.