FOR PUBLICATION
UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT
CHRISTOPHER J. SPREITZ, No. 09-99006
Petitioner-Appellant,
D.C. No.
v. 4:02-CV-00121-JMR
CHARLES L. RYAN, Warden,
Respondent-Appellee. OPINION
Appeal from the United States District Court
for the District of Arizona
John M. Roll, District Judge, Presiding
Argued July 11, 2013
Submitted March 4, 2019
San Francisco, California
Filed March 4, 2019
Before: Richard A. Paez, Marsha S. Berzon,
and Richard C. Tallman, Circuit Judges.
Opinion by Judge Paez;
Dissent by Judge Tallman
2 SPREITZ V. RYAN
SUMMARY *
Habeas Corpus / Death Penalty
The panel reversed the district court’s denial of habeas
corpus relief with respect to Christopher J. Spreitz’s death
sentence, and remanded, in a case in which Spreitz argued
that the Arizona Supreme Court violated Eddings v.
Oklahoma, 455 U.S. 104 (1982), by refusing to consider, as
a matter of law, mitigating evidence of Spreitz’s
longstanding alcohol and substance abuse on the ground that
he did not establish a causal connection between this
mitigating evidence and the crime.
The panel held that the district court erred in concluding
that Spreitz’s claim that the Arizona Supreme Court violated
Eddings is procedurally defaulted. The panel explained that
the first opportunity Spreitz had to raise that claim was
before the post-conviction-relief (PCR) court, at which time
he did so.
Because the decision of the PCR court – which first
declared the claim waived, but proceeded to adjudicate the
claim on the merits – was contrary to clearly established
Supreme Court precedent, the panel accorded that decision
no deference and reviewed Spreitz’s Eddings claim de novo.
The panel concluded that the Arizona Supreme Court
violated Eddings by impermissibly requiring that Spreitz
establish a causal connection between his longstanding
substance abuse and the murder before considering and
*
This summary constitutes no part of the opinion of the court. It
has been prepared by court staff for the convenience of the reader.
SPREITZ V. RYAN 3
weighing the evidence as a nonstatutory mitigating factor.
The panel concluded that the error was not harmless.
The panel affirmed the district court’s judgment denying
relief with respect to Spreitz’s conviction in a concurrently
filed memorandum disposition.
Dissenting, Judge Tallman wrote that the record does not
establish that either the sentencing court or the Arizona
Supreme Court unconstitutionally refused to consider
relevant mitigating evidence; and that even if the Arizona
courts did violate Eddings, Spreitz cannot show that this
error had a “substantial and injurious effect or influence” on
his ultimate sentence.
COUNSEL
Timothy M. Gabrielsen (argued), Assistant Federal Public
Defender; John M. Sands, Federal Public Defender; Office
of the Federal Public Defender, Tucson, Arizona; Susan B.
Fox and Sean Bruner, Law Office of Sean Bruner Ltd.,
Tucson, Arizona; for Petitioner-Appellant.
Lacey Stover Gard (argued) and Jeffrey A. Zick, Section
Chief Counsel; Kent E. Cattani, Chief Counsel, Criminal
Appeals/Capital Litigation Section; Mark Brnovich,
Attorney General; Office of the Attorney General, Tucson,
Arizona; for Respondents-Appellees.
4 SPREITZ V. RYAN
OPINION
PAEZ, Circuit Judge:
In 1994, an Arizona jury convicted Christopher J. Spreitz
(“Spreitz”) of first-degree murder. The victim was thirty-
nine year old Ruby Reid (“Reid”). Finding that the cruelty
of the murder outweighed any mitigating circumstances, the
trial judge sentenced Spreitz to death. Spreitz appeals the
district court’s denial of his petition for a writ of habeas
corpus challenging his conviction and sentence. We affirm
the district court’s judgment with respect to Spreitz’s
conviction, 1 and reverse with respect to his sentence. 2
In challenging his sentence, Spreitz argues that the
Arizona Supreme Court unconstitutionally affirmed his
death sentence by failing to consider mitigating evidence of
his longstanding alcohol and substance abuse. He contends
that the state court refused to consider, as a matter of law,
this evidence in mitigation because he did not establish a
causal connection between the crime and his long-term
alcohol and substance abuse. In Eddings v. Oklahoma,
455 U.S. 104, 110 (1982), the Supreme Court held that under
both the Eighth and Fourteenth Amendments, a sentencer in
a capital case may not “refuse to consider, as a matter of law,
1
We affirm the judgment with respect to Spreitz’s conviction in a
concurrently filed memorandum disposition.
2
On February 13, 2015, we vacated submission of Spreitz’s case
pending final resolution of the en banc proceedings in McKinney v. Ryan,
730 F.3d 903 (9th Cir. 2013). In December 2015, the en banc court
issued an opinion in McKinney. 813 F.3d 798 (9th Circ. 2015) (en banc).
After the Supreme Court denied the State’s petition for a writ of
certiorari, Ryan v. McKinney, 137 S. Ct. 39 (2016) (mem), we ordered
supplemental briefing on the applicability of McKinney to Spreitz’s case.
SPREITZ V. RYAN 5
any relevant mitigating evidence” offered by the defendant.
Id. at 114. Although a sentencer “may determine the weight
to be given relevant mitigating evidence . . . they may not
give it no weight by excluding such evidence from their
consideration.” Id. at 114–15 (footnote omitted). In
interpreting and applying Eddings, the Supreme Court has
explained that “full consideration of evidence that mitigates
against the death penalty is essential if the [sentencer] is to
give a reasoned moral response to the defendant’s
background, character, and crime.” Penry v. Lynaugh
(Penry I), 492 U.S. 302, 328 (1989), abrogated on other
grounds by Atkins v. Virginia, 536 U.S. 304 (2002) (internal
quotation marks and citation omitted). Moreover, the
Supreme Court has been clear: requiring a defendant to
prove a causal nexus between his mitigating evidence and
the crime is “a test we never countenanced and now have
unequivocally rejected.” Smith v. Texas, 543 U.S. 37, 45
(2004) (per curiam).
At the time of Spreitz’s sentencing, Arizona Revised
Statute Annotated § 13-703(G)(1994) 3 listed five mitigating
factors, and Arizona case law additionally recognized
nonstatutory mitigating factors, including, for example, a
defendant’s difficult family background or mental condition
not severe enough to qualify as a statutory mitigating factor.
In an en banc decision of our court, McKinney v. Ryan,
813 F.3d 798 (9th Cir. 2015), cert denied, 137 S. Ct. 39
(2016) (mem), we explained:
For a period of a little over 15 years in capital
cases, in clear violation of Eddings, the
3
Arizona has since revised its death penalty sentencing scheme. All
references to Arizona’s Revised Statute Annotated are to those
provisions in effect at the time of Spreitz’s sentencing.
6 SPREITZ V. RYAN
Supreme Court of Arizona articulated and
applied a “causal nexus” test for nonstatutory
mitigation that forbade as a matter of law
giving weight to mitigating evidence, such as
family background or mental condition,
unless the background or mental condition
was causally connected to the crime.
Id. at 802. As a result, we held in McKinney that
“[a]pplication of the causal nexus test to nonstatutory
mitigating factors violated Eddings, for it resulted in Arizona
courts being entirely forbidden, as a matter of state law, to
treat as a mitigating factor a family background or a mental
condition that was not causally connected to a defendant’s
crime.” Id. Spreitz argues that the Arizona Supreme Court
applied its causal nexus test in his case, refusing to consider
evidence of his long-term substance and alcohol abuse
because he did not adequately establish a causal connection
between that history of abuse and his crime.
As in McKinney, “the precise question before us is
whether the Arizona Supreme Court applied its
unconstitutional causal nexus test in affirming [Spreitz]’s
death sentence on de novo review.” 4 Id. at 804 (emphasis
4
The Arizona Supreme Court is required by statute to undertake an
independent review of a death sentence. Ariz. Rev. Stat. § 13-703.01(A).
In conducting that review for crimes that occurred before 2002, the court
“independently review[s] the trial court’s findings of aggravation and
mitigation and the propriety of the death sentence. In doing so, [the
court] review[s] the record de novo, considering the quality and the
strength, not simply the number, of aggravating and mitigating factors.”
State v. Lynch, 357 P.3d 119, 141 (Ariz. 2015) (internal quotation marks
and citations omitted), rev’d on other grounds, Lynch v. Arizona, 136 S.
Ct. 1818 (2016).
SPREITZ V. RYAN 7
and internal quotation marks omitted). For the reasons that
follow, we conclude that it did.
I.
A. Spreitz’s Crimes, Conviction, and Sentence
On May 25, 1989, the police arrested Spreitz after
discovering Ruby Reid’s body in the desert. Upon
questioning, Spreitz confessed to murdering Reid. We
briefly provide the facts of the murder.
On the evening of May 18, after drinking heavily and
being rejected by the woman he was dating, Spreitz “picked
up” Reid at a convenience store. State v. Spreitz (Spreitz I),
945 P.2d 1260, 1264–65 (Ariz. 1997). 5 In his confession,
Spreitz claimed that Reid voluntarily left with him and that
his understanding was that they would have sex later that
evening. Id. at 1265. Spreitz further claimed that he drove
her out to the desert, where Reid decided she no longer
wanted to have sex. Id. The two fought as a result. Id.
Spreitz explained that Reid slapped him and that he
responded by punching her in the mouth. Id. Spreitz then
sexually assaulted Reid—“remov[ing] her clothing and
ha[ving] vaginal intercourse with her.” Id. Spreitz also
recounted that he hit Reid in the head multiple times with a
rock “to make her stop yelling.” Id. He explained that he
left Reid without knowing whether she was alive or dead.
Id.
Shortly after leaving Reid in the desert, Spreitz was
stopped by a Tucson police department officer. Id. at 1264.
5
Our recitation of the facts is adopted from the Arizona Supreme
Court’s opinion affirming Spreitz’s conviction and death sentence.
8 SPREITZ V. RYAN
The officer observed that Spreitz had a ripped shirt, smelled
of feces, and appeared to be covered in blood and fecal
matter. Id. In addition, when detectives later searched
Spreitz’s car, they found blood spatter in the trunk, some of
which was inconsistent with Spreitz’s blood characteristics.
Id. at 1265.
On Monday morning, May 22, Reid’s naked and
decomposing body was discovered on the outskirts of
Tucson. Id. At trial, “the medical examiner testified that,
due to the advanced state of decomposition, he could not
determine the full extent and nature of [Reid]’s injuries.” Id.
Even so, he was able to observe “bruising on the legs, arms,
and back; bruising and abrasions on the buttocks; several
broken ribs; internal bleeding; a broken jaw; several head
lacerations; and a skull fracture where the skull had been
‘shoved in.’” Id. The medical examiner concluded that Reid
had been killed by “blunt-force trauma to the head.” Id.
In addition to finding Reid’s body “[a]t the scene, police
detectives observed tire tracks leading back to the pavement,
oil stains in the dirt, footprints, and drag marks in the dirt
leading away from the body. They also found feces-stained
pants, tennis shoes, socks, a used tampon, and a torn
brassiere. Two blood-stained rocks lay next to the body.”
Id. A few days later, police arrested Spreitz. Id.
On June 2, 1989, a grand jury indicted Spreitz for first-
degree murder, Ariz. Rev. Stat. Ann. §§ 13-1105, 13-703;
sexual assault, Ariz. Rev. Stat. Ann. §§ 13-406(A) & (B);
and kidnapping, Ariz. Rev. Stat. Ann. §§ 13-304(A)(3) &
(B). Spreitz I, 945 P.2d at 1265. After five years of pre-trial
proceedings mostly regarding the admissibility of DNA
evidence, a seven-day jury trial began on August 9, 1994.
Id. at 1266. After the conclusion of the trial, the jury
returned guilty verdicts on all three counts: first-degree
SPREITZ V. RYAN 9
murder (both premeditated and felony murder), sexual
assault, and kidnapping. Id.
Prior to both his aggravation-mitigation and sentencing
hearings before the trial judge, Spreitz submitted evidence
and a memorandum in support of certain mitigating
circumstances. As noted earlier, at the time of Spreitz’s
sentencing, Arizona’s death penalty statutes provided a list
of five specific mitigating factors; Arizona case law
recognized nonstatutory mitigating factors as well. See Ariz.
Rev. Stat. Ann. § 13-703(G); McKinney, 813 F.3d at 802.
Spreitz argued as nonstatutory mitigating factors: “(1) his
dysfunctional family life and lack of socialization; (2) a
history of alcohol and drug abuse; (3) his expressions of
remorse; [(4)] his good behavior while incarcerated; [(5)] his
lack of adult convictions; [and (6)] no prior record of violent
tendencies.” Spreitz I, 945 P.2d at 1279. Spreitz argued as
statutory mitigating factors: (1) his age at the time of the
murder, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and (2) that
his “capacity to appreciate the wrongfulness of his conduct
or to conform his conduct to the requirements of law was
significantly impaired [due to alcohol use], but not so
impaired as to constitute a defense to prosecution,” id.
§ 13-703(G)(1); see Spreitz I, 945 P.2d at 1279.
Spreitz provided evidence of and argued for all the
foregoing mitigating circumstances but focused heavily on a
combination of his relationship with his mother and his long
history of alcohol and substance abuse. To that end, Spreitz
submitted a written report by and presented testimony from
an examining psychologist, Dr. Todd Flynn, Ph.D. After
conducting interviews and research, Dr. Flynn concluded
that Spreitz’s longstanding alcohol and substance abuse
should be considered as both a statutory and nonstatutory
mitigating factor. In his report, which was admitted into
10 SPREITZ V. RYAN
evidence at the aggravation and mitigation hearing, Dr.
Flynn repeatedly emphasized Spreitz’s longstanding
substance abuse 6:
By age twelve or thirteen, Chris Spreitz
began drinking alcohol and smoking
marijuana. By age 15, he drank steadily on
weekends and would have a shot of vodka
before school.
The collateral information shows that the
alcohol abuse continued to intensify after he
left home. A variety of persons . . . described
him as a heavy drinker. This includes a
second cousin, Scott [Jouett], who saw him to
be intoxicated, “a majority of the time,” when
he was visiting Santa Barbara a week before
the current offense. To the interviewing
investigator, Mr. [Jouett] also described,
“several different occasions when Chris has
blackouts,” while drinking alcohol.
It appears completely clear from the available
information that Chris Spreitz had a long-
standing problem with alcohol which
probably reached the level of physical
dependence. He described himself as
drinking in the morning as early as age 15.
Virtually everyone else who spent much time
with him described him as a heavy drinker.
6
Dr. Flynn testified consistently with his report. For ease of
reference, we refer only to his report.
SPREITZ V. RYAN 11
Ultimately, Dr. Flynn concluded that Spreitz’s alcohol
abuse, childhood home life, and stunted development,
combined with rejection by his girlfriend on the day of the
crime, and intoxication at the time of the crime, led to the
murder.
At the end of his report, Dr. Flynn summarized his
findings and opinions with respect to both statutory and
nonstatutory mitigating factors. He opined that both
Spreitz’s age, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and
impaired capacity to appreciate the wrongfulness of his
conduct or conform his conduct to the law, Ariz. Rev. Stat.
Ann. § 13-703(G)(1), were mitigating statutory factors. As
to § 13-703(G)(5), specifically, Dr. Flynn opined that a
combination of a disturbing upbringing in a “pathogenic,
emotionally neglectful home environment,” and “[y]ears of
alcoholism intoxication” combined to cause major deficits in
Spreitz’s social and emotional development. As to
§ 13-703(G)(1), Dr. Flynn offered a similar conclusion. In
light of Spreitz’s “history of alcoholism . . . , a significant
but unknown degree of alcohol intoxication is likely” on the
night of the crime. In addition, Spreitz’s “history strongly
suggests years of early experiences likely to have caused a
build-up of pent-up angry, aggressive feelings toward
women generally (and older women especially) which may
have burst forth with uncontrollable intensity with or without
alcohol intoxication.” Dr. Flynn concluded that Spreitz’s
intoxication on the night of the crime coupled with his early
childhood experiences “likely . . . contributed to an
uncontrollable outburst of aggression” and inability to
control his conduct.
Dr. Flynn also concluded that certain nonstatutory
mitigating factors were present. The nonstatutory mitigating
factors included, in his opinion: Spreitz’s low potential for
12 SPREITZ V. RYAN
future violence, the “failure of [Spreitz’s] parents to provide
treatment for alcohol abuse in [his] teenage years,” and the
“emotionally deprived, physically punitive home
environment” of Spreitz’s upbringing. Dr. Flynn also
emphasized that even if both statutory mitigating factors
failed to satisfy the statutory threshold, they may still “be
appropriately considered . . . as nonstatutory mitigation.”
On November 28, 1994, the trial court conducted an
aggravation-mitigation hearing. At the hearing, Spreitz
called three mitigation witnesses: Dr. Flynn and two
correctional officers from the Pima County jail, where
Spreitz was incarcerated. The State did not offer any
witnesses or evidence at the hearing. Dr. Flynn testified
consistently with his report, as detailed above.
On the same date, a probation officer filed a Pre-
Sentence Report, which concluded:
It appears the defendant became involved in
the senseless commission of the instant
offense due to his alcohol and drug abuse.
After five years in custody, he now admits his
substance abuse problem; however, this does
not condone his involvement in the offense.
It is unfortunate the victim died before the
defendant had his revelation.
Prior to the sentencing hearing, Spreitz also submitted
several letters from friends, family, and jail personnel.
On December 21, 1994, the trial court conducted a
sentencing hearing. Spreitz, Spreitz’s counsel, the victim’s
SPREITZ V. RYAN 13
sister, and the prosecutor each addressed the court. 7 The
prosecutor first disputed the mitigating evidence presented
by Spreitz at the mitigation-aggravation hearing. With
respect to Spreitz’s alcohol abuse, the prosecutor argued that
Spreitz’s intoxication at the time of the crime did not meet
the statutory definition for mitigation under § 13-703(G)(1)
because his conduct both during the murder and afterwards
demonstrated that “he knew what he was doing.” The
prosecutor emphasized that the evidence revealed that Reid
had been forcibly abducted and stuffed into the trunk of a
car, thus forcing her to spend time “contemplat[ing] the
uncertainty of her fate.” The prosecutor argued that the
evidence revealed signs of struggle and that the serious
injuries to Reid’s body belied any notion that Spreitz did not
know what he was doing. The prosecutor further argued that
the fact that Reid defecated on herself revealed that she was
terrified prior to her murder. In sum, the prosecutor argued
that given both the heinous nature of the crime and the
manner in which Reid suffered, a finding in aggravation that
Reid was murdered in an “especially cruel manner” was
warranted.
After a short recess, the sentencing judge rendered oral
findings addressing the aggravation and mitigation issues.
He then imposed a sentence of death. 8 Following the
7
The trial court did not entertain argument at the conclusion of the
aggravation-mitigation hearing. As a result, counsel addressed the
aggravation-mitigation evidence and sentence options at the subsequent
sentencing hearing.
8
Ring v. Arizona, 536 U.S. 584 (2002), the Supreme Court decision
holding judge-sentencing in capital cases unconstitutional, had yet to be
decided.
14 SPREITZ V. RYAN
hearing, the sentencing judge filed a judgment setting forth
written findings.
The judge first found one aggravating circumstance—
that the offense was committed in an “especially cruel
manner.” Ariz. Rev. Stat. Ann. § 13-703(F)(6). The judge
proceeded to review the “many factors [that were] submitted
in mitigation,” ultimately finding five mitigating
circumstances of various levels of significance:
[1] A mitigating circumstance defense
submitted was that the defendant had an
extremely disruptive childhood . . . . The
court finds the home was sub-normal, not
even a minimally healthy one for developing
children; and it is obvious the defendant
suffered a disruptive middle childhood—had
a punitive, controlling, cold mother, who he
could not please, no matter what he did.
The defendant in his life turned to substance
abuse—alcohol and some suggestion he was
using cocaine and other drugs. However, the
court does not find such is a mitigating
circumstance that impaired his ability to
make a judgment on whether he was acting
rightfully or wrongfully in the death of the
victim.
The defendant’s history of intoxication is
longstanding. He had been abusing
substances for close to ten years of his life at
the time of this offense when he was twenty-
two. Again, the court does not believe that
the substance abuse or intoxication impaired
SPREITZ V. RYAN 15
the defendant’s ability and capacity to
appreciate the wrongfulness of his conduct to
any significant degree. . . . The court does not
believe intoxication is any sort of mitigating
circumstance.
...
[2] The court acknowledges that the
defendant has begun a process of
improvement and emotional growth while
confined at the Pima County Jail, where he
has taken part in education programs.
Correctional officers have testified he was a
prisoner who caused no problems.
[3] The age of the defendant at the time of the
offense (twenty-two) is not a mitigating
circumstance in and of itself. Immaturity
probably is, but the court does not believe
immaturity was a significant mitigating
circumstance.
[4] The court finds that Mr. Spreitz has no
criminal history of a felony nature—there is
no history or propensity for acts of violence.
[5] The court believes the defendant is
capable of being rehabilitated. The court
does not know whether he has a good
prognosis for the future, but the court
believes he can be rehabilitated.
In conclusion, the judge found that “the mitigating
circumstances [were] not sufficient to balance the
16 SPREITZ V. RYAN
aggravating circumstances, nor [were] they sufficiently
substantial to call for leniency.” He thus imposed a death
sentence as to the first-degree murder conviction.
B. Spreitz’s Appeal and Post-Conviction Proceedings
Spreitz appealed his conviction and sentence to the
Arizona Supreme Court with the assistance of new counsel.
The Arizona Supreme Court affirmed Spreitz’s conviction
and sentence after conducting its own independent review of
the record. 9 In reviewing the mitigating evidence, the court
found four of the five mitigating circumstances that the
sentencing judge had found, declining to find Spreitz’s good
behavior while in jail awaiting sentencing as a mitigating
circumstance. Spreitz I, 945 P.2d at 1280–81. The court
explained:
[1] “We agree with the sentencing judge that
defendant’s upbringing was subnormal.” Id.
at 1280.
[2] “We also find that the sentencing judge
. . . properly found that his emotional
immaturity was not a significant mitigating
factor.” Id. at 1281.
[3] “We agree that the record supports the
sentencing judge’s findings that defendant
had no previous adult felony convictions, no
prior acts of violence, and
9
See note 4, supra. Because the Arizona Supreme Court conducted
a de novo review, we focus on its analysis, rather than, as the dissent
does in part, the analysis by the trial court.
SPREITZ V. RYAN 17
[4] that the defendant is capable of
rehabilitation.” Id.
The Arizona Supreme Court also added that it considered
remorse as an additional nonstatutory mitigating factor. Id.
The court addressed Spreitz’s history of alcohol and
substance abuse as follows:
The record demonstrates defendant’s
longtime substance abuse problems. We
note, however, that defendant’s general
problems with substance abuse are not
essential to our decision here. We therefore
decline to conclude that defendant was
impaired by alcohol consumption to an extent
that it interfered with his “capacity to
appreciate the wrongfulness of his conduct or
to conform his conduct to the requirements of
the law.” A[riz]. R[ev]. S[tat]. [Ann.]
§ 13-703(G)(1); see also State v. Medrano,
185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)
(citing [State v.] Stokley, 182 Ariz. [505,]
520, 898 P.2d [454,] 469 [1995]).
Id. at 1280–81. In the only other portion of its opinion
addressing Spreitz’s history of alcohol and substance abuse,
the court said:
The sentencing judge found that defendant’s
ability to appreciate the wrongfulness of his
conduct was not impaired on the night of the
murder to any significant extent by substance
abuse, emotional disorders, situational stress,
or by a combination of these. Our review of
18 SPREITZ V. RYAN
the record convinces us that the trial court’s
finding was proper.
Id. at 1281.
The Arizona Supreme Court agreed with the trial court’s
aggravation finding that the murder was done in an
especially cruel manner, id. at 1279, and then proceeded to
reweigh the applicable aggravating and mitigating factors,
finding that “the aggravating circumstance of especial
cruelty in defendant’s murder of Ruby Reid outweigh[ed] all
factors mitigating in favor of leniency.” Id. at 1282. It
affirmed his death sentence. Id. at 1283. Spreitz filed a
petition for a writ of certiorari, which the United States
Supreme Court denied. Spreitz v. Arizona, 523 U.S. 1027
(1998) (mem).
In March 2000, represented again by new counsel,
Spreitz filed a petition for post-conviction relief in the
Arizona Superior Court (“PCR court”). See Ariz. R. Crim.
P. 32.1. Spreitz alleged, inter alia, that the sentencing judge
and the Arizona Supreme Court committed error by failing
to consider his history of alcohol and substance abuse as a
nonstatutory mitigating factor apart from its causal
connection to the murder—i.e., Eddings error. He also
alleged that his appellate counsel was ineffective for failing
to raise the sentencing judge’s Eddings error on appeal to the
Arizona Supreme Court. Spreitz additionally raised several
new claims of trial-counsel-ineffectiveness.
The PCR court denied all of Spreitz’s claims and entered
an order dismissing his petition for post-conviction relief. In
Part 3 of its order, the PCR court discussed Spreitz’s claims
of “nexus-error”—that both the trial court and the Arizona
Supreme Court erred when they failed to consider Spreitz’s
longstanding alcohol and substance abuse on the basis that
SPREITZ V. RYAN 19
Spreitz had failed to establish a causal nexus between the
long-term substance abuse and the murder. 10 The PCR court
dismissed the claim 11 as waived, reasoning that Spreitz had
failed to raise the issue on direct appeal. Nonetheless, the
PCR court addressed the merits of the nexus-error claim in
the course of analyzing Spreitz’s argument that his appellate
counsel was ineffective for failing to raise it. In doing so,
the PCR court explained that the claim failed because:
[I]t must be demonstrated, under A[riz].
R[ev]. S[tat]. [Ann.] § 13-703(G)(l), that
there is a causal link between the history of
alcohol or substance abuse and the offense
itself. E.g., State v. Stokley, [] 182 Ariz.
[505,] 523 [898 P.2d 454, 472 (Ariz. 1995)].
Without some basis for explaining or
defining the individual’s behavior at the time
of the offense, the Petitioner’s history of
alcohol or substance abuse would be
inconsequential (which is exactly what the
trial court and Supreme Court concluded).
State v. Kayer, 194 Ariz. 423, 984 P.2d 81
([Ariz.] 1999).
At times, the court can and should consider
an individual’s long-term alcoholism and
10
For ease of reference, we refer to the Arizona courts’ alleged error
as “nexus-error.”
11
Although Spreitz raised, and the PCR court recognized, two
distinct claims—one with respect to the sentencing court and one with
respect to the Arizona Supreme Court—the PCR court’s analysis treats
the claims as one and the same and refers to them in the singular. To
reflect accurately the PCR court’s discussion, our summary here likewise
refers to a singular “claim.”
20 SPREITZ V. RYAN
substance abuse, usually in conjunction with
other factors or diagnosis, as non-statutory
mitigation. However, the impact or effect of
the alcoholism or substance abuse must be
substantial and of such severity that it
provides a sufficient basis for explaining the
defendant’s conduct, character or ability to
control his behavior at the time of the offense.
[Citations omitted].
As previously discussed, there is no evidence
in Petitioner’s case to suggest that he suffered
any long-term effects from his alcohol or
drug abuse that precluded him from
controlling his behavior. Petitioner did not
suffer from any cognitive or emotional
deficits that rendered him incapable of
controlling his conduct. Therefore, the trial
court did not err in failing to find Petitioner’s
history of alcohol or substance abuse as a
separate, non-statutory mitigating factor.
[Citation omitted].
Spreitz filed a petition for review in the Arizona
Supreme Court challenging the PCR court’s judgment. The
Arizona Supreme Court summarily affirmed the PCR court’s
merits determination with respect to Spreitz’s Eddings
claims. State v. Spreitz (Spreitz II), 39 P.3d 525, 527 (Ariz.
2002).
Spreitz filed his federal habeas petition in February
2003. He alleged in claim seven that the sentencing judge
and the Arizona Supreme Court had both committed nexus-
error with respect to his long-time alcohol and substance
abuse, and that appellate counsel was ineffective for failing
SPREITZ V. RYAN 21
to argue the sentencing judge’s nexus-error on direct appeal
to the Arizona Supreme Court.
The district court denied all of Spreitz’s claims. With
respect to the alleged nexus-error by both the sentencing
court and the Arizona Supreme Court, the district court
found the claims procedurally barred, relying on the PCR
court’s determination that Spreitz had waived the claims
because he could have raised them on direct appeal but failed
to do so. The district court concluded that Spreitz had
properly exhausted his ineffective assistance of appellate
counsel claim, and thus addressed the alleged nexus-error in
that context, ultimately concluding it was not meritorious.
Accordingly, the district court concluded that “appellate
counsel’s failure to raise this issue on appeal does not
constitute ineffectiveness.”
Spreitz timely appealed.
II. Jurisdiction and Standard of Review
We have jurisdiction pursuant to 28 U.S.C. §§ 1291
and 2253(a), and review de novo the district court’s denial
of a writ of habeas corpus. Poyson v. Ryan, 879 F.3d 875,
887 (9th Cir. 2018). Because Spreitz filed his federal habeas
petition after April 24, 1996, he must satisfy the
Antiterrorism and Effective Death Penalty Act (“AEDPA”).
Under AEDPA, we may not grant habeas relief unless the
state’s adjudication of Spreitz’s claim (1) “was contrary to
. . . clearly established federal law[] as determined by the
Supreme Court,” (2) “involved an unreasonable application
of” such law, or (3) “was based on an unreasonable
determination of the facts in light of the evidence presented
in the State court proceeding.” (28 U.S.C. § 2254(d)). “In
making this determination, we look to the last state court
decision to address the claim,” White v. Ryan, 895 F.3d 641,
22 SPREITZ V. RYAN
665 (9th Cir. 2018) (citing Wilson v. Sellers, 138 S. Ct. 1188,
1192 (2018)), which for Spreitz’s nexus-error claim is that
of the PCR court.
Spreitz argues that the PCR court’s decision was
“contrary to” Eddings v. Oklahoma, 455 U.S. 104 (1982).
“A state court’s decision is contrary to clearly established
federal law if it ‘applies a rule that contradicts the governing
law set forth in [U.S. Supreme Court] cases’ or arrives at a
different result in a case that ‘confronts a set of facts that are
materially indistinguishable from a decision of [the
Supreme] Court.’” Castellanos v. Small, 766 F.3d 1137,
1146 (9th Cir. 2014) (alteration in original) (quoting
Williams v. Taylor, 529 U.S. 362, 405–06 (2000)). “If the
state court applies a legal standard that contradicts clearly
established [Supreme Court] law, we review de novo the
applicant’s claims, applying the correct legal standard to
determine whether the applicant is entitled to relief.” Id.
(citation omitted).
III. Discussion
As discussed supra, the precise question we must decide
is whether the Arizona Supreme Court applied its
unconstitutional causal nexus test in violation of Eddings
when it affirmed Spreitz’s death sentence. To answer that
question, we must first determine whether the claim is
properly before us. After concluding that Spreitz’s claim
could not have been procedurally defaulted, we turn to the
level of deference we must accord the PCR court’s ruling on
Spreitz’s Eddings claim under AEDPA. Because the PCR
court’s decision was contrary to clearly established Supreme
Court precedent, we accord that decision no deference and
review Spreitz’s Eddings claim de novo. We conclude that
the Arizona Supreme Court violated Eddings by
impermissibly requiring that Spreitz establish a causal
SPREITZ V. RYAN 23
connection between his longstanding substance abuse and
the murder before considering and weighing the evidence as
a nonstatutory mitigating factor. Finally, having determined
that the Arizona Supreme Court applied an unconstitutional
causal nexus test in its sentencing procedure, we turn to
whether the error was harmless under Brecht v. Abrahamson,
507 U.S. 619 (1993). We conclude that the error was not
harmless and therefore reverse the district court’s judgment
with respect to Spreitz’s sentence.
A.
Under the doctrine of procedural default, a federal
habeas court “will not review the merits of claims, including
constitutional claims, that a state court declined to hear
because the prisoner failed to abide by a state procedural
rule.” Martinez v. Ryan, 566 U.S. 1, 9 (2012). “We review
de novo a district court’s conclusion that a claim is
procedurally defaulted.” Cooper v. Neven, 641 F.3d 322,
326 (9th Cir. 2011).
Here, the district court concluded that both of Spreitz’s
Eddings claims—that the sentencing court and the Arizona
Supreme Court applied an unconstitutional causal nexus
test—were procedurally defaulted, because the PCR court
found each claim waived for failure to raise them on direct
appeal. This conclusion, however, is erroneous with respect
to Spreitz’s claim that the Arizona Supreme Court violated
Eddings. After all, “[t]he Arizona Supreme Court reviews
capital sentences de novo, making its own determination of
what constitute[s] legally relevant aggravating and
mitigating factors, and then weighing those factors
independently. Ariz. Rev. Stat. Ann. § 13-755. The Arizona
Supreme Court ‘conducts a thorough and independent
review of the record and of the aggravating and mitigating
evidence to determine whether the sentence is justified.’”
24 SPREITZ V. RYAN
McKinney, 813 F.3d at 819 (emphasis removed) (quoting
State v. McKinney, 917 P.2d 1214, 1225 (Ariz. 1996)).
Spreitz could not have raised on direct appeal his claim that
the Arizona Supreme Court violated Eddings in performing
its de novo review of Spreitz’s death sentence. The first
opportunity he had to raise that claim was before the PCR
court, at which time he did so. Thus, Spreitz’s claim is not
procedurally defaulted and is properly before us. 12
B.
Because Spreitz’s claim that the Arizona Supreme Court
applied an unconstitutional causal nexus test to his
nonstatutory mitigating evidence of long-time alcohol and
12
The district court declined to issue a certificate of appealability
(“COA”) on Spreitz’s substantive Eddings claim, which he raised as
claim 1.7 on pages 108 through 111 in his federal habeas petition.
Although Spreitz did not present his Eddings claim as a separate
uncertified issue, we may grant a COA regarding any uncertified issue
discussed in a petitioner’s opening brief. See 9th Cir. R. 22-1(e).
Spreitz’s opening brief discusses the Arizona courts’ Eddings errors. We
therefore may exercise our discretion to expand the COA to encompass
the Eddings claim.
To obtain a COA under 28 U.S.C. § 2253(c), Spreitz must
demonstrate that “reasonable jurists could debate whether . . . the
petition should have been resolved in a different manner or that the issues
presented were adequate to deserve encouragement to proceed further.”
Slack v. McDaniel, 529 U.S. 473, 484 (2000). Reasonable jurists could
debate the merits of Spreitz’s Eddings claim. We thus exercise our
discretion and expand the COA to encompass Spreitz’s substantive
Eddings claim with respect to the Arizona Supreme Court. We
acknowledge that under Ninth Circuit Rule 22-1(f), the State would
normally be afforded an opportunity to submit supplemental briefing on
the uncertified Eddings issue. Here, however, the Eddings issue has been
exhaustively briefed, in the context of Spreitz’s appellate counsel’s
alleged ineffectiveness before the Arizona Supreme Court. Additional
briefing is unwarranted.
SPREITZ V. RYAN 25
substance abuse is properly before us, we next determine the
level of deference to accord the PCR court’s adjudication of
that claim. As discussed above, although the PCR court first
declared the claim waived, it proceeded to adjudicate the
claim on the merits. In denying the claim, the PCR court
explained:
Without some basis for explaining or
defining the individual’s behavior at the time
of the offense, [Spreitz]’s history of alcohol
or substance abuse would be inconsequential
(which is exactly what the trial court and
Supreme Court concluded).
At times, the court can and should consider
an individual’s long-term alcoholism and
substance abuse, usually in conjunction with
other factors or diagnosis, as non-statutory
mitigation. However, the impact or effect of
the alcoholism or substance abuse must be
substantial and of such severity that it
provides a sufficient basis for explaining the
defendant’s conduct, character, or ability to
control his behavior at the time of the offense.
....
As previously discussed, there is no evidence
in [Spreitz]’s case to suggest that he suffered
any long-term effects from his alcohol or
drug abuse that precluded him, from
controlling his behavior. Petitioner did not
suffer from any cognitive or emotional
deficits that rendered him incapable of
controlling his conduct. Therefore, the trial
26 SPREITZ V. RYAN
court did not err in failing to find [Spreitz]’s
history of alcohol or substance abuse as a
separate, non-statutory mitigating factor.
In other words, the PCR court concluded that Spreitz’s
longstanding alcohol and substance abuse could only be
considered a nonstatutory mitigating factor if it “provide[d]
a sufficient basis for explaining the defendant’s conduct,
character, or ability to control his behavior at the time of the
offense.” This discussion both accurately described the
causal nexus test and approved of it.
Such reasoning is contrary to clearly established federal
law. As explained in McKinney, “the causal nexus test
clearly violates Eddings.” 813 F.3d at 810. Therefore,
“[b]ecause the state court used the wrong standard, we need
not defer to that decision.” Hardy v. Chappell, 849 F.3d 803,
820 (9th Cir. 2016).
C.
Because the PCR court’s decision was “contrary to”
clearly established law, we review de novo the merits of
Spreitz’s Eddings claim. 28 U.S.C. § 2254(d)(1). We
therefore consider whether the Arizona Supreme Court
applied its unconstitutional causal nexus test in reviewing
and affirming Spreitz’s death sentence. In so doing, “we
look only to the decision of th[e Arizona Supreme Court],
. . . only [considering] . . . the decision of the sentencing
judge . . . to the degree it was adopted or substantially
incorporated by the Arizona Supreme Court.” McKinney,
813 F.3d at 819.
SPREITZ V. RYAN 27
1.
Our decision in McKinney frames our consideration of
the Arizona Supreme Court’s decision and Spreitz’s claim.
McKinney makes clear that the Arizona Supreme Court
consistently articulated and applied its unconstitutional
causal nexus test during the period in which Spreitz’s death
sentence was litigated in the trial court and reviewed by the
Arizona Supreme Court. Id. at 824. Moreover, McKinney
makes plain that although on habeas review, we generally
apply a “presumption that state courts know and follow the
law,” “the Arizona Supreme Court’s consistent articulation
and application of its causal nexus test . . . make such a
course impossible.” Id. at 803 (internal quotation marks and
citation omitted). In appeals heard from the late 1980s until
at least 2002, see State v. Canez, 42 P.3d 564, 594 (Ariz.
2002), it was unmistakably clear that “the Arizona Supreme
Court did not know and follow federal law.” Id. (internal
quotation marks omitted). 13 Therefore, just as in McKinney
and subsequent cases applying McKinney, see, e.g., Poyson,
879 F.3d at 889, “the presumption” that the Arizona
Supreme Court knew and followed the law “is rebutted” here
as well. McKinney, 813 F.3d at 804.
That said, McKinney does not dispose of Spreitz’s claim.
We must still review the Arizona Supreme Court’s decision
to determine whether it did, in fact, apply its unconstitutional
causal nexus test in Spreitz’s case. In doing so, we are
mindful of the fact that at the time the Arizona Supreme
Court decided Spreitz’s appeal, “if there is to be a
presumption, it is that the Arizona Supreme Court violated
13
In 2005, the Arizona Supreme Court repudiated its earlier nexus
test. See State v. Anderson, 111 P.3d 369, 391–92 (Ariz. 2005).
28 SPREITZ V. RYAN
the dictates of Lockett and Eddings.” Greenway v. Ryan,
866 F.3d 1094, 1095 (9th Cir. 2017).
2.
Against this backdrop, we are convinced that the Arizona
Supreme Court applied its causal nexus test with respect to
Spreitz’s evidence of longstanding alcohol and substance
abuse in violation of Eddings.
We quote again the Arizona Supreme Court’s discussion
of Spreitz’s history of alcohol and substance abuse:
The record demonstrates defendant’s
longtime substance abuse problems. We
note, however, that defendant’s general
problems with substance abuse are not
essential to our decision here. We therefore
decline to conclude that defendant was
impaired by alcohol consumption to an extent
that it interfered with his “capacity to
appreciate the wrongfulness of his conduct or
to conform his conduct to the requirements of
the law.” A[riz]. R[ev]. S[tat]. [Ann.] § 13-
703(G)(1); see also State v. Medrano,
185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)
(citing Stokley, 182 Ariz. at 520, 898 P.2d at
469).
....
The sentencing judge found that defendant’s
ability to appreciate the wrongfulness of his
conduct was not impaired on the night of the
murder to any significant extent by substance
abuse, emotional disorders, situational stress,
SPREITZ V. RYAN 29
or by a combination of these. Our review of
the record convinces us that the trial court’s
finding was proper.
Spreitz I, 945 P.2d at 1280–81. This discussion
demonstrates that the Arizona Supreme Court primarily
concluded that Spreitz failed to show statutory mitigation
under Ariz. Rev. Stat. Ann. § 13-703(G)(1) based on his
intoxication on the night of the murder. “When applied
solely in the context of statutory mitigation under
§ 13-703(G)(1), the causal nexus test does not violate
Eddings.” McKinney, 813 F.3d at 810.
The plain text of the Arizona Supreme Court’s decision,
however, also clearly demonstrates that the court did not
solely apply its causal nexus test to Spreitz’s evidence of
statutory mitigation. The court initially recognized Spreitz’s
“general problems with substance abuse”—evidence that
should have been relevant as nonstatutory mitigation—but
concluded those problems were “not essential to [its]
decision” because they did not show Spreitz was “impaired
on the night of the murder.” Spreitz I, 945 P.2d at 1280–81.
In other words, the court did not acknowledge the relevance
of longtime substance abuse as nonstatutory mitigation in the
absence of a causal relationship to the crime. Instead, the
court held that because there was no causal relationship,
Spreitz’s long-term alcohol abuse was not of significance—
i.e. “not essential”—to the court’s overall determination of
either statutory or nonstatutory mitigation.
The Supreme Court, however, has been clear that “‘full
consideration of evidence that mitigates against the death
penalty is essential if the sentencer is to give a reasoned
moral response to the defendant’s background, character,
and crime.’” McKinney, 813 F.3d at 812 (first emphasis
30 SPREITZ V. RYAN
added) (quoting Penry I, 492 U.S. at 319, 328).
Accordingly, “Eddings requires that ‘[t]he sentencer must
also be able to consider and give effect’” to any relevant
mitigation evidence proffered by the defendant, which
includes evidence going toward nonstatutory mitigation. Id.
(quoting Penry I, 492 U.S. at 319, 328). The Arizona
Supreme Court’s refusal to consider Spreitz’s alcohol and
substance abuse beyond its connection, if any, to Reid’s
murder constitutes application of its unconstitutional causal
nexus test to relevant nonstatutory mitigation evidence.
Our understanding of the Arizona Supreme Court’s
opinion is bolstered by the fact that in its discussion of
Spreitz’s longstanding alcohol and substance abuse, the
court cited to a prior decision that likewise applied the
unconstitutional causal nexus test, State v. Medrano,
914 P.2d 255 (Ariz. 1996). In McKinney, we explained that
in Medrano, the Arizona Supreme Court upheld the
sentencing judge’s determination that the defendant’s
cocaine use failed as a nonstatutory mitigating circumstance
because he did not show that it contributed to his conduct on
the night of the murder. McKinney, 813 F.3d at 825–26. In
so doing, “[t]he Arizona Supreme Court applied the causal
nexus test.” Id. at 825. Although the Arizona Supreme
Court’s citation to Medrano is not dispositive, it certainly
corroborates our understanding of the court’s opinion. See
also Poyson, 879 F.3d at 890 (emphasizing, among other
considerations, that the Arizona Supreme Court cited a
passage from a case specifically identified in McKinney as
applying the unconstitutional causal nexus test in order to
conclude that the state court also applied that test to Poyson’s
evidence). Moreover, as we explained in McKinney, “the
Arizona Supreme Court applied its unconstitutional nexus
test consistently” for fifteen years because it had “a mistaken
understanding of Eddings.” 813 F.3d at 826. It is thus
SPREITZ V. RYAN 31
entirely logical that in citing the exact portion of Medrano
where it previously applied its causal nexus test, the Arizona
Supreme Court was applying the same test in Spreitz’s case.
See id. at 821; see also Poyson, 879 F.3d at 890.
We find further support for our understanding of the
Arizona Supreme Court’s unconstitutional treatment of
Spreitz’s alcohol and substance abuse when we compare it
with the manner in which the court discussed other
mitigating factors. As detailed earlier, in evaluating the
evidence of Spreitz’s general history of alcohol and
substance abuse, the court stated that this evidence was “not
essential” to its decision. Spreitz I, 945 P.2d at 1280. By
contrast, when the court considered other mitigating factors,
it specifically discussed the weight it would give those
factors, as opposed to simply dismissing them outright. For
example, when discussing Spreitz’s subnormal upbringing,
the court stated, “Although we recognize defendant’s
upbringing as a mitigating circumstance, we accord it little
weight.” Id. at 1280 (emphasis added). As another example,
after “finding that since his arrest, [Spreitz] has
demonstrated remorse,” the court stated that his “remorse for
his actions does little to counterbalance especial cruelty as a
serious aggravating circumstance.” Id. at 1281 (emphasis
added). And finally, when discussing the sentencing judge’s
determination regarding Spreitz’s emotional immaturity, the
court stated, “We also find that the sentencing judge . . .
properly found that his emotional immaturity was not a
significant mitigating factor.” Id. (emphasis added). In
other words, when the Arizona Supreme Court wanted to
assign weight to a given factor, it said so. This pattern stands
in stark contrast to the court’s conclusion that anything
related to Spreitz’s long-term alcohol or substance abuse that
fell short of the statutory definition could not serve as
mitigation. Id. at 1280–81.
32 SPREITZ V. RYAN
Finally, as also discussed earlier, the PCR court
distinctly interpreted the Arizona Supreme Court’s decision
as having applied a causal nexus test. The PCR court
explained that “[w]ithout some basis for explaining or
defining the individual’s behavior at the time of the offense,
[Spreitz]’s history of alcohol or substance abuse would be
inconsequential (which is exactly what the trial court and
Supreme Court concluded).” Moreover, in articulating how
the Arizona Supreme Court had applied its causal nexus test
in Spreitz’s direct appeal, the PCR court, as had the Arizona
Supreme Court, cited to a decision of the Arizona Supreme
Court indisputably applying an unconstitutional nexus test,
State v. Kayer, 984 P.2d 31 (Ariz. 1999). See McKinney,
813 F.3d at 816 (explaining that Kayer held “that the
defendant’s mental impairment ‘was not established as a
nonstatutory mitigating factor’ in part because ‘defendant
offered no evidence to show the requisite causal nexus that
mental impairment affected his judgment or his actions at the
time of the murder’” (quoting Kayer, 984 P.2d at 46)). The
fact that the PCR court interpreted the Arizona Supreme
Court’s opinion as we do—and then itself applied the
constitutionally erroneous nexus test—lends further support
to our reading.
We recognize that the Arizona Supreme Court’s decision
on the particular issue before us was framed by language that
might, in a decisional vacuum, suggest the court knew,
understood, and applied the law with respect to Eddings.
The court acknowledged that Spreitz argued that his history
of alcohol and drug abuse served as both a statutory and
nonstatutory mitigating factor. Spreitz I, 945 P.2d at 1279.
It then stated that it “must consider any aspect of the
defendant’s character or record . . . relevant to determining
whether the death penalty should be imposed.” Id. And
finally, in reaching the conclusion to affirm Spreitz’s death
SPREITZ V. RYAN 33
sentence, the court explained that it had “examin[ed] the
entire record and reweigh[ed] the applicable aggravating and
mitigating factors.” Id. at 1281–82.
The Arizona Supreme Court’s acknowledgment of what
Spreitz argued does not indicate that it relied on or accepted
that argument as correct. Further, we cannot, given
McKinney, Poyson, and Greenway, proceed in a decisional
vacuum.
Considered in light of the baseline that the Arizona
Supreme Court was consistently applying its
unconstitutional causal nexus test in death penalty cases at
the time Spreitz’s appeal was decided, as well as its citations
to cases held in McKinney to be applying that standard, we
are confident that the manner in which the Arizona Supreme
Court disposed of Spreitz’s evidence of his longstanding
alcohol and substance abuse confirms that it applied the
causal nexus test it was using during that period.
Styers v. Schriro, 547 F.3d 1026 (9th Cir. 2008) supports
this conclusion. Styers stated that the mere fact that “the
Arizona Supreme Court stated that it had ‘considered all of
the proffered mitigation,’” is not sufficient to contradict the
fact that “its analysis prior to this statement indicated
otherwise.” Id. at 1035 (internal citation omitted). In Styers,
notwithstanding assertions to the contrary, the Arizona
Supreme Court refused to consider evidence of the
defendant’s Post Traumatic Stress Disorder (“PTSD”)
because he had failed to connect it causally to his crimes. Id.
Here, too, the Arizona Supreme Court’s actual analysis of
Spreitz’s longstanding alcohol and substance abuse reveals
that it refused to consider this particular aspect of Spreitz’s
character free from any connection to the crime. Rather, the
court’s discussion demonstrates that in evaluating Spreitz’s
long-term alcohol and substance abuse, it applied a causal
34 SPREITZ V. RYAN
nexus test in violation of Eddings and therefore refused, as a
matter of law, to consider it in assessing Spreitz’s evidence
of nonstatutory mitigating circumstances.
To the extent the state argues the Arizona Supreme
Court’s opinion is ambiguous as to whether it applied an
unconstitutional causal nexus test or not—and we believe
that it is not—McKinney requires resolving that ambiguity
in favor of Spreitz. McKinney held it clear that the Arizona
Supreme Court “consistently articulated and applied its
causal nexus test” at the time the trial court imposed
Spreitz’s death sentence and the Arizona Supreme Court
reviewed and affirmed that sentence. McKinney, 813 F.3d
at 803 (emphasis in original). Again, “if there is to be a
presumption, it is that the Arizona Supreme Court violated
the dictates of Lockett and Eddings.” Greenway, 866 at
1095.
In sum, because (1) the Arizona Supreme Court clearly
stated that Spreitz’s longstanding alcohol and substance
abuse was “not essential” to its decision, which was in line
with the court’s consistent practice of applying a causal
nexus test; (2) the Arizona Supreme Court cited to Medrano,
in which it had previously articulated a causal nexus test;
(3) the Arizona Supreme Court carefully laid out the weight
it would accord other mitigating factors as opposed to simply
ignoring them; and (4) the PCR court likewise interpreted
the Arizona Supreme Court’s opinion as applying a causal
nexus test, we conclude that the Arizona Supreme Court
held, as a matter of law, that Spreitz’s long-term alcohol and
substance abuse was not a nonstatutory mitigating factor and
therefore refused to consider it. This holding was in clear
violation of Eddings.
SPREITZ V. RYAN 35
3.
In light of our conclusion that the Arizona Supreme
Court failed to comply with Eddings, we must next decide
whether Spreitz was prejudiced by this error. On federal
habeas review, relief is warranted if an error “had substantial
and injurious effect or influence” on the challenged decision.
See Brecht v. Abrahamson, 507 U.S. 619, 623 (1993). The
petitioner is “not entitled to habeas relief” unless he can
establish that the error “resulted in actual prejudice.” Davis
v. Ayala, 135 S. Ct. 2187, 2197 (2015) (internal quotation
marks omitted). Here, because we are “in grave doubt about
whether [the Arizona Supreme Court’s] error of federal law
had substantial and injurious effect or influence in
determining [whether to affirm Spreitz’s death sentence],
that error is not harmless. And, [Spreitz] must win.”
McKinney, 813 F.3d at 822 (quoting O’Neal v. McAninch,
513 U.S. 432, 435 (1995)).
We previously have required resentencing based on
prejudicial Eddings error where the Arizona Supreme Court
refused to consider, as a matter of law, mitigation evidence
“central to” the petitioner’s “plea for leniency.” See id. at
823; see also Coleman v. Calderon, 210 F.3d 1047, 1051
(9th Cir. 2000) (concluding that a penalty-phase jury
instruction error was not harmless under Brecht because “it
undermined the very core of Coleman’s plea for life”). 14 For
the reasons that follow, it is clear that Spreitz’s long-term
14
We do not suggest that Brecht prejudice may only be established
if inappropriately considered mitigating evidence was at “the very core
of [a defendant’s] plea for life.” Coleman, 210 F.3d at 1051. It is
conceivable that other forms of Eddings error could have “a substantial
and injurious effect” on a defendant’s sentence if they would be
significant to the sentencer, despite the defendant’s failure to rely on
them as central to his or her argument. Brecht, 507 U.S. at 623.
36 SPREITZ V. RYAN
alcohol and substance abuse was important mitigation
evidence, and was so closely related as to become central to
his plea for leniency. The Arizona Supreme Court’s refusal
to consider that evidence had a substantial and injurious
effect on its sentencing determination that requires
resentencing.
While a defendant’s evidence of longstanding alcohol
and/or substance abuse can be probative mitigation evidence
on its own, a “history of substance abuse [is] substantially
mitigating when . . . combined with other mitigating
evidence.” Henry v. Ryan, 720 F.3d 1073, 1090 (9th Cir.
2013) (emphasis added) (listing cases and holding that the
defendant’s alcoholism “would not have had a substantial
and injurious effect” because it “st[ood] alone, was similar
to evidence already considered by the sentencing courts, and
was of limited probative value”) (footnote omitted). 15 Here,
Spreitz’s evidence regarding his history of alcohol and
substance abuse—spanning nearly half his life by the time
he committed the crime at the age twenty-two—was the
15
The dissent makes much of Henry, suggesting it dictates the
outcome of our prejudice analysis. Dissent at 29. Not so. In that case,
the state court had found that Henry’s intoxication on the day of the
murder did constitute a statutory mitigating factor but was not
sufficiently substantial to outweigh the two aggravating factors. Henry,
720 F.3d at 1090. Henry argued that the state court erred by failing to
consider evidence of his “historical alcoholism” in addition to his
intoxication on the day of the crime. Id. We concluded that any error
was harmless because the history of substance abuse “was similar to
evidence already considered by the sentencing courts and was of limited
probative value.” Id. In other words, because the state court had found
insufficient evidence of alcohol use that was tied to the murder, we could
not conclude that general alcohol use that was not tied to the murder
could have been determinative. Here, by contrast, the Arizona Supreme
Court found that Spreitz’s evidence of intoxication did not rise to the
level of statutory mitigation. Accordingly, there is not the same
contextual harmlessness problem as in Henry.
SPREITZ V. RYAN 37
factor on which much of his other mitigation evidence
turned.
In both his report and his testimony, Dr. Flynn discussed
at length the extent to which alcohol consumed Spreitz’s life
in the decade leading up to Reid’s murder. Spreitz began
drinking when he was twelve or thirteen; by fifteen, he was
drinking steadily on the weekends and drinking vodka before
school. According to Dr. Flynn, “[d]rug and alcohol abuse
dominated his teenage years.” After he left home, Spreitz’s
drinking intensified: Spreitz experienced blackouts multiple
times and, a week before the murder, Spreitz was
“intoxicated a majority of the time.” Dr. Flynn concluded
that “[i]t appears completely clear from the available
information that . . . Spreitz had a long-standing problem
with alcohol which probably reached the level of physical
dependence.”
Critically, Spreitz’s decade-long struggle with alcohol
and other addictive substances reflects the true gravity of his
“subnormal upbringing,” a mitigating circumstance
recognized by the Arizona Supreme Court. See Spreitz I,
945 P.2d at 1280. Both Spreitz’s sentencing memorandum
and Dr. Flynn detailed how his difficult childhood led to
Spreitz’s early dependence on alcohol. For instance, Dr.
Flynn explained that Spreitz turned to alcohol as a coping
mechanism to deal with the “pervasive subabusive
emotional battering and neglect that he received from his
mother,” whom he described as “punitive, controlling, [and]
emotionally cold.” Dr. Flynn found that Spreitz “slipped
quietly into alcoholic numbness” after his “failure to live up
to his mother’s standards or earn her love.” Reflecting
parental neglect, Spreitz’s parents “failed to arrange . . . for
an appropriate rehabilitation program” even though
“[v]irtually everyone . . . who spent much time with him
38 SPREITZ V. RYAN
described him as a heavy drinker.” “Given the intensity and
duration” of Spreitz’s reliance on alcohol, “the mother (and
step-father) were either so uninvolved with him to have
failed to notice the problem, or they were aware of it and
failed to arrange (or attempt to arrange) for an appropriate
rehabilitation program.”
Because Spreitz did not receive treatment during his
teenage years, his dependence on alcohol only deepened. In
his sentencing memorandum, Spreitz, too, argued that his
long-term alcohol abuse reflects the impact that his
upbringing by an emotionally unavailable, manipulative,
punitive mother had on him: it was “[n]o wonder that by 12
or 13 [h]e began drinking alcohol and smoking marijuana.”
Although the Arizona Supreme Court recognized that
Spreitz’s “subnormal” upbringing was “a mitigating
circumstance,” it “accord[ed] it little weight.” Spreitz I,
945 P.2d at 1280. Had the court properly considered
Spreitz’s longstanding history of alcohol abuse in
conjunction with and as a manifestation of his disturbing
childhood, both mitigating circumstances—and most
importantly, the two in combination—would have been
more compelling.
The dissent argues that Spreitz’s long-term history of
alcohol abuse cannot be central to his plea for leniency
because much of his mitigation evidence focused on his
“abusive and dysfunctional childhood, upbringing, and
interactions with his mother,” Dissent at 66, not his history
of alcohol and substance abuse. This argument, however,
misses the point, as it overlooks the interlinked relationship
between Spreitz’s alcoholism and his subnormal childhood:
Spreitz’s history of alcohol and substance abuse is central to
evaluating the actual significance of his childhood as a
mitigating factor, and vice versa.
SPREITZ V. RYAN 39
In addition to being a symptom of his dysfunctional
childhood, Spreitz’s alcohol and substance abuse can be seen
as linked to his emotional immaturity, another nonstatutory
mitigating circumstance recognized by the Arizona Supreme
Court but described as not “significant.” Spreitz I, 945 P.2d
at 1282. In his report, Dr. Flynn explained that Spreitz was
“very likely to have been socially and emotionally
immature” due, at least in part, to “[y]ears of alcoholism
intoxication [that] wipe[d] out many of the . . . healthy
developmental processes requisite to age-appropriate social
and emotional maturity.” Similarly, in his sentencing
memorandum, Spreitz connected “[h]is developmental
failure” with his “[n]umbing by use of alcohol.” Again, had
the Arizona Supreme Court properly considered Spreitz’s
longstanding alcohol and substance abuse, another
mitigating factor—Spreitz’s emotional immaturity—would
have been brought into proper perspective.
Thus, by failing to consider Spreitz’s longstanding
alcohol and substance abuse, the Arizona Supreme Court
was left with a critical void in Spreitz’s narrative: Spreitz’s
subnormal childhood was so emotionally disturbing that it
led him to drink by the age of twelve or thirteen, which, in
turn, disrupted his normal development and contributed to
his emotional immaturity. As both a symptom and cause of
other mitigating factors, Spreitz’s longstanding alcohol
abuse was thus central to his plea for leniency. 16 Without
16
Contrary to the dissent’s suggestion, Spreitz did not focus on his
intoxication at the time of the crime to the exclusion of his history of
alcohol abuse. Dissent at 22–24. Rather, both he and Dr. Flynn
distinguished between Spreitz’s history of alcohol abuse as nonstatutory
mitigation and his intoxication at the time of the crime as statutory
mitigation. It is Spreitz’s history of alcohol abuse that underlies and
amplifies much of his other mitigation evidence, not his intoxication at
the time of Reid’s murder.
40 SPREITZ V. RYAN
consideration of the evidence of his long-term substance and
alcohol abuse, the Arizona Supreme Court’s conclusions
regarding all of Spretiz’s mitigation evidence were
fundamentally flawed.
We come to this conclusion fully recognizing, as the
dissent emphasizes, that “the severity of an aggravating
circumstance must be considered when determining whether
an Eddings error is harmless.” Dissent at 71. The Arizona
Supreme Court described what Spreitz’s “confession and
physical evidence” revealed: that “Spreitz beat and raped
[Reid] in a brutal assault that lasted many minutes before he
crushed her skull.” Spreitz I, 945 P.2d at 1279. For this
reason, the Arizona Supreme Court affirmed the sentencing
court’s finding that Reid’s murder was especially cruel. Id.
Even so, we found prejudicial nexus-error in McKinney
where the murder was done in a cruel manner for pecuniary
gain as part of a double murder. McKinney, 813 F.3d at 823.
“When the choice is between life and death,” “the risk that
the death penalty will be imposed in spite of facts which may
call for a less severe penalty. . . . is unacceptable.” Abdul-
Kabir v. Quarterman, 550 U.S. 233, 264 (2007) (quoting
Lockett v. Ohio, 438 U.S. 586, 605 (1978)). Accordingly,
the sentencer “must be able to give meaningful consideration
and effect to all mitigating evidence that might provide a
basis for refusing to impose the death penalty on a particular
individual, notwithstanding the severity of his crime.” Id. at
247 (emphasis added).
Here, the Arizona Supreme Court failed to give any
meaning to Spreitz’s longstanding alcohol and substance
abuse, an oversight which, in turn, minimized the value of
other mitigating evidence as well. Because the court’s
analysis was missing a core component, we “cannot say,
with fair assurance” that the Arizona Supreme Court’s
SPREITZ V. RYAN 41
Eddings error did not affect its decision. McKinney,
813 F.3d at 822. As a result, the Arizona Supreme Court’s
refusal to consider the evidence was not harmless under
Brecht.
IV. Conclusion
We reverse in part the district court’s judgment denying
a writ of habeas corpus. We remand with instructions to
grant the writ with respect to Spreitz’s sentence unless the
state, within a reasonable period, either corrects the
constitutional error in his death sentence or vacates the
sentence and imposes a lesser sentence consistent with the
law.
REVERSED in part and REMANDED.
TALLMAN, Circuit Judge, dissenting:
The court declares that Christopher Spreitz’s death
sentence should be vacated because, applying de novo
review, the sentencing court and the Arizona Supreme Court
violated Eddings v. Oklahoma, 455 U.S. 104 (1982), and that
this error was not harmless. I respectfully dissent.
First, the record does not establish that either court
unconstitutionally refused to consider relevant mitigating
evidence. They were simply not persuaded by it. What the
record clearly shows is the Arizona courts did not find that
Spreitz’s proffered evidence outweighed the aggravating
factor of cruelty in the way by which he brutally murdered
Ruby Reid. Second, even if we determined that the Arizona
courts did violate Eddings, Spreitz cannot show that this
error had a “substantial and injurious effect or influence” on
42 SPREITZ V. RYAN
his ultimate sentence. Brecht v. Abrahamson, 507 U.S. 619,
623 (1993). Given the savage brutality of his crimes and the
weakness of all the mitigation evidence he presented, any
remand for a redetermination of the appropriate penalty
ought to result in the same decision.
Nonetheless, the majority attempts unsuccessfully to
navigate the course that Spreitz’s Eddings claim requires
him to sail to warrant federal habeas relief. In doing so, the
majority has once again foundered on the shoals of the
Antiterrorism and Effective Death Penalty Act of 1996
(AEDPA), specifically 28 U.S.C. § 2254—where so many
of our prior habeas corpus decisions are wrecked—failing to
afford the respect due the decisions of our sister state courts
and taunting the Supreme Court for another capsizing in lieu
of safe passage.
I
We review the Arizona post-conviction relief (PCR)
court’s decision because it is the last reasoned state court
decision on Spreitz’s Eddings claim. See Barker v. Fleming,
423 F.3d 1085, 1091 (9th Cir. 2005). Under AEDPA we first
must decide the level of deference to give to that decision by
determining whether it “was contrary to, or involved an
unreasonable application of, clearly established Federal law,
as determined by the Supreme Court of the United States.”
Id. § 2254(d)(1).
I do not quarrel with the majority’s determination that
the PCR court held a mistaken view of what Eddings
requires, and that we therefore must review Spreitz’s
Eddings claim de novo. See Castellanos v. Small, 766 F.3d
1137, 1146 (9th Cir. 2014). However, “we may not grant
habeas relief simply because of [the PCR court’s]
§ 2254(d)(1) error . . . .” Frantz v. Hazey, 533 F.3d 724, 735
SPREITZ V. RYAN 43
(9th Cir. 2008) (en banc); see also id. at 728, 735–37
(explaining our “approach to reviewing state court decisions
that rely on legal principles contradicting clearly established
Supreme Court law but do not necessarily reach the wrong
result”).
On de novo review, Spreitz can meet neither of the two
requirements to obtain habeas relief on his Eddings claim.
First, he cannot demonstrate that the state sentencing courts
actually violated Eddings. And second, even if there was a
violation, he cannot show that any Eddings error by those
state courts was prejudicial under Brecht.
II
Spreitz’s Eddings claim rests in large part on the
erroneous assumption that the sentencing court and the
Arizona Supreme Court failed to consider his history of
alcohol and substance abuse, unrelated to the offense, as
non-statutory mitigation when determining his sentence. He
contends that, by failing to consider all mitigating
circumstances, the state courts violated Eddings, which
provides that in capital cases “the sentencer may not refuse
to consider or be precluded from considering ‘any relevant
mitigating evidence.’” Skipper v. South Carolina, 476 U.S.
1, 4 (1986) (quoting Eddings, 455 U.S. at 114). Eddings also
provides that the sentencer “may determine the weight to be
given relevant mitigating evidence.” 455 U.S. at 114–15.
Recently, we adopted an erroneous presumption that the
Arizona Supreme Court repeatedly violated the dictates of
Eddings by “consistently” applying an unconstitutional
causal nexus test to all capital cases between 1989 and 2005.
See McKinney v. Ryan, 813 F.3d 798, 803 (9th Cir. 2015)
44 SPREITZ V. RYAN
(en banc), cert. denied, 137 S. Ct. 39 (2016). 1 “We did not
say, however, that the Arizona [Supreme Court] always
applied it.” See Greenway v. Ryan, 866 F.3d 1095, 1095 (9th
Cir. 2017) (per curiam). Notably, in listing the cases in
which the Arizona courts erroneously applied the causal
nexus test, McKinney did not include Spreitz’s case. See
813 F.3d at 815–16, 824–26. And in McKinney, our holding
resolved only the “precise question” whether the state court
in that specific case had applied the causal nexus test. Id. at
804; see also Hedlund v. Ryan, 854 F.3d 557, 586 n.22 (9th
Cir. 2017) (“We express no opinion as to how to apply
McKinney in future Arizona capital cases from the suspect
time period.”). Each case must be reviewed on its unique
facts. We therefore must examine the state court decisions
in Spreitz’s case to determine whether they properly took
into account all mitigating factors. 2
1
The real damage inflicted by the en banc decision in McKinney is
the broad assumption it adopts that now infects every death sentence
imposed by Arizona courts for horrendous murders adjudicated in the
sixteen-year period. It is regrettable that the United States Supreme
Court denied certiorari in McKinney. But this case presents yet another
opportunity to right the error in Ninth Circuit death penalty habeas
jurisprudence.
2
In reviewing the last reasoned state court decision, we may
consider “the record that was before the state court that adjudicated the
claim on the merits.” Cannedy v. Adams, 706 F.3d 1148, 1156 (9th Cir.
2013) (citation omitted). The record before the PCR court in this case
included the sentencing judge’s decision and the Arizona Supreme
Court’s direct appeal ruling. Thus, despite the majority’s preference for
cabining its view of the record to only the Arizona Supreme Court’s
decision, we must consider both decisions here.
SPREITZ V. RYAN 45
A
At all relevant times in this case, the operative Arizona
death penalty statute required a sentencing court to “impose
a sentence of death if the court finds one or more of the
aggravating circumstances enumerated in [the statute and
determines] that there are no mitigating circumstances
sufficiently substantial to call for leniency.” Ariz. Rev. Stat.
§ 13-703(E) (1993) (current version at Ariz. Rev. Stat.
§ 13-751(E) (2012)). Arizona law provided five statutory
mitigating factors, see § 13-703(G), including a modified
form of diminished capacity: “The defendant’s capacity to
appreciate the wrongfulness of his conduct or to conform his
conduct to the requirements of the law was significantly
impaired, but not so impaired as to constitute a defense to
prosecution.” Id. § 13-703(G)(1). Even our en banc court
acknowledged, “When applied solely in the context of
statutory mitigation under § 13-703(G)(1), the causal nexus
test does not violate Eddings.” McKinney, 813 F.3d at 810.
But Arizona law also provided for consideration of
catch-all, non-statutory mitigating evidence that
encompassed “any factors proffered by the defendant or the
state that are relevant in determining whether to impose a
sentence less than death, including any aspect of the
defendant’s character, propensities or record and any of the
circumstances of the offense.” Ariz. Rev. Stat. § 13-703(G).
The dictates of Eddings apply to such non-statutory
mitigating evidence. See McKinney, 813 F.3d at 810
(“When applied in the context of nonstatutory mitigation, the
causal nexus test clearly violates Eddings.”). Lastly,
Arizona law requires the Arizona Supreme Court to
“independently review the trial court’s findings of
aggravation and mitigation and the propriety of the death
46 SPREITZ V. RYAN
sentence.” Ariz. Rev. Stat. § 13-703.01(A) (1994) (current
version at Ariz. Rev. Stat. § 13-755(A) (2009)).
B
Here, Spreitz cannot show that the sentencing judge, the
Honorable William N. Sherrill, violated Eddings in
determining Spreitz’s sentence. At sentencing, Spreitz
submitted a memorandum that specifically discussed his
history of substance abuse and cited Eddings; defense
psychologist Dr. Todd C. Flynn provided the court with a
psychological evaluation report and testimony specifically
discussing Spreitz’s history of substance abuse as both
statutory and non-statutory mitigating factors; and Judge
Sherrill differentiated between the two types of mitigation
and ultimately decided that the mitigating evidence did not
outweigh the aggravating factor of cruelty based on how
Spreitz murdered Ms. Reid.
In his written judgment, Judge Sherrill acknowledged
that “many factors have been submitted in mitigation.”
Among the mitigating evidence, he observed that Spreitz “in
his life turned to substance abuse—alcohol and some
suggestion he was using cocaine and other drugs”—and that
Spreitz’s “history of intoxication [was] longstanding.” He
also found that Spreitz “had been abusing substances for
close to ten years of his life.” Judge Sherrill’s discussion of
Spreitz’s history of substance abuse unrelated to the crime
establishes that he considered that evidence in his sentencing
decision. See Williams v. Stewart, 441 F.3d 1030, 1057 (9th
Cir. 2006) (stating that a sentencing court does not violate
Eddings where it is clear that “the sentencing court
considered all relevant mitigating evidence that was
offered”).
SPREITZ V. RYAN 47
There is an important distinction between Judge
Sherrill’s consideration of Spreitz’s evidence of his long-
term substance abuse, and Judge Sherrill’s subsequent
findings regarding the weight he ultimately decided to give
that evidence. After discussing Spreitz’s proffered
mitigating evidence, Judge Sherrill correctly observed that
the issue was, in weighing the imposition of capital
punishment, “whether any of these mitigating
circumstances, individually or cumulatively, can balance or
outweigh the aggravating circumstances.” And, he finally
determined that “the mitigating circumstances [were] not
sufficient to balance the aggravating circumstances, nor
[were] they sufficiently substantial to call for leniency.” In
reaching that conclusion, Judge Sherrill specified that he did
“not believe that intoxication [was] any sort of mitigating
circumstance.” (Emphasis added.) One should fairly read
that conclusion to mean that Spreitz’s long history of
substance abuse was neither a statutory nor a non-statutory
mitigating factor “sufficiently substantial to call for
leniency.” Ariz. Rev. Stat. § 13-703(E).
Under Eddings, Judge Sherrill was free to conclude that
Spreitz’s mitigating evidence, whether statutory or non-
statutory, simply did not warrant leniency given the
aggravating circumstance of how Spreitz cruelly murdered
Ruby Reid. See Mann v. Ryan, 828 F.3d 1143, 1159–60 (9th
Cir. 2016) (holding that defendant “failed to show that the
mitigating circumstances he had presented,” including a
history of substance abuse, “outweighed the aggravating
circumstances of his crimes”), cert. denied, 137 S. Ct. 1091
(2017), 137 S. Ct. 1205 (2017); Henry v. Ryan, 720 F.3d
1073, 1090 n.11 (9th Cir. 2013) (recognizing that “historical
alcoholism might [be] considered aggravating as well as
mitigating, depending on the perspective of the sentencing
court”). “It is sufficient that a sentencing court state that it
48 SPREITZ V. RYAN
found no mitigating circumstances that outweigh the
aggravating circumstances.” Williams, 441 F.3d at 1057
(quotation omitted).
Eddings also did not prevent Judge Sherrill from giving
less weight to Spreitz’s substance abuse history based on a
lack of nexus to the murder. See Hedlund, 854 F.3d at 587
n.23 (stating that under Eddings, “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”); Styers v.
Ryan, 811 F.3d 292, 298–99 (9th Cir. 2015) (holding that the
Arizona Supreme Court did not violate Eddings in assigning
little weight to the petitioner’s PTSD in the absence of a
causal connection to the crime), cert. denied, 137 S. Ct. 1332
(2017). Nor did Eddings prevent Judge Sherrill from
assigning no weight to Spreitz’s long-term substance abuse
unrelated to the offense, so long as Judge Sherrill considered
that evidence, which the record shows he definitely did. See
Williams, 441 F.3d at 1057 (“Once mitigating evidence is
allowed in, a finding that there are ‘no mitigating
circumstances’ does not violate the Constitution.”); Ortiz v.
Stewart, 149 F.3d 923, 943 (9th Cir. 1998) (“[T]he
sentencing court did not refuse to consider mitigating
evidence; it considered the evidence and found it inadequate
to justify leniency. That assessment did not violate the
Constitution.”), overruled in part on other grounds by
Martinez v. Ryan, 566 U.S. 1 (2012).
In addition to the above findings, Judge Sherrill also
added that he did not “believe that [Spreitz’s] substance
abuse or intoxication impaired [his] ability and capacity to
appreciate the wrongfulness of his conduct to any significant
degree” under § 13-703(G)(1). At the time, Arizona courts
were required by law to examine whether evidence of
alcohol and drug use bore a causal connection to the crime.
SPREITZ V. RYAN 49
Ariz. Rev. Stat. § 13-703(G)(1) (requiring the sentencer to
consider leniency if the defendant proved that his “capacity
to appreciate the wrongfulness of his conduct or to conform
his conduct to the requirements of law was significantly
impaired, but not so impaired as to constitute a defense to
prosecution”). And, there is no question that Spreitz pursued
a mitigation strategy at sentencing that attempted to link his
long-term substance abuse with the kidnap/murder to show
statutory mitigation under § 13-703(G)(1).
For example, Spreitz’s sentencing memorandum stated
that Dr. Flynn’s insights about him were of “the nature, type
and character contemplated by A.R.S. Section
13-703(G)(1),” and that “Dr. Flynn’s report [was], indeed,
relevant and should be considered by the Court as statutory
mitigating circumstances.” Dr. Flynn’s report and expert
testimony went to great lengths to argue that Spreitz’s
alcohol abuse was “very strongly correlated with violent
behavior . . . .” Judge Sherrill could not respond to Spreitz’s
argument that § 13-703(G)(1) called for leniency without
analyzing whether Spreitz had shown a connection between
his substance abuse and the murder. We cannot infer from
that causal nexus analysis that Judge Sherrill flatly refused
to consider any evidence of substance abuse that was not
causally related to the crime.
Rather, this record shows that Judge Sherrill
constitutionally applied a causal nexus test in the context of
an Arizona statutory mitigating factor, § 13-703(G)(1). See
McKinney, 813 F.3d at 810 (“When applied solely in the
context of statutory mitigation . . . , the causal nexus test
does not violate Eddings.”). Judge Sherrill’s other
statements repeatedly demonstrate that he understood
Eddings’ mandate and considered all of Spreitz’s proffered
mitigating evidence, but ultimately found that evidence
50 SPREITZ V. RYAN
insufficient to warrant leniency. We are in the same
situation here that recently confronted our Court in Hedlund.
See 854 F.3d at 591 (Bea, J., concurring) (“[Although] Judge
Sheldon constitutionally applied a causal-nexus test in the
context of an Arizona statutory mitigating factor[,
§ 13-703(G)(1), that] statement does not show that Judge
Sheldon excluded mitigating evidence from his
consideration, and Judge Sheldon’s other statements
repeatedly demonstrate otherwise.”).
Nor does the record before us support the conclusion that
Judge Sherrill failed to consider evidence of Spreitz’s long-
term substance abuse as a non-statutory mitigating factor.
What we can say for sure is that he was not persuaded by
that evidence. Judge Sherrill listened to testimony and
arguments at trial and in the penalty phase, read everything
submitted, and then deliberated for three days before
concluding that Spreitz’s mitigation evidence was not
sufficient to outweigh the evidence in aggravation
warranting death. There is no Eddings error here.
C
Similarly, in its “detailed independent review” of
Spreitz’s sentence, the Arizona Supreme Court “examined
the entire record to weigh and consider the aggravating and
mitigating circumstances.” State v. Spreitz, 945 P.2d 1260,
1278 (Ariz. 1997). Consistent with Eddings, the Arizona
Supreme Court recognized “that the sentencing judge must
consider any aspect of the defendant’s character or record
and any circumstance of the offense relevant to determining
whether the death penalty should be imposed.” Id. at 1279
(quotation omitted). And, it recognized “that the weight
accorded such evidence is within the sentencing judge’s
discretion.” Id.; see Greenway, 866 F.3d at 1097 (citing
similar language in the Arizona Supreme Court’s review of
SPREITZ V. RYAN 51
a capital sentence to conclude that the court “did not reject
any mitigating factor, as a matter of law, on the theory that
it was not related to the commission of the crime”).
In that review, the Arizona Supreme Court specifically
observed that the “record demonstrates defendant’s longtime
substance abuse problems.” Spreitz, 945 P.2d at 1280.
Nonetheless, the court found that Spreitz’s “general
problems with substance abuse [were] not essential to [its]
decision” because, ultimately, “the aggravating
circumstance of especial cruelty in [Spreitz]’s murder of
Ruby Reid outweigh[ed] all factors mitigating in favor of
leniency.” Id. at 1280, 1282. The Arizona Supreme Court
thus rejected, on the merits, Spreitz’s argument that his long-
term substance abuse constituted either a statutory or non-
statutory mitigating factor warranting leniency in his case.
There is no mitigation evidence it refused to consider.
To the extent that the majority cites Poyson v. Ryan,
879 F.3d 875 (9th Cir. 2018), for the proposition that
McKinney, 813 F.3d at 802–03, adopts a per se rule that the
Arizona Supreme Court applied an unconstitutional causal
nexus test in all capital cases over a 15-year period, I fully
agree with Judge Ikuta’s reluctant concurrence in Poyson.
879 F.3d at 897–900. McKinney was wrongly decided and
flips the “presumption that state courts know and follow the
law.” Id. at 897–98 (citing Woodford v. Visciotti, 537 U.S.
19, 24 (2002)). The record here rebukes even the McKinney
presumption and in that respect is factually distinguishable
from both McKinney and Poyson.
The majority nonetheless concludes that the Arizona
Supreme Court’s use of the term “not essential” proves that
the court “require[d] that Spreitz establish a causal
connection between his longstanding substance abuse and
the murder before considering and weighing the evidence as
52 SPREITZ V. RYAN
a nonstatutory mitigating factor.” Op. at 22–23, 29. Not so.
We can reasonably interpret “not essential” to mean
inconsequential, which we have previously defined in a
similar context as having little effect on the outcome. See
Landrigan v. Stewart, 272 F.3d 1221, 1230 (9th Cir. 2001)
(“At any rate, any error in failing to consider Landrigan’s
use of alcohol and drugs would have been inconsequential;
it would have had no effect whatsoever on the outcome.”),
adopted sub nom. Landrigan v. Schriro, 501 F.3d 1147 (9th
Cir. 2007) (en banc).
The majority reasons that, unlike the Arizona Supreme
Court’s consideration of Spreitz’s substance abuse history
unrelated to the murder, “when the court considered other
mitigating factors, it specifically discussed the weight it
would give to those factors . . . .” Op. at 31. But the law has
never imposed an obligation on the sentencer to explicitly
disclose the value it assigns to every piece of mitigating
evidence. “The U.S. Supreme Court has ‘never held that a
specific method for balancing mitigating and aggravating
factors in a capital sentencing proceeding is constitutionally
required.’” Styers, 811 F.3d at 298 (quoting Kansas v.
Marsh, 548 U.S. 163, 175 (2006)). Nor did McKinney
abrogate the principle that a sentencing court is not required
to “itemize and discuss every piece of evidence offered in
mitigation.” Jeffers v. Lewis, 38 F.3d 411, 418 (9th Cir.
1994) (en banc). “We have determined that a sentencing
court need not specifically discuss each individual item of
mitigating evidence so long as it appears to have considered
all relevant evidence.” Ortiz, 149 F.3d at 943. This is
particularly important when reviewing Eddings claims
because otherwise we could infer that a sentencing court
refused to consider any piece of evidence to which it did not
expressly assign a value, even though it sufficiently
discussed that proffered evidence. That is not the law.
SPREITZ V. RYAN 53
For example, in Jeffers, our en banc court held that the
state court’s “failure to list and discuss each item of evidence
[that the defendant] offered in mitigation” did not
demonstrate that the “court neglected to consider all of his
mitigation evidence.” 38 F.3d at 417–18. Instead, we held
that there was no clear requirement under either Arizona law
or the federal Constitution that the sentencer itemize and
discuss each piece of evidence offered in mitigation. Id. at
418. And, we held that the sentencing court there clearly
considered all such evidence. Id. In addition, a sentencing
court “need not exhaustively analyze each mitigating factor
as long as a reviewing federal court can discern from the
record that the state court did indeed consider all mitigating
evidence offered by the defendant.” Moormann v. Schriro,
426 F.3d 1044, 1055 (9th Cir. 2005) (quotation omitted); see
also Lopez v. Schriro, 491 F.3d 1029, 1039 (9th Cir. 2007)
(observing that there is no “clearly established Supreme
Court precedent setting forth the record the sentencing court
must make in order to permit sufficient appellate review”
(quotation omitted)). We can easily make that determination
here.
The majority also focuses on the Arizona Supreme
Court’s additional finding that Spreitz “was [not] impaired
by alcohol consumption to an extent that it interfered with
his ‘capacity to appreciate the wrongfulness of his conduct
or to conform his conduct to the requirements of the law.’”
Op. at 28 (quoting Spreitz, 945 P.2d at 1280–81). But this
statement does not mean that the Arizona Supreme Court
refused to consider Spreitz’s long-term substance abuse
unrelated to the murder. On the contrary, as the majority
recognizes, the Arizona Supreme Court’s “decision . . .
suggest[ed] the court knew, understood, and applied the law
with respect to Eddings.” Id. at 32. For example, the
Arizona Supreme Court stated: “In our review, we have
54 SPREITZ V. RYAN
been mindful that the sentencing judge must consider any
aspect of the defendant’s character or record and any
circumstance of the offense relevant to determining whether
the death penalty should be imposed.” Spreitz, 945 P.2d at
1279 (quotations omitted) (emphasis added). After
independently “weighing the mitigating circumstances both
individually and cumulatively against the aggravating
circumstance,” it ultimately found that “the aggravator of
especial cruelty outweighed all other circumstances.” Id. at
1280. Those findings do not violate Eddings. They reflect
a careful consideration of all of the evidence, ultimately
concluding that the death sentence was warranted here
despite what Spreitz offered in mitigation. Thus, there was
no Eddings violation.
III
Even if we were to determine that the state courts
committed Eddings error by failing to consider Spreitz’s
long-term substance abuse unrelated to the offense, Spreitz
cannot show that this error was prejudicial because the
violent manner in which he abducted and murdered Ruby
Reid trenchantly outweighed all evidence proffered in
mitigation. See Greenway, 866 F.3d at 1100 (denying
habeas relief on petitioner’s Eddings claim because the
Arizona Supreme Court’s Eddings error was harmless).
Thus, even assuming that the state courts unconstitutionally
excluded that evidence, their factual determinations make
clear that the evidence would have been entitled to little
mitigating weight in their ultimate sentencing
determinations.
“The harmless-error standard on habeas review provides
that ‘relief must be granted’ only if the error ‘had substantial
and injurious effect or influence in determining the [sentence
imposed].’” McKinney, 813 F.3d at 822 (quoting Brecht,
SPREITZ V. RYAN 55
507 U.S. at 623). The United States Supreme Court has
distinguished this harmless error standard from the standard
applicable to cases on direct review, which require merely a
“reasonable possibility” that the error affected the outcome
of the proceedings. See Brecht, 507 U.S. at 634–38 (citing
Chapman v. California, 386 U.S. 18, 24 (1967)). Unlike
cases on direct review, “granting habeas relief merely
because there is a ‘reasonable possibility’ that [the] error
contributed to the [outcome of the proceedings] is at odds
with the historic meaning of habeas corpus.” Id. at 637
(citation and quotations omitted). As such, “an error that
may justify reversal on direct appeal will not necessarily
support a collateral attack on a final judgment.” Id. at 634
(citation and quotations omitted).
In McKinney, we clarified that a state court’s Eddings
violation is harmless if we can “say[] with fair assurance”
that “the judgment was not substantially swayed by the
error.” 813 F.3d at 822 (quoting Kotteakos v. United States,
328 U.S. 750, 765 (1946)) (emphasis added). On the other
hand, an Eddings error is not harmless if we are in “grave
doubt” about whether it had a substantial and injurious effect
or influence in determining the sentence imposed. Id.
(quoting O’Neal v. McAninch, 513 U.S. 432, 436 (1995)).
McKinney held that the excluded evidence there “was
important mitigating evidence” because it was “central to
[McKinney’s] plea for leniency.” 813 F.3d at 823 (emphasis
added). It cited Coleman v. Calderon, 210 F.3d 1047 (9th
Cir. 2000), where we held that an unconstitutional jury
instruction at sentencing was not harmless because “it
undermined the very core of Coleman’s plea for life.” Id. at
1051 (emphasis added).
Here, based on what the Arizona courts repeatedly said,
there is no “grave doubt” as to whether any Eddings error
56 SPREITZ V. RYAN
substantially influenced or swayed Spreitz’s ultimate
sentence. The courts at every level consistently found that
“the aggravating circumstance of especial cruelty in
[Spreitz’s] murder of Ruby Reid outweigh[ed] all factors
mitigating in favor of leniency.” Spreitz, 945 P.2d at 1282.
In considering the entire record, the Arizona Supreme Court
explained:
We agree with the sentencing judge that
defendant’s upbringing was subnormal. The
record supports the judge’s conclusion that
defendant’s home life was sadly lacking and
that his mother’s erratic behavior toward
defendant inhibited his emotional
development and social skills . . . . Although
we recognize defendant’s upbringing as a
mitigating circumstance, we accord it little
weight. While defendant’s inadequate
upbringing may have contributed to his
emotional immaturity and undeveloped
humanitarian skills, we concur with
defendant’s statement at his sentencing
hearing that “people that have had as bad a
background or worse haven’t killed. And I
don’t want what everyone has said about my
background to be an excuse for what’s
happened.”
....
The record demonstrates defendant’s
longtime substance abuse problems. We
note, however, that defendant’s general
problems with substance abuse are not
essential to our decision here. We therefore
SPREITZ V. RYAN 57
decline to conclude that defendant was
impaired by alcohol consumption to an extent
that it interfered with his “capacity to
appreciate the wrongfulness of his conduct or
to conform his conduct to the requirements of
the law.” A.R.S. § 13-703(G)(1) . . . .
[W]e find that defendant expressed remorse
for the victim’s death on more than one
occasion . . . . We recognize remorse as a
nonstatutory mitigating factor ....
However, defendant’s remorse for his actions
does little to counterbalance especial cruelty
as a serious aggravating circumstance in Ms.
Reid’s murder. According to defendant’s
confession, when he left Ms. Reid in the
desert early the morning of May 19, 1989, he
did not know whether she was alive or dead.
He confessed that he rode his bicycle out to
the murder site several days later to see if her
body was still there, hoping that it would not
be, that she was still alive. We would find
defendant’s remorse a more compelling
mitigating factor if, for example, it had
prompted him to report his actions toward
Ms. Reid to the authorities.
The sentencing judge found that defendant’s
ability to appreciate the wrongfulness of his
conduct was not impaired on the night of the
murder to any significant extent by substance
abuse, emotional disorders, situational stress,
or by a combination of these. Our review of
the record convinces us that the trial court’s
finding was proper . . . .
58 SPREITZ V. RYAN
We agree that the record supports the
sentencing judge’s findings that defendant
had no previous adult felony convictions, no
prior record of acts of violence, and that
defendant is capable of rehabilitation. We
also find that the sentencing judge correctly
rejected defendant’s age of twenty-two as a
mitigating circumstance and properly found
that his emotional immaturity was not a
significant mitigating factor.
After examining the entire record and
reweighing the applicable aggravating and
mitigating factors, we find that the
aggravating circumstance of especial cruelty
in defendant’s murder of Ruby Reid
outweighs all factors mitigating in favor of
leniency.
....
We have conducted an independent review of
defendant’s aggravating and mitigating
circumstances as required by A.R.S. section
13-703.01 and find that the mitigating
circumstances cumulatively are not
sufficiently substantial to warrant leniency in
relation to the aggravating circumstance of
cruelty.
Id. at 1280–83.
SPREITZ V. RYAN 59
Despite this explanatory language summarizing its
analysis of all of the evidence from the sentencing hearing,
the majority nonetheless insists that the Arizona Supreme
Court refused to consider Spreitz’s long-term substance
abuse as a non-statutory mitigating factor, and concludes this
refusal “had a substantial and injurious effect on [Spreitz’s
sentence].” Op. at 35–36. My colleagues reason that
“Spreitz’s long-term alcohol and substance abuse was
important mitigation evidence . . . central to his plea for
leniency.” Id. The record belies that characterization.
A
Unlike McKinney and Coleman, Spreitz’s history of
substance abuse unrelated to the murder was not “central” to
his plea for leniency or at the very core of his plea for life.
Instead, Spreitz argued at sentencing that he deserved
leniency because of, among other things, (1) his intoxication,
age, and emotional immaturity at the time of the crime;
(2) his disruptive childhood; (3) his abusive home
environment; (4) his emotionally cold mother; (5) his poor
social adjustment; (6) a persistent pattern of rejection; (7) the
absence of a healthy role model; (8) a lack of economic and
emotional support; (9) his lack of a high school education;
(10) the absence of adult convictions and violent behavior;
(11) his low risk of future danger; (12) his good character;
(13) his good behavior while in jail; and (14) his
rehabilitative nature. See, e.g., Spreitz, 945 P.2d at 1279
(“At sentencing, defendant argued as statutory and
nonstatutory mitigating factors: (1) his dysfunctional family
life and lack of socialization; (2) a history of alcohol and
drug abuse; (3) his expressions of remorse; (4) his impaired
capacity to appreciate the wrongfulness of his conduct,
A.R.S. § 13-703(G)(1); (5) his good behavior while
incarcerated; (6) his lack of adult convictions; (7) no prior
60 SPREITZ V. RYAN
record of violent tendencies; and (8) his age at the time of
the murder, A.R.S. § 13-703(G)(5).”). Specifically,
Spreitz’s sentencing memorandum posited that:
The Defendant’s background and upbringing
are ripe with numerous mitigating factors:
1. He faced a disruptive middle childhood.
2. Punitive and abusive behavior toward
[Spreitz].
3. An emotionally cold mother.
4. Poor social adjustment with peers.
5. No healthy role model.
6. Drug and alcohol abuse.
7. [That] Spreitz was forced to live on his
own without economic or emotional support.
8. Persistent pattern of rejection.
In addition, Spreitz devoted a large portion of his
sentencing memorandum to explaining how his childhood
abuse, dysfunctional upbringing, developmental immaturity,
and emotional problems contributed to his commission of
the murder:
SPREITZ V. RYAN 61
For Chris, as with other chronically rejected,
neglected and devalued children, there was
no escaping [his] deep seated anger and
resentment. The desperate, quiet, battle
wrestling inside of Chris as he sought to
please [his mother] and was rejected could
not be numbed forever by alcohol. His
developmental failure included the non-
development of personal insight and conflict
resolution skills. He could not control,
understand or deal with the anger. Numbing
by use of alcohol and trysts with older women
were but stop-gap measures. His fragile
facade was ready to crack. Each failure
brought Chris closer to the edge. He crossed
that point with the victim. The result was
inevitable [as] Chris had no way to
compensate for a lifetime of misdirected
development, through a dysfunctional life
and family, especially when combined with
arrested and distorted development and
alcoholism . . . .
Chris’s background is a relevant and
powerful mitigating factor with a
determinative factor culminating in the
instant offense. When a person’s background
has an impact [on] his behavior beyond his
control, as here, such a circumstance is
clearly mitigation entitled to great
[deference]. Add to that the family and
abusive environment and [this] mitigation
clearly outweighs any possible claim of
aggravation . . . . The situational stress build
up, including rejection by his mother, his
62 SPREITZ V. RYAN
girlfriend, and again that night, coupled with
poorly developed coping skills and alcohol
certainly mitigates the act.
(Emphasis added.) 3
But the majority nonetheless wishes to recast the defense
theories and reweigh the facts, as if we were conducting a
direct review. The majority singles out Dr. Flynn’s
observations that (1) Spreitz began drinking alcohol and
smoking marijuana at age twelve or thirteen, (2) Spreitz
drank steadily by age fifteen, (3) Spreitz’s “alcohol abuse
continued to intensify after he left home,” (4) Spreitz’s
alcoholism reached the level of physical dependence, and
(5) a “variety of persons” described Spreitz as a heavy
drinker, as intoxicated most of the time, and as having
blackouts while drinking alcohol. Those observations,
however, do not make Spreitz’s evidence of his longstanding
alcoholism, with no connection to the crime, central to his
plea for leniency.
Instead, Spreitz presented his long-term substance abuse
primarily to establish the statutory mitigating factors under
§ 13-703(G)(1) and § 13-703(G)(5). As to § 13-703(G)(1),
Dr. Flynn repeatedly emphasized that Spreitz’s lifetime of
substance abuse contributed to his murder of Ms. Reid.
Notably, Dr. Flynn’s report opined that:
The research on violent behavior shows a
strong correlation with alcohol intoxication.
3
Likewise, Spreitz claims in this appeal that his lawyer was
ineffective at sentencing by failing to present witnesses who had
extensive evidence about his early development, family history, head
injuries, substance abuse, and emotional problems. We have rejected
that claim in a contemporaneously filed memorandum disposition.
SPREITZ V. RYAN 63
Given [Spreitz’s] history, I can see a
perceived rejection by two women in the
same evening as especially disturbing to him,
particularly when intoxicated, to the point at
which he lost aggressive control, disinhibited
by the alcohol intoxication . . . .
My best guess is that he was drunk, hurt and
angry at the unwillingness of his girlfriend to
let him into her apartment, startled into an
alcoholic rage at the angry confrontation by
the victim, and ended up venting the years of
stored up rage at her.
And although several factors that normally
predict violence were not present, there is
alcoholism—one of the strongest correlates
of violent behavior. And, there is the
emotional, sexual maladjustment that comes
from his lifetime of misdirected development
in a dysfunctional family, combined with
arrested and/or distorted adult development
because of the alcoholism . . . .
1. Given the history of alcoholism . . . a
significant but unknown degree of alcohol
intoxication is likely. Alcohol intoxication
has a well-documented disinhibiting effect
which frequently includes losses of control of
angry emotions and aggressive behavior.
2. [Spreitz’s] history strongly suggests years
of early experiences likely to have caused a
build-up of pent-up angry, aggressive
feelings toward women generally (and older
64 SPREITZ V. RYAN
women especially) which may have burst
forth with uncontrollable intensity with or
without alcohol intoxication. Only trivial
provocation is required for this type of
aggression explosion, termed an Intermittent
Explosive Disorder by DSM-IV.
3. Still more likely is that a combination of
1. and 2. above contributed to an
uncontrollable outburst of aggression.
(Emphasis added.)
Dr. Flynn also gave extensive testimony on how
Spreitz’s substance abuse contributed to the impairment of
his mental state “in terms of conforming his behavior to his
understanding of the law and right and wrong generally.” At
the aggravation-mitigation hearing, Dr. Flynn went to great
lengths to opine that Spreitz’s alcohol abuse was “very
strongly correlated with violent behavior.” Dr. Flynn also
testified that Spreitz’s parents’ failure to treat his alcoholism
during his teenage years was significant because it increased
his violence potential: “[A]lcohol is probably the most
common, strongest correlative . . . with violent behavior
. . . , doubling the risk of violent behavior and I think the
absence of successful treatment for alcoholism is likely to
have been a significant and maybe a sine qua non that went
on in this offense.” (Emphasis added.)
In light of the evidence presented by Dr. Flynn, it is hard
to say that the Arizona courts did not consider the effect of
both his intoxication and his history of substance abuse as
required under § 13-703(G)(1). They just weren’t persuaded
by it. Rather, they reasonably and properly found that
Spreitz’s history of alcohol abuse was not sufficiently
SPREITZ V. RYAN 65
influential on his ability to conform his conduct to the law or
his ability to appreciate the consequences at the time he
viciously murdered Ms. Reid. Contrary to the majority’s
suggestion, Op. at 28–29, it cannot be said that the Arizona
courts never considered his longstanding alcoholism under
§ 13-703(G)(1) in the first place.
As to the statutory mitigating factor under
§ 13-703(G)(5), Dr. Flynn explained how Spreitz’s history
of substance abuse contributed to his emotional immaturity:
“When you spend most of your teenage years drunk or
stoned, it has not only been my observation in criminal
populations but my observations clinically . . . that virtually
all heavily drugged teenagers [are] socially and emotionally
immature as having substance abuse problems.” (Emphasis
added.) Dr. Flynn also testified that the lack of a treatment
effort in Spreitz’s home to deal with his alcoholism was
significant because it showed a lack of proper supervision
and parenting during his developing years, which
contributed to his emotional immaturity under
§ 13-703(G)(5):
[The lack of any treatment effort for Spreitz’s
alcoholism] is consistent with an uncaring
parent to have a child which is stealing booze
out of the house from the early teenage years,
who shows up drunk in his early teenage life
and everybody in the person’s life recognizes
there’s a serious alcoholic problem and yet
the parent either does one of two things,
doesn’t care enough to notice or doesn’t care
enough to do anything about it. That’s
important in and of itself in my opinion from
the standpoint of an evaluation of parenting.
66 SPREITZ V. RYAN
Thus, where Spreitz did not present his long-term
substance abuse to explain his conduct at the time of the
murder under § 13-703(G)(1), he did so to plead for leniency
based on his emotional immaturity under § 13-703(G)(5).
Any evidence that Spreitz presented of his long-term
substance abuse as non-statutory mitigation was cursory and
only ancillary to the considerable evidence he adduced to
explain how his past addiction affected his conduct and
immaturity at the time of the murder.
Moreover, the significant evidence presented regarding
Spreitz’s abusive and dysfunctional childhood, upbringing,
and interactions with his mother stands as further support
that Spreitz’s history of alcohol and substance abuse was not
central to his plea for leniency. Dr. Flynn testified to those
matters at length:
[Spreitz grew up in] a fully pathological
home environment, [which] include[d]
physical and emotional neglect, physical
abuse and emotional neglect and possibly
emotional abuse by the mother according to
which there was no hint of acceptance of him
in spite of ongoing efforts on his part to
please her. In fact of all the people that I
interviewed and all the interviews that I read,
the only favorable, the only person who had
a single favorable thing to say about the
mother was Mr. Spreitz himself. From all
other persons they described her in the most
negative terms. His sister described herself
as being saved from the pathological effects
of the home by the fact she ran away and was
placed in a group home. Mr. and Mrs. Spreitz
described most of the mother’s behavior in a
SPREITZ V. RYAN 67
manner that would be considered pathogenic
by anyone who knew anything about
psychology and parenting . . . .
[There is] massive evidence from a
psychological point . . . that there was a cold
uncaring mother who was anything but
sensitive to his needs, whom he was never
able to please. By this hypothesis it is most
common for children so treated to harbor a
store of anger, and it is consistent with this
train of thought that the anger was breaking
down partially by chronicity and partially by
series of emotionally unsuccessful
relationships with women, a tendency to
orient toward older women in the months
leading up to the offense[.] [S]hortly before
the offense . . . there had been a rejection by
a current girlfriend . . . . [T]his by itself and
probably especially this under the
disinhibiting influence of alcohol possibly
result[ed] in an explosion of aggressive
behavior especially toward a woman and
especially toward an older woman. The
recently pent-up rage accumulated over the
year aggravated the more recent months
which may be considered by the Court to
constitute an outburst of behavior that means
an inability to conform one’s behavior [to the
requirements of the law].
(Emphasis added.)
And, when Judge Sherrill asked Dr. Flynn what he
considered were non-statutory mitigating factors in this
68 SPREITZ V. RYAN
case, Dr. Flynn responded: “[T]he likely relationship
between the alcoholism and the violence generally and this
offense specifically and . . . the deprived pathogenic home
environment [are what] I would consider non-statutory
factors.” (Emphasis added.) In the face of this record, I am
at a loss to understand how my colleagues can say that
Spreitz’s history of substance abuse as a non-statutory
mitigating factor was central to or at the very core of
Spreitz’s plea for leniency. 4
B
Two other cases where we have found Arizona courts’
Eddings errors to be harmless, Henry v. Ryan, 720 F.3d 1073
(9th Cir. 2013), and Stokley v. Ryan, 705 F.3d 401 (9th Cir.
2012), are instructive here.
1
In Henry, the Arizona courts found a statutory mitigating
factor under § 13-703(G)(1)—namely, that “Henry’s
capacity to appreciate the wrongfulness of his conduct or to
conform his conduct to the requirements of law was
significantly impaired . . . because of his consumption of
alcohol on the day of the murder.” 720 F.3d at 1089–90.
Nonetheless, the courts imposed the death penalty because
they “agreed that the mitigation was not sufficiently
4
The majority does not take issue with the fact that the Arizona
courts considered Spreitz’s substance abuse history in the statutory
mitigating context or that they found it insufficient to outweigh the
aggravating factor of cruelty. “When applied solely in the context of
statutory mitigation,” the “causal nexus test does not violate Eddings.”
McKinney, 813 F.3d at 810.
SPREITZ V. RYAN 69
substantial to call for leniency” in light of two aggravating
factors. Id. at 1090.
As with Spreitz here, on habeas review, we assumed that
the state courts violated Eddings by failing to consider
Henry’s history of alcohol abuse as a non-statutory
mitigating factor. However, we denied habeas relief because
Henry failed to show that the Eddings error had a
“substantial and injurious effect or influence” on his
sentence. Id. at 1087–89 (quoting Brecht, 507 U.S. at 623).
We reasoned that, “in imposing the death penalty, the
Arizona courts already considered Henry’s intoxication at
the time of the murder as a mitigating factor” and concluded
that it was “not sufficient to call for leniency.” Id. at 1090.
And:
Given the similar nature of the mitigation, the
additional evidence of Henry’s historical
alcoholism would have had minimal
mitigating value. If the state courts
concluded that intoxication with a causal
connection to the crime was not sufficient to
call for leniency, it is highly doubtful that
they would have considered alcoholism
without a causal connection to be sufficient.
Id. 5
Similarly, here, the Arizona courts agreed that Spreitz’s
substance abuse at the time of the offense was not sufficient
to call for leniency. In addition, they found that Spreitz
could not show statutory mitigation under § 13-703(G)(1)
5
Although Henry involved different facts, as the majority points out,
its reasoning applies fully to Spreitz’s case. Op. at 36, n.15.
70 SPREITZ V. RYAN
because his “ability to appreciate the wrongfulness of his
conduct was not impaired on the night of the murder to any
significant extent by substance abuse.” Spreitz, 945 P.2d at
1281. The majority does not take issue with these findings.
As in Henry, the Arizona courts’ consideration of Spreitz’s
long-term substance abuse unrelated to the offense would
have had minimal mitigating value because of its similarity
to the evidence already rejected. Henry, 720 F.3d at 1090;
see also Landrigan, 272 F.3d at 1230 & n.9 (holding that,
where the sentencing court rejected the petitioner’s “alleged
intoxication and past history of drug use as a statutory
mitigating factor” and “did not go on to consider them as a
nonstatutory mitigating factor,” any “error in failing to
consider Landrigan’s use of alcohol and drugs would have
been inconsequential; it would have had no effect
whatsoever on the outcome”); Wong v. Belmontes, 558 U.S.
15, 23 (2009) (“The sentencing jury was thus ‘well
acquainted’ with Belmontes’ background and potential
humanizing features. Additional evidence on these points
would have offered an insignificant benefit, if any at all.”
(citation omitted)).
The majority thus errs by failing to apply Henry’s
reasoning to conclude that any Eddings error was harmless
here. See Hart v. Massanari, 266 F.3d 1155, 1172 (9th Cir.
2001) (“[I]f a controlling precedent is determined to be on
point, it must be followed.”). If direct evidence of
substance-induced intoxication on the night of the murder
was insufficient to outweigh the aggravated circumstances
of the crime, a long history of drug and alcohol abuse would
fare no better.
2
In Stokley, we assumed that the Arizona Supreme Court
committed Eddings error by failing to consider Stokley’s
SPREITZ V. RYAN 71
abusive childhood and good behavior in jail because of a
lack of causal nexus. 705 F.3d at 403–04. We held,
however, that this error was harmless because there were
three aggravating circumstances, including that Stokley—
like Spreitz—had “committed the offense in an especially
heinous, cruel, and depraved manner.” Id. at 404. We noted
the gruesome details of Stokley’s offense, including (1) that
he had sexual intercourse with his victims; (2) that the
victims’ bodies (two young girls) were violently beaten; and
(3) that the victims’ bodies were dragged to and thrown
down a mineshaft. Id. at 405 n.1. We also observed that
both the sentencing court and the Arizona Supreme Court
consistently determined that “leniency was inappropriate.”
Id. at 405. We found no likelihood “that, but for a failure to
fully consider Stokley’s family history or his good behavior
in jail during pre-trial incarceration, the Arizona courts
would have come to a different conclusion” as to Stokley’s
sentence. Id.
Although Stokley’s facts were relatively more egregious
than the facts here—Stokley involved three aggravating
circumstances instead of one and Stokley’s victims were
under the age of fifteen—Stokley nonetheless informs us that
the severity of an aggravating circumstance must be
considered when determining whether an Eddings error is
harmless. Here, both the sentencing court and the Arizona
Supreme Court consistently ruled that leniency was
inappropriate because of the especially cruel manner in
which Spreitz kidnapped and murdered Ruby Reid. Spreitz,
945 P.2d at 1266, 1278–79 (citing Ariz. Rev. Stat. § 13-
703(F)(6)). The crime scene photographs alone give
compelling support to that ruling, showing Ms. Reid’s body
beaten beyond recognition by heavy rocks. The pathologist
underscored the extensive damage to her body revealed on
autopsy.
72 SPREITZ V. RYAN
Arizona courts consider five factors in “determining
whether a murder was especially heinous, cruel or depraved:
(1) relishing the murder, (2) infliction of gratuitous violence,
(3) needless mutilation of the victim, (4) senselessness of the
crime, and (5) helplessness of the victim.” Gulbrandson v.
Ryan, 738 F.3d 976, 984 (9th Cir. 2013) (citing State v.
Gretzler, 659 P.2d 1, 11–12 (Ariz. 1983) (en banc)). “A
finding of cruelty is warranted when the defendant inflicts
on the victim mental anguish or physical abuse before the
victim’s death.” Spreitz, 945 P.2d at 1278. “Cruelty is found
when the victim is conscious at the time of the offense in
order to suffer pain and distress.” Id. (citation and alterations
omitted).
Here, the Arizona courts found that Ms. Reid suffered
tremendous mental and physical pain when she was forced
into Spreitz’s car trunk and transported to the desert where
she was beaten, sexually assaulted, and eventually murdered.
Id. at 1265, 1278. Judge Sherrill observed: “At the scene of
the murder the victim was thoroughly beaten. The autopsy
showed, in addition to the death-causing blows to the head,
five ribs broken, eighteen bruises and grab marks, internal
bruising and bleeding, lacerations[,] and a broken jaw.”
Spreitz did not confess to kidnapping Ms. Reid, but did
admit to beating her as she fought back, removing her
clothes, having intercourse with her, and smashing her in the
head with a rock when she would not stop yelling. Id.
Physical evidence found at the scene of the murder
corroborated Spreitz’s confession: Ms. Reid’s clothing,
including her torn brassiere, was strewn in one area of the
scene, and photographs of the area vividly depicted drag
marks running from that area to the spot where Ms. Reid’s
body and the bloody rocks were found. According to
Spreitz’s confession, when he left Ms. Reid lying in the
desert on the morning of May 19, 1989, he did not know
SPREITZ V. RYAN 73
whether she was still alive or dead. Also significant to Judge
Sherrill’s finding of mental anguish was that Ms. Reid had
defecated in and on her clothing. See id. at 1265, 1278–79.
In its independent review of these facts, the Arizona
Supreme Court agreed that Spreitz had murdered Ms. Reid
in an especially cruel manner. Id. at 1278. The Arizona
Supreme Court highlighted Spreitz’s “own admission that he
beat her as she fought back and hit her with the rock when
she would not stop yelling,” which it found was “clear
evidence of [Ms. Reid’s] conscious suffering.” Id. at 1279.
The court found that Spreitz’s confession and the physical
evidence all pointed to the conclusion that Spreitz “beat and
raped [Ms. Reid] in a brutal assault that lasted many minutes
before he crushed her skull.” Id. My colleagues in the
majority do not quarrel with that finding.
Thus, even assuming that the Arizona courts failed to
consider Spreitz’s evidence of long-term substance abuse
unrelated to the crime, it is unlikely their consideration of
that evidence would have substantially altered their ultimate
decision to sentence Spreitz to death based on the aggravated
manner in which Ms. Reid met her death. And we cannot
say “with fair assurance” that the sentence they imposed was
substantially swayed by their failure to consider that
evidence. McKinney, 813 F.3d at 822 (quoting Kotteakos,
328 U.S. at 765).
Further, Eddings allows the Arizona courts “to consider
the absence of a causal connection when assessing the
quality and strength of [mitigating] evidence.” Henry,
720 F.3d at 1090. It does not prevent the Arizona courts
from “assign[ing] less weight to mitigating factors that did
not influence a defendant’s conduct at the time of the crime.”
Hedlund, 854 F.3d at 587 n.23. The Arizona Supreme Court
has recognized this principle in its more recent cases. See,
74 SPREITZ V. RYAN
e.g., State v. Newell, 132 P.3d 833, 849 (Ariz. 2006) (en
banc) (“We do not require that a nexus between the
mitigating factors and the crime be established before we
consider the mitigation evidence. But the failure to establish
such a causal connection may be considered in assessing the
quality and strength of the mitigation evidence.” (citation
omitted)). Thus, we can confidently say that if Spreitz were
resentenced, the Arizona courts would give even less weight
to his history of substance abuse unrelated to the crimes in
balancing that evidence against what he did to Ruby Reid.
C
The majority cites Henry for the broad proposition that
“a ‘history of substance abuse [is] substantially mitigating
when . . . combined with other mitigating evidence.’” Op.
at 36 (emphasis omitted) (quoting Henry, 720 F.3d at 1090).
But the majority has not offered up any authority that reaches
this conclusion under factually-similar circumstances.
Indeed, unlike the three cases Henry relies upon in making
this statement, Spreitz’s defense counsel presented
considerable evidence at sentencing of his lifelong history of
substance abuse, his mental and emotional problems, and his
family dysfunction. Counsel retained Dr. Flynn, a
psychotherapist and forensic consultant, who personally
evaluated Spreitz, performed psychological assessments,
and interviewed people who were familiar with him. Dr.
Flynn submitted a report and testified extensively regarding
Spreitz’s alcohol and drug addictions, his family history of
substance abuse, his disruptive childhood, and his emotional
problems. Thus, the majority’s suggestion that Spreitz’s
case is comparable to those in which we have found evidence
of longstanding substance abuse to be substantially
mitigating is misplaced. This is particularly so in light of the
horrific way in which Ruby Reid died, the impact of which
SPREITZ V. RYAN 75
was not lost on all of the Arizona judges who reviewed this
record.
D
Lastly, Spreitz’s own arguments on appeal belie the
conclusion that his long-term substance abuse unrelated to
the crime was central to his plea for leniency. Spreitz argues
that whether he “was intoxicated at the time of the murder
was critical” to his case, and that “the issue of [his]
intoxication [at the time of the crime] was critically
important.” He contends that he “was highly intoxicated at
the time of the murder” and that “this was the primary cause
of him committing the murder.” And, he urges that there
“was ample evidence that [he] was highly intoxicated at the
time of the murder and that his intoxication contributed
directly to his actions.” “This was highly relevant
mitigation,” he says: “[h]ad the judge heard the evidence
described herein, it is highly probable that he would not have
sentenced [Spreitz] to death.” But the judge did hear all of
that evidence. He just didn’t buy the argument based on the
testimony of officers who interacted with Spreitz before and
after he killed Ms. Reid. That credibility determination is
not assailable on appeal.
Spreitz’s defense theories have shifted on collateral
appeal. Contrary to his position in supplemental briefing
ordered by this Court post-McKinney, Spreitz does not make
any similar arguments about his long-term substance abuse
as a non-statutory mitigating factor. Instead, he claims that
evidence of his childhood head injuries that his counsel
failed to present were vital to his mitigation evidence.
Spreitz argues that the “importance of counsels’ failure to
investigate [his] history of head injuries cannot be
overstated;” that the “connection between childhood head
injuries and impulsive murder” was “well established” at the
76 SPREITZ V. RYAN
time of sentencing; that “there was a correlation between
[his] head injuries as a child and his impulsive behavior in
the instant case;” and that “[expert] testimony could have
enlightened the court as to the connection between childhood
head injuries and impulsive behavior in murderers in
general, and could have developed the link between
[Spreitz]’s own head injuries and his impulsive behavior
during this murder in particular.” He contends that:
The murder in this case was a classic
impulsive act. It was not thought out or
premeditated beforehand. Appellant beat the
victim with a rock; he did not even bring a
weapon. The victim refused to have sex with
him, fought him, and he became enraged and
impulsively killed her. That is the classic
behavior of a murderer who suffered
childhood head injuries.
(Emphasis added.) Spreitz’s shifting defense theories
undermine his latest argument that long-term substance
abuse was central to his plea for leniency all along.
We will never know how Ruby Reid, forced into the
trunk of Spreitz’s car and pondering her fate as he drove her
into the desert, would have answered any of Spreitz’s
contentions. But the forensic evidence of the way in which
she died stands in silent, but compelling refutation of that
defense. Simply put, Spreitz cannot show on this record that
his long-term substance abuse, as a non-statutory mitigating
factor and without any connection to the crime, was either
“central to his plea for leniency,” McKinney, 813 F.3d at
823, or at “the very core of [his] plea for life,” Coleman,
210 F.3d at 1051. Therefore, he cannot demonstrate that the
Arizona courts’ alleged failure to consider that evidence
SPREITZ V. RYAN 77
substantially influenced or swayed his ultimate sentence.
Nor can he show that, had the Arizona courts further
considered its marginal mitigating weight, the evidence
would have substantially affected the outcome of his
sentence. See Brecht, 507 U.S. at 623. Because Spreitz
cannot show that the error he alleges was not harmless, the
district court’s denial of habeas relief on Spreitz’s Eddings
claim should be affirmed.
From the majority’s conclusion otherwise, I respectfully
dissent.