People v. Cervantes

Filed 3/9/17

CERTIFIED FOR PARTIAL PUBLICATION*

IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA

FIRST APPELLATE DISTRICT

DIVISION FOUR

THE PEOPLE,

Plaintiff and Respondent,

A140464

v.

ALEXANDER CERVANTES, (Solano County

Super. Ct. No. FCR281334)

Defendant and Appellant.

Alexander Cervantes was 14 years old when he attacked a 13-year-old girl and her

20-month-old brother, who were the younger siblings of one of his friends. After

breaking into their home in the middle of the night, he stabbed them repeatedly as they

slept, raped and sodomized the girl, forced her to orally copulate him, and ultimately

passed out during the attack. He had been drinking heavily that evening and his defense

rested on voluntary intoxication to negate specific intent. He was convicted of 15

charges, including various sex offenses, first-degree burglary, and two counts each of

attempted murder, torture, and aggravated mayhem. He received a prison sentence of 50

years to life under the one-strike law (Pen. Code,1 § 667.61), a consecutive 11-year

determinate term for one attempted murder (§§ 187, 664), plus a consecutive life term for

the other attempted murder.

We divide our discussion into three parts. Only an overview of the first part shall

be published, but the remaining two parts are certified for publication in their entirety

*

Pursuant to California Rules of Court, rules 8.1105(b) and 8.1110, this opinion is

certified for publication with the exception of part II.A.2 through II.A.7.e.

1

Statutory references, unless otherwise indicated, are to the Penal Code.

1

First, Cervantes argues that the representation he received was so far below the

minimum threshold of constitutionally effective assistance of counsel as to amount to no

defense at all. Pointing to dozens of shortcomings—beginning with an incomplete

investigation of his mental state, which he says guaranteed his counsel either had no basis

for strategic choices she made or simply failed to recognize choices she should have

made—he asks that we reverse outright, and remand for a new trial. While we reject that

argument, we agree there were a number of serious deficiencies in counsel’s

performance, enough to leave us without confidence in the outcome of the trial on most

of the specific intent crimes. We therefore reverse on eight specific intent counts, while

affirming as to the remaining seven, including the convictions for burglary and all of the

general intent crimes (four of the six sex offenses, and two counts of assault with a

deadly weapon).

Second, our conclusion that we must reverse and remand for retrial on eight of 15

counts presents some novel issues under recently-passed Proposition 57, the Public

Safety and Rehabilitation Act of 2016 (Prop 57). Cervantes argues Prop 57 requires that

the case be remanded for a “fitness hearing” to the juvenile court, which he contends has

“exclusive jurisdiction” over any trial of the offenses charged in this case until and unless

it determines that the case should be transferred to adult criminal court, and further, that a

remand for retrial on any of the counts of which he was convicted requires vacatur of all

the convictions and retrial of all charges. He argues that Prop 57 is retroactive, but that

this result is mandated even applying Prop 57 prospectively to any proceedings on

remand after a partial reversal. We do not agree that Prop 57 is retroactive. Nor do we

agree that a partial reversal requires that all convictions must be vacated. But we do

agree that Prop 57 requires a remand to the juvenile court for a “fitness hearing,” and that

the outcome of that hearing will determine which department of the Superior Court—

adult criminal court, or juvenile court—will handle any retrial on the reversed counts and

sentencing.

Third, Cervantes argues that the sentence imposed on him is the functional

equivalent of life without possibility of parole and therefore violates the Eighth

2

Amendment under Graham v. Florida (2010) 560 U.S. 48 (Graham) and People v.

Caballero (2012) 55 Cal.4th 262, 268 (Caballero). Since the sentencing choices made on

the convictions we affirm, even without conviction on the counts we reverse, could

produce another indeterminate life sentence with a lengthy minimum term, this Eighth

Amendment issue will be relevant to the proceedings on remand whether Cervantes’s

case is handled in adult criminal court or stays in juvenile court. Thus, we address it and

conclude that a sentence requiring Cervantes to serve at least 66 years in prison before he

would first become eligible for parole, is constitutionally infirm. Because that term

exceeds his life expectancy, it is the functional equivalent of life without parole and

violates the Eighth Amendment under Graham, at p. 74 and Caballero, at p. 268. Where,

exactly, the constitutional line lies below the 66 years to life imposed in this case has yet

to be addressed by our Supreme Court, although we note that some guidance on the issue

will likely be forthcoming in a case now pending before it. In the meantime, the Prop 57

“fitness hearing” we order today will rectify any constitutional concerns about the length

of whatever term of confinement is imposed on Cervantes for the convictions we affirm

here, as well as for any other offenses that may be tried on remand.

I. FACTUAL AND PROCEDURAL BACKGROUND

A. Facts Disclosed at Trial

1. Prosecution’s Case

On the evening of December 11, 2010, Cervantes attended a party, where

witnesses saw him drinking from 6:00 p.m. to 9:00 p.m. One of the partygoers described

him as “really drunk” at 10:20 p.m.

At approximately 1:30 or 2:00 a.m. on December 12, 2010, Cervantes entered the

Vacaville home of his sometime friend, Gabriel T. (Gabriel).2 He carried a steak knife he

2

The true nature of the relationship between Cervantes and Gabriel was

ambiguous. There was evidence that they had a “very strong relationship,” Cervantes

“worship[ped]” Gabriel, and Gabriel taught Cervantes about taking drugs. But

Cervantes, who was younger, was also described as “sort of the lackey,” and Gabriel told

3

had brought from home and had a condom with him. Gabriel was not at home because he

was in juvenile hall. The only occupants were Gabriel’s younger sister, 13-year-old A.P.,

and their 20-month-old brother, I.A. A.P. was babysitting I.A. while their mother and

another sister were out of the house.

As the evening progressed, A.P. and I.A. fell asleep on their mother’s bed. A.P.

awoke to the sensation of a sharp pain in her back like someone was punching her, and

she saw blood underneath her on the bed. Looking in the mirror, she saw someone else

in the room. Turning, she recognized Cervantes as a friend of her older brother’s. A.P.

testified he was wearing black gloves and a black jacket.3 She asked Cervantes why he

was there. “He said that it was revenge for my brother, against Richard—something with

an R,” or “revenge for [her] brother for snitching on . . . Richard . . . .” A.P. did not know

what he was talking about. As A.P. tried to unlock her cell phone to call 911, she fell off

the bed. While she was on the floor, Cervantes stabbed her repeatedly in her head, back,

and shoulders. A.P. suffered defensive wounds on her wrists and elbow when she tried to

shield her head. She begged Cervantes to stop and eventually he did. After Cervantes

stopped stabbing her, A.P. heard her baby brother “squeal,” but she could not move to

help him. Then the screaming stopped.

Cervantes came back over to A.P. and “started getting into [her] pants.” He put

his hand down the back of her pants and “was feeling on [her] butt.” Cervantes told A.P.

to get on the bed and she complied out of fear. Cervantes got on top of her and put his

penis in her vagina. After a couple of minutes, he told A.P. to moan and call him

“daddy.” When A.P. did not comply, Cervantes threatened to kill her brother, by which

the police he did not like Cervantes very much and called him “hella stupid,” “retarded,”

and a “pussy.”

3

Defense counsel elicited testimony on cross-examination of a police witness that

Cervantes was not wearing gloves at the time of his arrest, and that the officer did not

know of any gloves found on the premises. An investigating detective did not recall A.P.

saying the suspect wore gloves when she spoke to police near the time of the crime. The

implication was that A.P. may have embellished her story and was not trustworthy.

4

A.P. thought he meant Gabriel. A.P. complied with Cervantes’s demands. Cervantes

“finally stopped” and told her to “turn around.” A.P. rolled onto her stomach, and

Cervantes sodomized her. Cervantes “finally stopped,” then told A.P. she “was going to

suck his dick.” Cervantes “kind of put his penis up into [her] mouth” and “pulled [her]

head down.” When A.P. saw Cervantes starting to doze off, she tried to get away, but he

grabbed her hair and “yanked [her] back down.” Eventually, Cervantes fell asleep.

A.P. grabbed her phone and her little brother and quickly got out of the room.

Entering the kitchen, she noticed the door to the garage was standing open, as was the

door leading from the garage to the outside of the house. Both doors had been closed

earlier and the external door was normally kept locked. Once outside, A.P. called 911

and went across the street to a neighbor’s house.

Shortly after 4:00 a.m. a SWAT team responded to A.P.’s home, eventually

entered, and arrested Cervantes in the master bedroom. He was naked and smeared with

blood. A large bloodstain and what appeared to be urine were found on the sheets.

When the crime scene was processed three empty beer cans were found: one was

on the couch, one by the sink in the kitchen, and the third on the driveway in front of the

garage. Cervantes’s DNA was found on two of the beer cans. The parties stipulated that

both victims’ blood was found on a knife retrieved from the bedroom where the children

were attacked, and the investigation showed the knife matched another steak knife found

at Cervantes’s mother’s house. The parties also stipulated that (1) DNA testing could not

exclude Cervantes as the source of semen found on anal swabs taken from A.P., and (2)

A.P.’s dried blood was found on Cervantes at the time of his arrest.

After the attack, the children were taken to the hospital. A.P. had been stabbed a

total of 42 times, and I.A. had been stabbed 13 times. A.P. had a collapsed lung and a

lacerated liver and spleen. I.A. had four fractured ribs and internal injuries. A.P. was

hospitalized for three days, and both children were scarred by the knife wounds they

suffered that night.

5

2. The Charges, the Defense Case, and the Verdicts

Cervantes was charged as an adult by information filed on March 29, 2011.

(Welf. & Inst. Code, former §§ 602, subd. (b)(2), 707, subd. (d)(2)(A).) An amended

information filed on September 17, 2012 charged Cervantes with two counts of attempted

willful, deliberate, and premeditated murder (§§ 664, 187) (counts 1, 2), first degree

burglary (§ 459) (count 3), two counts of torture (§ 206) (counts 4, 5), two counts of

aggravated mayhem (§ 205) (counts 6, 7), two counts of assault with a deadly weapon

(§ 245, subd. (a)(1)) (counts 8, 9), assault with intent to commit rape, sodomy, or oral

copulation on a person under 18 years (§ 220, subd. (a)(2)) (count 10), forcible rape of a

child under 14 years (§§ 261, subd. (a)(2), 264, subd. (c)(1)) (count 11), forcible sodomy

of a child under 14 years (§ 286, subd. (c)(2)(B)) (count 12), two counts of forcible oral

copulation of a child under 14 years (§ 288a, subd. (c)(2)(B)) (counts 13, 14), and

forcible lewd act on a child (§ 288, subd. (b)(1)) (count 15). The information further

alleged as to counts 1 through 7 and 10 through 15 that Cervantes personally used a

deadly and dangerous weapon, a knife (former § 12022, subd. (b)(1)); as to counts 1

through 3 and 6 through 15 that Cervantes personally inflicted great bodily injury (former

§ 12022.7, subd. (a)); and as to counts 10 through 15 that Cervantes inflicted great bodily

injury (§ 12022.8). The information included allegations as to counts 11 through 15

under section 667.61, known as the “one-strike” law, which calls for life sentences,

including mandatory minimums, for individuals convicted of certain sex offenses in

particularly blameworthy circumstances. The one-strike allegations against Cervantes

included that the sex crimes were committed during a first-degree burglary (§ 667.61,

subds. (d)(4), (e)(2)), that Cervantes also committed aggravated mayhem against and

tortured A.P. (§ 667.61, subd. (d)(3)), personally inflicted great bodily injury on her

(§ 667.61, subd. (d)(6)), personally used a dangerous or deadly weapon (§ 667.61,

subd. (e)(3)), and committed the crimes against a child under age 14 (§ 667.61, subds.

(d)(7), (j)(1), (j)(2)).

Trial was held in mid-September 2012. Building on the testimony that Cervantes

had been drinking during the evening before the attacks, defense counsel’s strategy was

6

to defend against all of the crimes requiring specific intent based on voluntary

intoxication. The parties stipulated that Cervantes’s blood alcohol level at 8:57 a.m. on

December 12, 2010 was 0.12 percent. A forensic toxicologist, Jeffery Zehnder, testified

for the defense that Cervantes’s blood alcohol level could have been as high as .25

percent at the time of the crimes. Zehnder testified about increasing levels of impairment

that occur with increasing levels of blood alcohol. At .12 percent, “ability to do certain

things, like speak” would be impaired, and at .15 percent there would be not only “slurred

speech,” but also “impaired information processing in the brain.” Zehnder also testified

that alcohol reduces inhibitions and can cause an individual to behave and speak

recklessly and inappropriately, and a .20 percent blood alcohol level could cause severe

impairment, memory loss, and black-outs.

Defense counsel argued to the jury that Cervantes was so intoxicated that he never

formed the specific intent to kill and never premeditated or deliberated, and thus could

not be convicted of attempted murder. She also argued, due to his inebriated state,

Cervantes had not formed the intent required for burglary, torture or the other specific

intent crimes.

On September 21, 2012, the jury found Cervantes guilty of all the substantive

offenses, many of the enhancements, and found true some but not all of the one-strike

allegations.4 With respect to all the sex crimes, it found he personally used a deadly

4

As to count 1 (attempted murder of A.P.) the jury found there was premeditation

and deliberation, but as to count 2 (attempted murder of I.A.) it did not find premeditation

or deliberation (attempted second-degree murder). (§§ 187, 664.) As to both attempted

murders it found personal use of a deadly weapon and personal infliction of great bodily

injury. (§§ 12022, subd. (b)(1), 12022.7.) It also found Cervantes guilty of the following

crimes and enhancements: first-degree burglary with use of a deadly weapon and

personal infliction of great bodily injury on both children as well as burglary in the

commission of the sex crimes. (§§ 459, 460, 667.61, subds. (d)(4), (e)(2), 12022,

subd. (b)(1)); torture of both children and aggravated mayhem with respect to both, with

great bodily injury and personal deadly weapon use on those four counts (§§ 205, 206,

12022, subd. (b)(1), 12022.7); personal infliction of great bodily injury in the commission

of two counts of assault with a deadly weapon (§ 245, subd. (a)(1); 12022.7); assault with

7

weapon and committed the five alleged as one-strike offenses against a child under 14 in

the course of a first-degree burglary (§ 667.61, subds. (d)(4), (e)(2), (e)(3), (j)(1), (j)(2)),

but found the infliction of great bodily injury, torture and aggravated mayhem one-strike

allegations not true. 5 (§ 667.61, subds. (d)(3), (d)(6), (e)(3).) The true findings on the

one-strike allegations subjected Cervantes to a sentence of 25 years to life on each of the

five one-strike sex offenses. (§ 667.61, subds. (a), (e)(2), (e)(3), (j)(1), (j)(2).) The jury

deliberated for approximately seven hours before reaching its verdicts.

On November 16, 2012, Cervantes’s trial counsel filed a three-page motion for

new trial, arguing the evidence was insufficient to prove he had the required specific

intent to support the convictions on counts 1 through 4, 6, 7, 10 and 15. Counsel also

filed a thick stack of letters from family, friends, mentors, and community members in

support of Cervantes, to be used at sentencing.

B. Facts and Theories Developed by Pro Bono Counsel after Trial

After Cervantes was convicted, Peter Obstler, a partner in the San Francisco office

of Bingham McCutchen (and later of Arnold & Porter), took an interest in the case and

attended the hearing on the new trial motion. With Obstler in the lead, Bingham

McCutchen substituted in as Cervantes’s counsel on a pro bono basis, and Obstler and

Arnold & Porter continue to represent him on appeal. A new date was set for the hearing

of the motion for a new trial.

intent to commit rape, forcible rape on a person under age 14, sodomy by force on a

person under age 14, two counts of forcible oral copulation on a person under age 14, and

forcible lewd act on a child (§§ 220, subd. (a)(2), 261, subd. (a), 264, subd. (c)(1), 286,

subd. (c)(2)(B), 288, subd. (b)(1), 288a, subd. (c)(2)(B)).

5

The jury’s findings were not sufficient to support a one-strike sentence under

section 667.61, subdivision (d)(4), which requires not only that the sex crimes occurred

during a first degree burglary, but that the burglary was committed “with intent to

commit” one of the sex crimes specified in subdivision (c). (§ 667.61, subd. (d)(4).) The

jury’s findings do not include the intent element. But the findings do support a 25-to-life

sentence on each of the qualifying sex offenses based on burglary coupled with personal

use of a deadly weapon, and because the victim was under age 14. (§ 667.61, subds. (a),

(e)(2), (e)(3), (j)(1), (j)(2).)

8

As he came up to speed on the case, Obstler learned that three different attorneys

from the Conflict Defender’s Office had represented Cervantes following his arrest. First

was William Pendergast, who represented Cervantes from December 2010 to March

2011, when he was appointed as a commissioner. Then Kathryn Barton took over

Cervantes’s defense until she resigned from the Conflict Defender’s Office around the

end of 2011 or January of 2012. Beginning in February 2012, Erin Kirkpatrick, who was

new to the Conflict Defender’s Office, took over the representation and continued in that

role through trial. Kirkpatrick had more than ten years of experience as a criminal

defense attorney and entered the Conflict Defender’s employ at a high level of experience

and responsibility.

After reviewing the transcripts and conducting his own investigation, aided by

experts, Obstler came to advocate a different theory of the crime than Kirkpatrick had

pursued at trial. He noted and built upon the fact that, early in the defense investigation

of the case, Cervantes reported to Pendergast that, in addition to drinking, he had also

smoked marijuana6 and taken psilocybin mushrooms on the night of the crimes, which he

had purchased from a “Northsider” at the party he had attended before the attacks.

Cervantes specifically told Pendergast he did not get the mushrooms from Gabriel.

Cervantes told Pendergast he had taken Ecstasy (MDMA), which he purchased from

Gabriel, on numerous occasions but had not used MDMA on the night of the attacks.

Another recent criminal case in town also involved a defendant named Richard Calkins,

who took psilocybin mushrooms and then shot and killed two of his best friends and

maimed a third while they were watching television, with no preceding argument or other

provocation. Calkins’s crime and the role of psilocybin mushrooms in it had been

reported in the local newspapers.

Even though Cervantes and Calkins apparently did not know each other, Obstler

embraced the theory that the “Richard” mentioned by A.P. in her testimony was this very

6

A test of Cervantes’s blood or urine was positive for THC (the active ingredient

in marijuana). This was not brought out at trial.

9

same Richard Calkins. He argued at the new trial motion and again on appeal that

Cervantes never made the remark about “Richard,” and A.P. probably misunderstood him

or made it up from publicity about the Calkins case because her family had become

suspicious that they were the targets of someone seeking revenge against Gabriel for

having provided the mushrooms that resulted in such violent consequences.

Despite their lack of an actual acquaintance with one another, Obstler claims there

was a connection between Cervantes and Calkins. It turned out that Gabriel had sold the

psilocybin mushrooms to Calkins more than three weeks before Cervantes’s crimes, and

Gabriel admitted as much to the police shortly after Calkins was arrested. That is

apparently why he was in juvenile hall. Based on the reactions by Calkins and Cervantes

after eating the mushrooms, Dr. Alex Stalcup, a defense addiction expert, speculated that

the mushrooms in both cases may have been adulterated with MDMA or “angel dust”

(PCP); the expert admitted on cross-examination that he did not, in fact, know whether

the mushrooms were tainted. According to expert testimony in the Calkins case, some of

the mushrooms in his possession did test positive for MDMA. From these leads, Obstler

and his team have developed the theory that Gabriel also sold psilocybin mushrooms to

Cervantes at around the same time he sold them to Calkins and they, too, were tainted

with MDMA.7

Obstler also pointed to additional information about Cervantes’s developmental

and mental health history that could have been used at trial to help explain his behavior

7

Among the expert advice Obstler received was legal advice on competency of

counsel in criminal defense from John Coffer, an attorney with 34 years of experience in

both the prosecution and defense of criminal cases. Coffer prepared a lengthy written

report that was admitted into evidence at the hearing on the new trial motion. The report

is not signed under oath and is replete with hearsay, speculation and opinions well

beyond his expertise. For instance, at the time he filed his report, Coffer’s theory was

that the mushrooms had been adulterated with PCP based on speculation by Dr. Alex

Stalcup, an addiction expert hired by Obstler after trial. Despite this earlier theory,

Obstler has now focused his suspicion on MDMA because of the similarities with the

Calkins case.

10

that night. Cervantes’s umbilical cord was wrapped around his neck at birth, resulting in

oxygen deprivation. He was required to use oxygen machines to maintain breathing at

times in his childhood. He had learning difficulties in school and had an individualized

education program (IEP). He was in a car accident as a child and suffered some head

injuries, at once characterized as “minor,” but having led to “moderately severe brain

damage.” He was diagnosed with attention deficit hyperactivity disorder (ADHD) in

elementary school and had been on medication for it since second grade.

But the evidence of Cervantes’s developmental and mental health history was not

all helpful. In one IEP, it was noted that Cervantes had an atypical obsession with sex

and violence, so much so that the evaluator wondered if he had been sexually abused.

When he was in the fifth or sixth grade he was removed from school and taken to the

hospital after he said he was going to “kill [his] mother and burn [down the] house.”

To help interpret the mental health evidence, Obstler sought out an expert, Dr.

Michael Shore, a neuropsychologist, who conducted a battery of neurological tests and

examinations, which led him to conclude that Cervantes had a baseline mental status of a

ten-year-old. Among the tests he ordered performed was a single photon emission

computed tomography (SPECT) study, which is a nuclear imaging test that employs

glucose injected with radioactive material to measure brain activity. The SPECT test

itself was performed by Dr. Daniel Amen, and his report was attached as an exhibit to the

motion for a new trial. Dr. Amen concluded: “The most significant finding is decreased

activity in the orbitofrontal cortex . . . consistent with an ADHD like process.” He also

found scalloping, “often indicative of some form of toxic exposure,” and decreased

activity in the prefrontal pole, which “may indicate a history of brain injury.”

Based in part on Dr. Amen’s report, Dr. Shore concluded that Cervantes’s brain

was “diffusely damaged” and found specific “compromise” of the prefrontal cortex. He

opined that Cervantes was in a state of “acute organic psychosis” when he attacked the

children. Cervantes was also suffering from “longstanding and quite significant”

depression, “moderate brain damage,” and learning disabilities. Dr. Shore believed

Cervantes’s drug use partly explained his behavior that night, and he was allowed to

11

testify to Cervantes’s drug and alcohol history. He opined that Cervantes’s substance

abuse may have contributed to his brain damage, along with lack of oxygen at birth and

“multiple minor head injuries.” Dr. Shore accepted as a fact that Cervantes had taken

psilocybin mushrooms on the night of his crimes. His opinion was based on brain

damage, psychedelic drug ingestion, and depression; no single factor could be considered

the cause of the psychotic episode.

Dr. Stalcup, a physician specializing in addiction medicine, also conducted an

examination of Cervantes, focusing on the effect of hallucinogenic drugs on someone

with his specific cognitive impairments, as described by Dr. Shore. Assuming Cervantes

consumed psilocybin mushrooms on the night of the crimes, Dr. Stalcup endorsed Dr.

Shore’s opinion and concluded that Cervantes’s alcohol and drug use caused an organic

psychosis and Cervantes attacked his victims in a dissociative state. His opinion relied

on Dr. Shore’s report and on Cervantes’s report of having taken psilocybin mushrooms

that night. He was not allowed to testify that Cervantes was, in fact, using psilocybin

mushrooms that night. He also thought Cervantes was “possibly” under the influence of

MDMA. Dr. Stalcup also opined, however, even without psilocybin mushrooms,

Cervantes’s alcohol and marijuana use alone, combined with the deficits revealed in the

SPECT scan, would have led to organic psychosis.

C. The Motion for a New Trial

On April 16, 2013, Obstler filed a motion for a new trial. Evidentiary hearings

were conducted over the next several months, and on October 18, 2013, the court denied

the motion. In addition to expert psychological testimony from Drs. Shore and Stalcup,8

summarized above, Obstler called his legal expert, Coffer, to testify as an expert witness

on competency of criminal defense counsel. (See fn. 7, ante.) In a lengthy written report

8

Drs. Shore and Stalcup wrote reports summarizing their findings and opinions,

but the trial court would not admit either report into evidence because they both

contained hearsay. The judge’s hearsay rulings in connection with expert testimony in

these reports and in the hearing itself were, for the most part, consistent with People v.

Sanchez (2016) 63 Cal.4th 665, 674–686.

12

with multiple attachments, as well as in testimony on the new trial motion, Coffer

catalogued the errors and omissions he saw in Kirkpatrick’s preparation for trial and trial

strategy, and ultimately opined that Kirkpatrick had provided ineffective assistance of

counsel. Obstler also called as witnesses two mental health experts retained by the

defense before trial, as well as Kirkpatrick, Pendergast and Barton, and two of

Kirkpatrick’s superiors in the Conflict Defender’s Office. The testimony will be

summarized below, in the unpublished section of the opinion.

The trial court denied Cervantes’s motion for a new trial, finding Kirkpatrick

credible and her decisions “competently made based upon the information she had at the

time.” The court further found that, had the defense presented at trial the evidence

presented in the new trial motion, “there is no reasonable probability that a different

result would occur.” The mental health evidence had included an “[atypical] obsession

with sex and violence and lack of control and behavioral problems” and, had it been

presented to the jury along with “the actual horrific facts of the crime itself,” it “would

have been absolutely devastating to the defense.” The court also opined that “more

evidence of voluntary intoxication through marijuana and psychedelic mushrooms”

would not have resulted in a more favorable verdict for Cervantes.

D. Sentencing

On October 28, 2013, the court sentenced Cervantes to 50 years to life for two of

the sex crimes under the one-strike law (two consecutive terms of 25-to-life), plus 11

years imposed consecutively for the attempted murder of I.A., plus an additional

consecutive life term for the attempted murder of A.P. The remaining counts were either

imposed concurrently or stayed under section 654. Thus, Cervantes’s aggregate sentence

was 61 years to life, plus a consecutive life term. Importantly, the longest terms were

attributable to the sex offenses under the one-strike law, four of which were general

intent crimes (rape, sodomy and two counts of oral copulation). Under section 667.61,

Cervantes was not awarded any credits for good conduct before sentencing. (See section

II.C.2.c., post.) The probation officer had calculated Cervantes’s exposure at four life

sentences without possibility of parole, plus an indeterminate term of 155 years to life

13

and a determinate term of 15 years, 4 months. The judge stayed many of the potential

sentences on various counts of conviction and enhancements in an effort to bring the

sentence within the limits established by Graham, supra, 560 U.S. at page 74 and

Caballero, supra, 55 Cal.4th at page 268. Through its comments at sentencing and the

many sentencing choices it made, we believe the court fairly clearly expressed its

intention to impose a life sentence with the highest minimum term it could, while

avoiding an unconstitutional sentence under Caballero.

II. DISCUSSION

A. Ineffective Assistance of Counsel

1. Cervantes’s Claims of Ineffective Assistance

Still represented by Obstler and his firm on appeal, Cervantes contends

Kirkpatrick was constitutionally ineffective in representing him, pointing out numerous

errors she allegedly made. He claims that, in combination, these errors amounted to a

complete failure to subject the prosecution’s case to meaningful adversarial testing.

First, Cervantes argues trial counsel failed to adequately investigate certain details

of the case, specifically she (1) failed to consult with Cervantes’s two prior attorneys,

who could have informed her of their investigation to date and their theories of the case;

(2) failed to follow up on crime scene photographs that showed a bicycle in the garage of

the home where the attacks occurred, which defense counsel claims would have bolstered

Cervantes’s story that he entered Gabriel’s house to retrieve his bike; (3) failed to follow

up on crime scene photos that showed there was no forcible entry into the house; (4)

failed to investigate the route Cervantes took in arriving at Gabriel’s house; (5) failed to

investigate other forensic evidence, specifically five cigarette butts, a beer can, and

footprints outside the home that showed they were made before Cervantes entered the

house, not afterwards, and therefore tended to undercut the prosecution’s theory of the

crime; (6) failed to subject the prosecution’s revenge theory to adversarial scrutiny or

investigation; (7) failed to investigate Gabriel’s “exculpatory” statement to police about

Cervantes’s revenge motive; (8) failed to investigate or present evidence about who

“Richard” was; and (9) abandoned an involuntary intoxication/drug psychosis defense

14

based on consumption of psilocybin mushrooms laced with MDMA, based largely on a

negative urine test conducted after Cervantes’s arrest, which Cervantes claims was

unreliable and an insubstantial basis for rejecting the mushroom defense.

Second, he claims Kirkpatrick failed to adequately communicate with or prepare

her expert witnesses, which caused her to “abandon[] important mental health and

cognitive brain evidence that was not only material but crucial to the jury’s

determination” of the specific intent required for many of the alleged offenses.

Specifically, she cut off communication with Dr. Roderick Pettis, an expert psychiatrist

hired by the defense, without even securing a copy of his detailed notes of an interview

he had conducted with Cervantes. She also failed to obtain a recommended brain scan

for Cervantes. And, when informed shortly before trial by a different defense expert, Dr.

John Podboy, a clinical and forensic psychologist, that Cervantes may have been in a

“dissociative state” at the time of the crimes, she decided not to call that expert as a

witness and went to trial without investigating further.

And third, Cervantes contends the defense that Kirkpatrick did put on through

Zehnder’s testimony was not a legally viable voluntary intoxication defense, primarily

because she did not choose an appropriate expert to explain why his state of intoxication

would have negated Cervantes’s formation of the specific intent required for the various

crimes.9 Zehnder testified primarily about the physical effects of alcohol intoxication,

such as slurred speech and impairment of motor skills. We agree with Cervantes that he

would have been better served if Kirkpatrick had called a neuroscience expert10 who

9

Cervantes also argues there were evidentiary errors at the hearing on his motion

for a new trial that compromised the outcome of the motion. In light of our disposition

on his claims of ineffective assistance of counsel, we do not separately address those

claims, but we do address some of the claims of evidentiary error in our discussion of

ineffective assistance of counsel.

10

When we use the term “neuroscience expert” we include any professional who

has specialized knowledge of brain function and impairment.

15

could have testified about the mental effects of alcohol intoxication and perhaps

psilocybin intoxication on a person with Cervantes’s apparent cognitive deficits.

Where, as here, the trial court has denied a motion for a new trial based on an

ineffective assistance claim, we apply the standard of review applicable to mixed

questions of law and fact, upholding the trial court’s factual findings to the extent they

are supported by substantial evidence, but reviewing de novo the ultimate question of

whether the facts demonstrate a violation of the right to effective counsel. (People v.

Taylor (1984) 162 Cal.App.3d 720, 724–725.)

We agree with Cervantes that trial counsel should have done more to investigate

leads developed by Pendergast and Barton. Kirkpatrick’s limited investigation of a

psilocybin mushroom defense fell below the standard of objectively reasonable

representation. Her minimal and ineffective communication with defense-retained

experts caused her to make an uninformed decision in rejecting a theory of cognitive

impairment as part of her voluntary intoxication defense. She chose as her only expert

witness someone who was not qualified to opine on the mental impacts of extreme

alcohol intoxication, much less on the combined effects of alcohol and psilocybin

mushrooms on someone with Cervantes’s brain impairment.

Because Kirkpatrick’s performance leaves us without confidence in the outcome

of the trial on the specific intent crimes other than burglary, we hold she was

constitutionally ineffective. We therefore reverse the eight counts specified in the

disposition. On the general intent sex offenses, assault with a deadly weapon, and

burglary charges, however, we are satisfied Cervantes got a fair trial with a reliable

result, and we affirm.

2. This Case is Governed by Strickland, not Cronic.

In arguing that Kirkpatrick failed to subject the prosecution’s case to “adversarial

testing,” Cervantes attempts to bring himself within the exception identified in United

States v. Cronic (1984) 466 U.S. 648, 659 (Cronic) that allows reversal for ineffective

assistance of counsel without a showing of prejudice where counsel “entirely fails to

subject the prosecution’s case to meaningful adversarial testing” so as to make “the

16

adversary process itself presumptively unreliable.” Cronic recognized this exception

would apply only to an extreme deviation from the norm. (Id. at pp. 656–657.) To come

within the exception, the United States Supreme Court has explained, counsel’s failure to

represent the defendant must be “complete.” (Bell v. Cone (2002) 535 U.S. 685, 697.)

The California Supreme Court has also deemed the Cronic exception to be extremely

limited: “Defendants have been relieved of the obligation to show prejudice only where

counsel was either totally absent or was prevented from assisting the defendant at a

critical stage.” (In re Visciotti (1996) 14 Cal.4th 325, 353; see also, Florida v. Nixon

(2004) 543 U.S. 175, 190 [“narrow exception”].)

Cervantes’s case does not come close to meeting the requirements for the Cronic

exception. Defense counsel filed motions in limine, gave an opening statement, made

objections, cross-examined witnesses, presented two witnesses on behalf of the

defendant, requested jury instructions, and made a closing argument to the jury. She

succeeded in convincing the jury to make not true findings on several of the one-strike

allegations, including infliction of great bodily injury, torture and aggravated mayhem.

Kirkpatrick made a motion for a new trial based on insufficiency of the evidence on the

specific intent crimes. She also filed a sentencing memorandum and assembled a large

packet of letters of support for Cervantes in anticipation of sentencing. Cronic’s

reversible per se standard does not apply. In a case subject to the Cronic exception, these

hallmarks of legal representation would be markedly absent or almost absent. (People v.

Williams (2013) 56 Cal.4th 630, 690–691.)

Instead, Strickland v. Washington (1984) 466 U.S. 668 (Strickland) provides the

guiding principles, and it requires a showing of both deficient professional performance

and prejudice. (Id. at p. 687.) On the performance prong, we “indulge a strong

presumption that counsel’s conduct falls within the wide range of reasonable professional

assistance” (id. at p. 689), and measure counsel’s performance against an “objective

standard of reasonableness” (id. at p. 688). To establish deficient performance under

Strickland, a defendant must show his or her attorney “made errors so serious that

counsel was not functioning as the ‘counsel’ guaranteed the defendant by the Sixth

17

Amendment.” (Id. at p. 687.) To show prejudice, he must show “counsel’s errors were

so serious as to deprive the defendant of a fair trial, a trial whose result is reliable.”

(Ibid.) “The defendant must show that there is a reasonable probability that, but for

counsel’s unprofessional errors, the result of the proceeding would have been different.

A reasonable probability is a probability sufficient to undermine confidence in the

outcome.” (Id. at p. 694.)

Although we usually require that claims of ineffective assistance of counsel be

raised in a petition for writ of habeas corpus (People v. Mendoza Tello (1997) 15 Cal.4th

264, 266–267), because the ineffective assistance of counsel claim was fully developed in

the trial court on a motion for a new trial, we shall entertain the issue in this appeal,

applying the Strickland test.

18

3. Cervantes’s Claims of Ineffective Assistance Do Not Affect the

Jury’s Verdicts on the General Intent Crimes: Forcible Rape,

Forcible Sodomy, Forcible Oral Copulation, and Assault with a

Deadly Weapon.

By far the strongest of Cervantes’s ineffective assistance claims is that trial

counsel failed to bring certain evidence before the jury that might have caused one or

more jurors to doubt that Cervantes had formed the specific intent required for some of

the offenses, either because of alcohol and possibly psilocybin mushroom ingestion or

because of cognitive impairment or brain damage, or especially because of those factors

in combination. He claims, in essence, that Kirkpatrick could have and should have

presented a stronger case of diminished actuality.11

Diminished actuality evidence is limited by statute. “Evidence of mental disease,

mental defect, or mental disorder is admissible solely on the issue of whether or not the

accused actually formed a required specific intent, premeditated, deliberated, or harbored

malice aforethought, when a specific intent crime is charged.” (§ 28, subd. (a).)

Likewise, “[e]vidence of voluntary intoxication is admissible solely on the issue of

whether or not the defendant actually formed a required specific intent, or, when charged

with murder, whether the defendant premeditated, deliberated, or harbored express

malice aforethought.” (§ 29.4, subd. (b).)

Thus, diminished actuality is available only to disprove or cast doubt on mens rea

in specific intent crimes. Cervantes makes reference to involuntary ingestion of MDMA

in an apparent attempt to extend his challenge to the validity of the general intent

11

In the trial court, counsel referred repeatedly to a “mental capacity defense” and

in this court continues to use the same language, questioning whether Cervantes had the

mental capacity to form the specific intent required for the various crimes with which he

was charged. The defense of diminished capacity has been abolished in California.

(§§ 25, subd. (a), 28, subd. (b).) (See fn. 23, post.) We therefore construe Cervantes’s

arguments as they might apply to a defense of what is now typically called “diminished

actuality.” (See People v. Elmore (2014) 59 Cal.4th 121, 139.)

19

crimes,12 but we reject that argument as a matter of law. Since he claims to have taken

psilocybin mushrooms voluntarily, the fact that they may have been, unknown to him,

tainted with MDMA would not render the MDMA intoxication “involuntary.” (People v.

Gallego (1990) 52 Cal.3d 115, 183–184; People v. Velez (1985) 175 Cal.App.3d 785,

795-797; CALJIC No. 4.23.)

Kirkpatrick’s alleged failings would not have led to a different verdict on the

general intent crimes. It is significant that four of the one-strike crimes (forcible rape,

forcible sodomy, and two counts of forcible oral copulation) were general intent crimes.

(People v. DePriest (2007) 42 Cal.4th 1, 48 [forcible rape]; People v. Warner (2006) 39

Cal.4th 548, 557 [rape, sodomy, oral copulation]; People v. Hughes (2002) 27 Cal.4th

287, 341 [sodomy]; People v. Fox (2001) 93 Cal.App.4th 394, 398 [oral copulation with

a child under age 14].) Presenting more mental state evidence would not have affected

those verdicts. We therefore shall affirm the convictions on counts 11 through 14.

In addition, the two counts of assault with a deadly weapon (counts 8 and 9)

required only general intent. (People v. Colantuono (1994) 7 Cal.4th 206, 214.) Those

convictions shall also be affirmed.

4. Counsel’s Alleged Failures Were Not Prejudicial on the Burglary

Conviction or the Burglary One-Strike Findings.

Some of Cervantes’s claims of error by counsel are relevant solely to the burglary

charge. Indeed, the burglary charge was especially significant because the 25-to-life

terms were based in part on the fact that the sex crimes occurred during a burglary.

(§ 667.61, subds. (a), (d)(4), (e)(2).) But we find none of the purported errors of counsel

had any impact on the burglary conviction or the true findings on the burglary one-strike

allegations.

a. Photograph of Cervantes’s Bicycle in Gabriel’s Garage

Cervantes filed a declaration under oath attached to the new trial motion in which

he said that, after returning home from the party, he meandered about for a while and

12

Cervantes does not claim he was unconscious. (§ 26.)

20

ended up at Gabriel’s house. Once there, he remembered he had left his bicycle at

Gabriel’s house sometime earlier, so he decided to retrieve it. Relying on the declaration,

Cervantes claims his trial attorney was ineffective for failing to recognize the significance

of a photograph in evidence of Gabriel’s family’s garage, which shows the presence of

Cervantes’s bicycle. Cervantes claims this point could have been made without his

testimony, as family members could have testified it was his bike in the photograph. But

his intent to retrieve the bike could only have come to the jury through his own

testimony.

It is next to meaningless that Cervantes’s bike was, in fact, at Gabriel’s house. It

does not, as Cervantes suggests, negate revenge as a motive for the crimes. Even if the

bike photograph could explain his intent in entering the garage, it would not explain his

intent in entering the home or the master bedroom. (See People v. Thomas (1991) 235

Cal.App.3d 899, 903–906 [entry from garage into locked kitchen was burglary].) The

bike story was ultimately of only the slightest relevance and could not have been fleshed

out without putting Cervantes on the witness stand, which neither Kirkpatrick nor

Cervantes favored.

b. Photograph Showing There Was No Break-in

Similarly, we attach no significance to the fact that one photo of damage to the

frame of the door leading into the garage, examined closely, showed the damage had

been sustained at an earlier time. Cervantes claims this calls into question whether he

actually had to forcibly enter the family’s home or whether the door was unlocked. The

interior door from the garage to the house was not damaged and apparently had been

unlocked.

Cervantes overestimates the importance of the photograph. It does not matter

whether he broke into the house or walked in through an unlocked door. “[I]n California

no breaking or forcible entry is required in proof of the commission of a burglary.”

(People v. Beamon (1968) 268 Cal.App.2d 61, 65; accord, People v. Davis (1998) 18

Cal.4th 712, 720–721.) A burglary conviction requires that the defendant entered a

building or room with the intent to commit a theft or a felony. (People v. Wallace (2008)

21

44 Cal.4th 1032, 1077; CALCRIM No. 1700.) Unauthorized entry through an unlocked

door constitutes burglary. (People v. McCormack (1991) 234 Cal.App.3d 253, 254-257.)

Moreover, the illegal intent may be formed after entry into a building and before entry

into a room where the illegal act is committed. (Id. at pp. 255–257; see People v. Sparks

(2002) 28 Cal.4th 71, 88 [entry by houseguest into unlocked bedroom of victim with

intent to commit rape].) The lack of forcible entry was irrelevant.

c. Footprints, Beer Can and Cigarette Butts Outside the Home

Among the evidence found at the crime scene were a beer can and five cigarette

butts outside the home near some footprints in the ashes that had fallen from a barbecue

onto a pathway to the door leading into the garage. The prosecution used this evidence to

suggest that Cervantes, after entering the house, left again to drink a beer and smoke a

cigarette outside. A police officer testified the shoe prints in the ash matched Cervantes’s

shoes. The prosecutor suggested this was evidence of premeditation and deliberation

because it gave Cervantes time to reflect on the crime he was about to commit.

Cervantes pointed out in his motion for a new trial that footprints in the ash led toward

the door to the garage, not away from it, thus suggesting he smoked the cigarette and

drank the beer before entering the house, not afterwards. On appeal he argues the

footprints could have been left by someone other than Cervantes.

Even if trial counsel had made a point of trying to prove the officer was wrong,

that would not cast doubt on who committed the crimes. Cervantes obviously committed

the charged acts and never tried to point the finger at anyone else. Distracting the jury

with a largely irrelevant argument about when the footprints were made or by whom

would not have changed the outcome of the trial. There was far more evidence of

premeditation and deliberation than footprints in the ashes.

Whether he smoked and drank a beer before entering the home or afterwards, the

evidence still shows he spent appreciable time outside the home, which would have given

him an opportunity to contemplate his crimes. In fact, if he took this time before entering

the home, it strengthens the prosecution’s evidence of burglary because it tends to

increase the likelihood he entered with felonious intent. We fail to see how stressing to

22

the jury the direction in which the footprints were pointing, or questioning whether they

were made by a third party, could have significantly improved Cervantes’s defense, and it

might well have had the opposite effect.

d. Failure to Investigate Cervantes’s Route to Gabriel’s House

We likewise find it difficult to imagine the jurors’ verdicts would have been

different if they had known which route Cervantes followed when he walked to Gabriel’s

house. Cervantes argues the circuitous route he followed was more consistent with a

dazed “walkabout” than a deliberate trip to a known destination intent upon committing a

crime. He seems to suggest this would have helped to blunt the evidence that he had a

knife with him, suggesting it diminishes the inference he brought the knife along for the

purpose of stabbing someone. Evidence of his exact route would have been weak

evidence, at best, in his defense. And it could only have been presented through

Cervantes himself, who did not want to take the stand. Putting him on the stand to prove

this non-issue would have opened him up to cross-examination that could have had a

severe negative impact on his defense. The risk was great and the potential reward

trivial. This omission did not meet the showing required under the first prong of

Strickland.

e. Cervantes’s Own Declaration Shows He Entered the House to

Steal Alcohol.

Although Cervantes seems to assume his potential alcohol/mushroom intoxication

and cognitive impairment/brain damage defenses would apply equally to all the specific

intent crimes, we find an exception for the burglary conviction. His own declaration

explicitly describes how he entered the house, including the detail that the door stuck a

little bit as he opened it, though it was unlocked. The declaration also explains why he

entered: he entered the garage to retrieve his bicycle, and he entered the house to drink

some of Gabriel’s mother’s alcohol that he expected to find in the freezer. This shows

not only that he was clear-headed enough at the time of the burglary to form a specific

intent, but also that the specific intent he formed was larcenous. Even entering to steal

alcohol would amount to a burglary. (§ 459 [burglary is committed by entry “with intent

23

to commit grand or petit larceny or any felony”]; People v. Sanghera (2006) 139

Cal.App.4th 1567, 1572 [intent to commit petty theft sufficient]; People v. Nguyen (1995)

40 Cal.App.4th 28, 30–31 [petty theft by false pretenses].) Cervantes’s declaration

amounts to a confession of burglary, and we see no reason to reverse the conviction on

that count or on the burglary one-strike findings.

This conclusion does not pre-ordain his guilt of the later specific intent crimes,

however. The burglary was the first crime chronologically and was evidently committed

when Cervantes still had his wits about him sufficiently to form a specific intent.

Cervantes insists that he remembers virtually nothing after entering the house, stumbling,

and falling in the kitchen. He evidently stayed at the house quite some time before

attacking the children, time enough to consume three more beers. Thus, it is possible he

entered a dissociative state sometime after entering the house but before attacking the

children. This would not excuse him from the general intent crimes, or the burglary, or

the burglary one-strike findings. He still committed the sex offenses “during the

commission of a burglary” within the meaning of section 667.61, subdivision (e)(2).

(People v. Alvarado (2001) 87 Cal.App.4th 178, 191 [“during the commission of a

burglary” includes the entire duration of the burglary, from “initial entry with the

requisite intent [until] the burglar’s escape to a place of temporary safety”]; People v.

Palmore (2000) 79 Cal.App.4th 1290, 1295–1296 [same].) Therefore, based on

Cervantes’s own declaration, we affirm the burglary conviction and the burglary one-

strike findings accompanying the general intent crimes.

5. Failure to Investigate and Present Evidence Regarding the Calkins

Case and the Tainted Mushroom Defense Was a Legitimate Tactical

Decision.

Cervantes claims Kirkpatrick was ineffective because she failed to investigate and

present evidence regarding the Calkins case. He claims this line of investigation was

important to dispel the notion that the attack was committed as “revenge” for Gabriel’s

supposedly “snitching” on “Richard” because the revenge theory gave the prosecution a

basis for arguing that Cervantes harbored a felonious intent when he entered the house

24

(§ 459), supported the torture allegations (§ 206),13 and bolstered the People’s case for

premeditation and deliberation on the attempted murder charges (§§ 187, 664). He

claims Kirkpatrick performed below professional norms by failing to investigate his

claim that he ingested psilocybin mushrooms—specifically mushrooms tainted with

MDMA—shortly before committing the crimes. He draws a connection between his case

and Calkins, claiming that both he and Calkins got their mushrooms from Gabriel, and

the mushrooms were tainted with MDMA.14

Cervantes further claims that Kirkpatrick missed an opportunity to rebut the

prosecution’s revenge theory by failing to investigate and present to the jury evidence of

a litany of “facts” supported only by Coffer’s report: “(1) The ‘Richard’ referred to by

A.P. was Richard Calkins, a young man who was facing trial on a double homicide that

involved behavior as violent and bizarre as that of [Cervantes]; [¶] (2) Calkins was

defending on the theory that he consumed hallucinogenic mushrooms laced with MDMA

after purchasing the drugs from [Gabriel]; [¶] (3) [Gabriel] had been arrested in

connection with the Calkins case after the police found hallucinogenic mushrooms in his

house that were laced with MDMA; [¶] (4) Calkins and [Cervantes] had never met and

[Cervantes] had no idea who Calkins was or what his relationship was to [Gabriel]; [¶]

(5) [Cervantes] purchased hallucinogenic mushrooms from [Gabriel] around the same

time that Calkins purchased the drugs; [¶] (6) [Cervantes] told his lawyers that just before

leaving the house on his walk that night he ingested the mushrooms that he had

purchased from [Gabriel]; [¶] (7) [Gabriel] admitted and had written incriminating notes

13

Torture requires proof the defendant inflicted great bodily injury on someone,

and when inflicting the injury, he “inten[ded] to cause cruel or extreme pain and suffering

for the purpose of revenge, extortion, persuasion, or for any sadistic purpose.” (§ 206.)

Since Cervantes stabbed A.P. 42 times and I.A. 13 times, the jury reasonably could have

found he did so for a sadistic purpose or for revenge.

14

Barton also discussed with Calkins’s attorney sharing an expert: Dr. Stalcup.

Barton was proceeding on the theory that both Calkins and Cervantes got their

mushrooms from Gabriel. Her notes produced in discovery contained no mention of this

fact, but she testified they were not her complete notes.

25

about his dealing of mushrooms; [¶] (8) Photographs taken at the crime scene showed a

newspaper article about the Calkins crime next to the bed where the victims were

sleeping when [Cervantes] committed the acts;[15] [¶] (9) Prior to the night of the crime

the victims’ family had grown concerned about a report that men had been around their

house looking for [Gabriel]; and [¶] (10) Calkins was being represented by the Public

Defender and that representation provided the most reasonable basis for the Public

Defender’s decision to declare a conflict in [Cervantes’s] case.”

Drawing together the many strands of this theory of ineffective assistance of

counsel, the nub of it appears to be that Kirkpatrick should have investigated the Calkins

and tainted mushroom connections further to support a diminished actuality defense

based on voluntary intoxication (alcohol, marijuana and mushrooms), and perhaps

involuntary intoxication (MDMA), while simultaneously dispelling any inference he

attacked Gabriel’s siblings out of revenge against Gabriel on Calkins’s behalf. Although

the theory has some surface appeal and internal coherence, when considered as a whole,

its individual parts are not supported by admissible evidence. Besides, Kirkpatrick had

adopted a directly contrary strategy. When the prosecutor attempted to elicit information

from a police witness about who “Richard” was, Kirkpatrick successfully moved to strike

his comment that Richard had some connection to “an earlier Vacaville case.” She

evidently believed introducing the Calkins case into Cervantes’s trial would merely help

to flesh out the prosecution’s revenge theory.16

15

Cervantes cites this newspaper article as something the defense attorney should

have used to draw a connection with the Calkins case. We reject this claim along with

rejecting the entire Calkins connection as something Kirkpatrick was obligated to present

at trial as part of a competent defense.

16

Several hours after Cervantes’s arrest the police interviewed Gabriel at juvenile

hall and he told them various things that Cervantes now claims were “exculpatory.”

Cervantes claims Kirkpatrick’s response to this interview was ineffective assistance of

counsel. But Gabriel’s statements amounted to nothing more than his incredulity that

Cervantes would or could commit such crimes because he was normally nonviolent, his

inability to identify any motive for Cervantes’s attack on his siblings, and his suggestion

26

Kirkpatrick explained her reasons for not developing a Calkins defense: “[T]he

theory that was presented at trial was just [A.P.’s] statement, and that was it. So whoever

this Richard was not presented at trial or the relationship of whoever Richard was.”

Kirkpatrick did not investigate who Richard was because she “was not going to give them

a motive, no.” While we question how a failure to investigate can be justified by the

evidence actually presented at trial, we fail to see any prejudice flowing from

Kirkpatrick’s failure to investigate. We ultimately agree with her that there was no need

to rebut the revenge theory because the prosecution “didn’t present anything at trial, who

Richard was or Richard Calkins’s case.”

In cross-examining A.P., Kirkpatrick succeeded in getting her to admit she

“possibly” heard something about “Richard” on the news or read something on the

Internet. Exposing the jury to that much connection to the Calkins case, but no more,

tended to call into question A.P.’s testimony about Cervantes’s purported “revenge”

statement. We are convinced Kirkpatrick had a strategically logical approach to dealing

with the revenge theory.

that maybe Cervantes had been on “drugs.” His statements would have been

inadmissible lay opinion and speculation. (Evid. Code, §§ 702, 800.)

Kirkpatrick testified she did not interview Gabriel, but she did watch the video-

taped police interview of him. Kirkpatrick did not attempt to interview Gabriel because

he was the older brother of the two victims, who would have been unlikely to cooperate.

It also appears Gabriel was represented by counsel at the time, which was another reason

not to interview him.

Cervantes suggests Kirkpatrick should have called Gabriel as a witness, which

could have been disastrous. If she had, the prosecutor could have used other distinctly

unfavorable parts of the interview. (Evid. Code, § 356.) Gabriel told the police he

himself had never used mushrooms, did not know whether Cervantes used mushrooms,

and denied selling mushrooms to Cervantes. In addition to his other negative statements

about Cervantes (see fn. 2, ante), Gabriel told the police Cervantes “lies,” was “not

honest,” “not a trustworthy person,” and a “weirdo” who was “mean to his mom,” “steals

from friends” and was “always jacking off.” We cannot see the value of calling Gabriel

as a witness.

27

In closing argument, Kirkpatrick used the prosecution’s lack of evidence

connecting Cervantes to “Richard” to try to convince the jury to dismiss the prosecutor’s

revenge theory: “The District Attorney has said over and over again, ‘We know what his

intent was,’ right? ‘We know what his intent was because he told the victim, “This is for

your brother snitching. This is for your brother snitching.” ’ And she mentioned a name

by the name of Richard. [¶] Okay. Where is—where is the evidence of that? Where is—

where is the evidence of the snitching? Who is this Richard guy? Who is he in

relationship to Alex? It’s the rambling of a kid who’s really drunk. Where is that

evidence? They want you to rely solely on this statement, but it means nothing. Unless

you have something to show what it means, it means nothing. They’re just hoping you

held onto it and go, That’s it. That’s it. It’s revenge. You have no evidence of that.

Revenge for what?” We are not inclined to second-guess this mode of handling A.P.’s

testimony about Richard.

Whether to pursue the complicated, speculative Calkins/tainted mushroom defense

was a tactical decision lying within the realm of professional judgment. Kirkpatrick’s

decision not to pursue the defense was reasonable. The most significant evidence

underlying the Calkins/tainted mushroom defense was absent or conflicting. Gabriel

denied selling him the mushrooms, and Cervantes himself had given conflicting accounts

of where he got them. There was also a negative drug test for both psilocin (the active

ingredient in psilocybin mushrooms) and MDMA.17

That counsel on appeal has developed a new, alternative theory of defense does

not mean the course chosen by Kirkpatrick was professionally incompetent. The

Calkins/tainted mushrooms defense could not have been established without Cervantes’s

testimony. But based on his inconsistent prior statements about the source of the

mushrooms, Cervantes would have been vulnerable on cross-examination. Especially

given her client’s cognitive challenges, Kirkpatrick was justifiably worried about his

17

According to Dr. Stalcup, both MDMA and psilocin leave the body quickly and

are difficult to detect in available tests.

28

ability to withstand cross-examination. We also take judicial notice of the fact that

Calkins in his own trial was not allowed to introduce evidence of a connection to

Cervantes’s case. (Evid. Code, §§ 452, subd. (d), 459.) We have no faith that Judge

Kinnicutt would have been any more receptive to a “Calkins” defense, so prejudice from

deciding not to pursue such a defense would be difficult to establish.

“Judicial scrutiny of counsel’s performance must be highly deferential. It is all too

tempting for a defendant to second-guess counsel’s assistance after conviction or adverse

sentence, and it is all too easy for a court, examining counsel’s defense after it has proved

unsuccessful, to conclude that a particular act or omission of counsel was unreasonable.”

(Strickland, supra, 466 U.S. at p. 689.) A “court must indulge a strong presumption that

counsel’s conduct falls within the wide range of reasonable professional assistance.”

(Ibid.) “There are countless ways to provide effective assistance in any given case. Even

the best criminal defense attorneys would not defend a particular client in the same way.”

(Ibid.; see also Harrington v. Richter (2011) 562 U.S. 86, 106 (Richter) [Ninth Circuit’s

grant of habeas relief overturned by Supreme Court where California court had held a

defense attorney was not ineffective despite failure to consult forensic blood experts or

introduce expert evidence].)

Finally, presenting evidence of the Calkins connection might have hurt

Cervantes’s case by confirming the revenge motive. Gabriel had confessed to the police

that he sold mushrooms to Calkins, which the jury could have inferred was the

“snitching” that A.P. testified about. The jurors may have concluded that Cervantes

believed Gabriel had “snitched” on Calkins by admitting to the police that he sold

Calkins the mushrooms, and Cervantes may have been afraid Gabriel might also “snitch”

on him for his drug activity. Thus, developing a connection between the two cases may

have backfired by clarifying and strengthening the prosecution’s revenge theory. Such

potential downsides to following a given course are properly taken into account in

addressing a claim of ineffective assistance of counsel. (See Richter, supra, 562 U.S. at

pp. 107–110.) We cannot say defense counsel was ineffective for failing to follow the

path that Coffer would have taken.

29

6. Abandonment of the Mushroom Defense Without Further

Investigation Was Objectively Unreasonable, But Was Not By Itself

Prejudicial.

The mushroom defense standing alone is on different footing. The two Conflict

Defenders who represented Cervantes before Kirkpatrick took over the case had

contemplated raising a defense based on ingestion of psilocybin mushrooms. Pendergast

had a sample of Cervantes’s blood and urine sent to a laboratory for testing. The blood

had been drawn and urine collected some seven or eight hours after Cervantes claims he

ingested the mushrooms. Pendergast was awaiting the results when he left the office.

The lab results came in while Barton was defending Cervantes (in June 2011), and the

urine was negative for psilocin (the blood apparently having not been tested). According

to Coffer, Kirkpatrick had rejected the mushroom defense by late February 2012, just two

weeks after being assigned to the case. 18

A significant factor in her rejection of the mushroom theory appears to have been

the negative psilocin test. Based on Coffer’s Internet research and contacts with the lab,

he learned it is difficult to test for psilocin because it leaves the body rapidly after

ingestion, a fact confirmed by Dr. Stalcup. Such tests almost invariably show up

negative. Kirkpatrick never communicated with the laboratory, except to ask for a copy

of the lab report, and apparently never learned of the shortcomings of the test. We agree

with Cervantes that Kirkpatrick should not have ruled out a mushroom defense based

solely on the results of an unreliable urine test at a time when she had no other clear

alternative defense.

Kirkpatrick considered the evidence supporting a psilocybin mushroom defense to

be conflicting, but this seems to be largely because Cervantes did not tell her in their

initial interview that he had taken mushrooms. She acknowledged he later told her about

it, but Kirkpatrick seemed to doubt his credibility. Kirkpatrick testified Cervantes’s

18

Although we see a number of serious deficiencies in Kirkpatrick’s investigation,

as explained below, we reject the suggestion that her failure to communicate with former

counsel was in and of itself ineffective assistance of counsel.

30

statements “were inconsistent and all over the place.” We disagree with that

characterization. Where he got the mushrooms, it is true, was a subject of some

conflicting information from Cervantes, but the fact that he took them was consistently

reported.

We are especially concerned that Kirkpatrick failed to secure from Dr. Pettis, his

notes of an interview conducted on November 14, 2011 (before Kirkpatrick came onto

the case), in which Cervantes’s reported to Dr. Pettis that he had consumed mushrooms

on the night of the crimes and got them from Gabriel. These detailed notes, some 18

single-spaced, typed pages in length, prepared by Dr. Pettis while working for the

defense, surely should have been acquired by Kirkpatrick,19 yet Kirkpatrick never

received a copy because she did not ask for them. Dr. Pettis never told her he had

interview notes, but a reasonably competent defense attorney would have asked. To this

extent we find her performance fails the first prong of Strickland. Dr. Pettis’s detailed

notes of his interview with Cervantes would have given her greater insight into the case

and could have caused her to rethink the mushroom defense or even the Calkins

connection.

Kirkpatrick claimed she had “absolutely no evidence” that Cervantes had taken

mushrooms that night. That may be true in the sense that she had no admissible evidence

to produce at trial, but for purposes of investigation, her client’s repeated statements were

19

When Obstler attempted to move the notes into evidence, the court sustained a

relevance objection because Kirkpatrick had not been provided with the notes. The

relevance objection should have been overruled. The notes were admissible as evidence

of what Kirkpatrick could have discovered from a defense-retained expert if she had

asked. As such, the notes are relevant, not for the truth of what they say, but for the

added information Kirkpatrick would have gleaned had she exercised reasonable

professional diligence. We are aware we must “eliminate the distorting effects of

hindsight” in assessing Kirkpatrick’s performance. (Strickland, supra, 466 U.S. at

p. 689, accord, Richter, supra, 562 U.S. at pp. 106–107.) Still, an attorney cannot

immunize herself from a claim of ineffective assistance of counsel simply by avoiding

the acquisition of readily available information. Dr. Pettis’s notes are included in the

record before us, and we have reviewed them on appeal.

31

evidence enough. Cervantes signed a post-trial declaration under oath stating he ate a

piece of a psilocybin mushroom just before going to Gabriel’s house on the night of his

crimes and he got the mushroom from Gabriel at some earlier point in time.20 He also

told Kirkpatrick some version of a mushroom ingestion story before trial. In addition, his

bizarre conduct during the crime itself should have raised some concerns about drug

intoxication; the experts have since confirmed that his behavior was consistent with

ingestion of psilocybin mushrooms in combination with other factors. Following up on

her client’s statements, by consulting an expert or otherwise, was necessary for her to

learn: (1) the significance (or insignificance) of the negative psilocin test; and (2) the

effect psilocybin mushrooms might have had on Cervantes’s ability to form specific

intent, if in fact he had consumed them. Some form of further investigation was required

of competent counsel before rejecting the mushroom defense.

But Kirkpatrick had a second problem with the defense: proving Cervantes ate the

mushrooms. Since he took them at home, not at the party, there were no witnesses.

Barton, too, recognized that unless she could get a positive psilocin test from Cervantes’s

blood sample, Cervantes would have to testify. Kirkpatrick rejected the prospect of

putting Cervantes on the stand because (1) he did not want to testify, (2) she was

concerned about how he would hold up under cross-examination, and (3) she was

concerned he might perjure himself. These were all legitimate reasons for rejecting use

of the mushroom defense at trial after conducting an appropriate investigation and

consulting with her client. But they do not excuse her failure to investigate. An

appropriate investigation is prerequisite to providing competent counseling to the client

about his decision to testify or stand mute.21

20

The judge clearly placed little or no credence in that declaration unless

Cervantes was willing to testify at the hearing.

21

Coffer suggested trial counsel was ineffective for failing to call her own client

as a witness, going so far as to testify that not putting Cervantes on the stand was

“absolutely death for the defense.” Though Coffer criticizes Kirkpatrick, it is both ironic

and telling that Obstler himself did not call Cervantes as a witness at the hearing on the

32

We conclude that Kirkpatrick failed to conduct a reasonable investigation into the

mushroom defense, but we would not find that failure prejudicial standing on its own.

Because the mushroom defense could only be established by putting Cervantes on the

witness stand, and because there were good reasons for not doing so, the failure to

present a mushroom defense was not ineffective assistance of counsel. We doubt

Kirkpatrick’s advice to Cervantes about testifying would have changed even if she had

investigated further. Thus, the failure to investigate was not individually prejudicial.

When combined with the failure to adequately investigate other aspects of Cervantes’s

mental state at the time of the crimes, however, this failure contributes to our unease with

the verdicts on the specific intent crimes.

7. Failure to Adequately Investigate Potential Expert Testimony on

Mental State Was Prejudicial on the Specific Intent Crimes:

Attempted Murder, Torture, Aggravated Mayhem, and Two of the

Sex Offenses.

a. The Importance of Pretrial Investigation

Cervantes contends Kirkpatrick failed to communicate effectively with her

experts, which led her to sacrifice a viable diminished actuality defense to the specific

intent crimes based on his long-term cognitive difficulties and brain damage. This claim,

together with the mushroom defense or even with the alcohol intoxication defense, could

have helped Cervantes defend against the specific intent crimes. Cervantes contends, if

Kirkpatrick had been more receptive to and communicative with her hired experts, she

would have learned that Cervantes’s cognitive impairment might be linked to organic

brain damage, which would have led her to order a brain scan. Had she consulted with

Dr. Shore or a similar mental health expert, she would have learned Cervantes’s brain

was “diffusely damaged,” with impairment specific to his prefrontal cortex.

new trial motion, even though the judge told him repeatedly that only by putting

Cervantes on the stand could he prove mushroom ingestion. But, of course, the decision

whether to testify is ultimately the client’s. (People v. Robles (1970) 2 Cal.3d 205, 215;

People v. Gray (1998) 66 Cal.App.4th 973, 987–988.)

33

We cannot ignore or condone these shortcomings in Kirkpatrick’s investigation. It

appears Kirkpatrick repeatedly misheard or misinterpreted advice given by the experts,

often enough that we cannot help but conclude she did not serve her client competently.

It appears she was so concerned about the negative evidence in Cervantes’s IEP records

that she ruled out presenting a mental health expert without due consideration. Had she

been more receptive to mental health testimony—receptive enough to at least listen to her

own hired experts carefully, get copies of their notes, and follow up on their leads—it

might have made a difference in how she presented the alcohol intoxication defense.

That, in turn, is reasonably likely to have made a difference in the jury’s verdicts in each

of the specific intent crimes, except burglary.

The Supreme Court in Strickland emphasized the importance of pretrial

investigation as a prerequisite to any claim of tactical rejection of a potential defense.

“[C]ounsel has a duty to make reasonable investigations or to make a reasonable decision

that makes particular investigations unnecessary.” (Strickland, supra, 466 U.S. at p. 691;

accord, Richter, supra, 562 U.S. at p. 106.) “[S]trategic choices made after less than

complete investigation are reasonable precisely to the extent that reasonable professional

judgments support the limitations on investigation.” (Strickland, at pp. 690–691.)

“Criminal trial counsel have no blanket obligation to investigate possible ‘mental’

defenses, even in a capital case.” (People v. Gonzalez (1990) 51 Cal.3d 1179, 1244.) But

if an attorney knows or has reason to know of the defendant’s mental problems, the rule

is otherwise. (Ibid.) Where evidence exists tending to show a mental state defense, and

counsel fails to follow up on it, a finding of ineffective assistance of counsel has often

resulted.22 The question becomes whether there were enough red flags to impose upon

22

An attorney’s failure to investigate potentially helpful leads has frequently been

deemed deficient performance, especially failure to investigate defenses based on mental

illness, ingestion of drugs, or other factors going to state of mind. (See Liao v. Junious

(9th Cir. 2016) 817 F.3d 678, 682–683 [failure to secure evidence supporting sleep-

walking defense]; In re Cordero (1988) 46 Cal.3d 161, 181-182 [failure to investigate

PCP-laced cigarette found at crime scene]; In re Scott (2003) 29 Cal.4th 783, 825–826

34

trial counsel an obligation to investigate further. In this case, we think there were. (See

Wiggins v. Smith (2003) 539 U.S. 510, 527–528; Rompilla v. Beard (2005) 545 U.S. 374,

391, fn. 8 [counsel must not ignore “ ‘red flags’ ” relating to mitigation]; Tharpe v.

Warden (11th Cir. 2016) 834 F.3d 1323, 1335 [citing cases where habeas petition had

been successful because counsel ignored “red flags” relating to the petitioner’s

“upbringing or mental-health status”]; Pruitt v. Neal (7th Cir. 2015) 788 F.3d 248, 271–

274 [failure to investigate and present evidence of mental illness].)

b. Factual Background

Kirkpatrick reviewed Cervantes’s medical records and IEPs, but considered them

“very harmful” to Cervantes’s defense. The records showed Cervantes had consistent,

“horrible behavioral problems” even before the age of 12. Kirkpatrick elected to forgo a

mental state defense in part so she would not be required to turn over the IEP and mental

health records to the prosecution. Pendergast, too, acknowledged the IEP reports were

“obviously harmful,” as were “a lot of things in this case,” and that was one thing he

would have to “balance” in deciding how to defend Cervantes.

Kirkpatrick also consulted with three experts about a possible mental state

defense: Drs. Pettis and Podboy and Dr. Dale Watson. The defense use of mental health

experts developed as follows: Pendergast did not hire a mental health expert in the three

or four months he represented Cervantes. In approximately July 2011, Barton retained

Dr. Pettis to evaluate the case. In seeking authorization for funds to retain Dr. Pettis,

Barton described Cervantes as having “issues of substance abuse, organic brain damage

[whether defense counsel’s “ ‘minimal’ ” investigation of a mental defense was

professionally deficient was a “close question,” but alleged error was non-prejudicial]; In

re Sixto (1989) 48 Cal.3d 1247, 1258-1262 [failure to obtain blood alcohol test and to

investigate possible PCP ingestion]; People v. Ledesma (1987) 43 Cal.3d 171, 222–224

[failure to adequately investigate and research diminished capacity defense]; People v.

Frierson (1979) 25 Cal.3d 142, 162-164 [same]; cf. Wade v. Calderon (9th Cir.1994) 29

F.3d 1312, 1318–1319 [defense counsel not ineffective for declining to use PCP defense

where “there was no evidence that [the defendant] ingested PCP on the day of the

offense”), cert. denied, 513 U.S. 1120 (1995).)

35

& budding schizophrenia” and described Dr. Pettis as being “phenomenal on the witness

stand.” Kirkpatrick got in touch with Dr. Pettis early on in her representation of

Cervantes, but Coffer reported that she never contacted Pettis again after March 15, 2012.

Dr. Pettis recommended that she retain a neuropsychologist to assist in the case.

She did so by consulting Dr. Watson. Dr. Watson performed an IQ test on Cervantes,

who scored 80. Kirkpatrick was surprised by this result, which ruled out any claim of

mental retardation. According to Kirkpatrick’s account, both Dr. Pettis and Dr. Watson

were discouraging about the prospects of presenting a defense based on Cervantes’s

mental condition.

By August 2012, Kirkpatrick believed she had narrowed down the available

defenses to “alcohol induced blackout” but was concerned that she would need her

“client to testify that [h]e [d]idn’t [r]emember anything.” A few days later she contacted

two different experts who could possibly support an alcohol intoxication defense and

selected Dr. Podboy, a clinical and forensic psychologist, as the preferable expert. She

was specifically interested in learning from Dr. Podboy whether Cervantes had

experienced alcohol-induced blackouts. By mid-August 2012, she was also in touch with

Zehnder, the forensic toxicologist who testified at trial, to estimate Cervantes’s blood

alcohol level at the time of the crimes.

Kirkpatrick had originally planned to use Dr. Podboy to testify about the effects of

alcohol on Cervantes’s mental state but decided against it for reasons we shall discuss

below. Even on the first day of trial, Kirkpatrick represented to the court that she would

call Dr. Podboy as an expert witness. Instead Kirkpatrick decided to use Zehnder to

opine about Cervantes’s likely blood alcohol level at the time of the crimes, and to

explain to the jury the likely effects on people at various and increasing blood alcohol

levels.

c. Weaknesses in the Voluntary Alcohol Intoxication Defense

Actually Presented Could Have Been Resolved by Calling a

Neuroscience Expert

36

Cervantes claims the defense pursued by Kirkpatrick was not a viable one. He

points out especially that defense counsel did not call an expert witness to testify about

the effects of alcohol intoxication at Cervantes’s blood alcohol level. She did call

Zehnder, but Cervantes now claims that a different expert witness, one capable of

expressing an opinion on the effect of alcohol on a brain impaired as Cervantes’s is

allegedly impaired, and armed with more evidence of Cervantes’s history of cognitive

deficits, could have provided more helpful testimony.

Cervantes suggests that, had Kirkpatrick used a neuroscience expert instead of a

toxicologist, the alcohol intoxication defense would have been much more likely to

succeed. A neuroscience expert could have testified that Cervantes’s alcohol

consumption, coupled with Dr. Amen’s SPECT scan results and Cervantes’s mental

health history, caused him to lapse into a “dissociative state” or “organic psychosis” in

which he attacked the children. Indeed, Dr. Stalcup was evidently prepared to testify to

that effect. He opined at the new trial hearing that even without psilocybin mushrooms,

the alcohol and marijuana consumed by Cervantes alone, combined with the deficits

revealed in the SPECT scan, would have led to organic psychosis. Such testimony would

have been extremely helpful to the defense at trial.

Zehnder’s testimony focused mostly on physical effects of alcohol intoxication,

such as slurred speech, delayed reaction times, and loss of balance. Dr. Stalcup’s

testimony at the new trial motion, on the other hand, dealt specifically with what it means

to be in a dissociative state, how one’s brain functions in a dissociative state, and how

one perceives reality in such a state. Such information would have been more useful to

the jury in deciding a specific intent issue than Zehnder’s testimony about impaired

motor skills.

We are fully aware that testimony such as that offered by Drs. Shore and Stalcup

at the new trial motion might be somewhat more constrained in a jury trial.23 But we feel

23

Due to the elimination of the diminished capacity defense in California (§ 25)

(see fn. 11, ante), experts may not testify about a defendant’s “capacity” to form specific

37

certain that a similar expert would have been allowed to testify to his opinion that

Cervantes was in a dissociative state when he committed the crimes and to give a general

description of a dissociative state or organic psychosis in a manner that would have aided

the jury and could have been persuasive. (See People v. Herrera (2016) 247 Cal.App.4th

467, 474–480.) We are also aware Kirkpatrick was up against a trial date when Dr.

Podboy gave her the information about Cervantes possibly being in a “dissociative state”

at the time of the crimes, but with the extreme penal consequences Cervantes was facing

and the turnover in attorneys at the Conflict Defender’s Office, a request for a

continuance might well have been favorably received. Kirkpatrick did not ask for a

continuance and went forward with the trial using Zehnder as her only expert witness.

We agree with Cervantes that Kirkpatrick’s selection of experts was not optimal and we

find her reasons for deciding not to present a mental health expert, in light of the whole

record, do not reflect the level of representation reasonably expected of a competent

defense attorney in like circumstances.

mental states. (§ 28; People v. Elmore, supra, 59 Cal.4th at p. 139.) A mental health

expert, or an expert on intoxication or addiction, also may not give testimony that the

defendant did or did not, in fact, form the mental state required for conviction, for making

that determination is the jury’s role. (§ 29; see People v. DeHoyos (2013) 57 Cal.4th 79,

121 [assuming without deciding that section 29 would preclude an expert’s testimony

“tantamount to stating an opinion that defendant did or did not have the mental state

required for the crime charged”]; People v. Coddington (2000) 23 Cal.4th 529, 582–583;

People v. Larsen (2012) 205 Cal.App.4th 810, 826; People v. Cortes (2011) 192

Cal.App.4th 873, 908; People v. Nunn (1996) 50 Cal.App.4th 1357, 1364–1365.) What

is left as a legitimate subject of expert testimony is somewhat uncertain; applying the

evidentiary rules established by sections 25, 28 and 29 has been compared to venturing

into a “legal ‘bog.’ ” (Nunn, at p. 1364; see also, Larsen, at p. 828 [testimony about

hypothetical situations similar to the actual case would “come dangerously close to the

territory precluded by sections 28 and 29”].) But regardless of the uncertain contours of

allowable expert testimony on a defense of diminished actuality in a given case, this

much is certain: assuming the strategic calculus of advantages versus disadvantages

favors pursuit of a defense of diminished actuality, there is unquestionably a range of

allowable expert testimony that competent counsel would seek to present in support of

such a defense, if the evidence supports it, and an expert is available to present it.

38

d. Kirkpatrick’s Reasons for Not Using a Neuroscience Expert

or Further Investigating a Dissociation/Brain Damage

Defense Were Insufficient in Light of the Information

Provided to Her by Drs. Pettis and Podboy.

Kirkpatrick testified she dropped Dr. Pettis as a potential expert witness because

he told her Cervantes had antisocial personality disorder and was “basically bad.” She

recounted Dr. Pettis’s opinion that Cervantes had consistent “behavioral problems” as a

child and “always chose the wrong path.” Pettis observed that Cervantes “respects no

one” and frequently “loses his temper.” Kirkpatrick thought Dr. Pettis would be a

“horrible” witness.

Dr. Pettis denied making these statements and testified he told Kirkpatrick only

that Cervantes’s answers to questions might be taken as antisocial by a listener without

some further explanation. In fact, he testified it was not possible to diagnose antisocial

personality disorder in a person so young. He also told Kirkpatrick he did not think a

functional mental illness defense was likely, but frontal cortex problems could account

for Cervantes’s behavior. Dr. Pettis testified he told Kirkpatrick he suspected brain

impairment in Cervantes’s case and insinuated he had told her to get Cervantes “brain-

tested.” He had not diagnosed Cervantes while retained by Kirkpatrick because he was

awaiting the results of the neuropsychological testing. Dr. Pettis was surprised that

Kirkpatrick cut off communication with him.

Kirkpatrick decided against using Dr. Watson as an expert because he “did not see

any mental defenses.” Dr. Watson found no “obvious mental health problems or

anything that [Kirkpatrick] could use as a defense.” His “final evaluation” was that he

“did not have anything to help [her] further” on the case, and he told her “all [she] had

was alcohol.” Kirkpatrick evidently discounted him as a witness and decided not to use a

mental health defense on the basis of his opinion, together with Dr. Pettis’s.

Dr. Watson did not testify at the new trial hearing. Dr. Shore was asked about his

conversations with Dr. Watson by the district attorney on cross-examination, and he

suggested Dr. Watson had been investigating and was open to a “temporary insanity” (or

diminished actuality) defense.

39

Kirkpatrick rejected Dr. Podboy as an expert because she felt she could not pin

him down on exactly what his testimony would be. Besides, the doctor also told her

Cervantes was an “alcoholic,” which she did not consider helpful, though Coffer points

out it could have been turned to his advantage in a diminished actuality defense. And,

according to Kirkpatrick, Podboy performed a Minnesota Multiphasic Personality

Inventory (MMPI) test on Cervantes without Kirkpatrick’s request or permission, which

Kirkpatrick considered inappropriate and potentially harmful to the defense because she

would have to turn those documents over to opposing counsel if she used Podboy as an

expert witness.24

Dr. Podboy’s testimony at the hearing on the new trial motion was quite different

from Kirkpatrick’s. He was engaged to evaluate a defense based on alcohol intoxication,

and specifically alcohol-induced blackouts, but he testified Kirkpatrick “cut [him] off

pretty quickly.” He claimed she stopped responding to his phone calls and her

communication with him was so problematic that he reported the issue to her supervisor.

This apparently occurred after the trial.

Dr. Podboy testified he had reached the preliminary conclusion that Cervantes was

“in a dissociative state at the time of the incident” and included that finding in his notes,

which he faxed to Kirkpatrick shortly before the trial started. The court refused to admit

the notes into evidence based on a hearsay objection. This was evidentiary error, as

Podboy’s notes were relevant for the non-hearsay purpose of showing what information

Kirkpatrick possessed in making her decisions. The issue is whether there were red flags

in the documents such as to require further investigation, not whether the information in

the red flags was true. Due to the erroneous ruling, we do not have Dr. Podboy’s notes

24

Dr. Podboy testified he did not administer the complete MMPI test to Cervantes.

He used some of the questions from the MMPI as a “springboard” to understanding him.

He acknowledged, however, that the questions he posed to Cervantes could have possibly

been used to detect personality disorders.

40

before us, but we credit his testimony that he alerted Kirkpatrick to the possibility that

Cervantes was in a “dissociative state” when he committed the crimes.

Dr. Podboy was not finished with his investigation when he stopped working for

Kirkpatrick. In the course of the investigation he had learned Cervantes had chronic

alcohol and substance abuse problems, and he may have been under the influence of

psilocybin mushrooms at the time of the crimes. But, he testified, Kirkpatrick “blew

[him] off” and told him, “We’ve already looked into that” whenever he tried to talk to her

about issues other than alcohol-induced blackouts.

After considering Kirkpatrick’s explanation for not using a mental health defense,

we cannot help but conclude she rejected it primarily because of the negative comments

found in the IEP and medical histories. Though that may have been a legitimate tactical

decision if she had first investigated to find out what expert testimony might be available,

we find her investigation was inadequate. Kirkpatrick rejected using a mental health

expert before making a reasonable investigation that would have allowed a competent

weighing of the pros and cons. (Cf. Pruitt v. Neal, supra, 788 F.3d at pp. 271–274

[mitigating evidence].)

Finally, Cervantes criticizes Kirkpatrick for not ordering a brain scan performed.

She discussed that possibility with her supervisor, who suggested she ask Dr. Watson’s

opinion. Dr. Watson did not give her a clear answer. Kirkpatrick was not asked about

any advice by Dr. Pettis on this point, and in his testimony Dr. Pettis implied he told her

Cervantes needed to be “brain-tested.” Combined with Dr. Podboy’s advice that

Cervantes was in a “dissociative state” when he committed the crimes, we conclude

counsel should have followed up with further investigation. Further consultation, with

Podboy, Watson or some other expert, was necessary before rejecting a diminished

actuality defense based on organic brain damage as a synergistic cause of Cervantes’s

behavior, together with alcohol intoxication. A SPECT scan was one option in

investigating further.

41

We realize the SPECT brain scan has been subject to criticism by some experts,

who reject its use for diagnostic and forensic purposes.25 We certainly do not suggest

that a SPECT scan is required to comply with an attorney’s objective standard of

reasonable competence whenever mental problems are suspected in a client. But here,

there were clear red flags that there might be organic brain issues. Counsel’s failure to

secure a SPECT scan, coupled with her failure to consult further with experts about next

steps, contributes to our concern about the reliability of the verdicts. Regardless whether

SPECT scans are universally accepted as a forensic tool, whether to accept or reject

SPECT results would best be left to a jury.

e. Kirkpatrick’s Faulty Investigation Requires Reversal of the

Specific Intent Convictions, Except Burglary.

Wherever the truth lay in the communications between Kirkpatrick and her

experts, the trial court found Kirkpatrick’s testimony to be credible, and we normally

defer to trial court findings on credibility. (See People v. Jones (2013) 57 Cal.4th 899,

917; People v. Carasi (2008) 44 Cal.4th 1263, 1290–1291; Shamblin v. Brattain (1988)

44 Cal.3d 474, 479.) We do not doubt she truthfully reported her conversations with

her experts as she recalled them. Nevertheless, we cannot ignore the disturbingly

different testimony of Drs. Pettis and Podboy.

Judge Kinnicutt did find that Kirkpatrick’s decisions were “competently made

based upon the information she had at the time.” The judge ultimately rejected the

ineffective assistance of counsel claim on the basis that the jury would not have reached a

different conclusion even if the mushroom and dissociation evidence had been admitted.

25

See People v. Banks (2014) 59 Cal.4th 1113, 1143 [expert testified “SPECT

imaging . . . shows whether all areas of the brain are getting proper blood perfusion, but

cannot be used to make a diagnosis or predict behavior” and could not be used to prove

defendant had “brain damage”]; Yum v. Carey (C.D. Cal. 2009) 2009 U.S. Dist. LEXIS

120284, at pp. *65–*71 [denying federal habeas relief where California trial court

excluded defendant’s SPECT evidence].

42

That is a conclusion about prejudice, which we review de novo.26 (People v. Ault (2004)

33 Cal.4th 1250, 1264, fn. 8, citing Strickland, supra, 466 U.S. at p. 698.)

We have serious doubts about the quality of the defense Kirkpatrick provided in

the realm of mental state evidence due to her insufficient investigation and resulting

incompetent selection of an expert on the alcohol intoxication defense. (Cf. People v.

Frierson, supra, 25 Cal.3d at p. 159 [presentation of an “undeveloped theory” of

diminished capacity].) Four specific failings trouble us: (1) her failure to secure a copy

of Dr. Pettis’s notes; (2) her failure to conduct further investigation of the psilocybin

mushroom defense; (3) her failure to follow up when Dr. Podboy told her Cervantes was

probably in a “dissociative state” when he committed the crimes; and (4) her selection of

Zehnder as the sole expert to testify in support of the alcohol intoxication defense.

We shall reverse all the convictions involving specific intent, except the burglary

conviction and burglary special findings, as to which, for reasons already discussed, we

do not feel the same concerns. Because the trial court had the discretion to grant the

motion for a new trial as to some counts but not others (People v. Drake (1992) 6

Cal.App.4th 92, 99), we find it erred in denying the motion in its entirety. We reverse,

however, only eight of the convictions that required specific intent, as specified in the

disposition.27

26

Since we reverse the convictions involving specific intent, we do not address

Cervantes’s argument that he was prejudiced in plea bargaining to a determinate term.

The parties were far apart, and it is utterly speculative to suggest the prosecutor would

have agreed to a term of years that Cervantes would have found acceptable, even if

defense counsel had prepared more thoroughly for trial.

27

Cervantes also claims Kirkpatrick violated her duty of loyalty by refusing to

turn over her files to Obstler’s firm after trial or otherwise cooperate in preparing the

motion for a new trial. We agree she was under an ethical duty to turn over her files

(State Bar of California, Standing Committee On Professional Responsibility And

Conduct, Issue: To What Extent Must a Criminal Defense Attorney, Having Been

Relieved by Successor Counsel, Cooperate with New Counsel?, State Bar Formal Opn.

No. 1992-127, 1992 WL 166235), but we question Obstler’s conclusion that she violated

that duty. It appears she turned over her files, then stopped cooperating at the direction of

43

B. Proposition 57

1. Overview of this Section

Having concluded in the unpublished portion of our opinion that we must reverse

eight counts of conviction that required specific intent and will affirm seven counts,

including the burglary conviction, we now turn to the impact on our disposition of Prop

57, which was passed by the voters on November 8, 2016.28 Six weeks later, Cervantes

sought and was granted permission to file a supplemental brief arguing that Prop 57

should be applied retroactively to all non-final cases under the rationale of In re Estrada

(1965) 63 Cal.2d 740 (Estrada), where the Supreme Court held that a legislative

reduction in the statutory penalty for a crime must be applied to all non-final cases. He

contends Prop 57 requires us to vacate the judgment entered below and remand the case

for a “fitness” or “transfer” hearing29 in juvenile court, after which Cervantes would

either be dealt with through the juvenile justice system, or else would be transferred by

her supervisor, and eventually cooperated fully after her supervisor changed his directive.

Even if she was temporarily out of compliance with the ethical duty, such a failure does

not give rise to a claim of ineffective assistance of counsel. (See Nix v. Whiteside (1986)

475 U.S. 157, 165; Roe v. Flores-Ortega (2000) 528 U.S. 470, 479; People v. Williams,

supra, 56 Cal.4th at p. 692.)

28

Prop 57 also included provisions relating to prison inmates, prison credits, and

eased access to parole consideration. We deal in this opinion solely with the changes

implemented with respect to juveniles.

29

Under Welfare and Institutions Code former section 707, subdivision (a)(1) and

(c), the determination to be made was whether the minor was a “fit and proper subject to

be dealt with under the juvenile court law.” (See Stats. 2015, ch. 234, § 2, p. 2138.) The

current section, as amended by Prop 57, eliminates the phrase beginning “fit and proper

subject”; instead, the juvenile court is simply empowered to “decide whether the minor

should be transferred to a court of criminal jurisdiction.” (Welf. & Inst. Code, § 707,

subd. (a)(2).) Given this change, it is perhaps more correct to refer to the hearing as a

“transfer” hearing. The criteria to be used, however, correspond closely to the criteria

used to determine fitness. (Compare Welf. & Inst. Code, § 707, subd. (a)(2)(A)–(E) with

Welf. & Inst. Code, former § 707, subds. (a)(1)(A)–(E), (a)(2)(B)(i)–(v), (c)(1)–(5)

(Stats. 2015, ch. 234, § 2, pp. 2138–2139).) We use the terms “fitness hearing” and

“transfer hearing” interchangeably.

44

the juvenile court30 to adult court under Welfare and Institutions Code, section 70731 for a

new trial. In fact, he contends this procedure is mandated for any counts reversed on

appeal and remanded for a new trial, even if Prop 57 is not given retroactive effect. He

then suggests that such a disposition would result in “split” jurisdiction between two

courts, with reversed counts going to juvenile court and affirmed counts remaining in

criminal court, which would violate the rule of In re Dennis J. (1977) 72 Cal.App.3d 755,

760–762 (Dennis J.) that the same juvenile cannot have charges pending in both criminal

court and juvenile court that might lead to conflicting orders from the two divisions. (But

see Welf. & Inst. Code, § 707.01, subd. (a)(1) [allowing simultaneous jurisdiction].)

Hence, he reasons, we may not reverse some counts and affirm others. A new trial of all

counts is required.

We reject the argument that Prop 57 gives Cervantes, in effect, a full “do over” on

all counts despite the fact he has already been tried, convicted and sentenced, but we

conclude that Prop 57, even if not given retroactive effect—which we conclude is not

mandated—nonetheless requires a fitness hearing. That hearing will, in effect, determine

which department of the superior court, the juvenile court or the adult criminal court, will

try any remaining counts on remand (should the People elect retrial) and will decide the

consequences Cervantes faces for the offenses of which he stands convicted or that are

found to be true following remand.

In arriving at this conclusion, we specifically reject Cervantes’s argument that he

is entitled to a fitness hearing on retroactivity or equal protection grounds because (1) the

30

We refer to “juvenile courts” and “criminal courts” throughout this section of

the opinion, rather than the more unwieldy “court of criminal jurisdiction” (Welf. & Inst.

Code, § 707, subd. (a)) or “superior court sitting as a juvenile court” (see Welf. & Inst.

Code, § 245.) We use “adult court” and “criminal court” (or a combination of the two)

interchangeably.

31

In this section of our opinion we shall refer to both former and current Welfare

and Institutions Code section 707 as “Section 707” and to both former and current

Welfare and Institutions Code section 602 as “Section 602.” The former sections will be

designated as such.

45

juvenile division and criminal division of superior court both have subject matter

jurisdiction over statutorily specified crimes committed by minors; (2) the statutory

amendments under Prop 57 do not amount to a reduction of a penalty and are not subject

to retroactive application under Estrada; and (3) failing to extend the new hearing

procedure to Cervantes does not deprive him of equal protection because a prospective

procedural change in the law that treats offenders differently depending upon when their

crimes were committed does not violate equal protection. We ultimately agree, however,

based on both the text of Prop 57 and the ameliorative purpose behind it, that Cervantes

is entitled to a fitness hearing before he may be retried or resentenced in adult court. The

initiative measure fundamentally favors rehabilitation over punishment32 for juveniles

and requires procedural changes to ensure that juveniles facing trial in adult criminal

court will receive a judicially conducted fitness hearing first. Because Cervantes once

again faces a trial or resentencing on remand, he is entitled under Prop 57 to a transfer to

juvenile court for a fitness hearing, as described in Section 707, before retrial or

resentencing.

2. Procedure for Transferring a Minor to Adult Court under Prop 57

Historically, California required a judicial determination of unfitness for juvenile

court before a minor could be prosecuted in adult court. (See Stats. 1961, ch. 1616, § 2,

p. 3485; Stats. 1975, ch. 1266, § 4, p. 3325; Juan G. v. Superior Court (2012) 209

Cal.App.4th 1480, 1489, & fn. 4 (Juan G.).) Beginning in March 2000 and continuing

until the adoption of Prop 57, however, the district attorney was authorized, as a matter of

executive discretion, to file a criminal action against a juvenile in certain defined

32

This change of focus corresponds to a legislative change in section 1170,

subdivision (a)(1), which recently adopted the language “the purpose of sentencing is

public safety achieved through punishment, rehabilitation, and restorative justice.”

(§ 1170, subd. (a)(1), as amended by Stats. 2016, ch. 696, § 1, effective January 1, 2017.)

Beginning in 1977 and for nearly 40 years thereafter, the same statute reflected a pure

punishment model: “The Legislature finds and declares that the purpose of imprisonment

for crime is punishment.” (Former §1170, subd. (a)(1), added by Stats. 1976, ch. 1139,

§ 273, operative July 1, 1977.)

46

circumstances, rather than filing the case in juvenile court, a practice known as “direct

filing” or “discretionary direct filing.” (Former § 707, subd. (d); see Voter Information

Guide, Primary Elec. (Mar. 7, 2000) Proposition 21, the Gang Violence and Juvenile

Crime Prevention Act of 1998, § 26, pp. 126–127 (Prop 21); see generally, Manduley v.

Superior Court (2002) 27 Cal.4th 537, 548–550 (Manduley); Juan G., at p. 1489 & fn. 4.)

Some crimes, like Cervantes’s, were considered so serious by the voters that, if

committed by a minor age 14 or older, juvenile court was not an option; filing in adult

criminal court was mandated by statute (“mandatory direct filing”).33 (Former § 602,

subd. (b); see Prop 21, § 18, p. 125; Juan G., at pp. 1488–1489.) Mandatory direct filing

had first been instituted by the Legislature for juveniles age 16 or older effective January

1, 2000 (Stats. 1999, ch. 996, § 12.2, p. 7560), but Prop 21 reduced the age to 14.

Prop 57 was designed to undo Prop 21. After the passage of Prop 57, the charging

instrument for all juvenile crimes must be filed in juvenile court. (See § 602.) While

prosecuting attorneys may move to transfer certain categories of cases to criminal court

(§ 707, subd. (a)(1)), they have no authority to directly and independently file a criminal

complaint against someone who broke the law as a juvenile, even by committing the

crimes that previously qualified for mandatory direct filing. In cases where transfer to

adult court is authorized (§ 707, subd. (b)) (and not all cases qualify),34 the juvenile court

33

Mandatory direct filing existed before Prop 21 was passed, but it was authorized

only for youths age 16 or older. (Stats. 1999, ch. 996, § 12.2, p. 7560; Manduley, supra,

27 Cal.4th at pp. 549–550; Juan G., supra, 209 Cal.App.4th at pp. 1488–1490.)

34

Under former Section 707, for minors age 16 and over, transfer was authorized

for any felony listed in former Section 707, subdivision (b) (former § 707, subd. (d)(1)),

or any other felony if committed in the circumstances described in former section 707,

subdivision (d)(3). (Former § 707, subd. (a)(1).) Fourteen and 15-year-old minors could

be transferred to adult court only if they were charged with one of 30 crimes listed in

former section 707, subdivision (b), which, as relevant to this case, included attempted

murder, forcible rape, sodomy, oral copulation, forcible lewd act with a child under age

14, torture and aggravated mayhem. (Former § 707, subd. (b)(4)–(7), (12), (23) & (24).)

Cervantes would continue to be eligible for transfer to adult court under the new law, but

he would be entitled to a fitness hearing before transfer. (§ 707, subds. (a)(2), (b)(4)–(7),

47

now has sole authority to determine whether the minor should be transferred. (§ 707,

subd. (a)(2); see Brown v. Superior Court (2016) 63 Cal.4th 335, 340–341 [describing

history and general provisions of the initiative measure].) Thus, Prop 57 effectively

guarantees a juvenile accused felon a right to a fitness hearing before he or she may be

sent to the criminal division for prosecution as an adult.

3. Prop 57 Is Not Retroactive.

Cervantes relies on two provisions of Prop 57 in advancing his claim of

retroactivity: (1) Section 602 was amended to provide that jurisdiction in the juvenile

court is what Cervantes calls “exclusive”; and (2) Section 707 was amended to require a

fitness hearing in juvenile court as a prerequisite for transfer of a juvenile to adult

criminal court. (§ 707, subd. (a).) He relies on voter intent to invest greater efforts in

rehabilitation—“especially for juveniles”—in order to stop the “revolving door” of crime.

(Voter Information Guide, Gen. Elec. (Nov. 8, 2016) text of Prop. 57, § 2, p. 141, (2016

Voter Guide).) Because a juvenile court has a time limit on jurisdiction over its wards

(e.g., Welf. & Inst. Code, §§ 607, 607.1, 1769), and because its focus is on

rehabilitation,35 Cervantes argues the new transfer provisions effectively mandate more

lenient treatment—or a reduction in punishment—such that Prop 57 must be applied

retroactively under Estrada.

We reject Cervantes’s position on retroactivity because (1) jurisdiction is

concurrent in crimes that subject the juvenile offender to adult prosecution; (2) Estrada’s

reasoning does not support a retroactive application; and (3) subsequent Supreme Court

cases interpreting Estrada have limited it to situations in which there is, strictly speaking,

(12), (23) & (24).) Under prior law he was not entitled to adjudication in juvenile court

under any circumstances and was not entitled to a fitness hearing because at age 14 he

was alleged to have committed the above-specified sex crimes in circumstances

warranting one-strike treatment. (Former § 602, subd. (b)(2)(A), (D) & (F).)

35

E.g., Welf. & Inst. Code, §§ 202, 1700; Manduley, supra, 27 Cal.4th at p. 593;

In re Greg F. (2012) 55 Cal.4th 393, 416–417; In re Kasaundra D. (2004) 121

Cal.App.4th 533, 539.

48

a “ ‘ “legislative mitigation of the penalty for a particular crime.” ’ ” (People v. Brown

(2012) 54 Cal.4th 314, 325 (Brown).)

a. Section 602 Does Not Deprive the Criminal Division of

Concurrent Subject Matter Jurisdiction Over Statutorily

Specified Crimes Committed by Juveniles.

With the passage of Prop 57, Section 602 now reads: “Except as provided in

Section 707, any person who is under 18 years of age when he or she violates any law of

this state or of the United States or any ordinance of any city or county of this state

defining crime other than an ordinance establishing a curfew based solely on age, is

within the jurisdiction of the juvenile court, which may adjudge such person to be a ward

of the court.” Cervantes contends this gives the juvenile court “exclusive jurisdiction”

over his case and the cases of all juvenile offenders36 whose cases are not yet final on

appeal—at least until a fitness hearing is conducted.

Yet, the statute does not use the term “exclusive,” and it assigns jurisdiction over

all juvenile criminal matters to the juvenile court explicitly subject to the exceptions in

Section 707. Section 707 applies to “a minor . . . alleged to be a person described in

Section 602 by reason of the violation, when he or she was 16 years of age or older, of

any felony criminal statute, or of an offense listed in subdivision (b) when he or she was

14 or 15 years of age.” Section 707, subdivision (b) lists 30 serious crimes that subject

even the youngest juveniles eligible for criminal prosecution under section 707, ages 14

and 15, to adult prosecution.

We conclude, for crimes that qualify the juvenile offender for transfer to adult

court, subject matter jurisdiction is concurrent between the criminal division and the

juvenile division. “The juvenile court and the criminal court are divisions of the superior

court, which has subject matter jurisdiction over criminal matters and civil matters,

36

We use the term “juvenile offender” to refer to “any person who is under 18

years of age when he or she violates any law of this state or of the United States or any

ordinance of any city or county of this state defining crime other than an ordinance

establishing a curfew based solely on age.” (§ 602.)

49

including juvenile proceedings. (See Cal. Const., art. VI, § 10.) When exercising the

jurisdiction conferred by the juvenile court law, the superior court is designated as the

juvenile court. (Welf. & Inst. Code, § 245.) Accordingly, when we refer herein to the

jurisdiction of the juvenile court or the jurisdiction of the criminal court, we do not refer

to subject matter jurisdiction, but rather to the statutory authority of the particular

division of the superior court, in a given case, to proceed under the juvenile court law or

the law generally applicable in criminal actions. (See In re Harris (1993) 5 Cal.4th 813,

837.)” (Manduley, supra, 27 Cal.4th at p. 548, fn. 3.) To the extent Cervantes suggests

Prop 57 deprives the criminal court of fundamental subject matter jurisdiction over a

juvenile felon37 upon remand after appeal, we disagree.38

37

We use the term “juvenile felon” to denote “a minor . . . alleged to be a person

described in Section 602 by reason of the violation, when he or she was 16 years of age

or older, of any felony criminal statute, or of an offense listed in subdivision (b) [of

Section 707] when he or she was 14 or 15 years of age.” (§ 707, subd. (a)(1).)

38

Manduley decided implementation of the direct file system under Prop 21 did

not violate the Equal Protection Clause. (Manduley, supra, 27 Cal.4th at pp. 567–573.)

Cervantes, too, contends denial of a fitness hearing in his case would deprive him of

equal protection. But Cervantes is not similarly situated with a juvenile felon who

committed his crimes after Prop 57 was enacted. (See Brown, supra, 54 Cal.4th at

pp. 328–330 [rejecting equal protection argument based on legislative changes in the

rates at which defendants earn presentence conduct credits].)

Moreover, we conclude if there was no equal protection problem in adopting the

direct file procedure, there can be no equal protection problem in abandoning it. Both

systems are merely different approaches or experiments in how best to deal with juvenile

delinquents, and a state may change course without creating equal protection problems.

(People v. Spears (1995) 40 Cal.App.4th 1683, 1688 [“ ‘It is perfectly proper for the

Legislature to create a new sentencing procedure which operates prospectively only.

Despite the disparity created by rendering different sentences after an admittedly

arbitrarily chosen date, prospective application of such a statute does not violate equal

protection principles, because of the legitimate public purpose’ ” motivating the change

of law.].) “ ‘[T]he 14th Amendment does not forbid statutes and statutory changes to

have a beginning, and thus to discriminate between the rights of an earlier and later

time.’ ” (People v. Floyd (2003) 31 Cal.4th 179, 191, quoting Sperry & Hutchinson Co.

v. Rhodes (1911) 220 U.S. 502, 505 [Floyd refused to apply drug treatment alternative

sentencing retroactively].)

50

Prop 57’s amendment to Section 602 does not govern where jurisdiction must be

sited permanently, nor does it either prescribe or prohibit revisiting the jurisdictional

issue when a case is remanded for a new trial. Being subject to Section 707, subdivision

(b), which lists the crimes that may qualify a 14-year-old for transfer to criminal court,

Cervantes is subject to concurrent jurisdiction in both criminal and juvenile divisions of

superior court. Here, the criminal division lawfully assumed jurisdiction under pre-Prop

57 law and retained jurisdiction throughout the trial; Section 602 does not oust the

criminal division of jurisdiction upon remand after an appeal.

b. Estrada’s Reasoning Does Not Support Retroactive

Application of Prop 57.

New statutes or changes in statutes ordinarily are applied prospectively only,

“absent an express declaration of retrospectivity or a clear indication that the electorate,

or the Legislature, intended otherwise.” (Tapia v. Superior Court (1991) 53 Cal.3d 282,

287 (Tapia); see also § 3; Brown, supra, 54 Cal.4th at pp. 319–323.) Nevertheless, a

well-recognized exception prevails when a criminal statute reduces the penalty for a

given crime; in that circumstance, the new, less punitive statute applies to all defendants

whose convictions are not yet final on appeal. (Estrada, supra, 63 Cal.2d at p. 745.)

Estrada involved an escape without force or violence from the California

Rehabilitation Center at a time when that offense required a minimum one-year

additional sentence and a two-year minimum after being returned to custody before

parole consideration. (Estrada, supra, 63 Cal.2d at p. 743.) Between the time of the

escape and the time Estrada was convicted, however, the governing statutes were

amended to reduce the minimum term to six months for nonviolent escape and to require

no minimum period before parole consideration. (Id. at pp. 743–744.) Estrada was being

held in custody solely because of the minimum terms required by the former version of

the statute and would have been entitled to release under the new version. (Id. at p. 744.)

Finding the reduction in penalty amounted to a legislative decision that the prior law had

been too harsh, the Court held the amended statutes applied to the petitioner.

“[L]egislative mitigation of the penalty for a particular crime” called for the retroactive

51

application of the reduced penalty, effectively establishing a rule that any law reducing

the penalty for a crime was intended to apply to all non-final judgments. (Id. at p. 745;

see also, People v. Conley (2016) 63 Cal.4th 646, 656 [Estrada’s holding “reflects a

presumption about legislative intent, rather than a constitutional command”].)

c. The Supreme Court Has Limited Estrada to Statutory

Changes That Mitigate the Penalty for a Particular Offense.

Cervantes argues that Prop 57 amounts to a reduction in punishment, requiring us

to find it retroactive under Estrada. But later Supreme Court cases have limited Estrada’s

retroactivity exception to statutory changes that mitigate the penalty for a particular

crime, which is not true of Prop 57.

In 1991, in response to Proposition 115 (Prop 115), which, among other things,

gave judges in criminal trials the power to conduct voir dire instead of attorneys (see

Code Civ. Proc., § 223), Tapia, supra, 53 Cal.3d 282 held most of the new procedures,

including that relating to voir dire, were not retroactive under Estrada. (Id. at pp. 287.)

Tapia emphasized that the retroactivity exception turns on the type of legal change

effectuated by the new or amended statute: changes in direct penal consequences like the

one under consideration in Estrada, would call for retroactive application, while those

like the one involved in Tapia that “address the conduct of trials which have yet to take

place, rather than criminal behavior which has already taken place” are to be applied

prospectively.39 (Id. at pp. 288–289.) Under that rubric, the transfer procedure dictated

39

In applying these rules, Tapia identified four categories of legal changes

introduced by Prop 115, including those that “change the legal consequences of criminal

behavior to the detriment of defendants” and others described as “provisions which

clearly benefit defendants” or “benefit only defendants.” (Tapia, supra, 53 Cal.3d at

pp. 297, 300.) The first category of changes could not be applied retroactively, but Tapia

saw no impediment to applying the latter category of rules in trials involving crimes that

had occurred before the effective date of Prop 115. (Id. at pp. 300–301.) Tapia suggests

such laws “may be applied to pending cases” if they “ ‘lessen the punishment’ ” for

crime. (Id. at p. 301.) We believe Brown has narrowed Estrada further, requiring that a

statute “ ‘mitigat[e] . . . the penalty for a particular crime’ ” before it is accorded

retroactive effect. (Brown, supra, 54 Cal.4th at p. 326.)

52

by Prop 57 is not one that addressed “criminal behavior which has already taken place,”

but is more correctly identified as one “address[ing] the conduct of trials which have yet

to take place.” (Ibid.) This suggests its application should be prospective only.

Brown, supra, 54 Cal.4th 314 even more strongly supports a prospective

application. Brown dealt with a situation more closely linked to the length of punishment

for an offense than Prop 57 is. Specifically, Brown addressed the legislative changes in

presentence conduct credits in 2010–2011, which first gave pretrial detainees more

generous conduct credits, then reverted to less generous credits, and finally switched back

again to more generous credits. (Brown, at pp. 317–318, 320.) The Supreme Court

unanimously decided that a jail inmate awaiting trial and sentencing should earn credits

at the rate in effect at the time he served the days in jail, not at the more generous rate

available when he was sentenced. (Id. at pp. 319–330; see also, In re Strick (1983) 148

Cal.App.3d 906, 909–914 [legislation to allow greater worktime credits in prison held not

to apply retroactively and not to violate equal protection].)

Brown recognized that a change in credits-earning eligibility would affect the

length of a prisoner’s time in custody, and in that sense would have a direct effect on

punishment (Brown, supra, 54 Cal.4th at p. 325), but nevertheless held Estrada did not

apply. The Brown court called the Estrada rule a “contextually specific qualification to

the ordinary presumption that statutes operate prospectively.” (Id. at p. 323.) It

reasoned, “the rule and logic of Estrada is specifically directed to a statute that represents

‘ “a legislative mitigation of the penalty for a particular crime” ’ (Estrada, at p. 745,

italics added) because such a law supports the inference that the Legislature would prefer

to impose the new, shorter penalty rather than to ‘ “satisfy a desire for vengeance” ’

(ibid.). The same logic does not inform our understanding of a law that rewards good

behavior in prison.” (Brown, at p. 325.)

We find the rationale underlying Estrada equally inapplicable to the procedural

changes implemented by Prop 57. While Prop 57 will have a substantive impact on time

in custody in some cases—sometimes a big impact—the transfer procedure required

under Section 707 does not resemble the clear-cut reduction in penalty involved in

53

Estrada. Although it is now the juvenile court, rather than the district attorney, that

makes the decision whether a juvenile felon will be tried as an adult, we may presume

that many cases filed in juvenile court will still end up in adult court (with adult

penalties) under Prop 57, after the fitness hearing is held. Prop 57 mitigates the penalty

for a particular crime even less directly than the jail credits at issue in Brown. More like

the voir dire procedure in Tapia, which affected who performed a particular function in

the judicial process, Prop 57 may or may not in some attenuated way affect punishment,

but it is not a direct reduction in penalty as required for retroactivity under Estrada.

(Brown, supra, 54 Cal.4th at p. 325.)

4. Applied Prospectively, Prop 57 Requires A Fitness Hearing Before a

Juvenile Felon Is “Tried in Adult Court” Initially or on Remand.

a. Textual Provisions of Prop 57 and Principles of Statutory

Construction Bearing on this Issue

Section 2 of Prop 57, titled “Purpose and Intent,” “[r]equire[s] a judge, not a

prosecutor, to decide whether juveniles should be tried in adult court.” (2016 Voter

Guide, supra, text of Prop. 57, § 2, p. 141.) The Legislative Analyst spoke even more

strongly: “the only way a youth could be tried in adult court is if the juvenile court judge

in the hearing [under Section 707, subdivision (a)(2)]] decides to transfer the youth to

adult court.” (2016 Voter Guide, supra, analysis of Prop. 57 by Legis. Analyst, p. 56.)

Thus, the phrase, “tried in adult court”—or the prospect of being tried in adult court—

appears to be the trigger for a juvenile’s right to a fitness hearing. This case requires us

to confront whether prospective application of Prop 57 requires a fitness hearing if a

juvenile felon has had one trial in adult court but is about to be tried again because his

case is being remanded for partial retrial or resentencing.

The key statutes amended to effectuate the juvenile provisions of Prop 57 were

Section 602 and Section 707, the latter of which provides for the mechanics of the fitness

hearing. (2016 Voter Guide, supra, text of Prop 57, §§ 4.1 & 4.2, pp. 141–142.) Section

707 describes the ordinary chronology of a fitness hearing when charges are filed after

Prop 57’s effective date. (See section II.B.2, ante, and section II.B.4.c., post.) It does not

54

purport to explore the conduct of or entitlement to such a proceeding on remand. Section

707 describes its reach broadly, indicating it applies “[i]n any case in which a minor is

alleged to be a person described in Section 602.” (§ 707, subd. (a)(1), italics added.) We

see nothing in Section 707 that requires the allegation to be pending in juvenile court at

the time of the fitness hearing. One of the allegations included in each of the counts in

the amended information in criminal court was that Cervantes was “14 years of age” at

the time of his crimes, citing former Section 602, subdivision (b). Thus, in facing retrial

on remand, Cervantes facially qualifies for a fitness hearing. Nothing in Prop 57

indicates the offense had to occur on or after the effective date of Prop 57 in order for the

new procedures to apply. (Cf. John L. v. Superior Court (2004) 33 Cal.4th 158, 169; In

re Chong K. (2006) 145 Cal.App.4th 13, 18–19.) The amendment to section 602 is also

critically important to understanding the availability of a fitness hearing because it strikes

out entirely former subdivision (b) and its reference to mandatory direct filing, thus

raising the question whether a criminal prosecution may be reinitiated on retrial where

the original filing was by authority of former subdivision (b). Together, these two

sections provide a general framework for a transfer hearing in all post-Prop 57 initiated

juvenile cases, but neither sheds much light on how Section 707 should be applied, if at

all, in the case of a remand.

“In interpreting a voter initiative . . . , we apply the same principles that govern the

construction of a statute.” (People v. Canty (2004) 32 Cal.4th 1266, 1276 (Canty).) Of

course, when the intent of a statute is clear from the plain meaning of its text, we follow

its mandates without looking elsewhere. (Greb v. Diamond Internat. Corp. (2013) 56

Cal.4th 243, 256; People v. Leal (2004) 33 Cal. 4th 999, 1008.) Clear intent, clearly

expressed, is decisive. But Section 707, to the extent it provides clues to the availability

of a fitness hearing on remand, does not explicitly or by clear implication either authorize

or preclude such a procedure. It does anticipate the fitness hearing will be conducted

“prior to the attachment of jeopardy.” (§ 707, subd. (a)(1).) As we shall discuss, that

language naturally raises a question as to whether, on a retrial after remand, jeopardy

attaches anew with the swearing of a new jury, or whether the reference to “attachment of

55

jeopardy” should be construed as a reference to the attachment of jeopardy at the first

trial. We find the language of Section 707 to be ambiguous on the issue before us, which

requires that we consult additional interpretive aids in understanding its meaning.

“ ‘[T]he “plain meaning” rule does not prohibit a court from determining whether

the literal meaning of a measure comports with its purpose or whether such a construction

of one provision is consistent with other provisions of the statute.’ ” (Canty, supra, 32

Cal.4th at p. 1276; accord, Lungren v. Deukmejian (1988) 45 Cal.3d 727, 735.) “When

construing a statute or an initiative measure, we are obliged to give effect to every phrase

and paragraph of the law, leaving no part useless or deprived of meaning.” (People v.

Nava (1996) 47 Cal.App.4th 1732, 1737.) “Because the most reasonable interpretation of

a [statutory] provision [enacted as a voter initiative] may be reflected, in part, by

evidence of the enacting body’s intent beyond the statutory language itself, in its history

and background [citation], [a court] also consider[s] the measure as presented to the

voters with any uncodified findings and statements of intent. In considering the purpose

of legislation, statements of the intent of the enacting body contained in a preamble,

while not conclusive, are entitled to consideration. [Citations.] Although such

statements in an uncodified section do not confer power, determine rights, or enlarge the

scope of a measure, they properly may be utilized as an aid in construing a statute.”

(Canty, supra, at p. 1280.) In determining voter intent, we may also look to the ballot

arguments favoring the measure, the Legislative Analyst’s interpretation, and the history

of the initiative measure. (Amwest Sur. Ins. Co. v. Wilson (1995) 11 Cal.4th 1243, 1256.)

Applying those guidelines, additional uncodified provisions of Prop 57 inform our

interpretation of its scope. Uncodified section 5 mandates that Prop 57 be “broadly

construed to accomplish its purposes,” and that section actually prohibits the Legislature

from amending Section 602 and Section 707 unless “such amendments are consistent

with and further the intent of this act.” (2016 Voter Guide, supra, text of Prop. 57, § 5,

p. 145.) Uncodified section 9 also advises us to “liberally construe[] [Prop 57] to

effectuate its purposes.” (2016 Voter Guide, text of Prop 57, § 9, p. 146.)

56

In reaching our conclusion about the availability of a fitness hearing on remand,

we employ the usual rules of construction to ascertain the intent of the electorate, which

of course is our ultimate objective in any endeavor to construe a voter initiative. (Canty,

supra, 32 Cal.4th at p. 1276.) We find especially significant, indeed controlling, the tenet

requiring us to read a statute in a manner to effectuate its underlying purpose. (Id. at

pp. 1276–1277 [“The intent of the law prevails over the letter of the law, and ‘ “the letter

will, if possible, be so read as to conform to the spirit of the act.” [Citation.]’ ”]; Amador

Valley Joint Union High Sch. Dist. v. State Bd. of Equalization (1978) 22 Cal. 3d 208,

245 [“The literal language of enactments may be disregarded to avoid absurd results and

to fulfill the apparent intent of the framers.”].) With these principles and provisions in

mind, we turn to the task of determining Prop 57’s impact on this case.

b. The Ameliorative Intent Underlying Prop 57 Is Implemented

Largely Through the Requirement of an Impartial Judicial

Fitness Hearing to Inquire into the Juvenile Felon’s

“Behavioral Patterns and Social History” and to Determine

His or Her Amenability to Rehabilitation Through Juvenile

Court.

Taking a broad view of the stated “Purpose and Intent” of Prop 57 in section 2 of

the initiative measure, we must first acknowledge it as a dramatic change of course and a

ringing endorsement of rehabilitation as opposed to pure punishment, especially for

youthful offenders. Indeed, Prop 57 expressly aimed to “Stop the revolving door of

crime by emphasizing rehabilitation, especially for juveniles.” (2016 Voter Guide, supra,

text of Prop. 57, § 2, p. 141.) The ballot argument in favor of Prop 57 noted that the

measure “[r]equires judges instead of prosecutors to decide whether minors should be

prosecuted as adults, emphasizing rehabilitation for minors in the juvenile system. [¶] We

know what works. Evidence shows that the more inmates are rehabilitated, the less likely

they are to re-offend. Further evidence shows that minors who remain under juvenile

court supervision are less likely to commit new crimes. Prop. 57 focuses on evidence-

based rehabilitation and allows a juvenile court judge to decide whether or not a minor

57

should be prosecuted as an adult.” (2016 Voter Guide, supra, Argument in Favor of

Proposition 57, p. 58.)

Prop 57 resulted in several important changes in the juvenile court law that,

considered in light of its increased emphasis on rehabilitation, convince us Cervantes

should be given a fitness hearing on remand. We are struck first with Prop 57’s

insistence that “a judge, not a prosecutor, [must] decide whether juveniles should be tried

in adult court.” (2016 Voter Guide, supra, text of Prop. 57, §2, p. 141.) This suggests

the electorate wanted an impartial decision-maker to weigh the juvenile felon’s prospects

for rehabilitation before allowing him or her to be “tried in adult court,” with its attendant

consequences. And judicial discretion was to be exercised in every individual case,

regardless of the juvenile’s crime.

Thus, Prop 57 eliminated the category of mandatory direct file crimes and

offenders that under prior law had foreclosed any prospect of the rehabilitative juvenile

model entirely to the most serious offenders, even as young as 14. (2016 Voter Guide,

supra, text of Prop. 57, § 4.1, pp. 141–142.) This amendment represented a fundamental

policy shift and a rejection of the concept of mandatory direct filing.

By excising subdivision (b) of Section 602, the voters determined that every

juvenile facing trial in criminal court should first have his or her fitness for treatment

under the juvenile court law. Under Prop 57, even those who have committed the most

heinous crimes, and even if they are nearer to the age of majority, are entitled to a fitness

hearing and the exercise of judicial discretion before transfer to adult court. This

suggests an underlying popular conviction that no minor should be considered

irredeemable or beyond rehabilitation without an individualized look at his or her

background and characteristics.40

40

We note that this expressed policy is broadly consistent with the rationale

adopted by the United States Supreme Court in of Graham, supra, 560 U.S. 48, and

Miller v. Alabama (2012) 567 U.S. __, 132 S.Ct. 2455 (Miller), discussed below in

connection with Cervantes’s Eighth Amendment contentions (see post, section II.C), that

58

Turning to the nature of the hearing provided, the probation department must first

prepare and provide to the court a report on the juvenile’s “behavioral patterns and social

history” (§ 707, subd. (a)(1); 2016 Voter Guide, supra, text of Prop. 57, § 4.2, p. 142.)

The report forms a partial basis for the court’s ultimate decision, when considered in

combination with any evidence the juvenile or the district attorney might present. (§ 707,

subd. (a)(2).) In deciding whether to retain jurisdiction or transfer the minor to adult

court, the court must consider specific statutorily defined factors, namely, the degree of

criminal sophistication exhibited by the minor; whether the minor can be rehabilitated

prior to the expiration of the juvenile court’s jurisdiction; the minor’s previous delinquent

history; the success of previous attempts by the juvenile court to rehabilitate the minor;

and the circumstances and gravity of the offense alleged to have been committed by him

or her. (§ 707, subd. (a)(2)(A)–(E).)

Though Cervantes claimed at the new trial motion that he had certain cognitive

impairments and was drug-addled and in a dissociative mental state at the time of the

crimes, he has never had such evidence presented on his behalf in a forum where his

amenability to rehabilitation was an issue to be decided. Because he was subject to

mandatory direct filing, no one ever exercised discretion to determine whether he was

suitable for treatment in juvenile court. Yet, being 14 at the time of his crimes, Cervantes

was in the youngest category of offenders subject to criminal prosecution under Section

707.

Thus, if we remand Cervantes for retrial without ordering a fitness hearing, he will

fall into the subset of juveniles in which no discretion has ever been exercised, either by

the district attorney or by the juvenile court, to determine whether the juvenile court law

might have something to offer for his rehabilitation. This new opportunity, while not a

“mitigation of penalty” in the Estrada sense, must be recognized as ameliorative in

a juvenile’s immaturity and potential for rehabilitation must be taken into account before

any sentence amounting to life imprisonment is imposed.

59

intent, and sometimes in substance, and not just a difference in identity of the person

performing identical procedural tasks, such as that involved in Tapia.

c. The Requirement That the District Attorney Make the

Transfer Motion “Prior to the Attachment of Jeopardy” Does

Not Negate the Possibility That the Voters Intended Prop 57

to Apply on Remand.

We must next determine whether the electorate intended to provide the same

procedural protections to a juvenile felon returning to superior court for retrial after an

appeal. Complicating the question is the fact that Prop 57 requires the district attorney’s

transfer motion to be brought “prior to the attachment of jeopardy.” (§ 707, subd. (a)(1).)

There was no occasion here for a transfer motion—which normally triggers a fitness

hearing—because this case was prosecuted under the old “direct file” regime. The “prior

to the attachment of jeopardy” language tends to suggest the electorate intended a Prop

57 fitness hearing to occur near the beginning of the case against a juvenile felon, before

the jurisdictional hearing in juvenile court. (See fn. 41, post.) Literally and narrowly

read, that language could be taken to mean that, if a case were transferred from criminal

court to juvenile court for a fitness hearing after jeopardy had attached in criminal court,

the case would be stuck in juvenile court, with no means for the district attorney to move

it back to criminal court. We have no trouble concluding the voters would not have

wanted a construction of Prop 57 that would leave the district attorney hamstrung by a

claim the motion was untimely. Deciding the applicability of Prop 57 to Cervantes on

remand forces us to grapple with the question of when jeopardy attaches in this context.

Ordinarily, in criminal cases, jeopardy attaches when the jury, including alternates,

is sworn.41 (People v. Hernandez (2003) 30 Cal.4th 1, 5.) Our task is to determine

41

“In the case of a jury trial, jeopardy attaches when a jury is empaneled and

sworn. [Citations.] In a nonjury trial, jeopardy attaches when the court begins to hear

evidence.” (Serfass v. United States (1975) 420 U.S. 377, 388.) In juvenile proceedings,

jeopardy attaches when the court “ ‘ “enters upon” ’ ” jurisdictional proceedings, or when

the first witness is sworn. (People v. Malveaux (1996) 50 Cal.App.4th 1425, 1439.)

60

whether Cervantes may be tried on remand without a fitness hearing four years after his

first jury was sworn, but before the trial court will embark on empaneling a new jury and

conducting a new trial or resentencing on remand.

Of course, the question of the attachment of jeopardy usually arises in the context

of the double jeopardy prohibition of the Fifth Amendment. (U.S. Const., 5th Amend.)

“The double jeopardy rules are well known. The protection against double jeopardy

generally precludes retrial for the same offense after a conviction or an acquittal.

[Citation.] An exception to this rule applies if the judgment of conviction is reversed as a

result of a defendant’s appeal, motion for new trial, or other challenge by a defendant to

his or her conviction. [Citation.] Like other constitutional guarantees, double jeopardy

protections are not absolute, and may be waived by a defendant. A defendant who files a

motion for a new trial, like a defendant who moves for a mistrial, waives state and federal

double jeopardy protections. [Citations.] By seeking reversal of a judgment of

conviction on appeal, ‘ “[i]n effect [a defendant] assents to all the consequences

legitimately following such reversal, and consents to be tried anew . . . .” ’ (People v.

Sachau (1926) 78 Cal.App. 702, 706.)” (People v. Eroshevich (2014) 60 Cal.4th 583,

590-591.) Under a waiver theory, jeopardy apparently does reattach with the swearing of

a second jury on remand, but because the defendant initiated the appeal, he is deemed to

have waived any double jeopardy objection to a retrial. Because Eroshevich appears to

be the most recent statement on the subject from our state Supreme Court,42 we follow

the waiver theory adopted in that case.

Under that theory, we gather, Cervantes was no longer in jeopardy after he was

convicted and sentenced in the trial below. Nevertheless, when his jury is sworn to retry

the reversed counts in any upcoming trial upon remand, he will once again be “put in

42

We recognize there is some older authority to the contrary. (See Bunnell v.

Superior Court (1975) 13 Cal.3d 592, 606 [“jeopardy as to an offense of which a

defendant has been convicted continues during appellate proceedings and retrial

following reversal of the judgment”].)

61

jeopardy of life or limb.” (U.S. Const., 5th Amend.) His retrial is constitutionally

permissible only because he implicitly waived his right to challenge it on double jeopardy

grounds.

But does the waiver of a double jeopardy challenge by appeal necessarily

encompass a waiver of new and more protective pretrial procedures enacted since

Cervantes’s last trial? We think not. If Prop 57 was intended to give every juvenile felon

a right to a fitness hearing before being “tried in adult court,” as we have held, then we

think that right pertains to a second trial upon remand, as well. In context, the “prior to

attachment of jeopardy” provision governs only timing of the prosecutor’s transfer

motion, not the more fundamental question of the defendant’s entitlement to such a

hearing. In the absence of express guidance in the text of Prop 57 on this particular point,

we decline to adopt a reading of the “attachment of jeopardy” language that treats those

two questions as one and the same. We find it unlikely the voters read the jeopardy

phrase as disentitling a defendant to a fitness hearing on remand. (See also, 2016 Voter

Guide, supra, text of Prop. 57, § 9, p. 146.) By our understanding of jeopardy principles,

if we instruct the trial court to transfer the case to juvenile court for a fitness hearing

before commencing any retrial, the district attorney will have an opportunity to request a

transfer to adult court “prior to the attachment of jeopardy” within the meaning of Section

707, subdivision (a)(1), since by obtaining the benefit of section 707 in that circumstance,

the defendant implicitly waives any right to object to a transfer motion. At oral

argument, Cervantes’ counsel explicitly waived any such objection, but we make clear

that, having requested a fitness hearing at a point following the initial attachment of

jeopardy, the waiver is deemed made by his acceptance of the benefit of section 707.

5. Cervantes Is Eligible for A Fitness Hearing on Remand, Even If the

Reversed Counts Are Dismissed and He Is Only Resentenced on the

Affirmed Counts.

Because we realize the district attorney may opt not to retry Cervantes, we must

also decide whether Cervantes will be entitled to a fitness hearing on remand if the

district attorney decides not to retry him and the trial court then proceeds straight to

62

resentencing. In other words, when Prop 57 mandates a fitness hearing before a juvenile

felon is “tried in adult court” does it also mandate such a hearing before a defendant

remanded after appeal is simply resentenced, rather than being retried?

We have already discussed the voters’ strong support for providing a fitness

hearing for every juvenile facing the prospect of being “tried in adult court.” We now

explain our conclusion that the voters were equally committed to providing such a

hearing before a juvenile convicted of a felony could be sentenced in adult court.

The Supreme Court has found the word “trial” to be “ambiguous” in terms of

whether a “trial” encompasses the sentencing. “[W]hile the word ‘trial’ has long been

interpreted to refer to the process culminating in the determination of guilt, particularly in

bail cases [citations], the word has also been interpreted to include the sentence or

judgment in other cases [citations]. Accordingly, we recognize that the word is

ambiguous as to whether it includes proceedings following the determination of guilt

. . . .” (People v. Overstreet (1986) 42 Cal.3d 891, 896 [plur. opn.]; accord, In re Martin

(1987) 44 Cal.3d 1, 49–50 [opn. of Mosk, J., for a unanimous court].) For some

purposes, the word “trial” does not encompass the sentencing. (Peracchi v. Superior

Court (2003) 30 Cal.4th 1245, 1253 [remand for retrial treated differently from remand

for resentencing for motion under Code Civ. Proc., § 170.6]; People v. Stuckey (2009)

175 Cal.App.4th 898, 910–913 (Stuckey) [“trial” does not include sentencing for purposes

of Evid. Code, § 730, but acknowledging the word is “ambiguous”]; People v.

Domenzain (1984) 161 Cal.App.3d 619, 621–623, & fn. 4 [for purposes of time limits

under section 1382, “brought to trial” on remand does not include delay in resentencing].)

That it is “ambiguous” means the word “tried” or “trial” may have different

meanings in different contexts and different statutes. For instance, it has been recognized

that the right to a speedy trial, guaranteed by the Sixth Amendment as well as by the state

constitution (Cal. Const., art. I, § 15) and state statute (§§1191, 1202, 1381.5, 1382),

includes the right to a speedy sentencing. (See Pollard v. United States (1957) 352 U.S.

354, 361 [assuming arguendo the “sentence is part of the [criminal] trial for purposes of

the Sixth Amendment”]; People v. Mahan (1980) 111 Cal.App.3d 28, 32 [state and

63

federal constitutional rights to “speedy trial” include prompt sentencing]; People v.

Brown (1968) 260 Cal.App.2d 745, 750–751 [“Although in certain contexts the

expression ‘brought to trial’ might possibly encompass only that portion of the criminal

proceeding which results in a determination of the accused’s guilt or innocence, it is clear

that as used in section 1381.5 it includes the entry of a judgment or other final, appealable

order. The imposition of sentence is an essential part of the speedy trial guaranteed to all

accused.”]; see also, People v. Hsu (2008) 168 Cal.App.4th 397, 404 [declining to decide

whether defendant had a right to speedy sentencing, but citing authorities acknowledging

such a right]; cf. People ex rel. Dorris v. McKamy (1914) 168 Cal. 531, 535–536 [“the

word ‘trial,’ as commonly understood in our practice, includes nothing beyond

proceedings in the court in which the case originated”; appeal was not part of trial].)

Similarly, a penalty trial in a capital case is regarded as part of a “single, unitary criminal

proceeding” that includes a “retrial, be it of the entire proceedings or the penalty phase

only.” (People v. Superior Court (Mitchell) (1993) 5 Cal.4th 1229, 1233–1234

[discovery allowed “at trial” under § 1054.3, subd. (a)(1) encompassed the penalty phase

of a capital trial].) Sentencing is also recognized as a “critical stage” in the “criminal

prosecution[]” for purposes of the Sixth Amendment right to counsel. (U.S. Const., 6th

Amend.; Mempa v. Rhay (1967) 389 U.S. 128, 133 [“ ‘the right to counsel is not a right

confined to representation during the trial on the merits’ ”]; cf. People v. Arbee (1983)

143 Cal.App.3d 351, 356 [right to “fair trial” entitles defendant to be present at

sentencing because it “constitutes an essential and material phase of the criminal

proceeding”].)

Our task is to determine the meaning of “tried in adult court” in section 2 of Prop

57. Considering the material in the ballot pamphlet, we believe when the voters approved

Prop 57, they most likely understood being “tried in adult court” to encompass all the

proceedings in a criminal trial court, including sentencing. As demonstrated above, this

is not an unheard of construction.

64

In reaching this conclusion we are cognizant that, although juvenile court

jurisdiction would normally require Cervantes’s release from custody at age 2343 (Welf.

& Inst. Code, § 1769, subd. (c)), there are provisions allowing appropriate authorities to

petition to keep a youthful inmate committed under Section 602 in custody longer—up to

“the maximum term prescribed by law for the offense of which he or she was

convicted”—based on concerns about public safety and dangerousness. (Welf. & Inst.

Code, §§ 1780, 1782.) Upon petition by the Department of Youth Authority (now called

the Division of Juvenile Facilities (DJF)), a minor committed under Section 602, who is

about to reach the maximum age he can be held at DJF and who has not yet been

confined for the maximum term attributable to his offenses, may be held for the

remainder of the term applicable to the crimes of which he or she was convicted if the

petitioner can show “that unrestrained freedom for that person would be dangerous to the

public.” (Welf. & Inst. Code, § 1780; see generally, Chaparro v. Superior Court (1990)

218 Cal.App.3d 560, 562–568 (Chaparro).) In that situation, the inmate is entitled to a

judicial hearing in which he or she has rights to presence, appointed counsel, compulsory

process, and the right to produce evidence. (Welf. & Inst. Code, § 1781.)

Similarly, the prosecuting attorney may petition for extended detention if the

inmate “would be physically dangerous to the public because of the person’s mental or

physical deficiency, disorder, or abnormality that causes the person to have serious

difficulty controlling his or her dangerous behavior . . . .” (Welf. & Inst. Code, § 1800;

see Chaparro, supra, 218 Cal.App.3d at pp. 565, 568.) The inmate is entitled to a jury

trial on whether he or she represents a physical danger to the public. (Welf. & Inst. Code,

§ 1801.5.) In such circumstances, a juvenile offender over age 21 may be transferred to

the Department of Corrections and Rehabilitation for appropriate placement in order to

43

The juvenile court maintains jurisdiction over minors committed to the Division

of Juvenile Facilities until age 25 at the oldest, or age 23 for more recent commitments.

If the minor had not yet served a period of two years in custody, the commitment may be

extended until he or she has completed a “two-year period of control.” (Welf. & Inst.

Code, §§ 1769, subds. (b) & (c), 1771, subd. (b).)

65

protect other wards in the DJF. (Welf. & Inst. Code, § 1802.) Renewed petitions may be

filed every two years to keep the inmate in custody for two-year intervals. (Ibid.)

Thus, if the juvenile court were to retain jurisdiction after the fitness hearing, it

would not necessarily result in an early release from custody for Cervantes, but it would

give him a far greater opportunity for periodic inquiries into the necessity of his

continued incarceration. Thus, a juvenile disposition must realistically be considered a

more advantageous disposition than a criminal sentence for the same offense. And giving

Cervantes a fitness hearing upon remand, while not required under Estrada, removes one

obstacle to his obtaining an earlier discharge from custody.

We conclude the voters who approved Prop 57 intended a juvenile felon to have a

fitness hearing before being sentenced in adult criminal court. We arrive at that

conclusion based on the purposes underlying Prop 57, discussed above, and on the fact

that, for many juvenile felons, including many whose crimes were less serious than

Cervantes’s, the time in custody they face in criminal court is far longer than their likely

term of confinement in juvenile court. Indeed, it is the vastly more serious consequences

of trial in criminal court that makes the choice of tribunal so important.

That choice remains equally significant in the case of a retrial or resentencing.

Cervantes has already been tried in adult court under direct filing and has been sentenced

to literally a lifetime in prison. Though we shall reverse Cervantes’s sentence on

constitutional grounds, he faces almost equally severe consequences over again on

remand, even without a retrial. It is not primarily the prospect of enduring another guilt

determination that adversely affects Cervantes, however. Indeed, in some ways a trial in

adult court offers the juvenile

This text is long and has been trimmed here. Open the source document for the complete record.