Johnson v. Super. Ct.

• Court: California Court of Appeal
• Decided: October 27, 2016
• Docket: B266421
• Status: Published

Opinion

Filed 10/27/16 CERTIFIED FOR PUBLICATION IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA

SECOND APPELLATE DISTRICT

DIVISION THREE

CLEAMON DEMONE B266421 JOHNSON, (Los Angeles County Petitioner; Super. Ct. No. BA424006) v. DEATH PENALTY CASE THE SUPERIOR COURT OF LOS ANGELES COUNTY, Respondent;

THE PEOPLE,

Real Party in Interest.

ORIGINAL PROCEEDINGS in mandate. Sam O. Ohta, Judge. Petition granted in part, denied in part, and remanded with directions. Sanger Swysen & Dunkle, Robert M. Sanger and Stephen K. Dunkle, for Petitioner. No appearance for Respondent. Jackie Lacey, District Attorney, Phyllis C. Asayama and John Harlan II, Deputy District Attorneys, for Real Party in Interest. _______________________________________ INTRODUCTION

In 1997, Cleamon Demone Johnson was convicted of the first-degree murders of Peyton Beroit and Donald Ray Loggins, with multiple-murder special-circumstance findings as to both. The jury returned a verdict of death, which the trial court imposed. In 2011, the California Supreme Court reversed Johnson’s convictions and remanded for retrial. Before the second trial, the People investigated other murder and attempted murder cases from the early 1990s in which Johnson had been a suspect. Ultimately, they added four new charges to the case pending against him—for the capital murders of Albert Sutton, Georgia Denise (“Nece”) Jones, and Tyrone Mosley, and the attempted murder of Kim Coleman—all of which occurred in the early 1990s, and for two of which the People had previously tried and failed to convict Johnson. Where Johnson had been convicted of two capital-murder charges before his successful appeal, he now faces five capital-murder charges and an attempted murder charge—plus newly added gang enhancements—in the same case. Johnson challenged the filing of the new charges in a motion to dismiss for vindictive prosecution. Although the trial court agreed that Johnson made a prima facie showing sufficient to raise a presumption of vindictiveness, it determined the People rebutted the presumption, and denied the motion to dismiss. Johnson petitioned this court for writ of mandate/prohibition and we issued an order to show cause. We hold that the court erred in denying Johnson’s motion to dismiss the new charges concerning Jones, Mosley, and Coleman. We therefore grant Johnson’s petition as to those crimes and direct the court to dismiss counts 2, 5, 6, and their

2 related enhancements. We conclude that the court properly denied Johnson’s motion to dismiss the Sutton charges, however, and deny Johnson’s petition as to that crime. As for the gang enhancements added to the Beroit and Loggins counts, we remand for an evidentiary hearing to allow the People to present evidence to rebut the presumption of vindictiveness.

FACTUAL AND PROCEDURAL BACKGROUND

This case concerns six casualties of the gang wars between the Bloods and the Crips in the early 1990s. Johnson was a high-ranking member of the 89 Family Swans, a small, Bloods-affiliated gang.1 On August 5, 1991, Johnson told fellow gang member Michael Allen to shoot Peyton Beroit, a member of a rival Crips gang who was getting his car washed in 89 Family territory. Witnesses testified that Allen shot Beroit and Donald Ray Loggins as they sat in a parked car. On September 14, 1991, Freddie Jelks, another 89 Family member, alerted his colleagues that the rival 97 East Coast Crips were having a party nearby. Johnson, Jelks, and another member of their gang drove to the party and shot at the group. Tyrone Mosley was killed. Kim Coleman and Kenneth Davis were injured but survived. On September 12, 1992, Albert Sutton, a drug dealer and member of the 89 Family Bloods, took his brother to Johnson’s house. Sutton’s brother was a member of a rival Crips gang. A shootout ensued, and Sutton’s brother was shot; he survived but lost the use of one eye. The police arrested three men,

1 The gang is also referred to as 89 Family and 89 Family Bloods.

3 including Johnson. Sutton spoke to the police about the shooting, and Johnson was subsequently charged with attempted murder. Sutton planned to testify against Johnson at trial. When it became clear Sutton could not be dissuaded from testifying, Johnson ordered his cousin Leon Johnson (Leon) to kill Sutton. On September 16, 1992, Leon fatally shot Sutton in the back of the head. In 1994, the LAPD and FBI formed a joint task force to investigate the 89 Family. That spring, Charles Lafayette, a member of an allied gang, was brought to trial for the 1993 murder of Willie Bogan. Nece Jones testified that she saw Lafayette shoot and kill Bogan. On June 6, 1994, the case ended in a mistrial, and a second trial date was set. Two days later, members of the task force went to Ironwood State Prison, where they spent two hours interviewing Johnson. Detectives told Johnson they had formed a task force to investigate crimes committed by the 89 Family. They asked him “about murders that the LAPD was investigating[,]” and specifically asked about the Sutton killing. About an hour after detectives left the prison, Johnson called Reco Wilson and explained the task force investigation. He told Wilson to “clean up” and to “lock everything down around there.” Johnson continued, “[T]hem three smokers out there? Put a leash around their ass. By any means necessary.” “Smoker” is street slang for someone who smokes rock cocaine. Jones was such a person. The prosecution argued Johnson’s statement was a command to Wilson to kill Jones—and indeed, Jones was murdered one week later. The task force investigation bore fruit in the late 1990s when, over the course of four trials, the People prosecuted

4 a series of defendants for the crimes now at issue. In January 1997, Wilson was convicted of murdering Jones and sentenced to life in prison without the possibility of parole; the People’s theory was that Wilson killed her on Johnson’s orders.2 In September 1997, Johnson and co-defendant Allen were convicted of murdering Beroit and Loggins and were sentenced to death; the People’s theory was that Allen killed them on Johnson’s orders.3 In June 1998, after the jury was unable to reach a verdict, a mistrial was declared in Leon’s trial for the Sutton murder; Leon later pled guilty and was sentenced to 18 years to life. The People’s theory was that Leon killed Sutton on Johnson’s orders. Finally, in September 1999, another hung jury led to a mistrial in Johnson’s trial for the drive-by murder of Mosley and attempted murders of Coleman and Davis. In February 2000, the People dismissed those charges under Penal Code section 1382 (failure to proceed within the statutory period).4 In 2011, after an automatic appeal, the California Supreme Court reversed Johnson’s and Allen’s guilt and penalty judgments for the Beroit and Loggins murders, and remanded for retrial of both defendants. (People v. Allen and Johnson (2011) 53 Cal.4th 60, 79 [reversal]; § 1262 [reversal deemed order for

2 Our colleagues in Division Five affirmed Wilson’s conviction by unpublished opinion in People v. Wilson (Apr. 19, 1999, B111522 [nonpub. opn.]). On our own motion, we take judicial notice of that opinion. (Evid. Code, § 452, subd. (d)(1).) 3 Although Allen joined the vindictive-prosecution motion as to the new gang enhancements, he is not a party to this writ proceeding. 4 All undesignated statutory references are to the Penal Code.

5 new trial].) The People immediately began investigating other cases from the early 1990s in which Johnson had been a suspect. On March 2, 2012, the People notified Johnson of their intent to offer evidence of the Sutton, Jones, Mosley, and Coleman shootings at retrial. (Evid. Code, § 1101, subd. (b).) At some point, Johnson offered to plead guilty to the Loggins and Beroit murders, but the People declined to extend an offer of life in prison without the possibility of parole. In April 2014, Johnson moved to dismiss the still-pending Beroit and Loggins charges for outrageous government misconduct. The day the motion was set to be heard, the People moved to dismiss the original indictment in the Beroit/Loggins case (No. BA105846) and re-file the charges by criminal complaint under a new case number (No. BA424006). The court granted the motion over defense objection, and on April 25, 2014, the People filed a new complaint alleging five counts of capital murder (Loggins, Beroit, Sutton, Jones, and Mosley) and one count of attempted murder (Coleman). In addition to the special circumstances, the People also alleged various firearm and great bodily injury enhancements to four of the counts, and alleged a gang enhancement to all counts. Johnson was held to answer on the new charges, and the information in the new case (No. BA424006) was filed on May 13, 2014. The new information charged Johnson with five counts of premeditated murder (§ 187, subd. (a)) for the deaths of Sutton (count 1), Jones (count 2), Loggins (count 3), Beroit (count 4), and Mosley (count 5), with multiple-murder special-circumstance (§ 190.2, subd. (a)(3)) and street gang (§ 186.22, subd. (b))

6 allegations for each count.5 As to count 4, the information also alleged Johnson had furnished a firearm for the purpose of aiding or abetting a felony (§ 12022.4). As to count 5, the information alleged Johnson had personally used a firearm to commit the offense (§ 1203.06, subd. (a)(1); § 12022.5, subd. (a)). The information also charged Johnson with one count of attempted murder of Coleman (§ 664/187, subd. (a); count 6) and alleged great bodily injury (§ 12022.7, subd. (a)), personal use of a firearm (§ 1203.06, subd. (a)(1); § 12022.5, subd. (a)), and gang (§ 186.22, subd. (b)) enhancements for that count. Johnson argued the new charges constituted vindictive prosecution and moved to dismiss them. After a contested hearing, the court concluded Johnson made a prima facie showing sufficient to raise a presumption of vindictiveness. Additional briefing followed. After a second hearing, the court found the People had rebutted the presumption of vindictiveness and denied Johnson’s motion to dismiss. Johnson filed a timely petition for writ of mandate/prohibition in this court, and we issued an order to show

5 The complaint alleged the gang enhancements under section 186.22, subdivisions (b)(1)(C) (10-year enhancement) and (b)(5) (15-year minimum before parole). Subdivision (b)(1)(C) did not exist at the time of the last murder. (Prop. 21, § 4, as approved by voters, Primary Elec. (Mar. 7, 2000) [adding subd. (b)(1)(C)].) The gang enhancement in the subsequent information was alleged under section 186.22, subdivision (b). We asked for clarification by letter brief, and it appears Johnson has been charged with the enhancement under subdivision (b)(2) as it existed in 1991. (Stats. 1989, ch. 930, § 5.1; see People v. Valenti (2016) 243 Cal.App.4th 1140, 1172–1175 [discussing Ex Post Facto Clause].)

7 cause. We have original jurisdiction. (Cal. Const., art. VI, § 10 [original jurisdiction in mandate proceedings]; Code Civ. Proc., § 1085, subd. (a) [mandate]; Twiggs v. Superior Court (1983) 34 Cal.3d 360 (Twiggs) [petitioning for writ of prohibition/mandate proper to obtain pretrial review of denial of motion to dismiss for vindictive prosecution].)

ISSUES PRESENTED AND STANDARD OF REVIEW

In this writ proceeding, we are presented with the following issues: Does a defendant in a capital murder case raise a presumption of vindictiveness when, after his convictions are reversed and remanded for retrial but before that retrial can take place, the People add additional charges to the case pending against him? If so, have the People met their heavy burden of rebutting the presumption by showing that the newly added charges are justified by a change in circumstances or new evidence that legitimately influenced the charging process, and that they could not reasonably have discovered the new information before the defendant’s first trial? Because these issues present a mixed question of law and fact implicating an important constitutional right, our review is de novo. (People v. Cromer (2001) 24 Cal.4th 889, 893–903; In re Bower (1985) 38 Cal.3d 865, 872–873, 879 (Bower).) De novo review requires a two-part inquiry. First, we “determin[e] the historical facts[,]” which “will rarely be in dispute. When they are [in dispute, we] must, of course, apply a deferential standard of review to the trial court’s factual findings.” (People v. Cromer, supra, at p. 900.) Next, we apply “an objective, constitutionally based legal test to the historical facts.” (Ibid.) This is “a legal question given that the facts as found must support the conclusion reached. (Cf. People v. Superior Court (Day) (1985)

8 174 Cal.App.3d 1008, 1015 [noting distinction between a magistrate’s factual findings and a magistrate’s legal conclusion regarding the sufficiency of the evidence presented].)” (People v. Puentes (2010) 190 Cal.App.4th 1480, 1488.) We hold Johnson established a presumption of vindictiveness because the People charged him with additional crimes and enhancements in apparent response to his successful appeal. We also hold the People did not meet their heavy burden of rebutting the presumption of vindictive prosecution as to the Jones, Mosley, and Coleman charges because they did not establish how the purported new information or change in circumstances legitimately affected their charging decision. We conclude, however, that the People dispelled the presumption as to the Sutton murder because Leon’s testimony was unavailable in 1997 when the prosecution tried Johnson for the Beroit and Loggins murders.

DISCUSSION

“To punish a person because he has done what the law plainly allows him to do is a due process violation ‘of the most basic sort.’ [Citation.] In a series of cases beginning with North Carolina v. Pearce . . . , the [Supreme] Court has recognized this basic—and itself uncontroversial—principle. For while an individual certainly may be penalized for violating the law, he just as certainly may not be punished for exercising a protected statutory or constitutional right.” (United States v. Goodwin (1982) 457 U.S. 368, 372 (Goodwin).) Thus, the “due process clauses of the federal and state Constitutions (U.S. Const., 5th & 14th Amends.; Cal. Const., art. I, §§ 7, 15) forbid the prosecution from taking certain actions against a criminal defendant, such as increasing the charges, in

9 retaliation for the defendant’s exercise of constitutional rights.” (People v. Jurado (2006) 38 Cal.4th 72, 98.) The right to appeal is one such protected right. The Supreme Court has recognized the importance of protecting a criminal defendant’s right to appeal and has taken bold measures to preserve its unfettered exercise. (See, e.g., Douglas v. California (1963) 372 U.S. 353 [state must provide indigents with appointed counsel on their first appeal]; Griffin v. Illinois (1956) 351 U.S. 12 [state must supply trial transcripts to indigent appellants].) The California courts have gone further, holding a “ ‘defendant’s right of appeal from an erroneous judgment is unreasonably impaired when he is required to risk his life to invoke that right. Since the state has no interest in preserving erroneous judgments, it has no interest in foreclosing appeals therefrom by imposing unreasonable conditions on the right to appeal.’ ” (People v. Hanson (2000) 23 Cal.4th 355, 365.) In light of these principles, the vindictive-prosecution doctrine has developed as a prophylactic rule that “aims to free the defendant of the apprehension that the exercise of a right designed to guarantee that his or her trial is fair will be met with a retaliatory increase in the charge and potential period of incarceration to which he or she is subjected. [Citation.] Regardless of the actual motive of the individual prosecutor, a judicial process which permitted the prosecution to increase the charges against a defendant who successfully exercised a constitutional or procedural right at trial would have a chilling effect upon the assertion of those rights and could undermine the integrity of the entire proceeding.” (Bower, supra, 38 Cal.3d at pp. 877–878.)

10 To be sure, a defendant condemned to death in California cannot forego an appeal out of fear that his success will cause the State to increase the charges against him. (See People v. Massie (1998) 19 Cal.4th 550, 566; § 1239, subd. (b).) However, the vindictive-prosecution doctrine protects not only the defendant in an individual case, but also other defendants. (North Carolina v. Pearce (1969) 395 U.S. 711, 724–725.) Why? Because the imposition of more serious or additional charges in one case creates an apprehension among other defendants that they will be punished for exercising their right to appeal. (Id. at p. 724 [The State cannot “ ‘put a price on an appeal. A defendant’s exercise of a right of appeal must be free and unfettered.’ ”]; United States v. DeMarco (9th Cir. 1977) 550 F.2d 1224, 1227 [vindictive-prosecution doctrine designed to prevent chilling exercise of rights by other defendants making similar choice in the future].) This is a particular concern in capital cases, which tend to be highly publicized and rarely reversed.6 By guarding

6 Between 1987 and the end of 2015, the California Supreme Court reversed 11 guilt-phase verdicts on direct appeal—including Johnson’s and Allen’s—and affirmed 582, for a total reversal rate of 1.8% in death penalty cases. (See Cal. Comm’n on the Fair Admin. of Justice (CCFAJ), Report and Recommendations on the Administration of the Death Penalty in California (Jun. 30, 2008), p. 20, available at , pp. 120–121 [as of Oct. 25, 2016]; Cal. District Atty. Assn., Prosecutors’ Perspective on California’s Death Penalty (Mar. 2003), appen. A [as of Oct. 25, 2016]; Jones v. Chappell (C.D.Cal. 2014) 31 F.Supp.3d 1050, appen. A and related footnotes, reversed on other grounds by Jones v. Davis (9th Cir. 2015) 806 F.3d 538.) During the same period, 85 death-row inmates died awaiting execution—70 of them from natural causes.

11 against the fear of retaliation, the vindictive-prosecution doctrine works to safeguard the rights of the accused and promote the legitimacy of the courts as a check on the power of the State. To establish a presumption of vindictive prosecution, the defendant must show the State “increased the charges [against him] in apparent response to [his] exercise of a procedural right[.]” (Twiggs, supra, 34 Cal.3d at p. 371.)7 Then, the burden shifts to the prosecution to rebut this presumption by dispelling the appearance of vindictiveness. (Ibid.) To do so, the prosecution must show that new evidence or an objective change in circumstances legitimately influenced the charging decision and that they could not reasonably have discovered that

(See Cal. Dept. of Corrections and Rehabilitation, Condemned Inmates Who Have Died Since 1978 (Jun. 16, 2016) [as of Oct. 25, 2016].) 7 As we explain in detail in sections 1.2 and 1.3 post, the courts have framed the burden-shifting analysis in a variety of ways. Most recently, the California Supreme Court held that a presumption of vindictiveness “arises when the prosecutor increases the criminal charge against a defendant under circumstances [that] are deemed to present a ‘reasonable likelihood of vindictiveness.’ ” (Bower, supra, 38 Cal.3d at p. 879.) After examining the competing approaches, we conclude that in California, circumstances “are deemed to present” such a reasonable likelihood if they would appear vindictive to other criminal defendants. Put another way, if the prosecution increases the charges under circumstances that would appear vindictive to other criminal defendants, it has necessarily done so “under circumstances [that] are deemed to present a ‘reasonable likelihood of vindictiveness.’ ” (Ibid.) We therefore use the phrases interchangeably to describe the showing a defendant must make to raise a presumption of vindictive prosecution.

12 information before the first trial. (Bower, supra, 38 Cal.3d at pp. 873, 879.) If the prosecution does not meet this “ ‘heavy burden,’ ” the court must dismiss the new charges. (Twiggs, supra, at p. 371; see Bower, supra, at pp. 878–879 [noting that “when the cases discuss the possibility of rebutting a presumption of vindictiveness they refer only to a situation in which the prosecuting authority can show that ‘it was impossible to proceed on the more serious charge at the outset . . . .’ [Citation.]”].) Here, the People readily acknowledge they added new charges for the Sutton, Jones, Mosley, and Coleman crimes because the California Supreme Court reversed Johnson’s convictions for the Beroit and Loggins murders. But they argue that as a matter of law, there can be no presumption of vindictiveness in this case because they did not increase the original charges against Johnson. This argument takes two forms. First, the People argue that the new charges stemmed from different conduct than that at issue in the original case, and the new charges do not amount to an increase unless they concern the same conduct as the original counts. Second, they argue that they have not increased the charges because the new charges do not subject Johnson to additional punishment—he was already facing the ultimate punishment. Therefore, the People contend, Johnson has not made a prima facie showing sufficient to raise a presumption of vindictiveness. The People then argue they rebutted any presumption of vindictive prosecution because “useful statements from witnesses became both legally and practically available” after Johnson’s successful appeal.

13 To determine whether Johnson raised a presumption of vindictiveness, we first consider the limits of prosecutors’ charging discretion at various stages in the proceedings. We conclude the presumption of vindictiveness is triggered most easily after a conviction is overturned on appeal and the matter is set for retrial. We next examine the presumption itself. We conclude the People can “increase the charges” against a criminal defendant by charging him with new crimes if they do so under circumstances that would appear vindictive to other criminal defendants. And although we agree with the People that both the potential for increased punishment and the connection between the original and new charges are relevant to whether their actions appear vindictive, we conclude these factors are not dispositive. Based on a review of all the facts, we hold it would appear to other defendants that the People charged Johnson with the additional crimes and enhancements in response to his successful appeal. Therefore, Johnson has presented sufficient evidence to raise a presumption of vindictive prosecution. Because Johnson met his initial burden, we also examine whether the People have justified the addition of four new charges and six new gang enhancements by some objective change in circumstances or in the state of the evidence that legitimately influenced their charging process. We also examine why the People waited until after Johnson’s appeal was granted before filing the new charges and gang enhancements against him. Because the People have neither explained the importance of their new evidence, nor discussed in any meaningful way why the new evidence or changed circumstances legitimately affected their charging decision, we hold they did not rebut the presumption of vindictiveness as to the Jones, Mosley, or

14 Coleman charges. We are satisfied, however, that Leon’s testimony is both newly available and sufficiently important to justify the People’s reevaluation of their previous decision not to charge Johnson with the Sutton murder. 1. Johnson Established a Presumption of Vindictiveness. 1.1 Prosecutorial Discretion Before and After Appeal California and federal cases place great emphasis on when during the criminal proceedings the prosecutor’s allegedly vindictive action occurs. (Bower, supra, 38 Cal.3d at pp. 874–877, 879; Goodwin, supra, 457 U.S. at p. 381.) Thus, before determining whether Johnson has met his initial burden, we discuss the nature and scope of prosecutorial discretion and the interplay between vindictive prosecution and the government’s charging discretion. “It is well settled that the prosecuting authorities, exercising executive functions, ordinarily have the sole discretion to determine whom to charge with public offenses and what charges to bring. [Citations.] This prosecutorial discretion to choose, for each particular case, the actual charges from among those potentially available arises from ‘ “the complex considerations necessary for the effective and efficient administration of law enforcement.” ’ [Citations.] The prosecution’s authority in this regard is founded, among other things, on the principle of separation of powers, and generally is not subject to supervision by the judicial branch. [Citations.]” (People v. Birks (1998) 19 Cal.4th 108, 134; see e.g., Wayte v. United States (1985) 470 U.S. 598, 607 [subject only to constitutional restraints, prosecutors retain broad discretion in

15 deciding whom to prosecute].) This “broad discretion” extends to “selecting the cases to be subject to a capital charge.” (People v. Lucas (1995) 12 Cal.4th 415, 477–478.) Inherent in the prosecution’s charging discretion is its power not to bring charges. That decision is itself “deemed [to be] a discretionary charging decision[]” (People v. Mancebo (2002) 27 Cal.4th 735, 749), and courts are generally powerless to compel a prosecutor to proceed in a case he believes does not warrant prosecution (Inmates of Attica Correctional Facility v. Rockefeller (2d Cir. 1973) 477 F.2d 375, 379–380). Before trial, the State’s charging discretion is at its height. “While preparing for trial, new information may be discovered, the significance of possessed information may be realized and the proper extent of prosecution will crystallize.” (Barajas v. Superior Court (1983) 149 Cal.App.3d 30, 34.) “In contrast, once a trial begins—and certainly by the time a conviction has been obtained—it is much more likely that the State has discovered and assessed all of the information against an accused and has made a determination, on the basis of that information, of the extent to which he should be prosecuted.” (Goodwin, supra, 457 U.S. at p. 381.) “Thus, a change in the charging decision made after an initial trial is completed is much more likely to be improperly motivated than is a pretrial decision.” (Ibid.) At that point, prosecutors’ charging discretion decreases and judicial scrutiny increases. In other words, judicial oversight of the State’s charging discretion reaches its apex after a conviction is overturned on appeal and the matter is set for retrial, especially in a capital case. (See Blackledge v. Perry (1974) 417 U.S. 21, 27–28 (Blackledge); Goodwin, supra, 457 U.S. at pp. 376–377; Bower, supra, 38 Cal.3d at p. 877.)

16 1.2 The People’s Proposed Test We now turn to the People’s contention that we should use a two-factor test to assess whether Johnson has presented evidence that they “increased the charges” against him sufficient to raise a presumption of vindictiveness. (See Twiggs, supra, 34 Cal.3d at p. 371.) “[I]n the post-appeal context,” the People argue, “a presumption of vindictive prosecution arises if, and only if: [¶] 1. The prosecutor’s ability to bring the added charge or allegation was previously barred by double jeopardy, or jeopardy-related principles, such as those found in Kellett v. Superior Court [citations;] [¶] 2. The added charge or allegation increases punishment above what was sought by the prosecutor previously. [Citations.]” In evaluating this argument, we note that courts have variously described the prosecution’s suspect actions as “ ‘upping the ante’ ” (Blackledge, supra, 417 U.S. at p. 28), “substituting a more serious charge for the original one” (ibid.), “bringing a more serious charge against [the defendant] prior to the trial de novo[]” (id. at p. 29), “increas[ing] the charges so that the defendant faces a sentence potentially more severe than the sentence he or she faced at the first trial[]” (People v. Ledesma (2006) 39 Cal.4th 641, 731), bringing “increased or additional charges” (Bower, supra, 38 Cal.3d at p. 872), “ ‘ “upping the ante” ’ with more serious charges or a potentially greater sentence” (People v. Puentes, supra, 190 Cal.App.4th at p. 1484, quoting People v. Bracey (1994) 21 Cal.App.4th 1532, 1543), an “increase in charges or a new prosecution” (People v. Valli (2010) 187 Cal.App.4th 786, 802 (Valli)), a “decision to increase the stakes for the accused” (United States v. Griffin (9th Cir. 1980) 617 F.2d 1342, 1347), and an “increase in the severity or number

17 of charges” (Hardwick v. Doolittle (5th Cir. 1977) 558 F.2d 292, 301). We also note that the prosecutor’s actual motives are immaterial. (Bower, supra, at pp. 877–879.) Instead, the presumption of vindictiveness “is a legal presumption which arises when the prosecutor increases the criminal charge against a defendant under circumstances [that] are deemed to present a ‘reasonable likelihood of vindictiveness.’ The presumption is not based on the subjective state of mind of the individual prosecutor and does not imply that he or she individually harbors an improper motive.” (Id. at p. 879.)8 The People’s proposed test is quite novel. While they suggest California precedent supports their approach, their cited authorities do not stand for the bright-line rule they advocate. Each case cited to us involves new charges brought in a new case after the original case was over. The defendants in those cases were charged when prosecutorial discretion was highest and the presumption of vindictiveness was not easily triggered. (See Goodwin, supra, 457 U.S. at p. 381.) Contrary to the People’s contention, none involves “the post-appeal context[.]” And no case cited to us involves the scenario Johnson faced—new charges added to the old ones after a successful appeal but before retrial, when prosecutors have the least discretion and the presumption applies most strongly.

8 Accordingly, the People’s statements denying any improper motive—and asserting the decision to file additional murder and attempted murder charges was not a reaction or response to Johnson’s successful appeal—are not relevant at this stage of the proceedings. For purposes of meeting his initial burden, Johnson does not allege that the People were motivated by actual vindictiveness or spite.

18 For example, the People cite Valli for the proposition that “[e]ven where evidence of the newly charged offense was presented as evidence of a consciousness of guilt at the original trial on the original charge, no presumption of vindictiveness is raised by the subsequent prosecution for that different conduct.” But Valli is inapplicable here. In Valli, minutes after the defendant was acquitted of murder, the District Attorney charged him with two counts of evading arrest. (Valli, supra, 187 Cal.App.4th at pp. 790–791.) Though substantially similar evidence had been introduced at the earlier murder trial to show consciousness of guilt, the defendant was brought to trial and convicted in the second case. (Id. at p. 790.) Valli addressed two discrete issues—whether joinder was mandatory under Kellett v. Superior Court, and whether the second prosecution was impermissibly vindictive.9 The court affirmed on both grounds. While the Valli court addressed cross-admissibility of evidence in its Kellett analysis, the issue was irrelevant to the vindictive- prosecution holding, and was not addressed in that part of the opinion. (Id. at pp. 794–802.) As discussed, a criminal defendant raises a presumption of vindictiveness where he shows (1) “the prosecution has increased

9 Under Kellett, when “the prosecution is or should be aware of more than one offense in which the same act or course of conduct plays a significant part, all such offenses must be prosecuted in a single proceeding unless joinder is prohibited or severance permitted for good cause. Failure to unite all such offenses will result in a bar to subsequent prosecution of any offense omitted if the initial proceedings culminate in either acquittal or conviction and sentence.” (Kellett v. Superior Court (1966) 63 Cal.2d 822, 827 (Kellett).)

19 the charges” (2) “in apparent response to” (3) “the defendant’s exercise of a procedural right[.]” (Twiggs, supra, 34 Cal.3d at p. 371, quoted in Valli, supra, 187 Cal.App.4th at p. 803.) The People’s arguments focus on the first factor—whether they increased the charges against Johnson. Valli, on the other hand, focused on the second factor. In Valli, the People increased the charges after the defendant exercised a procedural right—the right to testify—but the People did not increase the charges because the defendant testified. (Valli, supra, at pp. 803–805; see id. at p. 805 [concluding the People’s decision “was a response to the acquittal, not to defendant’s testifying at trial.”].) Accordingly, Valli stands only for the proposition that the filing of “new charges after an acquittal on separate charges do[es] not, without more, give rise to a presumption of vindictiveness.” (Id. at p. 805; see People v. Guevara (2004) 121 Cal.App.4th 17, 27 [“cases may not be used for propositions not considered”].)10 The People’s reliance on People v. Tirado is also inapt. (People v. Tirado (1984) 151 Cal.App.3d 341 (Tirado).) In Tirado, the defendant pled guilty to robbery and successfully argued for a mitigated sentence. (Id. at pp. 346–347.) In response, the

10 The People also emphasize Valli’s reliance on United States v. Esposito (3d Cir. 1992) 968 F.2d 300, 306 (Esposito). Like Valli, Esposito involved new charges brought in a new case after the defendant was acquitted in the original case. And like Valli, Esposito concluded the new charges were filed in response to an acquittal. Esposito placed particular emphasis on the fact that the second indictment was brought in a second case. (Id., at pp. 303–304 [where “the prosecutor has done nothing to deter the exercise of one’s right during the case or proceeding, and the prosecution has come to a natural end, no presumption of vindictiveness applies.”], 306.)

20 People charged the defendant in a new case with a second robbery. (Ibid.) The defendant argued the prosecution’s decision to file the new case was a vindictive response to his exercise of the right to file a statement in mitigation. (Id. at pp. 348–349.) The Tirado court emphasized that case’s procedural posture— pretrial plea-bargaining—which the courts have consistently held is not typically subject to a presumption of vindictiveness. (Id. at pp. 350–352.) Indeed, the court spent a full page quoting Goodwin on that point, then concluded it was not required to follow Twiggs, Blackledge, or Pearce, since those cases involved post-trial conduct. (Id. at pp. 349–350.) Tirado thus stands for the proposition that the presumption of vindictiveness does not apply to a prosecutor’s pretrial charging decisions. (See People v. Bracey (1994) 21 Cal.App.4th 1532, 1544 [citing Tirado as holding presumption does not apply before jeopardy attaches].) Tirado does not support the rule the People advance in this case.11

11 The People stress Tirado’s reliance, in dicta, on United States v. Mallah, a 1974 case from the Second Circuit that predates Twiggs and Blackledge. (United States v. Mallah (2d Cir. 1974) 503 F.2d 971, 988.) Since Mallah primarily concerns the Double Jeopardy Clause, only a handful of courts have cited it for the proposition at issue here—most recently in 1984. (See People v. Lucious (1984) 153 Cal.App.3d 416, 422–423 [prosecutor increased the charges in plea-bargaining context].) To the extent Mallah stands for the People’s proposed relatedness rule, we find that approach is irreconcilable with California Supreme Court precedent that is binding upon this court. (Auto Equity Sales, Inc. v. Superior Court (1962) 57 Cal.2d 450, 455; see section 1.3, post.)

21 In support of their proposed test, the People argue that California’s mandatory joinder and double-jeopardy rules conclusively determine whether a defendant has raised an initial presumption of vindictiveness. They contend the vindictive- prosecution doctrine applies when the new charges are for “the same conduct” as the original charges but not when “new charges involve different acts or a different course of conduct.” The People also refer to “charges arising from separate offenses not part of the same act or course of conduct.” With this framing, the People ask us to apply the Kellett rule and consider whether the new charges were sufficiently related to require mandatory joinder in the original case. (See Kellett, supra, 63 Cal.2d at p. 827.) The People suggest that if Kellett did not require them to join the charges in the first instance, it cannot appear vindictive to add them to the case after a successful appeal. The People also attempt to equate vindictive prosecution with double jeopardy, which protects criminal defendants “from being consecutively charged with violation of the same law or violation of laws so related that conduct prohibited by one statute is necessarily included within conduct prohibited by the other.” (People v. Spicer (2015) 235 Cal.App.4th 1359, 1371; see People v. Hanson, supra, 23 Cal.4th at pp. 358–360, 363–367 [appellate reversal precludes more severe punishment after retrial].)12 In their final

12 California’s Double Jeopardy Clause (Cal. Const., art. I, § 15) is considerably broader than its federal counterpart (U.S. Const., 5th Amend.). (People v. Hanson, supra, 23 Cal.4th at pp. 358–360, 363– 367.) The People do not explain, however, whether they would like us to apply the federal constitution or the state constitution and instead refer only to “double jeopardy, or jeopardy-related principles[.]”

22 brief, the People combine these approaches and contend the relevant inquiry is whether the “prosecutor’s ability to bring the added charge or allegation was previously barred by double jeopardy, or jeopardy-related principles, such as those found in Kellett[.]” Certainly, it would appear vindictive for the prosecution, after a defendant’s successful appeal, to bring new charges that run afoul of Kellett or the Double Jeopardy Clause—but we disagree with the People that the opposite is also true. That is, while Kellett and double jeopardy may well be relevant to the ultimate burden-shifting analysis, they are not dispositive. First, because the defendant bears the initial burden of proof in a claim of vindictive prosecution, the People’s proposed test would effectively require the defendant, at an early stage of the proceedings, to prove the charged crimes are related. Obviously, since the prosecution has the burden of proof at trial, the defendant will not know what specific evidence the People will offer to support the charges. Yet under this rule, defendants would have to address the cross-admissibility of evidence and the extent to which the new and original charges require “separate proofs” simply to meet their initial burden. (See Valli, supra, 187 Cal.App.4th at pp. 797–802.) Such a rule would unfairly burden defendants by requiring them to produce evidence they may not have and to make arguments about the relatedness of criminal charges that may be adverse to their interests. In other words, to raise a vindictive prosecution claim, defendants would have to help the State prosecute them. Second, although the vindictive-prosecution doctrine stems from the Due Process Clause, which is concerned with “fundamental fairness,” the People’s approach would require

23 courts to evaluate vindictive-prosecution claims using narrower rules like the statutory rule of mandatory joinder and the Double Jeopardy Clause. (See 2 Modern Constitutional Law (3rd ed.) § 30:1 [discussing due process protections for criminal defendants].) “However, due process has always been understood to encompass more than the sum total of the other related constitutional guarantees. It promises fundamental fairness in the criminal justice process, whether or not another clause in the Constitution also addresses the question before the courts.” (Ibid.; see Rochin v. California (1952) 342 U.S. 165, 173 [“Due process of law, as a historic and generative principle, precludes defining, and thereby confining, these standards of conduct more precisely than to say that convictions cannot be brought about by methods that offend ‘a sense of justice.’ ”].) In short, we can think of no sound reason to limit application of the vindictive-prosecution doctrine to circumstances in which the newly added charges would be barred by Kellett or double jeopardy principles. 1.3 The Presumption of Vindictiveness and Increased Charges Having rejected the People’s attempt to limit the vindictive- prosecution doctrine to circumstances in which the newly added charges would be barred by Kellett or double jeopardy principles, we turn to their contention that as long as the new “charges involve different acts or a different course of conduct,” they have not increased the charges against the defendant. We have not found any published California case addressing the specific question and unique facts before us: Can a criminal defendant raise a presumption of vindictiveness where the “increased charges” are for different conduct than the original charges, but

24 the People have had most of the evidence underlying the newly added charges for years and only added them to the defendant’s pending retrial following a successful appeal? To evaluate the People’s contention notwithstanding the lack of California authority to support it, we turn to the federal courts for guidance. In doing so, we must reconcile the prosecutor’s broad discretion to file charges when there is probable cause to believe that someone has committed a crime with our duty to protect a criminal defendant’s unfettered right to appeal. As we shall explain, we hold that the People “increase the charges” against a criminal defendant when they bring new charges against him under circumstances that would appear vindictive to other criminal defendants—and those new charges may involve different facts, crimes, or victims. 1.3.1 Appearance of vindictiveness versus realistic likelihood of actual vindictiveness The presumption of vindictiveness “is a legal presumption which arises when the prosecutor increases the criminal charge against a defendant under circumstances [that] are deemed to present a ‘reasonable likelihood of vindictiveness.’ ” (Bower, supra, 38 Cal.3d at p. 879.) But what circumstances should be “deemed to present” such a likelihood and thereby warrant application of the presumption? To answer that question, the federal courts have adopted a variety of competing secondary tests. While these tests can appear ad-hoc, they can be grouped into two broad categories. One set of courts has emphasized the need to avoid an appearance of vindictiveness that may deter future defendants from exercising their rights. A second set of courts has emphasized the need to protect individual defendants from actual retaliatory conduct.

25 Courts in the First, Third, Fifth, Sixth, and Ninth Circuits belong to the first group and apply a presumption of vindictiveness when the prosecution’s actions are likely to have a chilling effect on other defendants. (See, e.g., United States v. DeMarco, supra, 550 F.2d at p. 1227; United States v. Krezdorn (5th Cir. 1983) 718 F.2d 1360, 1364–1365 (en banc); United States v. Andrews (6th Cir. 1980) 633 F.2d 449, 453–454 (en banc) [examining the prosecutor’s actions and stake in deterrence to determine whether a reasonable person would find a realistic likelihood of vindictiveness]; United States v. Schoolcraft (3d Cir. 1989) 879 F.2d 64, 68 [“The defendant bears the initial burden of proof in a vindictive prosecution claim and is required to establish the appearance of vindictiveness.”]; United States v. Young (1st Cir. 1992) 955 F.2d 99, 108 [following Krezdorn]; Lovett v. Butterworth (1st Cir. 1979) 610 F.2d 1002, 1005–1006 [citing DeMarco and Andrews, discussing prophylactic nature of the doctrine, and emphasizing irrelevance of prosecutors’ subjective motivations].) Because those courts are most concerned with the appearance of vindictiveness, they do not typically consider the subjective motivations driving the State’s actions. For example, in the Fifth Circuit, courts assess whether a defendant has presented evidence sufficient to establish a presumption of vindictiveness by examining the prosecutor’s actions in the context of the entire proceedings and asking whether “any objective event or combination of events in those proceedings should indicate to a reasonable minded defendant that the prosecutor’s decision to increase the severity of charges was motivated by some purpose other than a vindictive desire to deter or punish appeals[.]” (United States v. Krezdorn, supra, 718 F.2d at pp. 1364–1365.)

26 Courts in the Second, Fourth, Seventh, Eighth, Tenth, and District of Columbia Circuits, on the other hand, belong to the second group, and require defendants to prove a reasonable likelihood of actual vindictiveness. (See, e.g., United States v. King (2d Cir. 1997) 126 F.3d 394, 397; United States v. Wilson (4th Cir. 2001) 262 F.3d 305, 314–315; United States v. Falcon (7th Cir. 2003) 347 F.3d 1000, 1004 [to obtain evidentiary hearing on vindictive prosecution, defendant must offer sufficient evidence to raise a reasonable doubt that the government acted properly]; United States v. Chappell (8th Cir. 2015) 779 F.3d 872, 879–881 [“ ‘a defendant may, in rare instances, rely upon a presumption of vindictiveness,’ (citation)” if he establishes a reasonable likelihood of actual vindictiveness]; United States v. Raymer (10th Cir. 1991) 941 F.2d 1031, 1042 [court must determine whether “ ‘there is a realistic or reasonable likelihood of prosecutorial conduct that would not have occurred but for hostility or punitive animus towards the defendant because he exercised his specific legal right.’ ”]; United States v. Safavian (D.C. Cir. 2011) 649 F.3d 688, 692 [to establish a presumption of vindictiveness, defendant must show prosecutor’s actions were “more likely than not” attributable to actual vindictiveness].) Those courts therefore focus on whether the prosecutor harbored genuine animus toward the defendant. For example, to establish a presumption of vindictiveness in the Fourth Circuit, the defendant “must show that the circumstances ‘pose a realistic likelihood of [actual] vindictiveness.’ ” (United States v. Wilson, supra, 262 F.3d at pp. 314–315.) This is a “rigorous” standard that requires defendants to overcome “a significant barrier.” (Ibid.) That is, the showing must be “sufficiently strong to overcome the presumption of prosecutorial regularity.” (Ibid.)

27 Finally, courts in the 11th Circuit take a hybrid approach. They apply a presumption of vindictiveness when the State substitutes more serious charges for the original charges concerning the same conduct, but require the defendant to prove actual vindictiveness when the prosecution adds new and separate charges. (United States v. Jones (11th Cir. 2010) 601 F.3d 1247, 1260–1261 & fn. 5; United States v. Kendrick (11th Cir. 2012) 682 F.3d 974, 981–982.) Like other courts in the first group, the Ninth Circuit has focused on the concern expressed in Blackledge and Pearce for alleviating defendants’ apprehension that the government will retaliate against them if they exercise their legal rights. Thus, that court has concluded that the mere appearance of vindictive prosecutorial behavior offends due process. (United States v. Ruesga-Martinez (9th Cir. 1976) 534 F.2d 1367, 1369 [“Pearce and Blackledge seek to reduce or eliminate apprehension on the part of an accused that he may be subjected to retaliatory or vindictive punishment by the prosecution only for attempting to exercise his procedural rights. Hence, the mere appearance of vindictiveness is enough to place the burden on the prosecution.”].) As the court explained in United States v. DeMarco, “[i]t is irrelevant that a particular defendant exercises his statutory rights, despite his fear of vindictiveness and despite the lack of vindictiveness in fact in subsequent proceedings instituted by the prosecutor. The prophylactic rule is designed not only to relieve the defendant who has asserted his right from bearing the burden from ‘upping the ante’ but also to prevent chilling the exercise of such rights by other defendants who must make their choices under similar circumstances in the future.” (United States v. DeMarco, supra, 550 F.2d at p. 1227; see

28 United States v. Griffin, supra, 617 F.2d at p. 1347 [“It is now well established that the mere appearance of vindictiveness may give rise to a presumption of a vindictive motive sufficient to establish a due process violation.”].) In United States v. Jenkins, the Ninth Circuit relied on this reasoning when it rejected the premise that the vindictive- prosecution doctrine applies only when the old and new charges arise from the same nucleus of operative fact. (United States v. Jenkins (9th Cir. 2007) 504 F.3d 694, 700–701.) In that case, the defendant was “apprehended twice for attempting to cross the U.S.-Mexico border while driving a vehicle containing undocumented aliens. Both times, Jenkins stated that she had been paid to drive the car across the border. She was not charged with any crime. Almost three months later, Jenkins was apprehended while attempting to cross the border as a passenger in a vehicle containing marijuana. She stated that she had been paid to drive the car, which she believed contained illegal aliens, across the border. Jenkins was charged with importation of marijuana. At trial, she testified in her own defense and maintained that she believed the vehicle in which she had been a passenger contained illegal aliens because she had been paid on two previous occasions to smuggle aliens. While the jury was deliberating, the government filed alien smuggling charges against Jenkins [in a separate case] in connection with her first two border apprehensions. [¶] The district court found that the prosecutor’s conduct created the appearance of vindictive prosecution because the alien smuggling charges were brought only after Jenkins exercised her right to testify in her own defense at her separate marijuana smuggling trial.” (Id. at p. 697.) The Ninth Circuit affirmed. The court concluded that

29 because the government exercised its discretion not to prosecute Jenkins for alien smuggling until “she presented her theory of defense at the marijuana smuggling trial, the timing of the charges created the appearance of vindictiveness.” (Ibid.) Like the federal circuits in the first group, California courts are primarily concerned with the prophylactic nature of the vindictive-prosecution doctrine. Accordingly, California courts have adopted the approach favored by these courts—an appearance-of-vindictiveness test. For example, the Twiggs Court framed the presumption of vindictiveness as relating to the appearance of vindictiveness and emphasized that “the principles discussed in this opinion are designed to relieve the defendant of the ‘apprehension of vindictiveness.’ ” (Twiggs, supra, 34 Cal.3d at p. 374; see id. at pp. 369–370, 371, 374.) Likewise, in Bower, the Court relied on the Ninth Circuit’s rule that “ ‘the mere appearance of vindictiveness is enough to place the burden on the prosecution.’ ” (Bower, supra, 38 Cal.3d at p. 878; id. at pp. 873-874, 877 [actual vindictiveness is irrelevant]; see also People v. Puentes, supra, 190 Cal.App.4th at p. 1486 [the People’s actions gave “the appearance that defendant’s successful appeal changed the People’s mind about what charges were ‘in furtherance of justice.’ ”]; Tirado, supra, 151 Cal.App.3d at p. 350 [“The rationale supporting the Blackledge-Pearce-Twiggs rule is to reduce or eliminate the defendant’s apprehension he may be subjected to unilateral retaliation or vindictive punishment for attempting to exercise his procedural rights.”]; Barajas v. Superior Court, supra, 149 Cal.App.3d at pp. 33–34 [noting the holding in Twiggs “was made without regard for whether actual retaliatory motivation by the prosecutor existed on the rationale that the threat of such action deprives a defendant of due

30 process.”]; In re David B. (1977) 68 Cal.App.3d 931, 934–936 [discussing chilling effect]; People v. Welch (1993) 5 Cal.4th 228, 239–240 (conc. opn. of Arabian and Kennard, JJ.) [“any appearance of vindictiveness in this context is subject to scrutiny; and reviewing courts should not be reluctant to fashion appropriate remedies when warranted.”].) Even if California courts and the Ninth Circuit had not adopted an appearance of vindictiveness approach, however, we would not require a defendant seeking to raise a presumption of vindictiveness to demonstrate a reasonable likelihood that the prosecutor harbored actual animus toward him. First, such a requirement provides inadequate due process protection by straying too far from Blackledge’s concerns about the chilling effect on other criminal defendants. (Blackledge, supra, 417 U.S. at p. 28 [“The rationale of our judgment in the Pearce case . . . was not grounded upon the proposition that actual retaliatory motivation must inevitably exist. Rather, we emphasized that ‘since the fear of such vindictiveness may unconstitutionally deter a defendant’s exercise of the right to appeal or collaterally attack his first conviction, due process also requires that a defendant be freed of apprehension of such a retaliatory motivation . . . .’ ”].) Second, an emphasis on actual vindictiveness places courts in the untenable position of policing prosecutors’ subjective motivations. If defendants must prove a realistic likelihood of actual retaliatory motivation, a court must explicitly find prosecutorial bad faith before it can dismiss the improper charges. As explained in United States v. Andrews, supra, 633 F.2d at pp. 454–455, under that test, “a trial judge would have the Hobson’s choice of either not barring the extra charge or of saying that a prosecutor acted wrongly. In some

31 cases, a trial judge would, in effect, be calling a prosecutor a liar where the prosecutor claimed inadvertence and the judge ruled against him. We do not think that such confrontations before the judiciary and the executive branch are desirable.” We therefore hold that one way the People “increase the charges” is by bringing new charges against a defendant under circumstances that appear vindictive. As we discuss below, whether the circumstances appear vindictive is assessed by reviewing all the facts—even if the new charges stem from different events, conduct, or victims than those in the original case. (See United States v. Krezdorn, supra, 718 F.2d at pp. 1364–1365.) Our holding accounts for the concerns emphasized by the California Supreme Court and allows the courts to dismiss charges in appropriate situations without the need to find the prosecutor acted in bad faith. 1.3.2 Relatedness and the Totality of the Circumstances Having decided the People “increase the charges” when they bring new charges against a defendant under circumstances that would appear vindictive to other criminal defendants, we next examine whether the new charges must relate to the same criminal conduct at issue in the original trial. Since it is the appearance of vindictiveness that triggers the presumption, the People’s argument comes down to this: It can never appear vindictive for the prosecution to add new counts, for decades-old crimes, to a defendant’s pending case when that defendant is retried after exercising his right to appeal. Because the appearance-of-vindictiveness standard is incompatible with such a bright-line rule, we reject the People’s proposal. Instead, under the totality-of-the-circumstances approach favored by the Ninth

32 Circuit and the California Supreme Court, we hold that the addition of new charges based on different facts, conduct, and victims does not preclude a finding of vindictiveness. Federal courts that apply an appearance-of-vindictiveness test typically perform a fact-intensive analysis based on the totality of the circumstances. For example, the Fifth Circuit has explained that the “judicial history of decisions involving judicial and prosecutorial vindictiveness is now clear enough to teach that it is a mistake to measure cases in this area of the law against fixed gauges. The proper solution is not to be found by classifying prosecutorial decisions as changing or adding charges, as amending decisions already made, as covering the same basic conduct or spree of activity, or as being made pre- or post-trial. Nor is it determinative whether the procedural matrix is appeal and error or trial de novo. It is also unnecessary to seek to strike the delicate balance between the rights of defendant and prosecutor. The surer solution lies in applying a more familiar, less exact test. “If the defendant challenges as vindictive a prosecutorial decision to increase the number or severity of charges following a successful appeal, the court must examine the prosecutor’s actions in the context of the entire proceedings. If any objective event or combination of events in those proceedings should indicate to a reasonable minded defendant that the prosecutor’s decision to increase the severity of charges was motivated by some purpose other than a vindictive desire to deter or punish appeals, no presumption of vindictiveness is created.” (United States v. Krezdorn, supra, 718 F.2d at pp. 1364–1365.) The Ninth Circuit uses a similarly fact-intensive approach, which the California Supreme Court has implicitly adopted.

33 (Twiggs, supra, 34 Cal.3d at p. 371 [“The conclusion that a presumption of vindictiveness arose in this case is consistent with the rule developed in cases in the Ninth Circuit Court of Appeals.”].) For example, Twiggs relies on United States v. Groves, in which the Ninth Circuit concluded that the fact the government brought “two separate and distinct cases” involving “different crimes relating to completely separate fact situations” was not “controlling in any case” or “dispositive on the question of vindictiveness” (United States v. Groves (9th Cir. 1978) 571 F.2d 450, 453–454 (Groves); see Twiggs, supra, at p. 371.) Instead, Groves emphasized the totality of the circumstances—namely what the government knew, when they knew it, and when they decided to bring the new charges. (Groves, supra, at pp. 453– 454.) Twiggs also relied on United States v. Ruesga-Martinez, in which the Ninth Circuit emphasized the appearance of vindictiveness and held “that when the prosecution has occasion to reindict the accused because the accused has exercised some procedural right, the prosecution bears a heavy burden of proving that any increase in the severity of the alleged charges was not motivated by a vindictive motive.” (United States v. Ruesga- Martinez, supra, 534 F.2d at p. 1369, emphasis added; see Twiggs, at p. 371.)13

13 In United States v. Martinez, the Ninth Circuit appeared to adopt a bright-line rule that if “the second charge is unrelated to the first, the presumption does not arise.” (United States v. Martinez (9th Cir. 1986) 785 F.2d 663, 669 (Martinez).) In support of that statement, the court cited to United States v. Robison (9th Cir. 1981) 644 F.2d 1270, 1272 (Robison). But Robison held just the opposite. It concluded that the fact the “instant prosecution arose from events separate and distinct from those on which the earlier prosecutions

34 Indeed, Twiggs considered the totality of the circumstances throughout its burden-shifting analysis. (Twiggs, supra, 34 Cal.3d at pp. 371–372, 374.) In Twiggs, the trial court declared a mistrial after the jury was unable to reach a verdict. Rather than accept a plea bargain, the defendant exercised his right to a retrial. In response, the prosecutor amended the information to allege five additional prior-felony convictions. (Id. at p. 368.) The government argued the newly-charged status enhancements were not vindictive because they were “wholly unrelated to the underlying charges[.]” (Id. at p. 376 [summarizing government’s argument].) The Court rejected that argument and concluded the circumstances raised a presumption of vindictiveness. Prosecutors knew about the defendant’s prior convictions before the first trial, yet made no effort to verify

were based[]” was neither dispositive nor essential. (Robison, at p. 1272.) The Ninth Circuit noted this discrepancy in Jenkins, and explained that the defendants in both Robison and Martinez had “failed to demonstrate any connection between the exercise of procedural rights in prior prosecutions and the federal prosecution challenged” in those cases. (Jenkins, supra, 504 F.3d at p. 701.) In Robison, the second prosecution was initiated by a different sovereign; in Martinez, the new charges were initiated by a different sovereign after an acquittal. Thus, Jenkins explained, “we do not read Martinez as holding that a presumption of vindictiveness can never arise when the second charge is unrelated to the first.” (Jenkins, supra, at p. 701.) As Robison acknowledged, it is merely one of the factors. (See Jenkins, at p. 701 [“The government itself recognizes that it brought the alien smuggling charges only because Jenkins admitted to them during the marijuana importation trial. Therefore, to the extent that we consider the relatedness of the charges important to our analysis, this factor does not foreclose application of the doctrine of vindictive prosecution.”].)

35 them; the government developed no new facts at trial that could have legitimately influenced their charging decision; and the “prosecution showed no interest in charging the additional prior convictions until the defendant insisted on a retrial[.]” (Id. at p. 372.) These circumstances, the Court concluded, “plainly gave rise to a presumption of vindictiveness.” (Ibid.; see also Barajas v. Superior Court, supra, 149 Cal.App.3d at pp. 34–35 [evaluating claim under totality of circumstances].) To be sure, some courts decline to apply the vindictive- prosecution doctrine when the prosecution brings new charges in response to an acquittal. (See, e.g., Esposito, supra, 968 F.2d at p. 306; Valli, supra, 187 Cal.App.4th at pp. 803–805.) The People’s proposal is considerably broader than those holdings, however. Their rule would allow the prosecution to respond to a defendant’s successful appeal by charging him, in the just- reversed case, with an unlimited number of new counts and enhancements, as long as the new charges are not currently barred by Kellett, were not previously barred by double jeopardy, and do not increase the length of his potential sentence. (See sections 1.2, ante and 1.4, post.) We also acknowledge that some federal courts in the second group—courts that attempt to discern the likelihood that a given prosecutor acted out of actual animus toward a given defendant— appear to use a bright-line rule analogous to the one the People urge us to adopt. For example, in Williams v. Bartow, the Seventh Circuit affirmed a Wisconsin state court ruling that the presumption of vindictiveness applies only when the prosecution increases the charges against a defendant for the same conduct. (Williams v. Bartow (7th Cir. 2007) 481 F.3d 492.) The Wisconsin court acknowledged the “distinct possibility” that a defendant

36 would be chilled from exercising his appellate rights if he believed a successful appeal would result in a second trial on more serious charges, but concluded that concern was not present when the prosecutor brings charges based on different conduct. (Id. at p. 501.) The court reasoned that the appeal does not create the opportunity to charge the defendant with additional crimes; the prosecutor may proceed on the separate charges whether or not the defendant appeals the original conviction. (Ibid.) In the end, Williams v. Bartow does not expressly hold that the presumption of vindictiveness cannot apply when the State brings new charges for different conduct; it does conclude, however, that when the Wisconsin state court reached that conclusion, it did not unreasonably apply established federal precedent. (Id. at p. 503 [“We cannot conclude that Blackledge and Thigpen [v. Roberts (1984) 468 U.S. 27] clearly establish a different rule than that applied by the Wisconsin court.”]; see Williams v. Taylor (2000) 529 U.S. 362, 404–406 [federal review of state habeas claims].) We decline to follow those federal courts that have categorically declined to apply the vindictive-prosecution rule to situations in which the defendant is charged, post-appeal, with different criminal conduct, as opposed to a heightened charge for the same conduct. Instead, in light of California’s and the Ninth Circuit’s fact-intensive approach based on the totality of the circumstances, we hold that the prosecution’s challenged action must be viewed in context, taking into account when the crimes underlying the new charges were committed, when the defendant invoked a statutory or constitutional right, the nature of the protected right, when the prosecution added the new charges, and the nature, number, and severity of the new charges. Our

37 approach is especially necessary in this case given the importance of a defendant’s right to appeal, the high-profile nature of capital cases, and the need to guard against other defendants’ fears of retaliation. While we agree with the People that the degree of overlap between the old and new charges is relevant to whether the prosecution can successfully rebut the presumption of vindictiveness, it is not dispositive at this stage, and a criminal defendant need not prove relatedness to raise the presumption in the first instance. In reaching our conclusion, we do not mean to suggest that any new charge following a successful appeal would raise a presumption of vindictiveness. For example, it would be difficult for a defendant to meet his initial burden if the new charge is for a crime committed in prison during or after his appeal. And we do not suggest—because the issue is not before us—that a presumption of vindictiveness would arise where the prosecution files new charges in a new case after a defendant is acquitted. (See Valli, supra, 187 Cal.App.4th at p. 805 [the filing of “new charges after an acquittal on separate charges do[es] not, without more,” raise a presumption of vindictiveness, even where the old and new charges are related]; Esposito, supra, 968 F.2d at pp. 303–304 [no presumption of vindictiveness where prosecutor brings new charges in a second case in response to an acquittal].) Nor is the presumption the end of the analysis. A legal presumption may, of course, be rebutted. 1.3.3 The Presumption of Vindictiveness and Increased Punishment In the second part of their proposed test, the People argue that Johnson did not meet his initial burden because he “faces no increased punishment” since “he has always faced the death

38 penalty.” They insist “[n]o California case has found a presumption of vindictive prosecution where prosecutors have sought the exact same punishment at all stages.” This claim amounts to a contention that because the State cannot execute Johnson more than once, five capital charges are not more serious than two capital charges. In support of this argument, the People cite People v. Ledesma, which they describe as a case in which “the California Supreme Court rejected application of the presumption of vindictiveness where the defendant faced the death penalty at the initial trial and at the retrial following appeal.” (See People v. Ledesma, supra, 39 Cal.4th at p. 731 (Ledesma).) The People’s description is correct in a narrow, factual sense, but Ledesma does not stand for the broader rule the People imply. The Ledesma Court rejected the defendant’s argument on forfeiture grounds. (Id. at p. 730 [“Defendant did not preserve the issue because he did not make any motion in the trial court based upon a theory of vindictive prosecution.”].) And while it is true that in Ledesma, “the prosecution sought the same sentence upon retrial that it did at the initial trial” (id. at p. 731), it also brought the same charges upon retrial that it did at the initial trial. (Id. at pp. 655–656; see People v. Ledesma (1987) 43 Cal.3d 171, 176 [original appeal].) Unlike in this case, the prosecution in Ledesma did not add additional charges or enhancement allegations following the defendant’s successful appeal.

39 The People’s argument amounts to a claim that the prosecution cannot increase the charges in a death penalty case.14 First, the California Supreme Court has repeatedly held that the vindictive-prosecution doctrine applies to capital cases and non- capital cases alike. (See, e.g., People v. Jurado, supra, 38 Cal.4th at p. 98; People v. Lucas, supra, 12 Cal.4th at pp. 477–478; People v. Maury (2003) 30 Cal.4th 342, 438–439.) A contrary conclusion, of course, would provide less due process protection to capital defendants than to non-capital defendants. (See California v. Ramos (1983) 463 U.S. 992, 998–999 [recognizing that capital cases require a greater degree of judicial scrutiny than other criminal cases]; Radin, Cruel Punishment and Respect for Persons: Super Due Process for Death (1980) 53 S.Cal. L.Rev. 1143.) Second, the potential for increased punishment is not the only circumstance relevant to the appearance of vindictiveness. The number and seriousness of the new charges also matter. (See, e.g., Bower, supra, 38 Cal.3d at p. 872 [presumption raised where prosecution brings “increased or additional charges”].) Even if increased punishment were dispositive, however, the concurrent punishment sought in this case is still punishment. (In re Wright (1967) 65 Cal.2d 650, 654–655; see People v. Alford (2010) 180 Cal.App.4th 1463 [discussing § 654].) Likewise, consecutive sentences of life without the possibility of parole are “longer” than single sentences of life without parole even though

14 The People put a finer point on this view at oral argument, where they contended the vindictive-prosecution doctrine does not apply to death penalty cases because “practically speaking, you cannot give someone more than one sentence of death. ’Cause you only have one life to give.”

40 a defendant cannot begin to serve the second term until his death. (People v. Hardy (1999) 73 Cal.App.4th 1429, 1433–1435 [concluding Three Strikes law requires court to double LWOP base term]; § 669 [LWOP sentences “may be imposed to run consecutively with one another”].) The State can therefore increase a criminal defendant’s punishment without increasing the length of the sentence he actually serves—but in any event, as the court noted below, because attempted murder is not a capital offense, the Coleman count does subject Johnson to a lengthier sentence. Contrary to the People’s claim, therefore, five death sentences are indeed “more serious” than two death sentences, notwithstanding the State can carry out that sentence only once. 1.4 Johnson met his initial burden. We turn to the specific facts before us and view them in context of the entire proceedings. Here, after 14 years on death row, Cleamon Johnson won his appeal and was granted a new trial. But rather than returning to the status quo ante and receiving the fair trial that eluded him in 1997, Johnson found himself charged with three more capital murders, one attempted murder, and six new gang enhancements. All of the charges—old and new alike—were brought in the same pleading and would be resolved in the same trial, where they would result in one final judgment. The People have had most of the evidence underlying the new counts and enhancements for years as a result of a single investigation conducted decades ago. Indeed, they tested the evidence in four back-to-back trials in the late 1990s, where their theory was that Johnson ordered each killing. In every case, the People exercised their discretion not to charge Johnson with these crimes—initially or after a mistrial—until he won his

41 appeal. In fact, the People admit they added the new charges and enhancements because Johnson won his appeal, and because upon retrial “there is a potential that there will be no justice in any sense.” In light of these facts and admissions, we agree with the trial court that Johnson presented sufficient evidence that the prosecution “increased the charges in apparent response to” (Twiggs, supra, 34 Cal.3d at p. 371) his successful appeal “under circumstances [that] are deemed to present a ‘reasonable likelihood of vindictiveness.’ ” (Bower, supra, 38 Cal.3d at p. 879.) This is particularly so because we are especially concerned about prosecutorial actions that may appear vindictive to—and have a chilling effect on—other criminal defendants. As we have discussed in detail above, capital cases are high profile and rarely reversed. Increasing the charges in a case like this one serves as a potent deterrent to other criminal defendants considering whether an appeal is worth the risk. 2. The People have rebutted the presumption only as to the Sutton murder. The only remaining issue is whether the People have met their “ ‘heavy burden’ of dispelling the appearance of vindictiveness.” (Twiggs, supra, 34 Cal.3d at p. 371.) To rebut the presumption of vindictiveness, the prosecution must demonstrate that (1) the increase in charge was justified by some objective change in circumstances or in the state of the evidence that legitimately influenced the charging process and (2) the “new information could not reasonably have been discovered at the time the prosecution exercised its discretion to bring the original charge.” (Bower, supra, 38 Cal.3d at p. 879.)

42 The test is an objective one. (Goodwin, supra, 457 U.S. at p. 374.) Accordingly, “this legal presumption cannot be rebutted by the prosecutor’s declaration that he or she was motivated by a reassessment of the evidence against the defendant rather than by any desire to punish the exercise of a protected right.” (Bower, supra, 38 Cal.3d at p. 879.) “In this regard, the trial court should consider the prosecutor’s explanation in light of the total circumstances of the case in deciding whether the presumption has been rebutted.” (Twiggs, supra, 34 Cal.3d at p. 374.) The showing is the same on review. While a “petitioner normally bears the burden of proving the facts upon which he bases his claim for relief, [citation] where, as here, the possibility that increased or additional charges violated due process is at issue, he need only demonstrate facts giving rise to a presumption of vindictiveness at which time, even [in writ proceedings], the burden shifts to the People to rebut the presumption.” (Bower, supra, 38 Cal.3d at p. 872.) Here, the People’s writ return devotes fewer than three pages, exclusive of exhibits, to rebutting the presumption of vindictiveness. For the most part, the People have not summarized the facts of any newly-charged crime or explain the context or importance of any new evidence or changed circumstance. “By making only general denials of the allegations of the petition, alleging only conclusionary statements of ultimate facts, the People have indicated a willingness to rely on the record of proceedings in the superior court and the documentary evidence submitted by petitioner as exhibits to his petition.” (Bower, supra, 38 Cal.3d at p. 873; see Thigpen v. Roberts, supra, 468 U.S. at p. 33, fn. 6 [“The State had ample opportunity below to attempt to rebut [the presumption] but did not do so. Its only

43 argument has been that Blackledge should not apply.”].) As we shall explain, with the notable exception of the Sutton killing, the justifications offered by the People do not suffice to dispel the appearance of vindictiveness created by the timing and scope of the new charges. 2.1 The People have not rebutted the presumption as to Jones, Mosley, and Coleman. As a preliminary matter, we address the People’s assertion that it was not previously worth the expense to prosecute Johnson for the newly-added charges because he was already on death row. They argue: “When a defendant is already on death row, expending more resources to prosecute a defendant for other charges, charges that would add nothing to that punishment, does not make practical sense. Even considering the inherent justice that flows from guilty verdicts in the form of truth[,] validation and accountability . . . alone do not necessarily overcome the consideration of expending large amounts of resources to deem someone again to be deserving of death, who is already facing a death sentence.” Once the original convictions were reversed and he was no longer condemned to death, however, “Johnson’s change in status alone justified additional investigation and the subsequent decision to charge him with the murders of additional victims.” That is, the People acknowledge they added the additional charges because Johnson’s convictions were reversed, and contend the successful appeal itself is sufficient to establish changed circumstances. This echoes their argument below, in which the People asserted they “were under no obligation to re-investigate these additional murders after Johnson was convicted of the Loggins and Beroit murders and sentenced to death. Indeed, pursuing

44 these cases would have put witnesses’ lives at risk, expended limited resources, and could have resulted in no greater punishment than Johnson had already received on the Loggins and Beroit murders.” In short, “any further prosecution following Johnson’s conviction and death sentence for the Loggins and Beroit murders would have been futile, and unnecessarily placed witnesses’ lives in jeopardy.” As we did above (see section 1.4, ante), we reject the implication that the vindictive-prosecution doctrine does not apply to capital cases. We also reject the People’s circular argument that a successful appeal itself establishes changed circumstances that allow them to add new charges; such an argument ignores the importance of the right to appeal. We express no opinion, however, about whether an acquittal in the Beroit-Loggins case would constitute a sufficient changed circumstance. Next, the People argue the increase in charges was justified both by the discovery of new evidence and by a change in the law that rendered old evidence newly admissible. They note the court below “was familiar with the facts of the murder charges because the trial court also sat as the magistrate during the preliminary hearing.” The members of this panel, however, did not attend the preliminary hearing—and the People tell us little about the facts of any charged offense.15 As a result, it is difficult to discern

15 For example, Freddie Jelks was the chief prosecution witness at both of Johnson’s trials. In the Mosley/Coleman case, Jelks testified for approximately 215 transcript pages over three days. Yet we have received only 18 non-consecutive pages—about eight percent—of that

45 whether the proffered evidence matters. Put another way, even assuming the People have established the existence of new, previously unavailable information for each new charge, they do not explain whether the new evidence “legitimately influenced the charging process.” (Bower, supra, 38 Cal.3d at p. 879.) The People’s failure to explain the context or importance of the new evidence involving the Jones, Mosley, and Coleman crimes is fatal in light of Johnson’s contention that the new information is unimportant. (See People v. Duvall (1995) 9 Cal.4th. 464, 480, 483 [where the respondent files “a return that did not dispute the material facts alleged by the petitioner[,] . . . the respondent is deemed to have admitted those material factual allegations that they fail to dispute”].) In any event, most of the People’s purported new evidence does not support a legitimate change in the exercise of prosecutorial discretion, or explain why they could not have reasonably discovered the information before Johnson’s first trial.16 In reaching these conclusions, we discuss below the facts and circumstances underlying the Mosley, Coleman, Jones, and Sutton charges based on the limited record before us.

testimony. Nor does the record include transcripts or other evidence from Wilson’s trial; the facts we have been able to glean about the Jones murder come from our colleagues’ 1999 opinion in Wilson’s appeal. 16 We take no position on whether evidence supporting counts 2, 5 and 6 is otherwise admissible in the retrial of the Beroit and Loggins counts. And, because the issue has not been raised, we express no opinion as to whether there are circumstances in which the People could prosecute Johnson for these crimes in a separately filed case.

46 2.1.1 Mosley Murder and Coleman Attempted Murder—1991 On September 14, 1991, Jelks alerted members of 89 Family that the 97 East Coast Crips were having a party on 97th Street, less than a quarter-mile south of 89 Family territory. Johnson, Jelks, and an individual known only as Jelly Rock got into a car belonging to a local crack addict and set out on a gang mission.17 Jelks drove; Johnson sat either in the passenger seat or the back seat. As the car approached the party, they encountered two girls fighting in the street. Jelks slowed down and flashed his headlights. Without warning, Johnson and Jelly Rock began shooting at the group; the Crips fired back. Tyrone Mosley was killed. Kim Coleman and Kenneth Davis were injured but survived. On December 6, 1994, Jelks gave a statement to police in which he incriminated himself and Johnson in the Mosley murder and incriminated Johnson and Allen in the 1991 murders of Loggins and Beroit. On September 2, 1997, with the help of Jelks’s testimony, Johnson and Allen were convicted of the Loggins and Beroit murders. On September 30, 1997, the jury returned a verdict of death, which the court imposed. Four months later, Johnson was transported back to Los Angeles from death row to stand trial for the Mosley murder and the Coleman attempted murder. The People planned to seek another death sentence. However, the jury was unable to reach a verdict, the

17 Jelks explained that a mission “is when you go out and deal with rivals.”

47 court declared a mistrial,18 and the case was ultimately dismissed. The People contend the “newly available statement of Tarone Burnaugh, an eyewitness to the shooting, corroborated the testimony of other witnesses regarding the description of the car involved in the shooting and direction of travel.” Johnson contends, and the People do not dispute, that Burnaugh’s account conflicts with both physical and ballistics evidence—particularly the location of shell casings and direction of travel. Burnaugh was also mistaken about the number of people present, the number of shots fired, and the number of victims. Because the People do not “allege additional facts that contradict those allegations[,]” they are “deemed to have admitted” them. (People v. Duvall, supra, 9 Cal.4th at pp. 480, 483.) If Burnaugh’s testimony is important for some other reason—such as to resolve a particularly critical or disputed issue—the People do not disclose it. We thus conclude the People have not established that the Burnaugh testimony “legitimately influenced the charging process” for counts 5 and 6. (Bower, supra, 38 Cal.3d at p. 879.) Even assuming the evidence is both new and important, however, the People have not satisfied the second prong of the Bower test—they have not explained why “the new information could not reasonably have been discovered at the time the prosecution exercised its discretion to bring the original charge.”

18 According to Johnson, the jury was split 8–4 on two counts and 6–6 on one count, but there is no indication as to the nature of the split. According to the People, the split was 9–3 in favor of guilt.

48 (Bower, supra, 38 Cal.3d at p. 879.) Burnaugh lived with his mother directly across the street from the shooting. Detectives interviewed him on September 14, 1991, the night of the shooting. They noted his address, phone number, height, weight, and date of birth—all of which are apparently accurate. However, the People contend his statement “could not reasonably have been discovered” sooner because detectives misspelled Burnaugh’s name. The People do not account for any previous efforts to locate Burnaugh, explain why the misspelling frustrated those efforts, or explain why Burnaugh’s address, phone number, height, weight, and date of birth were insufficient to find him. In short, the People simply do not explain why the prosecution could not reasonably have discovered this corroborating evidence at some point during the many years that elapsed between their first conversation with Burnaugh in 1991 and Johnson’s trial for the Beroit and Loggins murders in 1997. We thus conclude the People have not rebutted the presumption of vindictiveness for counts 5 and 6. 2.2 Jones Murder—1994 “On May 25, 1994, Georgia Jones appeared as a witness in the trial of Charles Lafayette, who was charged with murdering Willie Bogan. Lafayette is a member of the 84 Swans. Jones testified that she saw Lafayette shoot and kill Bogan. On June 6, [1994,] a mistrial was declared. A second trial was set. The prosecutor intended to call Jones as a witness at this trial. On June 13, 1994, before the retrial began, Jones was shot and killed on the corner of Wadsworth and 87th Place.” (People v. Wilson, supra, (Apr. 19, 1999, B111522), at p. 2.) Reco Wilson was convicted of the murder in January 1997, and Division Five affirmed by unpublished opinion. (Ibid.)

49 At Wilson’s trial in January 1997, the People presented ample evidence of Johnson’s role in Jones’ murder. In September 1997, the prosecution introduced that evidence again in the penalty phase of the Loggins/Beroit trial. Central to both cases was a recorded phone call from Johnson to Wilson in which Johnson appeared to solicit Wilson to murder Jones. The People now claim their post-appeal reinvestigation uncovered new evidence—a statement Jones made to law enforcement in 1993— that placed existing evidence in an important new light and provided “evidence of Johnson’s motive to kill Jones.” On March 9, 1993, over a year before she was killed, Jones told Detective Rosemary Sanchez that she had seen Johnson, Johnson’s brother Timothy, and Sutton engaged in a shootout.19 Johnson was arrested for the shooting, and as police were taking him into custody, Jones heard him threaten to kill Sutton— a statement the arresting officers presumably heard as well. The People argue this information “placed in context a wiretap intercepted call between Reco Wilson and Johnson. Reco Wilson was convicted of killing Jones[;] with the new contextual information, the call provides evidence of Johnson’s motive to kill Jones.” Since the phone call was played for juries in Johnson’s trial and Wilson’s trial—and since Sanchez testified for the prosecution in both cases—the evidence is certainly not new. Without conceding the point, the People contend the statement is newly admissible under Evidence Code section 1390 and thus provides “new contextual evidence . . . of Johnson’s motive to kill

19 There is no evidence Jones witnessed the murder itself.

50 Jones.” But that evidence code section is irrelevant. The statement was always admissible as non-hearsay motive evidence. (People v. Bolden (1996) 44 Cal.App.4th 707, 714–715 [statement offered to prove motive is not hearsay].) Accordingly, the People have not dispelled the presumption of vindictiveness as to count 2. 2.3 The People rebutted the presumption as to Sutton. On September 12, 1992, Albert Sutton, a drug dealer and member of the 89 Family, took his brother to Johnson’s house. Sutton’s brother, Danny Foster Glass, was a member of a rival Crips gang. Johnson was furious. He said, “I can’t believe you brought this fool to our hood. He’s a Crip.” Gunfire erupted and Glass was shot; he survived but lost the use of one eye. Nece Jones and Officer Miller both witnessed the shootout. The police took three men into custody, including Johnson and his brother Timothy, an 89 Family member known as “Sinister.” As police arrested them, Johnson and Sinister threatened to kill Sutton. Officers interviewed Sutton about the shooting, and Johnson was subsequently charged in case No. TA020639 with attempted murder. Sutton, angry that Johnson shot his brother, planned to testify against him. In the days after the shooting, Sutton received two threatening phone calls, which he later discussed with his sister, Anita Terrell. Terrell overheard one call directly, and heard Sutton’s side of the other call. Sutton explained to Terrell, “[t]his fool Evil wants me not to testify, but I’m going to testify because he shot my brother.” When it became clear Sutton could not be dissuaded from testifying, Johnson ordered his cousin Leon to kill Sutton. Sutton was killed on September 16, 1992.

51 Leon confided his role in the killing to another gang member, who told the police. When police interviewed Leon in 1995, he admitted killing Sutton on Johnson’s orders. He repeated the claim before he was sentenced in 1998 and at his parole hearing in 2008. Thus, as of 2008—three years before Johnson’s convictions were overturned—Leon had directly implicated him at least three times. Other witnesses confirmed Leon’s account. For example, Johnson’s brother Earl Ray Johnson (“Silent”) told an anonymous police source that Johnson told Leon to kill Sutton because Sutton was going to testify against Johnson.20 Leon’s story did not change when police re- interviewed him in 2012, though his memory had faded in the 20 years since the shooting. As Leon stood by his story throughout the years, his credibility increased. As recently as 2008, Leon continued to tell parole boards that Johnson had ordered him to kill Sutton— a position that conflicted with his attempts to gain release because an inmate’s acceptance of responsibility and development of insight are “significant factors” in determining parole suitability. (In re Shaputis (2011) 53 Cal.4th 192, 218; Cal. Code Regs., tit. 15, § 2402, subd. (d)(3).) Put another way, by continuing to shift responsibility to Johnson while in prison, Leon increased the length of his incarceration—and his value to the prosecution.

20 The statements of the corroborating witnesses appear in a one- page excerpt of a police report from 1995. The record does not reveal whether these witnesses testified at Leon’s trial, whether Leon testified in his own defense, or what other corroborating evidence the People may have presented.

52 While Leon’s enhanced credibility is helpful to the prosecution’s argument, the critical factor in the Sutton case is timing. Johnson was tried for the Beroit and Loggins murders in 1997. Leon was not brought to trial for Sutton’s murder until 1998. Throughout 1997, Leon retained a Fifth Amendment privilege against self-incrimination, and the Confrontation Clause barred the prosecution’s use of his out-of-court statements against Johnson. Under Crawford v. Washington, testimonial hearsay is inadmissible unless the defendant has an opportunity to cross-examine the declarant. (Crawford v. Washington (2004) 541 U.S. 36, 53.) Leon’s statements were made during a police interrogation, and were thus testimonial. (Id. at p. 68, fn. 10.) Since, under the Fifth Amendment, Leon could not be compelled to testify, Johnson did not have an opportunity to cross-examine him. Following his 1998 guilty plea, Leon could no longer assert his Fifth Amendment privilege—and could be compelled to testify about the Sutton murder. By then, however, Johnson had been convicted and sentenced to death. By the time the People decided to charge Johnson with Sutton’s murder in 2014, they had key evidence that was not available to them when they brought Johnson to trial in the original case: Leon’s guilty plea and his subsequent statements that Johnson ordered him to kill Sutton—statements that may have cost him an early release. The availability of the shooter’s testimony—testimony that was not available to prosecutors when they tried Johnson in 1997—constitutes an “objective change in . . . the state of the evidence which legitimately influenced the charging process[.]” (Bower, supra, 38 Cal.3d at p. 879.) Given the timing of the charges and Leon’s role in the Sutton murder, we are also satisfied that the prosecution could not have

53 reasonably obtained this evidence when “it exercised its discretion to bring the original charge.” (Ibid.) 3. Gang Allegations Finally, we turn to the newly added gang allegations tethered to counts 3 and 4 for the Loggins and Beroit murders. We note that the People showed no interest in charging Johnson with any gang allegation for these crimes until his convictions were overturned on appeal. Below, the People stated that they “decided to add the gang allegation because these are quintessential gang crimes and gang evidence had been introduced at all four original trials” without explaining why they failed to do so when Johnson was first tried. After reviewing the supplemental briefing on this issue, since the gang allegations amount to “an increase in charges” (Valli, supra, 187 Cal.App.4th at p. 802; see People v. Fuentes (2016) 1 Cal.5th 218, 223 [“section 186.22, subdivision (b) increases the punishment for the underlying conviction in several different ways” and can “negatively impact defendant in future criminal cases as well. . . . [E]ven if the punishment is struck, an enhancement finding could impact defendant in a future case.”]), we conclude that the addition of gang allegations for counts 3 and 4 following Johnson’s successful appeal raises a presumption of vindictiveness under Twiggs and Bower. However, since the court below did not address whether the People rebutted the appearance of vindictiveness, we will direct it to hold a hearing to determine whether the People can rebut the presumption as to the gang allegations for these counts. If the People do not meet the heavy burden necessary to dispel the appearance of vindictiveness, the court should dismiss the gang allegations.

54 DISPOSITION

The petition for writ of mandate is granted in part. A peremptory writ of mandate shall issue directing the trial court to: (1) vacate the portion of its order denying Johnson’s motion to dismiss for vindictive prosecution as to the Jones, Mosley, and Coleman crimes; (2) enter a new order dismissing counts 2, 5, 6 and their related allegations; and (3) hold a hearing to allow the People the opportunity to rebut the presumption of vindictiveness raised by the newly-charged gang allegations (Pen. Code, § 186.22) attached to counts 3 and 4. The stay of proceedings issued on October 15, 2015 is vacated.

CERTIFIED FOR PUBLICATION

LAVIN, J. WE CONCUR:

EDMON, P. J.

ALDRICH, J.