People v. Vivar

May 3, 2021

S260270

Versions:11 Cal.5th 510

485 P.3d 425

278 Cal.Rptr.3d 2

Opinion of the Court by Cuéllar, J.

The population of the United States includes millions of immigrants who arrived as children, attended schools, and found work here. (See Dep‘t of Homeland Security v. Regents of the Univ. of California (2020) ___ U.S. ___, ___ [140 S.Ct. 1891, 1932] (conc. & dis. opn. of Kavanaugh, J.).) Whether they become citizens or not, these immigrants’ ties to our country are evident not only in their work and schooling, but in how they‘ve formed attachments and families of their own. In contrast, what ties they once had to their country of birth — from which they may lack even memories — often slip away. So when longstanding noncitizen residents of this country are accused of committing a crime, the most devastating consequence may not be a prison sentence, but their removal and exclusion from the United States. (See People v. Martinez (2013) 57 Cal.4th 555, 563 (Martinez).) Because the prospect of deportation “is an integral part,” and often even “the most important part,” of a noncitizen defendant‘s calculus in responding to certain criminal charges (Padilla v. Kentucky (2010) 559 U.S. 356, 364 (Padilla)), both the Legislature and the courts have sought to ensure these defendants receive clear and accurate advice about the impaсt of criminal convictions on their immigration status, along with effective remedies when such advice is deficient. (E.g., Pen. Code, §§ 1016.2 et seq., 1473.7; Lee v. United States (2017) ___ U.S. ___ [137 S.Ct. 1958] (Lee); Padilla, at p. 360; Martinez, at p. 559; People v. Superior Court (Giron) (1974) 11 Cal.3d 793, 798.)

How these provisions apply to people like defendant Robert Landeros Vivar — who came to the United States at age six and lacked any meaningful ties to his country of birth — is the problem we address in this case. Vivar was arrested in 2002 for attempting to steal Sudafed from a grocery store. Although he‘d spent four decades living in this country as a lawful permanent resident, he lacked American citizenship. What he nonetheless possessed were robust ties to the United States. His mother, wife, children, and grandchildren were all citizens. His son, who was serving in the United States Air Force, was about to be deployed to the Middle East.

Unfortunately, as the Court of Appeal held and the Attorney General concedes, Vivar was never properly advised about the immigration consequences of his plea options. He didn‘t know, for example, that pleading guilty to violating Health and Safety Code section 11383, former subdivision (c), would necessarily subject him to mandatory deportation, while pleading guilty to violating Penal Code section 459 would not. Vivar took the former plea offer and rejected the latter. His mistake soon became manifest: within days, Vivar was subjected to an immigration hold, and a few months later he was deported.

After Vivar made his way back into the United States by crossing the border without insрection, he sought expungement of his drug conviction. He succeeded and then tried to secure further relief by way of a petition for writ of error coram nobis. Neither had any effect on his immigration status, however. He was again deported in 2013.

In 2018, Vivar filed a motion to vacate his 2002 conviction under a recently enacted statute offering relief to those who had already served their sentences. (Pen. Code, § 1473.7 (section 1473.7).) A successful section 1473.7 motion requires a showing, by a preponderance of the evidence, of a prejudicial error that affected the defendant‘s ability to meaningfully understand the actual or potential immigration consequences of a plea. (Id., subds. (a)(1), (e)(1).) The Court of Appeal concluded that while counsel had failed to offer Vivar competent advice about immigration consequences in 2002, Vivar failed to demonstrate any prejudice from the error. (People v. Vivar (2019) 43 Cal.App.5th 216, 225–231 (Vivar).) Based on an independent review of the record, we disagree. Vivar has demonstrated a reasonable probability that if he had been properly advised by counsel about the immigration consequences of his plea, he wouldn‘t have pleaded guilty to an offense subjecting him to mandatory deportation. We therefore reverse the judgment of the Court of Appeal.

I.

In 1962, when Vivar was six years old, he and his family immigrated as lawful permanent residents from Mexico tо the United States. He now has two children and six grandchildren. All are American citizens and all reside here in California, along with Vivar‘s two siblings.

A.

Upon arrival, Vivar quickly adapted to life in the United States. Since early in his youth, his primary language has been English. In high school, he helped establish a Reserve Officers’ Training Corps program and hoped to serve his country in Vietnam like his older brother, Martin, but the war ended a few months after he graduated. Vivar instead began working at an airline and was soon promoted to a management position that required him to work a night shift at the airport and a day shift in the office. Limited to only a few hours of sleep a night, he turned to amphetamines to stay awake.

Vivar first entered — and successfully completed — a residential drug treatment program in the late 1990s. Unfortunately, he relapsed in 2001. The conviction under review arose from his methamphetamine addiction. In February 2002, he was caught trying to steal 12 boxes of Sudafed from a grocery store in Corona. Vivar told the store‘s loss prevention officer — and later, the police — that he planned to provide the Sudafed to someone who would manufacture methamphetamine and, in turn, share some of the finished product with him. The Riverside County District Attorney charged Vivar with possessing methamphetamine precursors with the intent to manufacture the drug (Health & Saf. Code, § 11383, former subd. (c); see id., § 11383.5, subd. (c)) as well as petty theft with a prior conviction (Pen. Code, § 666).

B.

The District Attorney offered Vivar several plea options. What happened next is in some dispute. Vivar recalls his attorney conveying an offer of an unspecified felony plea with a three-year sentence. He rejected that offer because of his mistaken belief — never corrected by his appointed attorney — that all felony convictions resulted in deportation and that the opposite was true for misdemeanors. (Cf. U.S. v. Graham (3d Cir. 1999) 169 F.3d 787, 792 [some misdemeanors can qualify as an aggravated felony under federal immigration law].) Based on this mistake, he asked counsel to secure a plea deal that could eventually be reduced to a misdemeanor. He also informed her he had a drug problem and wanted treatment, even if not required by the plea offer.

Following those discussions, counsel relayed an offer for Vivar to plead guilty to burglary (Pen. Code, § 459) with a low-term prison sentence. With good-conduct credits, he could‘ve served just a year in prison and avoided mandatory deportation. (See Pen. Code, §§ 461, former subd. (a), 2933, subd. (a).) According to Vivar, though, counsel never advised him about the immigration-related benefits of this plea, nor did she correct his misimpression about the respective immigration consequences of felonies and misdemeanors. Unaware the burglary plea offer could be deportation-neutral, he rejected it. He pleaded guilty instead to possessing methamphetamine precursors with intent to manufaсture in exchange for an agreed-on 365-day county jail sentence — with a stipulation that the court would recommend admission to a residential drug treatment facility — and that a low-term, two-year prison sentence would be imposed only if he failed to complete the treatment program. Vivar mistakenly believed this disposition would allow him both to get treatment and, once the conviction was reduced to a misdemeanor, avoid deportation.¹

Before entering his plea in March 2002, Vivar executed a form that required him to initial 17 separate paragraphs acknowledging that he understood the potential consequences of his plea. One paragraph stated, “If I am not a citizen of the United States, I understand that this conviction may have the consequences of deportation, exclusion from admission to the United States, or denial of naturalization pursuant to the laws of the United States.” Another paragraph provided, “I have had an adequate time to discuss with my attorney (1) my constitutional rights, (2) the consequences of any guilty plea, and (3) any defenses I may have to the charges against me.”

Vivar‘s counsel declined to submit a declaration in connection with the hearing on the section 1473.7 motion. Her recollection was presented to the court instead through unsworn email correspondence and her handwritten notes. Counsel stated that while she didn‘t specifically remember Vivar‘s case, her “standard practice” at the time was to “advise non-citizen clients of the potential for immigration consequences” of their convictions and that she “routinely followed that practice.” After reviewing her notes from the plea negotiations, counsel indicated that she was “confident that Mr. Vivar was ‘fully advised’ of the consequences of the plea,” which under the circumstances of the case “would have included the standard advisement of possible deportation.” She also stated that she “believe[d]” she “specifically cautioned” Vivar “that, in spite of his experience” in a prior criminal proceeding, residential treatment “would NOT determine whether or not he would be deported on the new offense,” and that if Vivar had any questions “he should consult an immigration attorney for clarification.” What she did not advise him was whether his understanding of felonies, misdemeanors, and immigration law was correct. Nor did she advise him as to the actual immigration consequences of a plea to the drug charge or any other plea.

C.

A few days after being sentenced, Vivar was informed that he couldn‘t be admitted to the recommended drug treatment program ” ‘due to an “immigration hold.” ’ ” (Vivar, supra, 43 Cal.App.5th at p. 221.) As he would‘ve known had he been properly advised, his conviction activated a tripwire in immigration law — it qualified as a controlled substance offense as well as an aggravated felony. (See 8 U.S.C. §§ 1101(a)(43)(B), 1227(a)(2)(A), 1227(a)(2)(B).)² Vivar promptly sent a series of letters to the court expressing confusion about the situation and requesting assistance with admission to the drug treatment program. (Vivar, at p. 221.) In those letters, he explained that he had been a legal resident for the past 40 years, that his family members were United States citizens, and that his son was currently serving in the United States Air Force and awaiting deployment to the Middle East. He made plain that “[i]f I would have been made aware of these facts I would never have plead[ed] Guilty to this Charge.” In the meantime, federal immigration authorities notified Vivar that he was subject to removal because of his recent criminal conviction and, in January 2003, deported him. (Vivar, at p. 221.)

Determined to rejoin and support his family — and unable to find work in Mexico — Vivar reentered the United States, without inspection, in May 2003. In 2008, he successfully obtained an order to expunge his conviction under Penal Code

section 1203.4. It took another three years for him to learn, when he was again detained by immigration authorities, that expungement did not mitigate the immigration consequences of his plea. (See Martinez, supra, 57 Cal.4th at p. 560.) Vivar then filed a petition for writ of error coram nobis, which was denied. In March 2013, he was again deported and has been living in Tijuana, Mexico ever since. There, he works full-time at a call center and founded a nonprofit organization to help deported mothers of United States citizen children as well as deported mothers of children lawfully residing in the United States under the Deferred Action for Childhood Arrivals program. He has also been volunteering with organizations to support deported United States veterans. If this conviction can be vacated, Vivar — who has remained drug-free since 2002 — may be able to seek reentry to the United States and be reunited with his family.

Vivar filed a motion in January 2018 to vacate his conviction under Penal Code section 1473.7 — the motion under review here. He asserted, among other things, that he would never have pleaded guilty to violating Health and Safety Code section 11383, former subdivision (c), if counsel had informed him it would result in his deportation. The trial court denied the motion, reasoning (1) that counsel made no affirmative misadvisement, and (2) that “nonadvisement” of immigration consequences didn‘t qualify as ineffective assistance under United States Supreme Court precedent. The trial court didn‘t consider whether Vivar suffered prejudice from counsel‘s failure to provide adequate advice.

The Court of Appeal affirmed, but on different grounds. Contrary to the trial court, the Court of Appeal determined that defense counsel provided ineffective assistance. (Vivar, supra, 43 Cal.App.5th at p. 228.) At a minimum, the court reasoned, Vivar had asked “a specific question about deportation” (ibid.), a question that “required an attorney to research and apprise their client of the immigration consequenсes of a plea” (id. at p. 227). To warn merely ” ‘that his plea might have immigration consequences,’ ” in circumstances where the consequences were “certain,” was “constitutionally deficient.” (Id. at p. 228.) What barred relief here, in the Court of Appeal‘s view, was Vivar‘s failure to demonstrate prejudice — in this context, a reasonable probability that he wouldn‘t have entered the same plea if he had been properly advised. (Id. at p. 229.) The court reasoned that Vivar‘s main priority seemed to be drug treatment, not immigration consequences, and asserted that this was corroborated by counsel‘s contemporaneous notes, by Vivar‘s rejection of the immigration-neutral burglary plea, and by the trial court‘s finding that Vivar was ” ‘was more willing to rely on his experiences than he was on his counsel‘s advice.’ ” (Id. at p. 230.) Despite the decades Vivar spent in this country, his family members’ American citizenship, and his prompt objection to the federal immigration hold, the Court of Appeal insisted there was “no contemporaneous evidence in the record” to corroborate Vivar‘s claim that he would‘ve preferred an immigration-neutral disposition. (Ibid.)

We granted Vivar‘s petition to review two rulings made by the Court of Appeal: first, its conclusion that he suffered no prejudice within the meaning of section 1473.7, subdivision (a)(1); and second, its conclusion that appellate courts must review deferentially factual findings made by the trial court concerning prejudice under section 1473.7, even if those findings are based on a cold record consisting solely of documentary evidence. Because no one sought review to challenge the Court of Appeal‘s finding that Vivar‘s counsel was ineffective, we assume for purposes of this proceeding that counsel failed to properly advise Vivar about the immigration consequences of his plea or of the plea offers he rejected. After we granted review — and after receiving a 30-day extension to file his brief on the merits — the Attorney General has undertaken a “fresh look” at the Court of Appeal‘s analysis and now concedes that the Court of Appeal erred in applying a deferential standard of review to the trial court‘s prejudice findings. He further concedes that, under independent review, Vivar has demonstrated prejudice and is entitled to relief. We have retained the case for decision to resolve a conflict in the Court of Appeal concerning the standard of review governing prejudice findings under section 1473.7, subdivision (a)(1) and to clarify more generally what demonstrates prejudice under that provision. (See People v. Maya (2020) 9 Cal.5th 239, 241.)

II.

It took less than a month for Vivar to realize the dire ramifications of his mistaken embrace of a felony drug possession plea. What ensued in the 18 years that followed underscores how much Vivаr consistently valued his presence on American soil, and how likely it is that — properly advised — he would have prioritized a resolution of his case allowing him to stay in the country. Mere weeks after entering his plea, when he learned that he was subject to an immigration hold and thus ineligible for a residential drug treatment program, he sent the sentencing judge a handwritten letter “to seek the court‘s mercy.” Vivar informed the court that he was “a legal resident and ha[s] been for the past 40 years“; that his mother and wife were American citizens; that his children and grandchildren, all born in Riverside County, were likewise citizens; and that his oldest child and only son was in the United States military awaiting deployment to the Middle East. He “fully accept[ed]” responsibility for his actions but “would like to change my life for good and become a productive member of society.” Two months later, after he was transferred to a federal immigration facility, he reiterated his willingness “to do whatever it takes to once again be an asset to my community and not a liability” and asked the court to reduce his conviction to a misdemeanor. (See Pen. Code, § 17, subd. (b).)

A few months later, in October 2002, Vivar asked that his case “be Re-opened” on due process grounds. The legal advice he received at the time of his plea never conveyed, Vivar insisted, that he was accepting responsibility for “an Aggravated Felony for Immigration purposes and thus would warrant Immediate Deportation.” Had he been so advised, he “would have never plead[ed] Guilty to this Charge.” He was deported a few months later.

In 2008, Vivar successfully moved to expunge his conviction through another pro se filing. Only later did he learn that expungement hadn‘t erased or even mitigated the immigration consequences of his plea. (See Martinez, supra, 57 Cal.4th at p. 560.) In 2012, a lawyer advised Vivar he could obtain relief on grounds of ineffective assistance of counsel by filing a petition for writ of error coram nobis. Vivar hired the lawyer to file such a petition. Yet this filing, too, proved fruitless: This court had already held, in 2009, that a defendant‘s ignorance of a plea‘s immigration consequences — or counsel‘s failure to negotiate a different plea — constituted a mistake of law and thus did not qualify as a ground for relief on coram nobis. (People v. Kim (2009) 45 Cal.4th 1078, 1102–1104 (Kim).) In a companion case, we also held that persons in federal immigration ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍custody after completing their state sentences, as well as any probation or parole period, are no longer in state custody. (People v. Villa (2009) 45 Cal.4th 1063.) So their state convictions are beyond the reach of habeas corpus.

But they are not beyond the reach of remedies recently enacted by the Legislature. As Vivar was running out of options, lawmakers considered the problem faced by Vivar and so many others who were unaware of the immigration consequences posed by a plea entered many years earlier. (See Kim, supra, 45 Cal.4th at p. 1107 [“the Legislature has been active in providing statutory remedies when the existing remedies . . . have proven ineffective“; “the Legislature remains free to enact further statutory remedies for those in defendant‘s position“].) They did so by enacting section 1473.7, which “create[d] an explicit right for a person no longer imprisoned or restrained.” (Legis. Counsel‘s Dig., Assem. Bill No. 813 (2015–2016 Reg. Sess.).) Under this new provision, a court “shall” vacate a conviction or sentence upon a showing, by a preponderance of the evidence, of “prejudicial error damaging the moving party‘s ability to meaningfully understand, defend against, or knowingly accept the actual or potential adverse immigration consequences of a plea of guilty or nolo contendere.” (§ 1473.7, subds. (e)(1), (a)(1).) A finding of prejudicial error under this provision may, but need not, be based on ineffective assistance of counsel. (Id., subd. (a)(1).) If the motion is meritorious, “the court shall allow the moving party to withdraw the plea.” (Id., subd. (e)(3).)

The Court of Appeal determined that trial counsel failed to advise Vivar of “the certain immigration consequences of his plea.” (Vivar, supra, 43 Cal.App.5th at p. 228.) This rendered counsel‘s representation “constitutionally deficient.” (Ibid.) Because no party challenged this finding — and the question of counsel‘s deficient performance falls outside the issues presented for review — we accept it as true for purposes of this proceeding. We review only the Court of Appeal‘s finding that Vivar suffered no prejudice on account of counsel‘s error. In examining that finding, we consider first what is the applicable standard of review. Then we apply that standard to the record here. Reviewing the record independently, we conclude Vivar was prejudiced within the meaning of section 1473.7, subdivision (a)(1).

A.

When a trial court grants or denies a motion to vacate a conviction under section 1473.7, the parties can appeal. (§ 1473.7, subd. (f).) Both parties acknowledge, though, that the standard for reviewing such orders is “unsettled.” (People v. Rodriguez (2019) 38 Cal.App.5th 971, 977.) In the Court of Appeal, the Attorney General analogized section 1473.7 motions to other statutes authorizing withdrawal of a plea — despite their different wording (see, e.g., Pen. Code, § 1018 [a trial court “may” permit a defendant to withdraw a plea “for a good cause shown“]) — and argued that denial of the motion should be reviewed deferentially for abuse of discretion. (See, e.g., Rodriguez, at p. 977.) Vivar disagreed, making the case that his prejudice claim raised a mixed question of law and fact that should be reviewed independently. (See, e.g., People v. DeJesus (2019) 37 Cal.App.5th 1124, 1133.) Unsatisfied with these two possibilities, the Court of Appeal proposed yet another option: a complicated framework in which the standard of review governing a trial court‘s section 1473.7 рrejudice ruling would vary depending on the basis of the claimed error. Under this option, the trial court‘s ruling would be reviewed independently where the prejudicial error consists of constitutionally ineffective assistance of counsel but would be reviewed for abuse of discretion where the claim rests merely on ” ‘statutory error.’ ” (Vivar, supra, 43 Cal.App.5th at p. 224.)

The Attorney General reversed course in the proceedings before us. He no longer advocates the abuse of discretion standard — even in the context of mere statutory error. Instead, he urges us to apply the independent standard of review to all prejudice determinations under section 1473.7, subdivision (a)(1). Although we are not “bound” to accept a party‘s concession on a question of law (Desny v. Wilder (1956) 46 Cal.2d 715, 729), after careful review we accept the Attorney General‘s concession. (See In re McKinney (1968) 70 Cal.2d 8, 14.)

Our case law has applied the independent review standard — which accords substantial weight to the trial court‘s credibility findings — in analogous circumstances. Whether counsel‘s advice regarding immigration was inadequate and whether such inadequacy prejudiced the defense, while mixed questions, are predominantly questions of law. (See In re Resendiz (2001) 25 Cal.4th 230, 248–249 (Resendiz) (lead opn. of Werdegar, J.).)³ Accordingly, we review such rulings independently (Resendiz, at p. 248), and rightly so, given the profound and substantial consequences of a prejudicial misadvisement on a defendant‘s life. (Cf. People v. Ault (2004) 33 Cal.4th 1250, 1265 (Ault) [“the proper review standard is influenced in part by the importance of the legal rights or interests at stake“]; id. at p. 1266 [“another important consideration in determining the appropriate standard of review

is the consequences of an erroneous determination in the particular case“].)

Nothing in section 1473.7, subdivision (a)(1) or elsewhere gives us a reason to deviate from this template. Indeed, prior to section 1473.7‘s amendment in 2018 — which clarified that the “legal invalidity” of a conviction or sentence “may, but need not, include a finding of ineffective assistance of counsel” (§ 1473.7, subd. (a)(1); Stats. 2018, ch. 825, § 2) — our courts had “uniformly assumed” that relief was available only to those who had demonstrated constitutionally ineffective assistance (People v. Camacho (2019) 32 Cal.App.5th 998, 1005 (Camacho)). And to the extent those courts considered the question, they applied a standard of independent review to such claims. (See, e.g., People v. Ogunmowo (2018) 23 Cal.App.5th 67, 75–76 (Ogunmowo); accord, People v. Tapia (2018) 26 Cal.App.5th 942, 950 (Tapia) [following Ogunmowo]; People v. Olvera (2018) 24 Cal.App.5th 1112, 1115–1116 (Olvera) [same].) A standard of independent review — the same standard governing our review of these claims on habeas corpus — is most consistent with section 1473.7‘s purpose: to offer relief to those persons who suffered “prejudicial error” but are “no longer imprisoned or restrained” and for that reason alone are unable to pursue relief on habeas corpus. (Legis. Counsel‘s Dig., Assem. Bill No. 813 (2015–2016 Reg. Sess.).)

When the Legislature amended section 1473.7 in 2018, nowhere did it “signal an intent to supersede” the standard of review the Court of Appeal had already articulated (In re W.B. (2012) 55 Cal.4th 30, 57; see, e.g., Ogunmowo, supra, 23 Cal.App.5th at pp. 75–76), nor did it propose that appellate courts adopt a more deferential standard of review. To the contrary: the Legislature explicitly stated its intended purpose was to make relief more broadly available to deserving defendants, given the critical interests at stake. (See Assem. Com. on Public Safety, Analysis of Assem. Bill No. 2867 (2017–2018 Reg. Sess.) as amended Apr. 5, 2018, pp. 2, 4 [this bill helps achieve the original goal of “creating a process for individuals to erase the catastrophic consequences . . . that can attach to even very old criminal convictions” by “clearing up minor discrepancies that have arisen since implementation“].) The 2018 amendment expanded the category of defendants who could obtain relief by eliminating any requirement that the defendant establish ineffective assistance of counsel. An uncodified section of the 2018 amendment declared that the expanded language in subdivision (a)(1) provided ”clarification to the courts regarding Section 1473.7 of the Penal Code to ensure uniformity throughout the state and efficiency in the statute‘s implementation.” (Stats. 2018, ch. 825, § 1, subd. (b), italics added.) Moreover, the Legislature instructed courts to interpret section 1473.7 “consistent with the findings and declarations made in section 1016.2 of the Penal Code” (Stats. 2018, ch. 825, § 1, subd. (c)) — which in turn articulated a purpose ”to codify . . . related California case law and to encourage the growth of such case law in furtherance of justice” (Pen. Code, § 1016.2, subd. (h), itаlics added). Under these particular circumstances — where legislators expressed keen awareness of how section 1473.7 was being implemented and viewed the 2018 amendment as a clarification and codification of existing law — we see no reason to disturb the prevailing independent standard of review.

The Court of Appeal posited that a different standard should apply when the moving party relies on a mistake of law under section 1473.7, subdivision (a)(1) that does not rise to the level of ineffective assistance of counsel. (See Vivar, supra, 43 Cal.App.5th at p. 224.) But that approach would cut against the Legislature‘s stated goals of codifying existing law and ensuring uniformity. (Stats. 2018, ch. 825, § 1, subd. (b).) What‘s more, it would endow with determinative significance the precise distinction — whether the asserted error constituted ineffective assistance of counsel — that the Legislature sought to erase by amending section 1473.7, subdivision (a)(1) to provide that “[a] finding of legal invalidity may, but need not, include a finding of ineffective assistance of counsel.” Overburdened trial courts might well choose to consider only whether there was “prejudicial error” damaging the moving party‘s ability to understand actual or potential immigration consequences without deciding whether such an error actually rose to the level of constitutionally deficient performance. (§ 1473.7, subd. (a)(1).) Or courts may fail to make the latter finding simply because of the happenstance that no party provided trial counsel with “timely advance notice of the motion hеaring,” which is a prerequisite to “a specific finding of ineffective assistance of counsel.” (Id., subd. (g).) It would make little sense to make the standard of review hinge on these trivial choices.⁴ (See People v. Bravo, supra, 58 Cal.App.5th at p. 1180 (conc. opn. of Raphael, J.).)

Our embrace of the independent standard of review also fits with how section 1473.7 motions generally arise. Only

defendants who have already completed their sentences may even seek relief under section 1473.7. So these motions — as the separate opinion acknowledges — are ordinarily brought many years after the plea. (Compare Pen. Code, § 1018 [allowing a plea to be withdrawn only “before judgment or within six months . . . if entry of judgment is suspended“].) Vivar, for example, brought his motion nearly 16 years after entering his plea. Years later, the judge adjudicating the resulting motion may never have participated in any of the underlying proceedings and must rely entirely on a cold record. (Cf. Haraguchi v. Superior Court (2008) 43 Cal.4th 706, 711, fn. 3; see id. at p. 713 [motions to recuse the prosecutor are reviewed for abuse of discretion because trial courts “are in a better position than appellate courts to . . . evaluate the consequences of a potential conflict in light of the entirety of a case, a case they inevitably will be more familiar with than the appellate courts“].)⁵ Indeed, that‘s what happened here: the judge hearing the section 1473.7 motion had no firsthand familiarity with the circumstances surrounding Vivar‘s plea.

So our еmbrace of independent review in this context is a product of multiple factors with special relevance here: the history of section 1473.7, the interests at stake in a section 1473.7 motion, the type of evidence on which a section 1473.7 ruling is likely to be based, and the relative competence of trial courts and appellate courts to assess that evidence. (See Ault, supra, 33 Cal.4th at pp. 1260–1261, 1265–1266.) The fact that

the motion is reviewed by way of appeal does not necessarily dictate a particular standard of review. (See id. at pp. 1266–1267.)

“[U]nder independent review, an appellate court exercises its independent judgment to determine whether the facts satisfy the rule of law.” (In re George T. (2004) 33 Cal.4th 620, 634.) When courts engage in independent review, they should be mindful that ” ‘[i]ndependent review is not the equivalent of de novo review . . . .’ ” (People v. Jackson (2005) 128 Cal.App.4th 1009, 1021.) An appellate court may not simply second-guess factual findings that are based on the trial court‘s own observations. (See In re Ernesto H. (2004) 125 Cal.App.4th 298, 306; cf. George T., supra, 33 Cal.4th at p. 634 [under a de novo standard, ” ‘a reviewing court makes an original appraisal of all the evidence’ “].) In reviewing the constitutional claim raised in Resendiz, we explained that factual determinations that are based on ” ‘the credibility of witnesses the [superior court] heard and observed’ ” are entitled to particular deference, even though courts reviewing such claims generally may ” ‘reach a different conclusion [from the trial court] on an independent examination of the evidence . . . even where the evidence is conflicting.’ ” (Resendiz, supra, 25 Cal.4th at p. 249 (lead opn. of Werdegar, J.).)⁶ In section 1473.7 proceedings, appellate courts

should similarly give particular deference to factual findings based on the trial court‘s personal observations of witnesses. (See, e.g., Tapia, supra, 26 Cal.App.5th at pp. 948–950 [deferring where the trial judge hearing the § 1473.7 motion also presided over the plea hearing].) Where, as here, the facts derive entirely from written declarations and other documents, however, there is no reason to conclude the trial court has the same special purchase on the question at issue; as a practical matter, “[t]he trial court and this court are in the same position in interpreting written declarations” when reviewing a cold record in a section 1473.7 proceeding. (Ogunmowo, supra, 23 Cal.App.5th at p. 79.)⁷ Ultimately it is for the appellate court to decide, based on its independent judgment, whether the facts establish prejudice under section 1473.7.

B.

The Legislature made relief available only to certain immigrants who accepted pleas without understanding ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍the immigration-related consequences of such decisions. What someone seeking to withdraw a plea under section 1473.7 must show is more than merely an error “damaging the moving party‘s ability to meaningfully understand, defend against, or knowingly accept the actual or potential adverse immigration consequences” of the plea. (§ 1473.7, subd. (a)(1).) The error must also be “prejudicial.” (Ibid.) Although the statute doesn‘t itself define what “prejudicial” means, we can glean the meaning from its context. (See Quintano v. Mercury Casualty Co. (2000) 11 Cal.4th 1049, 1055.)

In People v. Superior Court (Zamudio), supra, 23 Cal.4th 183, we considered what constituted prejudice when a trial court failed to advise a defendant of the plea‘s potential immigration consequences as required by Penal Code section 1016.5. Prejudice in such circumstances depended on ” ‘whether it is “reasonably probable” the defendant would not have pleaded guilty if properly advised.’ ” (Zamudio, at p. 210.) The focus on “what the defendant would have done, not whether the defendant‘s decision would have led to a more favorable result” derived from the fact that a defendant ” ‘may view immigration consequences as the only ones that could affect his calculations regarding the advisability of pleading guilty to criminal charges.’ ” (Martinez, supra, 57 Cal.4th at pp. 562, 563.) A decision to reject a plea bargain, we explained, “might be based either on the desire to go to trial or on the hope or expectation of negotiating a different bargain without immigration consequences.” (Id. at p. 567.) When a court weighs whether a defendant would have taken the latter path, it need not decide whether the prosecution would actually “have offered a different bargain” — rather, the court should consider “evidence that would have caused the defendant to expect or hope a different bargain would or could have been negotiated.” (Ibid., italics added.)

We embraced a similar approach when deciding whether а lawyer‘s deficient advisement on immigration consequences amounts to prejudicial ineffective assistance of counsel. A defendant in those circumstances must demonstrate a reasonable probability that, but for counsel‘s incompetence, the defendant ” ‘would not have pled guilty.’ ” (People v. Patterson (2017) 2 Cal.5th 885, 901 (Patterson), quoting Resendiz, supra, 25 Cal.4th at p. 253 (lead opn. of Werdegar, J.).) The United States Supreme Court, too, undertakes a similar analysis. In Lee, supra, ___ U.S. at page ___ [137 S.Ct. at page 1967], the prejudice prong of the ineffective assistance inquiry turned on whether the defendant had “adequately demonstrated a reasonable probability that he would have rejected the plea had he known that it would lead to mandatory deportation.”

Section 1473.7, subdivision (a)(1) fits this definition of “prejudicial error,” and we discern no reasons lurking in its provisions to concoct a different one. (See Camacho, supra, 32 Cal.App.5th at p. 1010.) Indeed, the current version of the statute acknowledges that prejudicial error “may, but need not, include a finding of ineffective assistance of counsel.” (§ 1473.7, subd. (a)(1).) The statutory findings for the 2018 amendment also declared that the statute “shall be interpreted in the interests of justice and consistent with the findings and declarations made in Section 1016.2” (Stats. 2018, ch. 825, § 1, subd. (c)), which in turn articulate the Legislature‘s intended purpose: to codify Supreme Court “and related California case law and to encourage the growth of such case law in furtherance of justice” (Pen. Code, § 1016.2, subd. (h)).

So: showing prejudicial error under section 1473.7, subdivision (a)(1) means demonstrating a reasonable probability that the defendant would have rejected the plea if the defendant had correctly understood its actual or potential immigration consequences. When courts assess whether a petitioner has shown that reasonable probability, they consider the totality of the circumstances. (Lee, supra, ___ U.S. at p. ___ [137 S.Ct. at p. 1966].) Factors particularly relevant to this inquiry include the defendant‘s ties to the United States, the importance the defendant placed on avoiding deportation, the defendant‘s priorities in seeking a plea bargain, and whether the defendant had reason to believe an immigration-neutral negotiated disposition was possible. (See id. at p. ___ [137 S.Ct. at pp. 1967–1969]; Martinez, supra, 57 Cal.4th at p. 568.)

The Court of Appeal found it “not reasonably probable that [Vivar] would have rejected the plea but for his counsel‘s failure to properly advise him.” (Vivar, supra, 43 Cal.App.5th at p. 229.) Two premises supported its conclusion: (1) that “no contemporaneous evidence” corroborated Vivar‘s claim that he wouldn‘t have entered the plea had he known the plea would lead to his deportation (id. at p. 230), and (2) that Vivar “prioritized drug treatment over potential immigration-neutral pleas.” (Id. at p. 229.) Neither premise, though, withstands scrutiny. What we find, reviewing the record independently, is that Vivar was prejudiced.

1

In a declaration submitted with his section 1473.7 motion, Vivar claims he would never have entered this plea hаd he understood that it would require his deportation. But when a defendant seeks to withdraw a plea based on inadequate advisement of immigration consequences, we have long required the defendant corroborate such assertions with ” ‘objective evidence.’ ” (Resendiz, supra, 25 Cal.4th at p. 253 (lead opn. of Werdegar, J.).) That‘s what Vivar has done here. Time and again, the record readily conveys how Vivar would have considered his immigration status “the most important part” of his decision to plead. (Padilla, supra, 559 U.S. at p. 364.) Vivar was brought to this country at age six as a lawful resident, and he attended schools, formed a family, and remained here for 40 years. At the time of his plea, he had two children, two grandchildren, and a wife, all of whom are citizens and all of whom resided in California. By the time he was deported, his wife was undergoing radiation treatment for a thyroid condition. By contrast, Vivar had virtually no ties to Mexico, spoke Spanish “like an American,” and found it “difficult to function in Mexican society because people treat [him] like an outsider.” Trial counsel‘s recollection and contemporaneous notes reflect that Vivar was indeed concerned about the “consequences” of his plea. All of these constitute contemporaneous objective facts that corroborate Vivar‘s concern about the immigration consequences of his plea oрtions. (See People v. Mejia (2019) 36 Cal.App.5th 859, 872.)

Also revealing is the objective evidence of Vivar‘s state of mind reflected in uncounseled letters he wrote to the court at or near the time of his plea. In his first letter, written just a month after his March 2002 plea, Vivar objected to his immigration hold and emphasized that “I am a legal resident and have been for the past 40 years“; noted that not only his wife and mother are citizens, but his children and grandchildren were all “born here in Riverside County“; and explained that his oldest child and only son was serving in the United States Air Force and awaiting deployment to the Middle East. He reiterated these concerns in another letter from federal immigration custody a month later, pleading that he be allowed to become “an asset to my community and not a liability.” Three months after that, Vivar said that counsel never advised him that his plea would result in his deportation and declared that “[i]f I would have been made aware of these facts I would have never plead[ed] Guilty to this Charge.”

The Court of Appeal neglected to explain why these facts at or near the time of Vivar‘s plea failed to provide adequate corroboration that he wouldn‘t have pleaded guilty had he known it would result in his deportation. Indeed, the court‘s analysis failed to mention these facts at all. This was error. In our view, these objective and contemporaneous facts corroborate, in a most convincing way, the statement in Vivar‘s declaration that he ”would never have pleaded guilty” if his attorney had informed him of the plea‘s consequences. (See Camacho, 32 Cal.App.5th at pp. 1011–1012 [finding prejudice where the defendant was brought to the United States as a child, had lived here for over 30 years, and his spouse and children were citizens]; accord, Lee, supra, ___ U.S at p. ___ [137 S.Ct. at p. 1968] [finding prejudice where the defendant was brought to the United States as a child, had lived here for nearly 30 years, and his parents were citizens].)

2

What the record also shows — and neither the Court of Appeal nor the Attorney General disputes — is that Vivar could have entered a plea avoiding mandatory deportation. Trial counsel‘s contemporaneous notes indicate the prosecution offered a deal under which Vivar would plead guilty to a single count of burglary (Pen. Code, § 459) with a recommendation that he serve the low term of two years in state prison. With credits, Vivar could have cut that term in half. (See Pen. Code, § 2933, subd. (a).) At the time of his plea, burglary in California was a deportable felony only in particular situations (see Kim, supra, 45 Cal.4th at pp. 1089–1090, 1098), and the uncontradicted declaration from Vivar‘s immigration expert stated that Vivar could‘ve entered such a plea without subjecting himself to mandatory deportation. Under these circumstances, we find at least ” ‘a reasonable probability’ ” that he could have tried “to obtain a better bargain that [did] not include immigration consequences.” (Martinez, supra, 57 Cal.4th at p. 567.)

Concluding otherwise, the Court of Appeal relied principally on the fact that Vivar rejected the burglary plea. The court pointed in particular to counsel‘s notes, where she had written that Vivar ” ‘[w]ants help w/ [his] drug problem.’ ” (Vivar, supra, 43 Cal.App.5th at p. 229.) Because Vivar “was offered and rejected a plea agreement that would have completely avoided any immigration consequences,” the court inferred “that immigration consequences were not defendant‘s primary consideration in accepting or rejecting any plea offer, and that further advice on this front was not reasonably probable to change his decisionmaking.” (Id. at pp. 229–230.)

The Court of Appeal‘s inference fails to persuade. Vivar‘s rejection of a potentially deportation-neutral plea can hardly serve as evidence that he didn‘t care about immigration consequences when it is undisputed that Vivar was not properly advised — and thus was ignorant — of the immigration consequences attached to his various plea options. So the fact that he unknowingly rejected an immigration-neutral option cannot, in itself, demonstrate that “immigration consequences were not defendant‘s primary consideration.” (Vivar, supra, 43 Cal.App.5th at pp. 229–230.)

Even less supports the Court of Appeal‘s contention that Vivar “prioritized drug treatment over potential immigration-neutral pleas.” (Vivar, supra, 43 Cal.App.5th at p. 229.) Indeed, it doesn‘t make sense to say that Vivar would‘ve chosen a plea that triggered mandatory deportation just so he could participate in drug treatment when that plea rendered him ineligible for the program.

According to his declaration, Vivar told counsel that he was interested in a drug treatment program even if it was not required by the terms of his plea — and counsel‘s notes corroborate his interest in such a program. Vivar, then, did not perceive a conflict or tradeoff between the goal of drug treatment and the goal of a deportation-neutral disposition. What stands out most clearly from the record is that he was never properly advised of the role his immigration status would play either in assessing the attractiveness of his plea options or in his eligibility for a drug treatment program. In fact, the record shows he was upset to learn, just a few days after his plea, that he was ineligible for the recommended treatment program precisely because of the plea‘s impact on his immigration status. And it was scarcely a month after his plea, having heard no response from his lawyer, that he wrote a letter to the court seeking its help. Had he been properly advised, it‘s reasonably probable Vivar would‘ve sought a disposition — like the burglary plea — where he could remain in this country and undergo drug treatment. The Court of Appeal failed to explain why Vivar, if properly advised, would‘ve viewed these goals as incompatible — or why, if properly advised, he would‘ve insisted on a strategy that prevented him from achieving either of his goals.

The Court of Appeal tried to buttress its conclusion that Vivar suffered no prejudice by highlighting “a factual inference the trial court was entitled to draw” and then deferring to that inference. (Vivar, supra, 43 Cal.App.5th at p. 230.) The trial court‘s ” ‘finding’ ” was that Vivar ” ‘was more willing to rely on his experiences than he was on his counsel‘s advice.’ ” (Ibid.) As we explained in part II.A., ante, the Court of Appeal was mistaken in believing the trial court‘s factual findings, which were based entirely on a cold record, “must be accorded deference.” (Vivar, at p. 231.) An appellate court should instead review such findings independently where, as here, the factual record consists entirely of written documents. Reviewing this cold record under that standard, we reject the trial court‘s finding. If Vivar acted under the misimpression that he could avoid immigration consequences so long as his ultimate sentence was a year or less, it likely was because he failed to receive adequate and accurate advice from counsel about the immigration consequences attached to his plea options. Without proper advice, Vivar had no choice but to rely on his own experiences and judgment, no matter how uninformed they might be. Had he truly been “unwilling to listen to the advice of counsel” (id. at p. 230), he never would‘ve expressed to her his concern about the consequences of his plea. And had he been correctly advised about those consequences, it‘s reasonably probable he wouldn‘t have entered the plea that triggered his deportation. The Court of Appeal erred in holding otherwise.

Finally, we conclude that the advisements in Vivar‘s plea form did not mitigate the prejudice from counsel‘s deficient immigration advice. What the plea form stated was that deportation was a possibility. (Vivar, supra, 43 Cal.App.5th at p. 228.) The problem for Vivar, though, was that deportation in these circumstances was mandatory — and when he accepted the plea deal, he remained ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍unaware of that crucial fact. (See Patterson, supra, 2 Cal.5th at pp. 896, 898.) In light of Vivar‘s extensive ties to the United States, the generic advisements in the plea form do not undermine our conclusion that he was prejudiced by counsel‘s failure to inform him that his plea would result in his deportation. (See In re Hernandez (2019) 33 Cal.App.5th 530, 547–548; People v. Espinoza (2018) 27 Cal.App.5th 908, 916–917; Ogunmowo, supra, 23 Cal.App.5th at pp. 80–81.)

III.

Defendants who lack United States citizenship sometimes face not only penal sanctions but also harsh immigration consequences if convicted. Because of this, pleas accepted in the shadow of deficient advice about the risks of deportation can have “dire” repercussions. (People v. Superior Court (Giron), supra, 11 Cal.3d at p. 798.) Section 1473.7 offers a remedy in the form of permission to withdraw a plea. But it‘s a remedy available only to some: those who have completed their sentences and who suffered a prejudicial error that damaged their ability to meaningfully understand, defend against, or knowingly accept the plea‘s actual or potential immigration consequences. (§ 1473.7, subds. (a)(1), (e)(3).) A moving party demonstrates prejudice by showing that in the absence of the error regarding immigration consequences, it‘s reasonably probable the moving party would not have entered the plea. Courts should subject the trial court‘s prejudice finding under this statute to independent review, a standard that heavily weighs trial court factual findings based on the court‘s own observations, but not trial court findings arising only from a cold record.

The Court of Appeal failed to review the record independently. Nor did it take into account the substantial contemporaneous evidence at or near the time of Vivar‘s plea corroborating his claim that he wouldn‘t have pleaded guilty if he‘d known it would result in his deportation from his home of 40 years. We reverse the judgment and remand the case to the Court of Appeal with directions that it remand the case to the trial court for it to enter an order granting Vivar‘s section 1473.7 motion to withdraw his plea.

CUÉLLAR, J.

We Concur:

LIU, J.
KRUGER, J.
GROBAN, J.
JENKINS, J.

Concurring and Dissenting Opinion by Justice Corrigan

I concur in the disposition and fully agree that defendant Robert Landeros Vivar should be allowed to withdraw his plea under Penal Code section 1473.7.¹ Defense counsel‘s failure to properly advise Mr. Vivar about the immigration consequences of his plea options was plainly prejudicial, entitling him to reliеf under the statute. I join in Justice Cuéllar‘s majority opinion to the extent it rejects the notion that the standard of review applied to rulings under section 1473.7 differs depending upon the nature of the claimed error. (Maj. opn., ante, pp. 16–17.)

However, I respectfully dissent from the majority‘s holding that a form of “independent review” typically applied in habeas corpus proceedings should be applied in assessing a trial court‘s prejudice finding under section 1473.7. (Maj. opn., ante, pp. 15, 29.) While the majority opinion endeavors to distinguish independent review in this context from de novo review (id. at p. 19), as a practical matter this will be a distinction without a difference in most cases brought under section 1473.7. That is so because, as the majority acknowledges, the evidence considered by the trial court will often consist entirely of declarations and documentary evidence. (See maj. opn., ante, pp. 17–18.) Under the version of independent review articulated in the majority opinion, no deference is owed to the trial court‘s factual findings except when credibility determinations are based upon live testimony. (Id. at pp. 19–20.) Consequently, in a great number of appeals challenging section 1473.7 rulings, reviewing courts will assume the role of fact finder, requiring courts to resolve factual conflicts, weigh evidence, and engage in the type of factual inquiry ordinarily reserved for trial courts. This would constitute a departure for review of a ruling on a statutory motion. I would hold that appеllate courts should apply a conventional substantial evidence standard when reviewing a trial court‘s factual findings that bear upon the prejudice analysis under section 1473.7.

This court granted review to resolve a conflict over the standard of review governing prejudice findings under the statute and to clarify what constitutes prejudice under section 1473.7. (Maj. opn., ante, p. 10.) In the Court of Appeal, the defense argued that all aspects of the trial court‘s ruling, including its factual findings, should be reviewed independently. The Attorney General contended that the trial court‘s ruling should be reviewed for abuse of discretion but that its factual findings were owed deference, even if made on a cold record consisting entirely of documentary evidence.²

The Court of Appeal created a hybrid standard turning on the asserted basis for relief. According to the appellate court, if the section 1473.7 motion raised a constitutional challenge due to ineffective assistance of counsel, a reviewing court should ” ‘independently review the order.’ ” (People v. Vivar (2019) 43 Cal.App.5th 216, 224.) This standard requires courts to ” ‘accord deference to the trial court‘s factual determinations if supported by substantial evidence in the record, but [to] exercise . . . independent judgment in deciding whether the facts demonstrate trial counsel‘s deficient performance and resulting prejudice to the defendant.’ ” (Ibid.) By contrast, the court held that the denial of a section 1473.7 motion is ” ‘reviewed for an abuse of discretion’ ” if thе basis for the motion is ” ‘statutory error or a deprivation of statutory rights’ ” not rising to the level of a constitutional violation. (Vivar, at p. 224.)

In my view, the majority opinion correctly rejects the bifurcated approach adopted by the Court of Appeal. (Maj. opn., ante, pp. 16–17.) Such an approach would afford undue significance to a distinction the Legislature sought to erase. It extended relief to all defendants when legal error prevented meaningful understanding of immigration consequences, regardless of whether the error constitutes ineffective assistance of counsel. (Id. at p. 16; see § 1473.7, subd. (a)(1).) The bifurcated approach would also have the standard of review turn on a specific finding of ineffective assistance of counsel, which a trial court might not otherwise reach for reasons unrelated to the merits of the claim. (Maj. opn., ante, p. 17.)

However, I part ways with the majority conclusion that the trial court‘s prejudice finding under section 1473.7 is subject to a form of “independent review” that does not defer to the trial court‘s factual findings under conventional substantial evidence review. (See maj. opn., ante, pp. 29–30.) The majority holds that the independent review standard affords deference to the trial court‘s factual determinations only if “based on ’ “the credibility of witnesses the [superior court] heard and observed.” ’ ” (Id. at p. 19.) Under that view, no deference is owed to the trial court‘s factuаl findings when the “facts derive entirely from written declarations and other documents.” (Id. at p. 20.) My disagreement is not with applying independent review to the trial court‘s ultimate legal ruling but with adopting a form of review that largely dispenses with the deference normally afforded to a lower court‘s factual findings. (See People v. Hernandez (2008) 45 Cal.4th 295, 298–299; People v. Alvarez (1996) 14 Cal.4th 155, 182.) A substantial evidence inquiry examines the record in the light most favorable to the judgment and upholds a finding “if the record contains reasonable, credible evidence of solid value upon which a reasonable trier of fact could have relied in reaching the conclusion in question. Once such evidence is found, the substantial evidence test is satisfied. [Citation.] Even when there is . . . significant . . . countervailing evidence, the testimony of a single witness that satisfies the standard is sufficient to uphold the finding.” (People v. Barnwell (2007) 41 Cal.4th 1038, 1052.) There is no reason to dispense with this conventional appellate approach to reviewing findings of fact.

The standard of review described in the majority opinion is not completely unheard of, but its extension to review of statutory claims may well be.³ The standard articulated by the majority derives primarily from the lead opinion in In re Resendiz (2001) 25 Cal.4th 230, 249 (lead opn. of Werdegar, J.) (Resendiz).⁴ (Maj. opn., ante, pp. 14, 19.) Resendiz considered an ineffective assistance of counsel claim involving affirmative misadvice about the immigration consequences of a plea. (Resendiz, at p. 235.) The claim was raised on habeas corpus. Accordingly, the lead opinion recited the standard applicable to review of faсtual findings in habeas corpus proceedings. (Id. at p. 249.) Of course, in such a context the inquiry goes beyond the trial record to consider facts and assertions not before the original trial court. In such circumstances, a court considering a habeas petition is not simply reviewing the decisions made at the trial level. It is exercising its own authority based on its own review of new facts and claims. Those are two very different tasks. We should be hesitant here to uncritically apply a habeas corpus standard of review to appellate review of statutory claims.

The review of factual findings in habeas corpus matters arises from the procedural posture of those cases. Constitutionally, the courts of review are granted original jurisdiction to consider habeas corpus claims. (Cal. Const., art. VI, § 10.) Because appellate courts are not well suited to conduct evidentiary hearings, however, a referee will typically be appointed to make recommended findings of fact. (See Cal. Rules of Court, rule 8.386(f)(2).) But reviewing courts are not required to accept the referee‘s recommended findings. (In re Hitchings (1993) 6 Cal.4th 97, 109.) While those findings are entitled to ” ‘great weight’ ” when supported by substantial, credible evidence (ibid.), they are not binding upon the court as they would be under the substantial evidence standard. A court may ” ’ “reach a different conclusion on an independent examination of the evidence produced at the [reference hearing] even where the evidence is conflicting.” ’ ” (Ibid.) Further, no deference is afforded to factual findings unless ” ‘based on the credibility of live testimony.’ ” (Resendiz, supra, 25 Cal.4th at p. 249 (lead opn. of Werdegar, J.), citing In re Arias (1986) 42 Cal.3d 667, 695; accord, In re Long (2020) 10 Cal.5th 764, 774.)

When an appellate court exercises original jurisdiction in a habeas matter, it makes sense to give limited deference to the referee‘s recommended findings. In keeping with its original jurisdiction, the reviewing court is the ultimate fact finder. The same review principles apply to a successive writ situation in which a petitioner files a new habeas corpus petition in the appellate court when the superior court has denied habeas corpus relief after an evidentiary hearing. (In re Wright (1978) 78 Cal.App.3d 788, 801.) In such a case, the appellate court again exercises original jurisdiction.

The situation is different, however, when the matter comes to the appellate court as an appeal. When the superior court grants habeas corpus relief and the People appeal, the Court of Appeal exercises its appellate jurisdiction over the superior court rulings. (See Cal. Const., art. VI, § 11.) “The posture of [a] case as a People‘s appeal is to be differentiated from a situation in which an appellate court, reviewing a petition for writ of habeas corpus as a matter of original jurisdiction, assigns a referee to take evidence on the matter.” (In re Pratt (1999) 69 Cal.App.4th 1294, 1314, fn. 16, italics added.) In an appeal from a habeas corpus grant, a reviewing court applies the conventional substantial evidence standard to questions of fact, just as with any other appeal. (Id. at p. 1314.) In other words, it applies ” ’ “basic principles of appellate review.” ’ ” (In re Butler (2020) 55 Cal.App.5th 614, 648.) Findings of fact are accorded due deference under the substantial evidence standard, while questions of law are reviewed independently. (Ibid.)

An appeal from a ruling under section 1473.7 is just that: an appeal. (§ 1473.7, subd. (f).) It is not an equitable habeas corpus proceeding in which the appellate court possesses original jurisdiction. Indeed, the statutory remedy in section 1473.7 is necessary because habeas corpus writ relief is not available when, as here, the defendant is no longer in actual or constructive custody. (See People v. Villa (2009) 45 Cal.4th 1063, 1066.) In creating an opportunity for legal relief under section 1473.7, the Legislature also provided for conventional appellate review. It did not expand the jurisdiction of the courts reviewing the trial court‘s ruling on the motion. Basic principles of appellate review should apply, not principles imported from writ proceedings.

“[A]ppellate court deference to the trial court‘s resolution of fact issues is warranted by jurisdictional considerations and a recognition of the distinctive roles of trial and appellate courts: Trial courts decide questions of fact and appellate courts decide questions of law.” (Eisenberg et al., Cal. Practice Guide: Civil Appeals ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍and Writs (The Rutter Group 2019) ¶ 8:42, p. 8-21.) Whether substantial evidence supports a judgment or ruling is a question of law reposing with the appellate court. (Ibid.) Further, as a general matter, because of the jurisdictional roles of the trial and appellate courts, deference to trial court credibility determinations is the same for both written declarations and oral testimony. (Haraguchi v. Superior Court (2008) 43 Cal.4th 706, 711 & fn. 3; Shamblin v. Brattain (1988) 44 Cal.3d 474, 479; Lebel v. Mai (2012) 210 Cal.App.4th 1154, 1159.)

The argument that the reviewing court is ” ‘in the same position’ ” as the trial court in assessing documentary evidence is inaccurate. (Maj. opn., ante, p. 20, quoting People v. Ogunmowo (2018) 23 Cal.App.5th 67, 79.) A reviewing court exercising appellate jurisdiction is not in the same position as the trial court. Their respective roles are different. The trial court decides questions of fact in the first instance. The reviewing court defers to those findings and only considers legal holdings de novo. By declining to give deference to the trial court‘s findings when based on documentary evidence, a reviewing court simply assumes for itself the role of fact finder. For this reason, we confirmed over a decade ago that even when “the trial court‘s findings were based on declarations and other written evidence[, that fact] does not lessen the deference due those findings.” (Haraguchi v. Superior Court, supra, 43 Cal.4th at p. 711, fn. 3.) Indeed, in clarifying that deference is owed “whether the trial court‘s ruling is based on oral testimony or declarations,” this court has expressly disapproved authority suggesting otherwise. (Shamblin v. Brattain, supra, 44 Cal.3d at p. 479; see id. at p. 479, fn. 4.)

In its adoption of independent review like that described in Resendiz, the majority opinion cites a number of factors, including the “history of section 1473.7.” (Maj. opn., ante, p. 18.) That history purportedly reflects that the standard articulated by the majority was the “prevailing independent standard of review” applied to section 1473.7 when the Legislature amended the law in 2018. (Maj. opn., ante, p. 16.) However, the precise contours of the standard were far from clear at that time. In People v. Olvera (2018) 24 Cal.App.5th 1112, although the court cited Resendiz and referred to independent review of the section 1473.7 ruling, it characterized the standard as follows: “We defer to the trial court‘s factual determinations if supported by substantial evidence, but exercisе our independent judgment to decide whether the facts demonstrate deficient performance and resulting prejudice.” (Olvera, at p. 1116, italics added.) Likewise, in another 2018 case involving section 1473.7, People v. Tapia (2018) 26 Cal.App.5th 942, the court cited the independent review standard but applied conventional substantial evidence review to the trial court‘s factual findings, even though the record apparently consisted entirely of declarations and documents.⁵ (Tapia, at pp. 946–948, 951, 953.) Indeed, the appellate court upheld the trial court‘s implied finding that the defendant‘s declaration was not credible, stating: “We do not reevaluate witness credibility.” (Id. at p. 953.)

The standard described in People v. Olvera and applied in People v. Tapia is not the standard advocated by the majority. The majority approach does not involve substantial evidence review of factual findings, even as applied to findings based on live testimony. According great weight to findings is not the same as being bound by findings supported by substantial evidence. A court applying the standard adopted by the majority ⁵

is free to reach its own conclusions even when the evidence is conflicting and ” ‘great weight’ ” is afforded to certain findings. (Resendiz, supra, 25 Cal.4th at p. 249 (lead opn. of Werdegar, J.).) Simply put, it is incorrect to say the independent review standard adopted by the majority was the “prevailing” one.

As further support for dispensing with deferential review of factual findings, the majority cites “the interests at stake in a section 1473.7 motion.” (Maj. opn., ante, p. 18.) It may be appropriate to apply de novo review to mixed questions of law and fact that raise constitutional сoncerns or that would constitute a final determination of a party‘s rights. (People v. Ault (2004) 33 Cal.4th 1250, 1266.) Nevertheless, simply because independent review should be applied to the ultimate ruling does not justify giving factual findings less deference than they are owed under the substantial evidence test. Further, courts should not be free to disregard factual findings because they conclude the “interests at stake” in a particular case justify that approach. In most criminal cases, the “interests at stake” are high. Questions of guilt or innocence or touching on personal freedom are profoundly consequential. But reviewing courts are not free to disregard settled authority, or to expand the nature of their jurisdiction, simply by pronouncing: “This is really important.”

The majority opinion asserts that ” ’ “[i]ndependent review is not the equivalent of de novo review. . . .” ’ ” (Maj. opn., ante, p. 19, quoting People v. Jackson, supra, 128 Cal.App.4th at p. 1021.) Yet it does little to explain how, in practice, the standards will differ as applied to section 1473.7 rulings. The majority opinion distinguishes the original appraisal of all the evidence under the de novo standard from deference given to “factual findings that are based on the trial court‘s own observations” under independent review. (Maj. opn., ante, p. 19.) But many section 1473.7 proceedings will be based on documentary evidence without live testimony. Under those circumstances, the form of independent review described by the majority will for all practical purposes be de novo review involving an original appraisal of all the evidence. Then, a reviewing court will be thrust into the role of fact finder, requiring credibility assessments and a weighing of the evidence. A simple statement that the standards will somehow be different provides no guidance and sows confusion. We should hesitate to adopt a rule placing the reviewing court into a fact finder‘s position.

As the majority opinion notes, in this case the trial court did not even consider whether Mr. Vivar suffered prejudice, instead basing its ruling on the finding that his counsel did not provide ineffective assistance. (Maj. opn., ante, p. 8.) Simply put, the court made no express or implied factual findings with respect to prejudice. As a result, there is no finding to which to defer. To the extent the trial court might arguably have made findings bearing on prejudice, they would be easily dismissed.

The prejudice question turns on whether Mr. Vivar would not have entered the plea had he been properly informed and advised. This is a credibility question. The trial court found his credibility wanting because, it determined, he ” ‘was more willing to rely on his experiences than he was on his counsel‘s advice.’ ” (People v. Vivar, supra, 43 Cal.App.5th at p. 230.) The appellate court concluded, “This was a factual inference the trial court was entitled to draw. . . .” (Ibid.) Perhaps, but appellate review of that inference is not meaningless. The Court of Appeal was required to examine whether the inference found substantial support in the record. It does not. There was no evidence that counsel ever gave Mr. Vivar advice regarding immigration. Indeed, there is no evidence counsel understood the potential consequences herself or that she had made it her “business to discover what impact his negotiated sentence would have on his deportability.” (People v. Soriano (1987) 194 Cal.App.3d 1470, 1480.) The public defender did not claim she gave him any substantive information at all about immigration consequences. Indeed, she made no assertion as to any advice she provided Mr. Vivar. She said that she customarily told her noncitizen clients about ” ‘possible’ ” immigration consequences. (People v. Vivar, supra, 43 Cal.App.5th at p. 222.) But here, the unrebutted evidence was that she never asked Mr. Vivar about his immigration status. (Ibid.) Even if she had told him that he “might” get deported and, if he had further questions he should consult an immigration attorney, essentially that was no advice at all. (See id. at pp. 222–223.) This evidence, together with the record as whole, does not support an inference that Mr. Vivar would have ignored his counsel‘s advice on the immigration consequences of his plea. He received no such advice, and if he had, the proper inference to be drawn from his concern about deportation is that he would have accepted it or at least given it a fair degree of consideration. He was forced to rely on his own experiences only beсause counsel gave him no alternative. Because the record does not support the trial court‘s factual conclusions, a conventional substantial evidence review suffices here.

CORRIGAN, J.

I Concur:

CANTIL-SAKAUYE, C. J.

See next page for addresses and telephone numbers for counsel who argued in Supreme Court.

Name of Opinion People v. Vivar

Procedural Posture (see XX below)

Original Appeal
Original Proceeding
Review Granted (published) XX 43 Cal.App.5th 216
Review Granted (unpublished)
Rehearing Granted

Opinion No. S260270
Date Filed: May 3, 2021

Court: Superior
County: Riverside
Judge: Bambi J. Moyer

Counsel:

Munger, Tolles & Olson, Joseph D. Lee, William Larsen and Dane P. Shikman for Defendant and Appellant.

Gibson, Dunn & Crutcher, Kahn A. Scolnick, Daniel R. Adler and Jason S. Kim for Alyssa Bell, Reuven Cohen, Ingrid V. Eagly, Gilbert Garcetti, Meline Mkrtichian, Ronald J. Nessim, Gabriel Pardo, Jennifer Resnik and David J. Sutton as Amici Curiae on behalf of Defendant and Appellant.

Jennifer L. Pasquarella, Eva L. Bitran; Vasudha Talla; and David Loy for ACLU Foundation of Southern California, ACLU Foundation of Northern California and ACLU Foundation of San Diego and Imperial Counties as Amici Curiae on behalf of Defendant and Appellant.

O‘Melveny & Myers and Catalina J. Vergara for The Immigrant Legal Resource Center, Public Counsel, University of California Irvine Law Immigrant Rights Clinic, University of California Irvine Law Criminal Justice Clinic, East Bay Community Law Center, Community Legal Services in East Palo Alto and University of California Davis Immigrant Rights Clinic as Amici Curiae on behalf of Defendant and Appellant.

Xavier Becerra, Attorney General, Michael J. Mongan, State Solicitor General, Lance E. Winters and Gerald A. Engler, Chief Assistant Attorneys General, Samuel P. Siegel, Deputy State Solicitor General, Julie L. Garland, Assistant Attorney General, Melissa Mandel and Adrian R. Contreras, Deputy Attorneys General, for Plaintiff and Respondent.

Counsel who argued in Supreme Court (not intended for publication with opinion):

Dane Shikman
31 Mullen Ave.
San Francisco, CA 94110
(415) 512-4092

Samuel P. Siegel
Deputy State Solicitor General
1300 I Street
Sacramento, CA 95814
(916) 210-6269

Notes

Vivar says counsel informed him that the court could reduce the felony conviction to a misdemeanor — and Vivar believed that this disposition carried no immigration consequences. Further statutory references are to the Penal Code.

According to Vivar‘s immigration law expert, this conviction “triggered the worst of all immigration consequences: mandatory deportation with a bar to almost all forms of immigration relief, and permanent ineligibility for U.S. citizenship.” In this court, the Attorney General takes the position that appellate courts should independently review rulings under section 1473.7.

Because Justice Mosk concurred in Justice Werdegar‘s lead opinion in all respects relevant here (see (conc. & dis. opn. of Mosk, J.)), we cite only to the lead opinion. It should be noted that the majority explicitly limits its holding to review under section 1473.7. (Maj. opn., ante, p. 20, fn. 7.)

Because we adopt an independent standard of review for all claims made under section 1473.7, subdivision (a)(1), we disapprove , , and to the extent they are inconsistent with this opinion. The “independent review” standard described in the majority opinion also is applied to cases raising First Amendment issues in which an appellate court is charged with ensuring that a ruling does not intrude on constitutional free speech rights. (See [cited by maj. opn., ante, p. 19]; see also .) Because this case does not involve a First Amendment issue, these cases do not bear upon the standard of review that should be applied here.

Despite the passage of time, a trial court nonetheless retains the discretion to conduct an evidentiary hearing to resolve disputes of fact. (See .) In Tapia, the trial judge who heard the section 1473.7 motion also presided over the plea hearing. (See , 26 Cal.App.5th at p. 948.) While the majority opinion notes this fact (maj. opn., ante, p. 20) and presumably would give some degree of deference to the trial court‘s factual findings in such ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍a case, that deference still would not be the equivalent of the substantial evidence standard applied in Tapia.

The separate opinion correctly characterizes substantial evidence review as “deferential.” (Conc. & dis. opn., post, at p. 10.) But it doesn‘t follow that every time a court extends deference to a trial court‘s factual findings, it‘s engaging in substantial evidence review. (See, e.g., (lead opn. of Werdegar, J.); [citing Resendiz];

Cal.App.5th at p. 950 [citing Resendiz and quoting Ogunmowo]; Olvera, supra, 24 Cal.App.5th at p. 1116 [citing Resendiz and Ogunmowo].) What‘s distinctive about substantial evidence review is that adequately supported factual findings not only merit deference, but are binding, on appeal. (See ; cf. [while courts ” ‘generally defer to the referee‘s factual findings,’ ” they “are not binding“].) Our decision addresses only the independent standard of review under section 1473.7. Nothing we say here disturbs a familiar postulate: when reviewing a ruling under the substantial evidence standard, “an appellate court should defer to the factual determinations made by the trial court,” regardless of “whether the trial court‘s rulings are based on oral testimony or declarations.” (; see , 43 Cal.4th at pp. 711, 713.)

Read the detailed case summary