People v. Garcia

BAKER, J., Dissenting

Justice Corrigan warned there would be cases like this. In a case of recent vintage, *1008People v. Villatoro (2012) 54 Cal.4th 1152, 144 Cal.Rptr.3d 401, 281 P.3d 390 ( Villatoro ), a majority of our Supreme Court held a trial court properly instructed jurors they could consider evidence a defendant committed one or more charged sex crimes as evidence the defendant was predisposed to commit additional charged sex crimes that the prosecution was tasked with proving in the very same criminal trial. ( Id. at pp. 1164-1165, 144 Cal.Rptr.3d 401, 281 P.3d 390.) Justice Corrigan disagreed, explaining the court's holding "sows the seeds for confusion and unintended consequences." ( Id. at p. 1169, 144 Cal.Rptr.3d 401, 281 P.3d 390 (conc. & dis. opn. of Corrigan, J.).) Justice Corrigan specifically cautioned that confusion was inevitable if jurors were told they should use the preponderance of the evidence standard to decide whether a defendant committed one or more charged crimes, so as to infer he or she had a propensity to commit other similar charged crimes, but simultaneously told to somehow evaluate whether the defendant was guilty of all the charged crimes by using the constitutionally required beyond a reasonable doubt standard of proof. ( Id. at p. 1181, 144 Cal.Rptr.3d 401, 281 P.3d 390 (conc. & dis. opn. of Corrigan, J.) ["If the same preponderance standard is applied to charged offenses, ... there is a serious risk of confusion. Requiring the jury to apply two standards of proof to evidence of the same crime would inevitably lead to confusion and could potentially erode the presumption of innocence"].)

The jurors in this case were told precisely what Justice Corrigan foresaw as inevitably confusing. And that is not all. The trial court gave a propensity instruction at the prosecution's behest even though the charges and evidence in this case were nowhere near as amenable to a propensity inference as the charges and evidence in Villatoro . The ultimate consequence was a hopelessly circular jury instruction, along with an erosion of the presumption of innocence and the beyond a reasonable doubt standard of proof.

The majority nevertheless affirms the judgment of conviction. To arrive at that result, the majority expressly disagrees with a recent Court of Appeal decision that-correctly, in my view-holds a propensity instruction of the type given in this case asks far too much of lay jurors and, for practical purposes, lowers the standard of proof to determine guilt in a criminal trial. ( People v. Cruz (2016) 2 Cal.App.5th 1178, 1186, 206 Cal.Rptr.3d 835 ( Cruz ).) The majority also does not account for another Court of Appeal decision that helps to highlight the problems with the trial court's propensity instruction when considering the evidence in this case-particularly, the dissimilar nature of the various *801charges and the jumbled chronology of the victim's testimony. ( People v. Nicolas (2017) 8 Cal.App.5th 1165, 214 Cal.Rptr.3d 467 ( Nicolas ).) And furthermore, the majority mistakenly invokes a 30-year-old United States Supreme Court case to cast aside our Supreme Court's direction just five years ago that "[a]n instruction that effectively lowers the prosecution's burden of proving guilt beyond a reasonable doubt is structural error" that is not subject to review for harmlessness. ( *1009People v. Aranda (2012) 55 Cal.4th 342, 365, 145 Cal.Rptr.3d 855, 283 P.3d 632 ( Aranda ).) For my part, I believe reversal is required, and I write separately to further explain these three areas of disagreement.

I

Evidence of a person's character, i.e., evidence that a person is predisposed to act in a certain way, is generally inadmissible to prove the person acted in conformity with that character trait on a given occasion. ( Evid. Code, § 1101, subd. (a) ; Villatoro , supra , 54 Cal.4th at p. 1159, 144 Cal.Rptr.3d 401, 281 P.3d 390.) Evidence Code section 1109, however, is one of the few exceptions the Legislature has made to the general ban on character evidence. The statute provides that "in a criminal action in which the defendant is accused of an offense involving domestic violence, evidence of the defendant's commission of other domestic violence is not made inadmissible by [Evidence Code] Section 1101 if the evidence is not inadmissible pursuant to [Evidence Code] Section 352." ( Evid. Code, § 1109, subd. (a)(1).) In other words, "evidence of a prior act of domestic violence is admissible to prove the defendant had a propensity to commit domestic violence when the defendant is charged with an offense involving domestic violence. The trial court has discretion to exclude the evidence if its probative value is outweighed by a danger of undue prejudice or confusing the jury, or would result in an undue consumption of time." ( People v. Rucker (2005) 126 Cal.App.4th 1107, 1114, 25 Cal.Rptr.3d 62.)

Prior to 2012, California criminal cases that confronted questions concerning the admission of propensity evidence most often involved evidence of other uncharged acts by a defendant, meaning acts that did not serve as the basis for the criminal charges in the prosecution at issue. (See, e.g., People v. Reliford (2003) 29 Cal.4th 1007, 1013-1014, 130 Cal.Rptr.2d 254, 62 P.3d 601 ( Reliford ); People v. Anderson (2012) 207 Cal.App.4th 1440, 1484-1485, 144 Cal.Rptr.3d 606 ( Anderson ); People v. Reyes (2008) 160 Cal.App.4th 246, 250-253, 72 Cal.Rptr.3d 586 ( Reyes ); see also Villatoro , supra , 54 Cal.4th at p. 1160, 144 Cal.Rptr.3d 401, 281 P.3d 390.) Many of these cases, like the Reliford and Anderson cases, were decided under Evidence Code section 1108, which permits introduction of propensity evidence in sex offense prosecutions in the same way Evidence Code section 1109 does in domestic violence prosecutions.

The rule these cases settled on is that the prosecution need only prove the defendant committed other uncharged acts by a preponderance of the evidence to permit the jury to consider those uncharged acts when deciding whether the defendant had been proven guilty of the charged offenses beyond a reasonable doubt. (See, e.g., Reliford , supra , 29 Cal.4th at p. 1016, 130 Cal.Rptr.2d 254, 62 P.3d 601 [rejecting argument that the preponderance standard for other uncharged acts lowered the prosecution's burden of proof as to the charged crimes because "jurors can grasp their duty-as stated in the instructions-to apply the *1010preponderance-of-the-evidence standard to the preliminary fact identified in the instruction and to apply *802the reasonable-doubt standard for all other determinations"]; Anderson , supra , 207 Cal.App.4th at p. 1484, 144 Cal.Rptr.3d 606 [noting the United States Supreme Court has adopted the preponderance standard for proof of uncharged bad acts]; Reyes , supra , 160 Cal.App.4th at pp. 250, 253, 72 Cal.Rptr.3d 586 [rejecting argument that instruction on use of uncharged acts of domestic violence impermissibly lowered the prosecution's burden to prove the charged offenses].)

In 2012, however, our Supreme Court decided Villatoro , supra , 54 Cal.4th 1152, 144 Cal.Rptr.3d 401, 281 P.3d 390. The question presented in that case was whether Evidence Code section 1108 permits juries to consider not only uncharged sexual offenses as evidence of guilt, but whether juries may also "consider the defendant's charged sexual offenses as evidence of his propensity to commit the other charged sexual offenses." ( Id. at p. 1156, 144 Cal.Rptr.3d 401, 281 P.3d 390.) Specifically, the defendant was charged with raping five women on separate occasions between 2005 and 2008, and the question was whether the trial court appropriately instructed the jury that it could "use evidence of [the] defendant's guilt of [any] one of the charged sexual offenses as evidence of his propensity to commit the other charged sexual offenses." ( Id. at pp. 1156, 1158, 144 Cal.Rptr.3d 401, 281 P.3d 390.)

The Villatoro court held Evidence Code section 1108 does authorize juries to "draw an inference of propensity from other charged offenses" and the trial judge properly instructed the jury it could do just that. ( Villatoro , supra , 54 Cal.4th at pp. 1164-1166, 1168, 144 Cal.Rptr.3d 401, 281 P.3d 390 ; Cruz , supra , 2 Cal.App.5th at p. 1185, 206 Cal.Rptr.3d 835 ["[O]ur Supreme Court held that it ... allows evidence of sexual offenses charged in the current prosecution to be used to show a propensity to commit other charged offenses in the same case"].) But key to the Court's holding in Villatoro was the specific wording of the jury instruction given by the trial judge in that case.14 The instruction informed the jury "all offenses must be proven beyond a reasonable doubt, even those used to draw an inference of propensity," such that there was "no risk the jury would apply an impermissibly low standard of proof." ( Villatoro , supra , at pp. 1167-1168, 144 Cal.Rptr.3d 401, 281 P.3d 390.)

*1011In response to the Villatoro decision, the CALJIC committee revised its pattern instruction numbers 2.50.01 and 2.50.02, which concern the admission of propensity evidence under sections 1108 and 1109, respectively. Before Villatoro , both instructions made reference only to the potential to use uncharged acts as propensity evidence; shortly after Villatoro , language was included in each of the instructions in *803an attempt (unsuccessful, in my view) to marry the standard for consideration of uncharged other acts and Villatoro 's holding permitting consideration of other charged crimes. ( CALJIC Nos. 2.50.01, 2.50.02 (Spring 2013 ed.) [citing Villatoro in the comment to the pattern instructions].)

The trial court in this case instructed the jury with CALJIC No. 2.50.02 as it existed at the time (it has recently been revised again in light of Cruz , as noted by the majority and post ). This was the instruction given to the jury, with emphasis on the portions that are relevant to the issues presented in this appeal:

"In determining whether defendant has been proved guilty of the any [sic] crime of domestic violence of which he she [sic] is charged, you should consider all relevant evidence, including whether defendant committed any other domestic violence crimes, whether charged or uncharged, about which evidence has been received. The crimes charged in Counts 1-8, may be considered by you in that regard.
" 'Domestic violence' means abuse committed against an adult or a fully emancipated minor who is a spouse, former spouse, cohabitant, former cohabitant, or person with whom the defendant has had a child or is having or has had a dating or engagement relationship.
[¶] ... [¶]
" 'Abuse' means intentionally or recklessly causing or attempting to cause bodily injury, or placing another person in reasonable apprehension of imminent serious bodily injury to himself or herself, or another.
"If you find by a preponderance of the evidence that the defendant committed any such offense involving domestic violence, you may, but are not required to, infer that the defendant had a disposition to commit another other [sic] offenses involving domestic violence. If you find that the defendant had this disposition, you may, but are not required to, infer that he she [sic] was likely to commit and did commit the crime or crimes of which he she [sic] is accused.
"However, even though you find by a preponderance of the evidence that the defendant committed a prior crime or crimes involving domestic violence, that is not sufficient by itself to prove beyond a reasonable doubt that he ...
*1012committed the offenses you are determining. If you determine an inference properly can be drawn from this evidence, this inference is simply one item for you to consider, along with all other evidence, in determining whether the defendant has been proved guilty beyond a reasonable doubt of the charged crime that you are determining."

Thus, as worded, the first paragraph instructed the jury that all of the crimes charged against defendant were "domestic violence crimes." That is because the instruction defined "other domestic violence crimes, whether charged or uncharged" to include each and every count of the information against defendant. Of course, this meant the instruction would permit the jury to use its findings on any individual count (e.g., dissuading a witness) as reason to convict defendant on any of the other counts (e.g., disobeying a domestic relations court order). The fourth paragraph of the instruction, as quoted above, explained the conditions under which the jury could do this. That is, paragraph four told the jury it could rely on its findings on one or more of the counts to influence its verdict on other counts (i.e., it could conclude defendant "was likely to commit and did commit" the charged crimes) if it found one or more of the counts proven by a preponderance *804of the evidence and used that finding to infer defendant had a disposition to commit the other offenses. At the same time, the final paragraph of the instruction cautioned the jury that if it found by a preponderance of the evidence that defendant committed "a prior crime or crimes involving domestic violence," that would not be sufficient by itself to prove beyond a reasonable doubt that defendant "committed the offenses you are determining."

The Cruz decision holds that giving this sort of instruction is error. In Cruz , supra , 2 Cal.App.5th 1178, 206 Cal.Rptr.3d 835, the defendant was charged with three counts of committing a lewd act against a child under age 14. ( Id. at pp. 1180-1181, 206 Cal.Rptr.3d 835.) The trial judge instructed the jury with the CALJIC-modeled sex crimes propensity instruction that is in all material respects identical to the domestic violence propensity instruction in this case. ( Id. at pp. 1183-1184, 206 Cal.Rptr.3d 835 [reciting the instruction given].)

Unlike the jury instruction given by the trial judge in Villatoro (but like the instruction here), the instruction in Cruz stated "the preponderance [of the evidence] standard applied to the determination of whether Cruz committed charged and uncharged offenses for the purpose of deciding whether he had a propensity to commit sexual offenses." ( Cruz , supra , 2 Cal.App.5th at p. 1185, 206 Cal.Rptr.3d 835.) As Cruz explains, this was problematic: "In effect, the instruction ... told the jury it should first consider whether the offenses charged in counts 1, 2, and 3 had been established by a preponderance of the evidence, while holding its ultimate decision on the same offenses in suspension. Then *1013the jury was required to decide whether the preponderance finding showed a propensity, and whether this propensity, in combination with the other evidence, proved those offenses a second time, this time beyond a reasonable doubt. [¶] We conclude the court was incorrect to instruct the jury in this way. Villatoro did not expressly hold that currently charged offenses must be proved beyond a reasonable doubt before they can be used to show a propensity under Evidence Code section 1108, but it strongly implied that rule. ... [¶] A robot or a computer program could be imagined capable of finding charged offenses true by a preponderance of the evidence, and then finding that this meant the defendant had a propensity to commit such offenses, while still saving for later a decision about whether, in light of all the evidence, the same offenses have been proven beyond a reasonable doubt. A very fastidious lawyer or judge might even be able to do it. But it is not reasonable to expect it of lay jurors. We believe that, for practical purposes, the instruction lowered the standard of proof for the determination of guilt." ( Id. at pp. 1185-1186, 206 Cal.Rptr.3d 835.)

The Cruz analysis, which largely parallels Justice Corrigan's earlier observation in Villatoro , should control here.15 As in *805Cruz , the instruction the trial court gave told the jury it should first consider whether all the charged offenses had been established by a preponderance of the evidence, then decide whether that preponderance finding demonstrated a propensity, and then consider whether this propensity, in combination with the other evidence, proved the same offenses again beyond a reasonable doubt. That "presented the jury with a nearly impossible task of juggling competing standards of proof during different phases of its consideration of the same evidence" and "the ultimate effect [was] to lower the prosecution's burden of proving guilt beyond a reasonable doubt." ( Cruz , supra , 2 Cal.App.5th at p. 1187, 206 Cal.Rptr.3d 835.) *1014II

Even putting aside the problem with the propensity instruction given as illuminated by Cruz , the instruction suffers from an additional weakness-one that appears when focusing on the evidence and charges in this case as contrasted with the evidence and charges at issue in other cases where a propensity instruction was given.

In Villatoro , our Supreme Court approved the use of a propensity instruction in a case where there were five alleged rapes, of five different victims, at five different times. ( Villatoro , supra , 54 Cal.4th at pp. 1156-1158, 144 Cal.Rptr.3d 401, 281 P.3d 390.) In Cruz , the trial judge gave a propensity instruction (albeit a defective one, for reasons already recounted) in a case where there were three child sex offenses involving separate conduct, against separate victims, on separate dates. ( Cruz , supra , 2 Cal.App.5th at pp. 1181-1182, 206 Cal.Rptr.3d 835.) Here, however, the offenses charged against defendant were not predicated on acts that are fully distinct, both factually and temporally.

The eight crimes charged against defendant were considered crimes of domestic violence, and were included in the trial court's propensity instruction, based on what was essentially a continuous course of conduct by defendant as described by victim Amanda during her trial testimony. The problem, however, is that Amanda's testimony (even if believed in its entirety) concerned alleged criminal acts that occurred over the course of mere hours within a single day, and her recounting of the alleged criminal acts was chronologically jumbled to the point that the order in which defendant committed the acts alleged was unclear. The Attorney General forthrightly acknowledges the problem, conceding that "Amanda was unable to remember the exact order of the incidents [that served as the basis for the charged crimes] because they occurred on the same day and [defendant] went back and forth to her apartment." Indeed, the manner in which Amanda testified at trial compels the Attorney General to disclaim that his attempt to even recite the facts of this case is "based on what appears to be the most accurate order based upon Amanda's disjointed testimony."

The logical principle that permits admission of propensity evidence-encapsulated in the text of Evidence Code section 1109 itself-is that a jury may draw an inference a defendant is prone to commit a charged crime from other incidents of a *806sufficiently similar nature. ( Evid. Code, § 1109, subd. (a)(1) ["[I]n a criminal action in which the defendant is accused of an offense involving domestic violence, evidence of the defendant's commission of other domestic violence is not made inadmissible ..."].) Considering Amanda's testimony in this case, there really was no "other" domestic *1015violence.16 Unlike Villatoro where there were five discrete rapes such that a jury could conclude, for instance, that the defendant had a propensity to commit a charged rape against victim N.G. in June 2006 because he raped another victim R.I. in May 2005 ( Villatoro , supra , 54 Cal.4th at pp. 1156-1157, 144 Cal.Rptr.3d 401, 281 P.3d 390 ), the jury in this case had no reliable basis to distinguish among the alleged criminal acts and instead was left with an inseparable tangle. Giving a propensity instruction under these circumstances permitted the jury to draw what, in effect, were self-reinforcing, circular inferences-allowing the jury to infer, for instance, that defendant was prone to disobeying a court order and beating Amanda based on the same evidence the jury would have to use to conclude he disobeyed a court order and beat her.

That is not the only problem on the specific facts of this case. Again unlike Villatoro , and even Cruz , the eight charges against defendant are not all similar in a way that the sex offenses in those cases very much were. ( Villatoro , supra , 54 Cal.4th at p. 1158, 144 Cal.Rptr.3d 401, 281 P.3d 390 [five counts of rape]; Cruz , supra , 2 Cal.App.5th at p. 1180 [three counts of committing a lewd act against a child].) Under the statutory definition of "domestic violence" ( Evid. Code, § 1109, subd. (d)(3) ), or at least the trial court's broad application of that definition, the propensity instruction in this case swept so broadly as to include every charge against defendant within its ambit-from burglary to injuring a cohabitant, robbery to disobeying a domestic relations court order.

While there is good reason to believe that a defendant who commits sex crimes or core domestic violence offenses may be disposed to commit similar crimes, that good reason is absent when some of the charges are as disparate as they are here. Only on the most aggressive application of propensity principles would a factfinder be justified in concluding that defendant's commission of a burglary, for instance, makes him particularly prone to dissuading a witness. That sort of aggressive application, however, is exactly what the trial court's propensity instruction permitted despite being inconsistent with the public policy and fundamental fairness reasons why courts have long recognized the danger of permitting juries to consider propensity evidence. (See People v. Falsetta (1999) 21 Cal.4th 903, 913, 89 Cal.Rptr.2d 847, 986 P.2d 182 [observing that the rule excluding criminal propensity evidence is nearly three centuries old].) The instruction at issue permitted the jury to use the prosecution's decision to charge defendant with multiple *1016crimes to infer defendant had a propensity to commit crimes generally, and to use that sort of inference for all seasons when deciding whether to find him guilty.

The Nicolas case, while not factually similar in all respects, helps to highlight these *807fact-specific problems with the trial court's propensity instruction. In Nicolas , the trial court gave the jury a propensity instruction that permitted using uncharged acts (sending text messages before a car collision resulting in death) to infer propensity if proven by a preponderance of the evidence, even though the text messages " 'were a continuous back-and-forth conversation leading up to [the] time of the collision ... [and] were an indivisible part of the offense itself.' " ( Nicolas , supra , 8 Cal.App.5th at p. 1178, 214 Cal.Rptr.3d 467.) Analogously, the course of conduct defendant engaged in, as described by Amanda, was such that the evidence at trial constituted an indistinguishable whole supporting all the charges, many of which bore little elemental similarity. The consequence was an unwieldy and erroneous jury instruction that, as in Nicolas , undercut the burden of proof and the presumption of innocence on each of the charged offenses. ( Id. at pp. 1181-1182, 214 Cal.Rptr.3d 467.)

III

Hedging its bets, the majority opinion holds that even if it was error to give the propensity instruction at issue, the error was harmless. I believe I must take our Supreme Court at its word, and that means I must conclude the instructional error here is not amenable to harmlessness review and is instead reversible per se.

Our Supreme Court in Aranda articulated the pertinent rule as follows: "An instruction that effectively lowers the prosecution's burden of proving guilt beyond a reasonable doubt is structural error" and reversible per se.17 ( Aranda , supra , 55 Cal.4th at p. 365, 145 Cal.Rptr.3d 855, 283 P.3d 632.) The courts in Cruz and Nicolas had no difficulty applying this rule to conclude the errors there-like the error here, in my view-required reversal without any effort to attempt an analysis of prejudice. ( Nicolas , supra , 8 Cal.App.5th at pp. 1181-1182, 214 Cal.Rptr.3d 467 ["Here, the trial court gave what the Attorney General concedes was an erroneous and unnecessary jury instruction. ... [¶] [U]nder the rule announced by our Supreme Court in Aranda ... and appropriately applied by the Court of Appeal in Cruz ..., we find that the trial court's erroneous jury instruction concerning 'uncharged offenses' constitutes structural error and is reversible per se"]; Cruz , supra , 2 Cal.App.5th at p. 1187, 206 Cal.Rptr.3d 835.)

*1017There is no basis to disregard the unqualified language of Aranda by resort to a case that had long been on the books when Aranda was decided: Rose v. Clark (1986) 478 U.S. 570, 106 S.Ct. 3101, 92 L.Ed.2d 460 ( Rose ). As an initial matter, I doubt that Rose is "instructive." Rose holds that a jury instruction that misallocates the burden of proof as to a single element of a charged crime can be amenable to harmlessness review ( id. at pp. 579-580, 106 S.Ct. 3101 ), but that is of course different from an instructional error that effectively lowers the burden of proof entirely. ( Aranda , supra , 55 Cal.4th at p. 365, 145 Cal.Rptr.3d 855, 283 P.3d 632 [citing Hedgpeth v. Pulido (2008) 555 U.S. 57, 129 S.Ct. 530, 172 L.Ed.2d 388 for the proposition that harmless-error analysis applies to instructional errors so long as the error does not categorically vitiate all the jury's findings].) In that circumstance, which is the case here, reliably assessing whether the jury would have properly found *808defendant guilty on the proof presented is prohibitively difficult, and a harmlessness inquiry runs the risk of degenerating into a usurpation of the role of a jury in a criminal trial.18 (See Kotteakos v. United States (1946) 328 U.S. 750, 763, 66 S.Ct. 1239, 90 L.Ed. 1557 ["[I]t is not the appellate court's function to determine guilt or innocence. ... Nor is it to speculate upon probable reconviction and decide according to how the speculation comes out"].) But my doubts aside, there is no direct conflict between Rose and Aranda , and settled stare decisis principles require this court to follow Aranda and leave to our Supreme Court the task of announcing it overlooked the implications of Rose when writing Aranda , if that is in fact what occurred. (See People v. Superior Court (Williams ) (1992) 8 Cal.App.4th 688, 702-703, 10 Cal.Rptr.2d 873 [" '[W]e are ... bound by [decisions of] the California Supreme Court ... unless the United States Supreme Court has decided the question differently' "], emphasis added.)

IV

Fortunately, the impact of today's decision is limited. As the majority recognizes, both the CALCRIM and CALJIC committees have revised their pattern jury instructions concerning the use of charged offenses to prove propensity. I submit that even those who are inclined to agree with the views espoused by the majority would see these clarifying revisions as a step in the right direction. The recent revisions also largely if not entirely eliminate the *1018issue of statewide importance this case would otherwise present. Nevertheless, the impact on this defendant remains, and I respectfully dissent from the affirmance of the judgment. As the revised pattern instructions help illustrate, the confusion generated by the earlier version of the propensity instruction at issue in this case was "entirely avoidable" ( Villatoro , supra , 54 Cal.4th at p. 1181, 144 Cal.Rptr.3d 401, 281 P.3d 390 (conc. & dis. opn. of Corrigan, J.)), and I see no reason to hold this instruction up as a model to which prosecutors and trial courts should aspire.

The instruction provided: " 'The People presented evidence that the defendant committed the crime of rape as alleged in counts 2, 4, 7, 9, 12 and 15 and the crime of sodomy as alleged in count 14. These crimes are defined for you in the instructions for these crimes. [¶] If you decide that the defendant committed one of these charged offenses, you may, but are not required to, conclude from that evidence that the defendant was disposed or inclined to commit the other charged crimes of rape or sodomy, and based on that decision also conclude that the defendant was likely to and did commit the other offenses of rape and sodomy charged. If you conclude that the defendant committed a charged offense, that conclusion is only one factor to consider along with all the other evidence. It is not sufficient by itself to prove the defendant is guilty of another charged offense. The People must still prove each element of every charge beyond a reasonable doubt and prove it beyond a reasonable doubt before you may consider one charge as proof of another charge.' " (Villatoro, supra, 54 Cal.4th at p. 1167, 144 Cal.Rptr.3d 401, 281 P.3d 390.)

Perhaps the most revealing sentence in the majority's opinion is this: "Echoing Justice Corrigan's dissent in Villatoro, our dissenting colleague essentially argues Cruz was correctly decided and Villatoro was not." (Maj. Opn. at p. 788-89.) That misses the point entirely. Justice Corrigan's opinion is significant here not because of the points on which it disagreed with the holding of Villatoro, but because it understood Villatoro "sow[ed] the seeds for confusion" in future cases if a jury received a propensity instruction concerning charged offenses that did not incorporate the beyond a reasonable doubt language that saved the instruction in Villatoro and instead incorporated preponderance of the evidence language used for other uncharged acts. (Villatoro, supra, 54 Cal.4th at pp. 1169, 1181, 144 Cal.Rptr.3d 401, 281 P.3d 390 (conc. & dis. opn. of Corrigan, J.).) The majority opinion does not recognize the distinction and, indeed, is founded on the notion that a charged offense is no different from an uncharged offense for purposes of crafting a propensity instruction. As the Cruz court understood, however, the distinction makes all the difference. Use of charged offenses to show propensity is what can introduce (absent the features of the instruction in Villatoro ) confusing, competing standards of proof that effectively lower the prosecution's burden to prove guilt.

In fact, because the jury had no reliable basis to determine in which order the alleged criminal acts by defendant occurred, the jury could not possibly have followed the propensity instruction's admonition that a finding "by a preponderance of the evidence that the defendant committed a prior crime or crimes involving domestic violence ... is not sufficient by itself to prove beyond a reasonable doubt that he ... committed the offenses you are determining." The jury had no way to determine which of the crimes were "prior."

Our Supreme Court's use of the word "effectively" indicates the rule applies beyond those circumstances in which a trial court gives a defective reasonable doubt instruction (see, e.g., Sullivan v. Louisiana (1993) 508 U.S. 275, 113 S.Ct. 2078, 124 L.Ed.2d 182 ).

For example, it is immaterial that "the jurors were told seven times that the burden of proof was on the prosecution to prove guilt beyond a reasonable doubt ...." (Maj. Opn. at p. 779.) While a court may properly conclude that a jury is likely to have followed a more specific instruction over a more general instruction, a court can only guess what burden of proof a jury applied simply by counting the number of references in a jury instruction packet and concluding the jury probably was swayed by those that are more numerous. When an error seriously threatens the beyond a reasonable doubt standard that is fundamental to the fairness of criminal trials, reviewing courts should require parties and trial courts to get it exactly right, not mostly right.