Opinion
No. D049993.
March 5, 2008. DEPUBLISHED: June 11, 2008
Appeal from the Superior Court of San Diego County, No. SCD195342, Albeit T. Harutunian III, Judge.
Christopher Blake, under appointment by the Court of Appeal, for Defendant and Appellant. Edmund G. Brown, Jr., Attorney General, Dane R. Gillette and Gary W. Schons, Assistant Attorneys General, Barry Carlton and Teresa Torreblanca, Deputy Attorneys General, for Plaintiff and Respondent.
OPINION
This appeal concerns two trials arising from Craig Smith's conduct of driving with alcohol in his system. Smith was arrested shortly after his vehicle was rear-ended by another vehicle. Approximately one hour after the accident, Smith's blood-alcohol level was 0.17 percent. Smith claimed this elevated level was the result of his drinking alcohol immediately after the accident. In the first trial, the jury acquitted Smith of driving with a blood-alcohol level of 0.08 percent or more (Veh. Code, § 23152, subd. (b), hereafter sometimes referred to as "per se DUI" (driving under the influence)), but could not reach a verdict as to whether he drove under the influence of alcohol (§ 23152, subd. (a), hereafter sometimes referred to as "generic DUI"). At the second trial, the jury convicted Smith of driving under the influence of alcohol. Challenging this conviction on appeal, Smith argues that at the second trial the court violated collateral estoppel principles by: (1) admitting the evidence that he had a postaccident 0.17 percent blood-alcohol level; (2) instructing the jury that a 0.08 percent or more blood-alcohol level creates a permissive presumption of driving under the influence; and (3) failing to instruct the jury that it should presume he did not drive with a blood-alcohol level of 0.08 percent or more. We reject Smith's argument that the 0.17 percent blood-alcohol evidence was inadmissible. However, we conclude that based on the interrelationship between the generic and per se DUI offenses, collateral estoppel principles were violated when the second jury was permitted to consider the issue of whether Smith drove with a 0.08 percent or more blood-alcohol level. Moreover, we agree with Smith that the jury should not have been instructed regarding the permissive presumption arising from a 0.08 percent or more alcohol level and should have been instructed to presume his blood-alcohol level was less than 0.08 percent while driving. We conclude the error requires reversal.
Subsequent statutory references are to the Vehicle Code unless otherwise specified.
FACTUAL AND PROCEDURAL BACKGROUND
On December 1, 2005, Smith's vehicle was rear-ended by another vehicle. Smith smelled of alcohol, had bloodshot, watery eyes, and performed poorly on field sobriety tests. He was arrested and taken to the police station, where a test performed approximately one hour after the accident showed a blood-alcohol level of 0.17 percent. Smith was charged with violating sections 23152, subdivision (a) (driving while under the influence of alcohol) and 23152, subdivision (b) (driving with a blood-alcohol level of 0.08 percent or greater). At trial, Smith called witnesses who testified that he had about two or three drinks at a bar; he was involved in an altercation with bar patron Alephonsion Deng; Deng followed Smith after Smith drove away from the bar; and shortly thereafter Deng rear-ended Smith's vehicle. Smith did not dispute that he had a 0.17 percent blood-alcohol level some time after the accident, but explained this elevated level by testifying that after the accident he was distressed and in response he drank brandy that he had in his vehicle. On cross-examination, the prosecution's expert acknowledged that a person's blood-alcohol level could rise to 0.17 percent one hour after rapidly consuming 12 ounces of brandy. A coffee cup containing alcohol was found in Smith's vehicle, and a liquid that appeared to be alcohol was spilled on the front passenger seat. In February 2006, a jury found Smith not guilty of the section 23152, subdivision (b) (per se DUI) offense, but could not reach a verdict on the section 23152, subdivision (a) (generic DUI) offense. The court declared a mistrial on the generic DUI count. In April 2006, retrial commenced on the generic DUI offense. Smith moved to limit the issues presented to the second jury. Based on the first jury's verdict acquitting him of driving with a blood-alcohol level of 0.08 percent or more, he argued that under collateral estoppel principles the second jury should not be allowed to consider whether he was driving with a blood-alcohol level of 0.08 percent or more. He requested that the court exclude evidence related to the 0.08 percent issue, including the evidence that his blood-alcohol level was 0.17 percent after the accident. Alternatively, he requested that the court (1) give the second jury a limiting instruction based on the first jury's acquittal, and (2) refrain from instructing the second jury regarding the permissive presumption of driving impairment arising from a blood-alcohol level of 0.08 percent or more. The trial court denied Smith's motions, ruling that the 0.08 percent issue could be considered by the second jury; the second jury should not be told about the first jury's acquittal on the 0.08 percent count; the 0.17 percent blood-alcohol test results were admissible; and the second jury would be instructed regarding the permissive presumption arising from a 0.08 percent or more blood-alcohol level. Accordingly, at the second trial the prosecution presented its case with no restrictions arising from the first jury's acquittal, and the jury was instructed that it could infer that Smith drove under the influence of alcohol if it found the prosecution proved he had a blood-alcohol level of 0.08 percent or greater. (See § 23610; Judicial Council of Cal. Crim. Jury Instns. (2007-2008) CALCRIM No. 2110.) The second jury convicted Smith of generic DUI under section 23152, subdivision (a). Smith appeals from this conviction, essentially reiterating the assertions he made at trial. He contends that based on collateral estoppel principles the 0.17 percent blood-alcohol evidence should not have been admitted at the second trial. Alternatively, he contends the second jury should not have been instructed regarding the permissive presumption of driving under the influence arising from a 0.08 percent or more blood-alcohol level, and the second jury should have been instructed to presume he did not drive with a 0.08 percent or more blood-alcohol level. To evaluate his contentions, we first summarize general collateral estoppel principles, and then apply these principles to the particular circumstances of this case.
DISCUSSION I. Collateral Estoppel Principles
The double jeopardy clause of the Fifth Amendment of the United States Constitution prohibits trying a defendant more than once for the same offense. ( People v. Santamaria (1994) 8 Cal.4th 903, 910 [ 35 Cal.Rptr.2d 624, 884 P.2d 81] ( Santamaria).) Collateral estoppel is a component of double jeopardy protection, prohibiting relitigation of factual issues when certain requirements are met. ( Id. at p. 912, fn. 3; Ashe v. Swenson (1970) 397 U.S. 436, 445 [ 25 L.Ed.2d 469, 90 S.Ct. 1189] ( Ashe).) The doctrine provides that "when an issue of ultimate fact has once been determined by a valid and final judgment, that issue cannot again be litigated between the same parties in any future lawsuit." ( Ashe, supra, at p. 443.) When applying the principle in criminal cases, the United States and California Supreme Courts have emphasized that "collateral estoppel . . . is not to be applied with [a] hypertechnical and archaic approach . . . but with realism and rationality. . . . The inquiry `must be set in a practical frame and viewed with an eye to all the circumstances of the proceedings.'" ( Id. at p. 444; accord, Santamaria, supra, 8 Cal.4th at pp. 920, 926.) Generally, collateral estoppel applies "`if (1) the issue necessarily decided at the previous trial is identical to the one which is sought to be relitigated; if (2) the previous trial resulted in a final judgment on the merits; and if (3) the party against whom collateral estoppel is asserted was a party or in privity with a party at the prior trial.'" ( Santamaria, supra, 8 Cal.4th at p. 916.) Additionally, "the issue to be precluded must be `an issue of ultimate fact. . . .'" ( Ibid.) To satisfy the "ultimate fact" requirement, the issue must pertain to a fact that the prosecution must prove beyond a reasonable doubt in the second trial. ( Dowling v. United States (1990) 493 U.S. 342, 348-349 [ 107 L.Ed.2d 708, 110 S.Ct. 668] ( Dowling); Santamaria, supra, 8 Cal.4th at p. 922; People v. Catlin (2001) 26 Cal.4th 81, 124 [ 109 Cal.Rptr.2d 31, 26 P.3d 357] ( Catlin); see U.S. v. Wells (8th Cir. 2004) 347 F.3d 280, 285 ["`A fact previously determined in a criminal case is not an `ultimate fact' unless it was necessarily determined by the jury against the government and, in the second prosecution, that same fact is required to be proved beyond a reasonable doubt in order to convict.'"].) A review of several United States and California Supreme Court decisions reveals that application of the collateral estoppel rule is heavily dependent on the factual and legal context of the particular case. In Ashe, supra, 397 U.S. 436, the defendant was charged with a robbery involving a group of people playing poker. The defendant was tried for the robbery of one of the victims, and found not guilty. ( Id. at pp. 438-439.) Thereafter, the defendant was tried for the robbery of another of the poker players, and found guilty. ( Id. at pp. 439-440.) The Ashe court concluded the second prosecution violated collateral estoppel. The court reasoned the only issue in dispute at the first trial was whether the defendant had been one of the robbers; the first jury had determined he was not one of the robbers; and accordingly under the principle of collateral estoppel the state could not "constitutionally hale him before a new jury to litigate that issue again." ( Id. at p. 446.) In Dowling, supra, 493 U.S. 342, the defendant was charged with robbery, and the prosecution sought to introduce evidence of an uncharged attempted robbery offense of which the defendant had been acquitted. The uncharged offense evidence was offered to prove identity based on the testimony of the uncharged offense victim identifying the defendant as the man who entered her house and revealing common features with the charged offense. ( Id. at pp. 345-346.) The trial court admitted the uncharged offense evidence, instructing the jury that the defendant had been acquitted of the uncharged crime and emphasizing the limited purpose for the admission of the uncharged crime evidence. ( Ibid.) On review, the United States Supreme Court found no collateral estoppel bar to the admission of the uncharged crimes evidence. The court reasoned that, unlike the situation in Ashe, the defendant's "prior acquittal did not determine an ultimate issue in the present case." ( Id. at p. 348.) The Dowling court observed that in Ashe, the "acquittal in the first trial foreclosed the second trial because, in the circumstances of that case, the acquittal verdict could only have meant that the jury was unable to conclude beyond a reasonable doubt that the defendant was one of the bandits. A second prosecution was impermissible because, to have convicted the defendant in the second trial, the second jury had to have reached a directly contrary conclusion." ( Id. at pp. 347-348.) In contrast, in Dowling, the prosecution in the current robbery case did not need to prove the defendant's commission of the uncharged crime beyond a reasonable doubt but only needed to present sufficient evidence from which a jury could reasonably find the defendant committed the uncharged offense. ( Id. at pp. 348-349.) Alternatively, the Dowling court found the defendant had not shown that the jury at the uncharged offense trial necessarily decided he was not the man who entered the victim's home, because the record suggested that he had conceded his presence at her home but claimed it was not for purposes of robbery. ( Id. at pp. 350-352.) Following Dowling, the California Supreme Court rendered two decisions that included the concept that an issue did not pertain to an "ultimate fact" barred by collateral estoppel unless it needed to be proven by the prosecution at the second trial to establish the charged offense. ( Santamaria, supra, 8 Cal.4th 903; Catlin, supra, 26 Cal.4th 81.) In Santamaria, the first jury convicted the defendant of murder and robbery, but found not true a personal-knife-use enhancement. ( Santamaria, supra, 8 Cal.4th at p. 909.) The principal witness against the defendant was a man who was with the defendant at the time of the murder and who pleaded guilty to being an accessory to the murder; this witness testified that he saw the defendant stab the victim. ( Id. at pp. 908-909.) After the judgment was reversed on appeal for trial error, the defendant was again charged with the same offenses, but without the weapon enhancement allegation. ( Ibid.) At the second trial, the trial court ruled that the defendant could not be retried on the theory that he personally used the knife during the killing. ( Ibid.) Because the evidence showed the cause of death was the knife wound, the prosecution stated it could not proceed with the case in light of the court's ruling; accordingly, the trial court dismissed the case. ( Id. at pp. 909-910.) On review, the California Supreme Court considered whether collateral estoppel should be applied to "mandate that after a judgment is reversed on appeal, the original jury's finding on a sentencing enhancing allegation affects retrial of a murder charge, even though the same jury convicted defendant of that murder. . . ." ( Santamaria, supra, 8 Cal.4th at p. 908.) Noting that collateral estoppel should be applied in a practical, realistic fashion, the Santamaria court found collateral estoppel did not bar the knife use theory from the second trial because (1) the defendant could have been culpable for murder either as a direct perpetrator or an aider and abettor, and the first jury may have rendered the not true finding on the weapon enhancement merely because it was not certain whether the defendant or the admitted accessory had used the knife, and (2) the defendant's personal knife use was not an ultimate issue that had to be proven beyond a reasonable doubt in the second prosecution for murder. ( Id. at pp. 918-922, 926.) In Catlin, the defendant was charged with two murders, and evidence of a third uncharged murder was admitted to show identity and common plan based on the common features shared in all three murders. ( Catlin, supra, 26 Cal.4th at pp. 98, 103-104, 120-121.) At the trial for the two charged murders, the defendant asserted that collateral estoppel required exclusion of evidence that he had received life insurance proceeds after the death of the uncharged murder victim, because at the trial on the uncharged murder the court had found not true the special circumstance allegation that he committed the murder for financial gain. ( Id. at p. 123.) On review, the California Supreme Court found that the trial court properly denied the defendant's motion to exclude the financial gain evidence associated with the uncharged offense. The Catlin court reasoned that the issue of whether the defendant murdered the uncharged murder victim for financial gain "was not an issue of ultimate fact to be determined in the present proceeding" because the defendant was not currently on trial for the murder of the uncharged murder victim, and "the prosecution was not required to establish [the financial gain circumstance alleged in the uncharged offense] beyond a reasonable doubt or, indeed, to prove it at all." ( Catlin, supra, 26 Cal.4th at pp. 124-125.) The court in Catlin also rejected the defendant's argument that it was inherently unfair to admit the financial gain evidence associated with the uncharged offense because he was in effect being retried on this issue. ( Id. at p. 126.) The court stated the financial gain evidence was being admitted to establish facts regarding the currently charged murders, not to relitigate the defendant's responsibility for the uncharged murder for financial gain. ( Id. at p. 127.)
II. Application of Collateral Estoppel in the Context of Per Se and Generic DUI Offenses
A.
Our task is to determine whether the collateral estoppel rule should be applied in a particular case with "`with an eye to all the circumstances of the proceedings,'" and with realism, rationality, and practicality. ( Ashe, supra, 397 U.S. at p. 444; see Santamaria, supra, 8 Cal.4th at pp. 920, 926.) A practical application of collateral estoppel in the context of this case necessarily requires a consideration of the manner in which the Legislature has defined offenses involving persons who drive with alcohol in their system. The Legislature has created two offenses to punish unsafe driving resulting from a driver's alcohol consumption: (1) driving while under the influence of alcohol (§ 23152, subd. (a) (generic DUI)), and (2) driving with a blood-alcohol level of 0.08 percent or more (§ 23152, subd. (b) (per se DUI)). (See People v. Bransford (1994) 8 Cal.4th 885, 888 [ 35 Cal.Rptr.2d 613, 884 P.2d 70].) The two offenses are related because they pertain to the same criminal event, but are distinct because they have different elements of proof — i.e., the generic DUI offense requires a showing of driving impairment but does not require a showing of any particular blood-alcohol level, and the per se DUI offense requires a showing of a 0.08 percent or more alcohol level but does not require a showing of driving impairment. Although both punish the same act — unsafe driving caused by alcohol consumption" that act can be established by proving either impairment or a blood-alcohol level of 0.08 percent or higher. The creation of an offense based on a 0.08 percent or greater blood-alcohol level (without requiring proof of actual driving impairment) passes constitutional muster because scientific evidence shows driving impairment at this level of alcohol. (See Burg v. Municipal Court (1983) 35 Cal.3d 257, 267-268 [ 198 Cal.Rptr. 145, 673 P.2d 732].) (4) Concomitantly, because scientific evidence shows impairment at the 0.08 percent level, when a defendant is charged with generic DUI the Legislature has authorized a jury instruction setting forth a permissive presumption allowing the jury to infer the ultimate fact of driving under the influence from the basic fact of a 0.08 percent or more blood-alcohol level. (§ 23610; People v. Milham (1984) 159 Cal.App.3d 487, 503-505 [ 205 Cal.Rptr. 688]; see CALCRIM No. 2110.) Thus, although the generic and per se DUI offenses are distinct, they are interrelated, and it is in this context that we must consider whether collateral estoppel principles apply. In the first trial the jury found the prosecution could not prove beyond a reasonable doubt that Smith had a blood-alcohol level of 0.08 percent or more while driving, but could not agree whether he was under the influence of alcohol while driving. Double jeopardy principles did not bar retrial on the generic DUI count because its elements are distinct from per se DUI. However, collateral estoppel principles could apply to bar issues (1) that were necessarily decided by the first jury, and (2) that are issues of ultimate fact in the second trial. ( Santamaria, supra, 8 Cal.4th at pp. 916, 922.) Here, the first collateral estoppel component is clearly met. The jury in the first trial necessarily decided that the prosecution could not prove beyond a reasonable doubt that Smith drove with a blood-alcohol level of 0.08 percent or more. Unlike the situation in Santamaria, where the jury could have rejected the knife use finding based on a doubt whether the defendant committed the murder by personally using the knife or as an aider and abettor, here the sole reason the jury could have rejected a guilty verdict on the per se DUI count was that it found that the prosecution had not proven Smith's blood-alcohol level was at least 0.08 percent while driving. As to the "ultimate fact" component, it would appear, at first blush, that the jury's rejection of the per se DUI count does not implicate collateral estoppel principles. At the second trial the prosecution had to prove impairment as a result of alcohol consumption; it did not need to prove that Smith had a blood-alcohol level of 0.08 percent or more while driving. Thus, the fact determined at the first trial (rejection of the 0.08 percent finding) was, at least facially, not an issue of ultimate fact in the second trial. But such a strict construction of the ultimate fact requirement would fail to consider the interrelated nature of the two offenses — i.e., (1) they involve the precise same criminal event, and (2) by virtue of the section 23610 presumption, a violation of per se DUI (0.08 percent or more blood-alcohol level) supports an inference of impairment establishing a violation of generic DUI. Additionally, such a strict construction ignores the evidentiary impact of the section 23610 presumption in a generic DUI trial where, as here, the only disputed issue is impairment. Although a jury deciding whether a defendant is guilty of driving while impaired is not required to rely on a 0.08 percent finding, that finding, coupled with the presumption instruction, plays a pivotal — if not decisive — role in the jury's determination of the charge. For all practical purposes, permitting the second jury to consider whether defendant had a 0.08 percent or more blood-alcohol level while driving allowed the second jury to decide an issue that was virtually dispositive of the ultimate fact, even though that predicate fact was necessarily adjudicated and rejected in the first trial. When considered in this context, collateral estoppel is clearly implicated. At the first trial, the prosecution proffered two different crimes in an effort to punish defendant for his alleged unsafe driving arising from the consumption of alcohol. The first jury rejected the crime premised on the 0.08 percent or higher blood-alcohol level allegation (§ 23152, subd. (b)), but could not reach a conclusion regarding the crime premised on an impairment allegation (§ 23152, subd. (a)). Thus, the whole purpose of the second trial was to determine whether the prosecution could prove "drunk driving" under the impairment theory, and to do so without relying on the 0.08 percent blood-alcohol level finding rejected by the first jury and without the benefit of the inference of impairment attendant to such a finding. Because collateral estoppel principles were not applied, instead of advising the jury of the limitation arising from the first jury's verdict, the second jury was told it could render a guilty verdict premised on the very finding rejected by the first jury: i.e., that Smith drove with a blood-alcohol level of 0.08 percent or more. Indeed, to the extent the second jury based its verdict on a finding that defendant drove with a 0.08 percent or more blood-alcohol level, the first jury's acquittal was effectively negated. Given that per se and generic DUI concern the same criminal event, and that a finding establishing per se DUI supports an inference that can establish generic DUI, this case strongly implicates the core collateral estoppel concern of "protect[ing] a man who has been acquitted from having to `run the gantlet' a second time." ( Ashe, supra, 397 U.S. at p. 446.) This case is distinctively different from Catlin, where the court found no unfairness in requiring relitigation of an issue resolved in an uncharged offense trial because the defendant was on trial for an entirely distinct criminal event. Here, it was inherently unfair to require Smith to relitigate an issue that a jury resolved in his favor in a trial involving the same criminal event, and which issue (if resolved against him) could have been highly influential or even dispositive on the ultimate issue of fact in the retrial. Further, unlike the situation in Santamaria, Smith was seeking to prevent the prosecution from pursuing a theory that was rejected in the context of an acquittal of the main criminal event, not in the context of a conviction of the main criminal event. Applying the collateral estoppel protection in a practical, realistic manner, we conclude that the first jury's finding that the prosecution did not prove defendant drove with a 0.08 percent or more blood-alcohol level is binding on the prosecution in the second trial and cannot be relitigated.B.
Having concluded that collateral estoppel applied at the second trial because the first jury necessarily decided the 0.08 percent issue and the 0.08 percent issue equated with an issue of ultimate fact at the second trial, we now address the three specific contentions of error raised by Smith: (1) the 0.17 percent blood-alcohol level evidence should have been excluded; (2) the jury should not have been instructed regarding the permissive presumption of driving under the influence arising from a 0.08 percent or more blood-alcohol level; and (3) the jury should have been instructed to presume Smith's blood-alcohol level was less than 0.08 percent while driving. Our holding that collateral estoppel barred relitigation of the issue of whether Smith drove with a 0.08 percent or more blood-alcohol level does not bar use of the evidence that his blood-alcohol level was 0.17 percent about one hour after the accident. Based on the expert testimony regarding alcohol absorption and Smith's own testimony that he drank alcohol after the accident, the first jury's finding that the prosecution could not prove Smith had at least a 0.08 percent blood-alcohol level while driving did not necessarily constitute a finding that he did not have a 0.17 percent blood-alcohol level about one hour after the accident. Indeed, Smith concedes that he had a 0.17 percent blood-alcohol level about one hour after the accident and attributes it to his claim that after the accident he consumed alcohol he had in his car. Further, even though the first jury rejected the 0.08 percent finding, the evidence that Smith's blood-alcohol level was 0.17 percent about one hour after the accident was still relevant to the prosecution's case on the issue of whether he was impaired at the time of driving. The jury's finding that the prosecution could not prove beyond a reasonable doubt that Smith drove with a blood-alcohol level of 0.08 percent or more does not preclude a finding that Smith drank a substantial amount of alcohol before the accident, even though the prosecution could not show it had reached the 0.08 percent level while he was driving. For example, the second jury could find that Smith's postaccident 0.17 percent blood-alcohol level supported an inference that he drank a substantial amount of alcohol both before and after the accident, and that (when considered with all the evidence) he had enough alcohol in his system while driving to cause impairment even though his blood-alcohol level was not shown to have reached 0.08 percent while driving. However, given the collateral estoppel bar operative because of the first jury's acquittal on the per se DUI count, the second jury should not have been instructed regarding the permissive presumption of driving under the influence arising from a 0.08 percent or more blood-alcohol level. The permissive presumption, which arises "[i]f the People have proved beyond a reasonable doubt that the defendant's blood-alcohol level was 0.08 percent or more" (CALCRIM No. 2110), is premised on a finding directly contrary to the finding rejected by the first jury. Additionally, because the jury was presented with the 0.17 percent blood-alcohol level evidence, the jury should have been instructed that it must presume Smith's blood-alcohol level was not 0.08 percent or more at the time he was driving. In order to consider the implications of the 0.17 percent blood-alcohol evidence, the jury necessarily needed to consider what level of blood-alcohol can cause impairment. Relevant to this issue, the jurors were aware from expert witness testimony and the prosecutor's argument that the legal blood-alcohol limit for driving is below 0.08 percent, and the jury was instructed on the permissive presumption of driving under the influence arising from a 0.08 percent or more blood-alcohol level. Based on the 0.17 percent blood-alcohol evidence and their knowledge of the 0.08 percent threshold, the jurors could readily draw an inference that Smith had a 0.08 percent or greater blood-alcohol level while driving, and in turn readily infer that he drove under the influence. A special instruction effectively advising the jury about the first jury's acquittal on the 0.08 percent count was necessary to ensure that the second jury did not improperly rest its verdict on a finding that was expressly rejected by the first jury.
As noted, CALCRIM No. 2110 and section 23610 refer to the 0.08 percent or more level at the time of the chemical analysis. (See fn. 2, ante.) However, because the prohibited conduct is driving under the influence, the jury must additionally infer that the 0.08 percent or more level existed at the time of driving to use the 0.08 percent or more chemical analysis evidence to support the generic DUI charge.
We are not persuaded by the Attorney General's argument that Smith forfeited his right to argue on appeal that the jury should have been instructed to presume his blood-alcohol level was less than 0.08 percent while driving. Although his trial counsel did not formulate language for such a special instruction, his trial counsel requested that the trial court give the jury a limiting instruction based on the acquittal. This was sufficient to preserve the issue.
For example, the prosecution's criminalist referred to the 0.08 percent level, and noted that the 0.17 percent blood-alcohol level was "more than twice the legal limit." Likewise, in closing argument the prosecutor noted the 0.17 percent level was twice the legal limit, and emphasized that the Legislature "feels so strongly about [0.08 percent]" that it created an inference of driving under the influence at this level.
Because of the easily drawn inference of driving impairment arising from a 0.08 percent or more blood-alcohol level, in the event the prosecution presents the 0.17 percent blood-alcohol evidence at a retrial the jury should be instructed that Smith's blood-alcohol level while driving was below 0.08 percent even though it will not be instructed regarding the permissive presumption. Regardless of what information is presented directly to the jury, many jurors will be aware of the 0.08 percent threshold (see Burg v. Municipal Court, supra, 35 Cal.3d at p. 272) and they need to be advised that this issue has already been resolved in Smith's favor.
C.
The erroneous failure to preclude the second jury's consideration of the 0.08 percent issue was prejudicial under any standard of review. (See Chapman v. California (1967) 386 U.S. 18, 24 [ 17 L.Ed.2d 705, 87 S.Ct. 824] [harmless beyond reasonable doubt standard for federal constitutional error]; People v. Watson (1956) 46 Cal.2d 818, 836 [ 299 P.2d 243] [no reasonable probability of different result standard for state law error].) Because a finding that the defendant drove with a 0.08 percent or more blood-alcohol level carries such a strong inference of driving impairment and the jury was expressly instructed regarding this inference, there is a reasonable probability that had this theory been foreclosed, the second jury would have reached a different verdict. Although there was sufficient evidence to support a finding that Smith was driving under the influence of alcohol, this is an issue that a jury must resolve without being permitted to rely on a finding that Smith drove with a 0.08 percent or more blood-alcohol level.DISPOSITION
The judgment is reversed. McConnell, P. J., and Benke, J., concurred.