120 Cal.Rptr.2d 776 (2002)
99 Cal.App.4th 289
The PEOPLE, Plaintiff and Respondent,
v.
Fernando NORIEGA, Defendant and Appellant.
No. B149785.
Court of Appeal, Second District, Division Seven.
June 12, 2002.
Rehearing Denied June 27, 2002.
Review Denied September 11, 2002.[**]
Vanessa Place, under appointment by the Court of Appeal, for Defendant and Appellant.
Bill Lockyer, Attorney General, Robert R. Anderson, Chief Assistant Attorney General, Pamela C. Hamanaka, Senior Assistant Attorney General, Mary E. Sanchez, Supervising Deputy Attorney General, and Myung Park, Deputy Attorney General, for Plaintiff and Respondent.
Certified for Partial Publication.[*]
JOHNSON, J.
A jury found Fernando Noriega guilty of felony corporal injury to spouse[1] and found true an allegation he inflicted great bodily injury upon his spouse. The trial court sentenced Noriega to the low term of two years on the offense and the low term of three years on the great bodily injury enhancement.[2] On appeal Noriega contends (1) the trial court's statements during voir dire about prospective jurors' affiliation with organizations attempting to change the law had a chilling effect on jury selection and encroached on the prospective jurors' First Amendment rights of expression and association, (2) the use of CALJIC 17.41.1 chilled jury deliberations and infringed on Noriega's rights to juror unanimity and a fair trial, (3) there was insufficient evidence to support the great bodily injury enhancement and (4) the trial court did not exercise its "informed discretion" when it sentenced Noriega under the great bodily injury enhancement. We affirm.
FACTS AND PROCEEDINGS BELOW
On August 25, 2000, Fernando Noriega ("Noriega") and his wife Sofia Noriega ("Mrs. Noriega") got into an argument. As a result of Noriega's actions during the argument, Mrs. Noriega lost consciousness, sustained physical injuries and went to the hospital for treatment. Because Noriega does not challenge his conviction for corporal injury to a spouse, we need not recount each side's version of how Mrs. Noriega's injuries occurred.[3] The issue on appeal is the nature and severity of the injuries.
After the argument, Noriega's roommate came out of his bedroom to find Mrs. Noriega unconscious. He testified "her whole face was bloody and her eyes were bruised and swollen."
Mrs. Noriega required stitches on her eyebrow, her cheek and the back of her head. She lost her bottom dental plate during the struggle. At trial, Mrs. Noriega testified she had not been able to replace the dental plates because her injuries had not healed and her right cheek was still numb.
When Mrs. Noriega returned home after she was released from the hospital, she "saw a lot of blood on the floor, and on the closet, and the couch."
The jury found Noriega guilty of corporal injury to a spouse under section 273.5, subdivision (a) and found true the allegation Noriega personally inflicted great bodily injury upon Mrs. Noriega "under circumstances involving domestic violence, within the meaning of ... [s]ection 12022.7(d)," now subdivision (e). The trial court sentenced Noriega to the low term of two years on the offense and the low term of three years on the great bodily injury enhancement.
DISCUSSION
I.-IV.[**]
V. THE TRIAL COURT ERRED IN DELIVERING AN EXTEMPORANEOUS ADMONISHMENT TO THE JURORS AGAINST JURY NULLIFICATION, BUT IN THE UNIQUE CONTEXT OF THIS CASE THE ERROR WAS HARMLESS.
In this case we are compelled to confront an issue far different from the propriety of giving a 17.41.1 instruction to the jury. It is even different from the question the Supreme Court expressly reserved in People v. Williams, that is, whether "a trial court may or must instruct a jury specifically that it has no power to render a verdict contrary to the law or the facts before it ...."[28] Here the trial judge went far beyond either 17.41.1 or any carefully-crafted instruction a CALJIC committee might conceivably devise to discourage jury nullification. Instead the trial court delivered what is best characterized as an extemporaneous admonishment, neither accurate nor evenhanded in its description of jury nullification, and full of threats against any juror or jurors who would even think about considering such a step.
The admonishment took place during the voir dire stage when the trial court told the assembled jury panel:
"We also want you to know about something we call jury nullification. This is a concept whereby the jury will nullify the law or the facts. An example is if you're on a juryusually it occurs in cases like maybe prostitution or gambling, involving circumstances where there's a lot of social debate on whether those should be even criminal activities. Even minor drug cases, there's a lot of discussion about whether certain thing should be criminalized. And let's say it's prostitution, and you're on the jury. And the prosecution puts absolute, slam-dunk, solid evidence on that proves beyond a shadow of a doubt that the prostitution did occur, but you Here's what jury nullification is: You go back, you start to deliberate. You say, you know what? Even though we promised to follow the law, we don't like the idea that we're wasting time on prostitution. Let people do what they want to do on their own time. Let's come back, tell the judge, tell the other jurors we don't believe in the law. We'll vote not guilty because we don't believe in the law. So you nullify the facts, you nullify the law. Now, we can't prevent that. We'll instruct you that you shouldn't do it, and we'll instruct you that if one or more of the jury members on this case goes back and starts talking like that, you know, I don't like these rules. I had my fingers crossed when I said I'd follow themwe'll ask you not to do that. The reason whyits okay to have those beliefs, but this is not the place to change the law. We don't make the law here. The law is made in Sacramento either by the Legislature or the Governor or through elections when we vote on laws. All we do is try to enforce the laws, follow the rules. We don't make the rules. So as jurors you are obligated to promise us that you will follow the law even if you disagree with it. If you disagree with it, tell us now so we can excuse you. Thewhat happens, if you do disagree with it, you'll get the other jurors mad, because obviously they don't want to do this. They want to make this trialwe all dosomething other than a dress rehearsal, and we want to do it right the first time. You'll make the other jurors mad and you'll have to come out and explain what you're doing. You'll get excused from the case, and, you know, in some instances it can result in a mistrial...."
This particular exhortation dramatically illustrates at least three dangers inherent in a judge's attempt to do extemporaneous variations on a theme the court believes is embodied in a CALJIC instruction.
First, like the many unsuccessful judicial attempts to "improve" on the reasonable doubt instruction, it is not a wise practice even when the message is delivered in well-thought-out and measured terms, which is not the most accurate characterization of the judge's off-the-cuff remarks in this case. The observation one court made about what happens when judges try to elaborate on the "reasonable doubt" instruction, applies with equal force here. "Well intentioned efforts to `clarify' and `explain' [the statutory reasonable doubt instruction] have had the result of creating confusion and uncertainty, and have repeatedly been struck down by the courts of review of this state."[29]
Second, the judge presented an erroneous interpretation of what we assume to be 17.41.1, characterizing it as an instruction expressly advising jurors they cannot engage in jury nullification. That term is not even mentioned in 17.41.1 and for good reason. This instruction instead speaks in terms of what jurors are expected to do, follow the law, and not in terms of a power they are prohibited from exercising. By labeling this instruction as an express admonishment against jury nullification the trial judge risked infecting it with additional issues and concerns beyond those raised by its present language. At this moment, we are awaiting a Supreme Court opinion whether 17.41.1 is an appropriate instruction, even in its current form. Coupling it with a judge's explanation to the jurors advising them it is really an instruction against those jurors exercising their power to engage in jury nullification adds a further level of complications to the question our high court has been struggling to resolve.[30]
Third, in choosing his illustrations the judge provided an unduly narrow and also one-sided view of what constitutes this prohibited practice of jury nullification, ignoring the possibility it can operate against the defendant not just in his favor. The illustrations all had to do with what might be considered "unpopular laws" by many people, including jurors. There are, of course, other possible reasons for jurors to engage in what is called "jury nullification" in favor of the defensea popular or sympathetic defendant, an unsympathetic victim, de minimis violations of otherwise popular laws, and the like.[31]
The trial court's failure to mention these other reasons for "jury nullification" was only a minor problem, however, compared to his failure to warn jurors it also would be "jury nullification" if they voted to convict the defendant of a charged crime despite not believing he had been proved guilty beyond a reasonable doubt of having committed that particular crime. This introduced bias into the trial court's exposition of what jurors should avoid and what they should report when one of their number proposed it. Admittedly, in some situations there are protections against jury nullification when it runs against the defendant. That is, trial judges or appellate courts are in a position to reverse unjust convictions when they are based on clearly insufficient evidence. But defendants are entitled not just to be freed where the evidence is so weak the jurors could not believe it proved their guilt beyond a reasonable doubt. They are entitled to be freed when the jurors do not believe the evidence was sufficient. For instance, in a one witness case it is not sufficient the jurors coidd believe the key prosecution witness, it is necessary they do believe that witness. If they vote to convict the defendant despite believing the evidence incredible or insufficient to prove he committed the charged crimeperhaps because they figure the defendant is a bad guy or probably committed some other crimes or is likely to do so in the future that qualifies as "jury nullification" just as much as if they voted to acquit because they did not like the crime with which he was charged. And it does happen. The classic study of jury behavior found that in three percent of trials the juries convicted where the judge would have acquitted the defendant.[32]
Neither trial judges nor appellate courts are well-equipped to deal with this sort of anti-defendant "jury nullification," however. Without sitting in the jury room and hearing all the deliberations it is difficult to pick up the clues the jurors disbelieve the evidence supporting the defendant's guilt of the charged crime yet decide to vote him guilty for other reasons. So long as a reasonable juror could convict the defendant on the evidence presented at trial, the judiciary has neither the knowledge of what occurred during jury deliberations nor is it their role to intervene and void the jury verdict. Indeed empirical evidence suggests trial judges rarely reverse convictions even when they believe the jury reached the wrong verdict and for the wrong reason, although they do reduce the sentence in such cases fairly frequently.[33] In any event, the trial judge here failed to even hint to these jurors they would suffer any of these promised punishments or consequences if they engaged in anti-defendant "jury nullification." Consequently, even if we were to approveor at least not condemnother aspects of the trial judge's anti-nullification speech to the jury panel during this trial we would find it objectionable as presenting a biased view of what constitutes this "jury nullification" phenomenon the jurors are to avoid.
In this case, we find it unnecessary to consider whether an instruction could be crafted to properly define and discourage jury nullification.[34] Like the Supreme Court we leave that question for another day. But we do disapprove the practice in which this trial judge engaged. Moreover, we conclude it could be grounds for reversal in a proper case. However, in this case Noriega does not even contest guilt of the main charge and, as discussed [in an unpublished portion of this opinion], overwhelming evidence supports the enhancement findings. It is not just improbable but nigh impossible this ill-advised extemporaneous speech influenced the jury verdict in any way or that had the trial judge omitted his "anti-nullification" remarks the outcome would have been different.
DISPOSITION
The judgment is affirmed.
We concur: LILLIE, P. J., and WOODS, J.
NOTES
[*] Pursuant to California Rules of Court, rules 976(b) and 976.1, this opinion is certified for publication with the exception of parts I. through IV.
[**] In denying review, the Supreme Court ordered that the opinion be not officially published. (See California Rules of CourtRules 976 and 977).
[1] Penal Code section 273.5, subdivision (a). All further statutory references are to the Penal Code.
[2] Penal Code section 12022.7, former subdivision (d), now subdivision (e). Statutes 2000, chapter 919, section 1.
[3] Noriega testified he was trying to restrain Mrs. Noriega, she tried to kick him, which caused her to trip over a table, the two of them fell down together and she hit her head on the closet door. Mrs. Noriega testified Noriega struck her on the back of the head, she fell to the ground and Noriega hit her face against a wall. The jury apparently believed Mrs. Noriega's testimony.
[**] See footnote *, ante.
[28] People v. Williams (2001) 25 Cal.4th 441, 459, footnote 8, 106 Cal.Rptr.2d 295, 21 P.3d 1209 (emphasis in original).
[29] People v. Garcia (1975) 54 Cal.App.3d 61, 63, 126 Cal.Rptr. 275. The Garcia court then proceeds to "point out some examples of innovative `reasonable doubt' instructions which have been held erroneous", describing seven such cases and a number of others where the appellate court cautioned the trial court to avoid any embellishments on the statutory language.
[30] Our research has not uncovered a case squarely holding a trial judge could tell a jury it must not engage in jury nullification nor one which ruled it could not do so. As mentioned above, our Supreme Court specifically reserved this issue in People v. Williams, supra, 25 Cal.4th 441, 459, footnote 8, 106 Cal. Rptr.2d 295, 21 P.3d 1209 saying: "[W]e express no view on whether, or under what circumstances, a trial court may or must instruct a jury specifically that it has no power to render a verdict contrary to the law or the facts before it, a question not being presented in this case." (Emphasis in original.)
Most cases discussing jury nullification have considered the opposite issuemust a trial court advise the jury it may engage in jury nullification. The nearly universal answer to this second question is no.
A few federal cases have implied an anti-nullification instruction may be proper. They did so in the course of approving trial court responses to jury inquiries about what "jury nullification" is after defense attorneys used the term and urged the jury to exercise that power (or right) during their closing arguments. The majority in U.S. v. Krzyske (6th Cir.1988) 836 F.2d 1013 approved the judge's brief reply: "There is no such thing as valid jury nullification. Your obligation is to follow the instructions of the Court as to the law given to you. You would violate your oath and the law if you willfully brought in a verdict contrary to the law given you in this case." (836 F.2d at p. 1021 [internal quotation marks omitted].) But on appeal, the majority treated the trial court's words as a "refusal to discuss jury nullification with the jury," in other words, a refusal to give an instruction describing their power (or right) to nullify. (836 F.2d at p. 1021.) The dissenter, on the other hand, found it was "simply error" for the trial court to tell "the jury in effect that it had no general authority to veto the prosecution.... [¶] The District Court gave short shrift to this legal tradition [of allowing juries to exercise their veto power] and made no effort to explain to the jury its historical role as the protector of the rights of the accused in a criminal case." (836 F.2d at pp. 1021-1022.)
Likewise, in U.S. v. Sepulveda (1st Cir.1993) 15 F.3d 1161, the defense attorney referred to jury nullification in open court and this time the jury asked the trial court to explain the law on this issue. The trial court told the jurors "Federal trial judges are forbidden to instruct on jury nullification, because they are required to instruct only on the law which applies to a case." (15 F.3d at page 1189.) This was followed by a recitation of the duty to convict if the prosecution proved all elements beyond a reasonable doubt or to acquit if it failed to do so. Thus, the circuit court only approved a refusal to instruct about jury nullification, not an instruction expressly denouncing the practice.
Meanwhile another circuit wrote a lengthy and scholarly opinion pointing out the virtues of jury nullification as a "safety valve." Nonetheless, it rejected a defendant's request and concluded it is better not to formally instruct jurors they possess this power. United States v. Dougherty (1972) 473 F.2d 1113. "What makes for health as an occasional medicine would be disastrous as a daily diet. The fact that there is widespread existence of the jury's prerogative, and approval of its existence as a `necessary counter to case-hardened judges and arbitrary prosecutors,' does not establish as an imperative that the jury must be informed by the judge of that power." (473 F.2d at p. 1136, fn. omitted.) If the Dougherty court's logic is sound, it also might be used to argue against the giving of an instructionbut especially an extemporaneous and misleading admonishmentdenouncing the power of jury nullification. Some might see such an instruction as throwing the "occasional medicine" out with the bath water.
In any event, this court finds it can disapprove what the trial court did in this case without resolving the difficult issue of whether and when a few well-chosen words may be appropriate on those relatively rare occasions where the topic of jury nullification arises during the proceedings.
[31] These and other explanations for pro-defendant jury nullification are discussed in the classic empirical study of jury behavior in criminal cases, Kalven and Zeisel, The American Jury (1966). The authors studied 3,576 criminal trials, including 16.7 percent of those held in California, during a single year. Judges were asked to identify the cases where their decision on guilt would have been the same as the jury returned and where it would not, and also to have their courts supply background information about the defendants and other characteristics of the case. While differences in the judge's and jury's evaluations of the evidence explained many disagreements, the study revealed a substantial percentage where the jury disagreed with the judge based on value judgments about the charged crime, certain defenses, the parties, and other factors unrelated to the evidence they heard. The study reported the judge and jury agreed the defendant should be convicted in 62 percent of the cases and agreed the defendant should be acquitted in 13.4 percent, for an overall agreement rate of slightly better than 75 percent. But the jury acquitted in 16.9 percent of the cases where the judge would have convicted and hung in another 4.4 percent of those cases. Moreover, in 2.2 percent of the cases the juries convicted when the judge would have acquitted and also hung in another 1.1 percent where the judge would have acquitted the defendant outright. (Id. at pp. 55-56.)
This does not mean these juries engaged in blatant "jury nullification" in nearly 25 percent of the cases they heard. Oftentimes the disagreements were based on honest differences in evaluating the credibility of witnesses and the like. (Kalven and Zeisel, The American Jury, supra, at pp. 115-117.) Even where extraneous, often impermissible factors led jurors to reach a different verdict than the judge, it may have been uncommon for the jurors to expressly agree to ignore the evidence and the judge's instructions. Instead their views on the law, the defendant, the victim, and the like could have affected the prism through which they evaluated the evidence. Unconsciously, the jurors may have given less credence to the evidence offered to prove propositions they didn't want proved.
[32] Kalven and Zeisel, The American Jury, supra, at pages 58-59.
[33] Kalven and Zeisel report trial courts only reversed convictions in 10 percent of the cases where they believed the jury had convicted the defendant in error. "In 56 per cent of the cases the judge lets stand a verdict he would not have reached himself. His intervention, when it occurs, comes chiefly in the form of reducing the sentence. In only 1 in 10 cases does he set the verdict aside." Kalven and Zeisel, The American Jury, supra, at page 412 (emphasis added).
[34] See discussion in footnote 30, ante.