Opinion

Under the “common-interest privilege,” codified in California in Civil Code section 47, subdivision (c) (hereafter section 47(c)), a defendant who makes a statement to others on a matter of common interest is immunized from liability for defamation so long as the statement is made “without malice.” In this case we must determine whether, when it is established that the statement in question was made on an occasion that falls within the parameters of the common-interest privilege, it is the plaintiff who bears the burden of proving that the statement was made with malice, or the defendant who bears the burden of proving that the statement was made without malice.

In the present case, the Court of Appeal held that the trial court erred in instructing the jury that, under section 47(c), defendants bore the burden of proof upon the issue of malice; in addition, the Court of Appeal found that this instructional error was prejudicial and required reversal of the judgment rendered by the trial court in favor of plaintiff. For the reasons discussed hereafter, we agree with the Court of Appeal that the trial court erred in instructing the jury that defendants, rather than plaintiff, bore the burden of proof upon the issue of malice under section 47(c), but we disagree with the Court of Appeal’s determination that this instructional error was prejudicial in the present case. As we shall explain, in view of the trial court’s instructions on the issue of punitive damages, and the jury’s award of punitive damages in favor of plaintiff, we believe there is no reasonable probability that the instructional error regarding the application of section 47(c) affected the outcome of the case. Accordingly, we conclude that the judgment of the Court of Appeal, overturning the judgment rendered by the trial court in favor of plaintiff, should be reversed.

I.

Plaintiff Vivienne Lundquist is an established breeder of Peruvian Paso horses, animals prized for their rarity, beauty, and smooth, elegant gait. She claimed defendants Heinz and Sylvia Reusser defamed her at a meeting of 30 to 40 owners and breeders of Peruvian Paso horses, held in Buellton, Santa Barbara County, on August 14, 1988. At this seminar, Heinz Reusser participated in a panel discussion with other recognized experts regarding the use of drugs and surgery to enhance the appearance and performance of show horses. Reusser told the audience that, prior to his acquisition of a “multiple-champion” mare, the animal had been altered surgically in order to conceal a neck bulge—a defect that he described as a “cosmetic conformational problem.”

Heinz Reusser did not identify the mare or the breeder, but asked his wife, Sylvia Reusser, to distribute to the audience six photographs depicting the mare and her foal. John Burges, a horse breeder who attended the seminar and had purchased several Peruvian Paso horses from Vivienne Lundquist, believed he recognized the mare. He asked Sylvia Reusser whether the photographs depicted a Peruvian Paso mare, raised by Lundquist, named Perla de Oro. Sylvia Reusser responded that the rules of the seminar prohibited her from identifying the mare by name, prompting Burges to ask, “Do I know that mare?” Sylvia Reusser’s reply suggested that Burges knew the mare. Burges and others in attendance deduced the mare was Perla de Oro, and Lundquist was the breeder. Following the seminar, rumors spread among Peruvian Paso horse owners and breeders that Vivienne Lundquist surgically altered her show horses to conceal conformational defects.

Upon ascertaining that the rumors originated from the seminar, Lundquist filed the present lawsuit, alleging that she had been defamed by the Reussers and that they had damaged her reputation and caused her to suffer emotional distress. In their answer, the Reussers raised the affirmative defense that the challenged communication was protected by the common-interest privilege set forth in section 47(c), because the statement was made without malice to other interested persons.

At trial, Vivienne Lundquist testified that, since the mid-1970’s, she had been involved in breeding and showing Peruvian Paso horses. In 1978, one of her horses, Perla de Oro, then eight months of age, developed a throat abscess, necessitating surgical removal of the abscess by a veterinarian. After the infected area healed, a 14-inch surgical scar remained, which nonetheless did not prevent Perla de Oro from winning multiple championships. The scar easily was visible during the summer months, becoming less visible during the winter, when the mare’s coat was thicker. Lundquist denied that Perla de Oro ever had had a conformational defect or undergone cosmetic surgery, testimony that was corroborated by the veterinarian who performed the operation to remove the abscess.

Perla de Oro’s success in the show ring attracted the attention of Heinz and Sylvia Reusser, who had established their own horse-breeding business in the mid-19 80’s and had purchased three other Peruvian Paso horses from Vivienne Lundquist. Prior to the sale of Perla de Oro to the Reussers, the mare had given birth to four foals, none of which exhibited conformational defects. In 1985, when the mare was pregnant with a fifth foal, Lundquist sold her to the Reussers for $25,000, without mentioning the surgery performed on Perla de Oro’s neck seven years earlier.

Heinz Reusser testified that he and his wife had been particularly interested in Perla de Oro because of Vivienne Lundquist’s representation to them, made shortly before the sale, that Perla de Oro was in foal to La Briego, a stallion Heinz Reusser described as “very charismatic.” Lundquist testified Perla de Oro indeed had been bred to La Briego, but that the effort did not produce a foal, leading Lundquist to “re-breed” the mare to another stallion, Romanesco—a circumstance Lundquist had neglected to mention when making her initial representations to the Reussers. Upon discovering her error as to the sire’s identity, and disclosing it to the Reussers prior to consummating the sale of Perla de Oro, Lundquist offered to rescind the sale, but the Reussers declined, preferring instead that Lundquist arrange a subsequent breeding of Perla de Oro to La Briego. Lundquist agreed to do so.

Following the sale of Perla de Oro to the Reussers, and the subsequent birth of her foal (Rio Allason), which displayed no conformational defects, the Reussers elected to forego the opportunity to have the mare bred to La Briego, instead breeding Perla de Oro to the mare’s own half-brother, HMS Domingo (a stallion renowned for his fluid stride). This union led to the birth, in January 1988, of Tanya, the foal depicted in the photographs (distributed by the Reussers at the seminar) as having been born with an abnormal bulge beneath her chin. The Reussers believed the bulge in Tanya’s neck constituted a genetic defect, although no veterinarian or other specialist in equine physiology had so advised them.

Following Tanya’s birth, and the shedding of Perla de Oro’s winter coat in the spring of 1988, the Reussers discovered the scar on the mare’s neck, located in the same area as the bulge beneath Tanya’s chin. The Reussers consulted a veterinarian, who told them that muscle tissue had been surgically removed from Perla de Oro’s neck, that he was unable to ascertain whether the surgery had been for medical or cosmetic purposes, and that he knew of no medical reason for such an operation. The Reussers compared “baby pictures” of Perla de Oro and Tanya and—without even attempting to contact Vivienne Lundquist—concluded that cosmetic surgery had been performed on Perla de Oro in order to conceal a genetic defect.

John Burges, one of the horse breeders who attended the seminar, testified that, upon hearing Heinz Reusser’s remarks attributing the scar on Perla de Oro’s neck to “secret” surgery performed to conceal a genetic defect, and upon viewing the photographs distributed at the seminar by Sylvia Reusser, he believed he had made a “huge mistake” in purchasing Peruvian Paso horses from Vivienne Lundquist. Shortly after returning to his residence in Texas, he telephoned Heinz Reusser to inquire whether the mare depicted in the photographs was Perla de Oro. Reusser confirmed that it was.

Vivienne Lundquist testified other breeders and owners of Peruvian Paso horses, including individuals who had not attended the seminar, subsequently informed her that they had heard she had been accused of surgically altering her horses. Lundquist testified the accusations damaged her reputation and caused her to suffer emotional distress. At trial, she waived her claim for economic losses directly attributable to the communications made by the Reussers at the seminar, and sought only general and punitive damages arising from the injury to her reputation and from the infliction of emotional distress.

The trial court instructed the jury that, whereas plaintiff bore the burden of proving that the statements in question were made by defendants and were defamatory, defendants bore the burden of proving that the statements were made on a privileged occasion and were made without malice. By contrast, in advising the jury upon the issue of punitive damages, the trial court instructed the jury that, in order to obtain such damages, plaintiff bore the burden of proving, by clear and convincing evidence, that defendants had acted with malice and oppression toward plaintiff.

The jury returned special verdicts awarding Vivienne Lundquist $20,000 on her cause of action for slander, $20,000 for intentional infliction of emotional distress, $25,000 in punitive damages against Heinz Reusser, and $15,000 in punitive damages against Sylvia Reusser. In reaching its verdict, the jury found that the Reussers’ communication to those persons in attendance at the seminar had defamed Lundquist and was not privileged, and, additionally, with regard to the jury’s award of punitive damages, that Vivienne Lundquist had established that the Reussers acted with malice. The trial court denied posttrial motions filed by the Reussers alleging instructional error and insufficiency of the evidence.

The Court of Appeal reversed the judgment of the trial court, holding that the trial court erred in instructing the jury that the Reussers bore the burden of proving that the statements they made to persons in attendance at the seminar had been made without malice. Relying upon language contained in our decision in Brown v. Kelly Broadcasting Co. (1989) 48 Cal.3d 711 [ 257 Cal.Rptr. 708 , 771 P.2d 406 ], as well as upon other decisions rendered by California appellate courts (Manguso v. Oceanside Unified School Dist. (1984) 153 Cal.App.3d 574, 580-581 [ 200 Cal.Rptr. 535 ]; Williams v. Taylor (1982) 129 Cal.App.3d 745, 752 [ 181 Cal.Rptr. 423 ]; Gantry Constr. Co. v. American Pipe & Constr. Co. (1975) 49 Cal.App.3d 186, 197-198 [ 122 Cal.Rptr. 834 ]), the Court of Appeal held that, where the uncontested facts establish that the allegedly defamatory statement was made on an occasion within the reach of the common-interest privilege, the plaintiff bears the burden of proving that the defendant acted with actual malice, in order to prevail upon the defamation claim.

We granted Vivienne Lundquist’s ensuing petition for review in order to resolve a conflict in the California decisions as to proper allocation of the burden of proof under section 47(c).

II.

The principal issue presented in this case is whether the trial court properly instructed the jury that, in the jury’s determination whether the common-interest privilege set forth in section 47(c) has been established, defendants bore the burden of proving not only that the allegedly defamatory statement was made upon an occasion that falls within the common-interest privilege, but also that the statement was made without malice. Defendants contend that, in California and throughout the United States, the general rule is that, although a defendant bears the initial burden of establishing that the allegedly defamatory statement was made upon an occasion falling within the purview of the common-interest privilege, once it is established that the statement was made upon such a privileged occasion, the plaintiff may recover damages for defamation only if the plaintiff successfully meets the burden of proving that the statement was made with malice. (See generally, Prosser & Keeton on Torts (5th ed. 1984) Defamation, § 115, p. 835; 2 Harper et al., The Law of Torts (1986) Defamation, § 5.27, pp. 232-242; Rest.2d Torts, § 613, subd. (1)(h), p. 307.) As stated above, the Court of Appeal agreed with defendants on this point. Although, as we shall explain, there are a few (primarily early) California decisions that state a contrary rule, both the legislative history of section 47(c) and the overwhelming majority of recent California decisions support the Court of Appeal’s conclusion. Accordingly, we agree with the Court of Appeal insofar as it concluded that the trial court erred in instructing the jury that defendants bore the burden of proof upon the issue of malice, for purposes of section 47(c).

A.

As we recently explained in Brown v. Kelly Broadcasting Co., supra, 48 Cal.3d 711 (hereafter Brown), defamation, which constitutes an abuse of the right of free speech, “has two forms—libel and slander.” (§ 44; Brown, supra, 48 Cal.3d at p. 723 .) Slander is defined as including “a false and unprivileged publication, orally uttered, . . . [ft] . . . [tending] directly to injure [any person] in respect to his office, profession, trade or business . . . (§ 46; see also Brown, supra, 48 Cal.3d at p. 723 .) For purposes of the law of defamation, “publication” does not require dissemination to a substantial number of individuals; it suffices that the defamatory matter “ ‘is communicated to a single individual other than the one defamed.’ ” (Brown, supra, 48 Cal.3d at 723, fn. 6 , quoting Rest.2d Torts, § 577, com. b, p. 202; Cunningham v. Simpson (1969) 1 Cal.3d 301, 306-307 [ 81 Cal.Rptr. 855 , 461 P.2d 39 ]; 5 Witkin, Summary of Cal. Law (9th ed. 1988) Torts, § 476, p. 561.)

Section 47(c) establishes that certain communications made between persons on a matter of common interest are privileged if the statements are made “without malice.” Section 48, in turn, provides that with respect to statements falling within section 47(c), “malice is not inferred from the communication." For purposes of this statutory privilege, malice has been defined as “‘a state of mind arising from hatred.or ill will, evidencing a willingness to vex, annoy or injure another person.’ ” (Brown, supra, 48 Cal.3d at p. 723 , quoting Agarwal v. Johnson (1979) 25 Cal.3d 932, 944 [ 160 Cal.Rptr. 141 , 603 P.2d 58 ].) We have.held that “[i]f section 47(3) applies to the occasion on which a communication is made and if it was made without malice, it is privileged and, cannot constitute á defamation under California law.” (Brown, supra, 48 Cal.3d at p. 723 , original italics, fn. omitted.)

The parties do not dispute that the allegedly defamatory statements at issue in the present case, made by defendants at a seminar to persons sharing a common interest in horse breeding, were made upon a “privileged occasion” for purposes of the common-interest privilege. Accordingly, the determination whether defendants could be held liable for such statements turned upon whether defendants made the statements with or without malice. As indicated above, in the present case the trial court instructed the jury that, for purposes of the application of section 47(c), defendants bore the burden of proving an absence of malice. On appeal, the parties dispute whether defendants’ assertion of the common-interest privilege as an affirmative defense requires that defendants prove an absence of malice in order to establish the privilege, or requires instead that plaintiff, in order to establish that she was defamed, prove that the defamatory communication was made with malice. In analyzing this issue, we turn first to the applicable statutes.

B.

Our primary task in construing a statute is to determine the Legislature’s intent. (Brown, supra, 48 Cal.3d at p. 724 ; Moyer v. Workmen’s Comp. Appeals Bd. (1973) 10 Cal.3d 222, 230 [ 110 Cal.Rptr. 144 , 514 P.2d 1224 ].) “ ‘The court turns first to the words themselves for the answer.’ ” (Brown, supra, 48 Cal.3d at p. 724 , quoting People v. Knowles (1950) 35 Cal.2d 175, 182 [ 217 P.2d 1 ]; see also Committee of Seven Thousand v. Superior Court (1988) 45 Cal.3d 491, 501 [ 247 Cal.Rptr. 362 , 754 P.2d 708 ].) In this instance, however, the language of the relevant statutes does not provide a ready answer. As we have seen, section 46, by its terms, defines slander as, inter alia, “a false and unprivileged publication”; section 47(c) defines a privileged publication as including one made “[i]n a communication, without malice, to a person interested therein” under specified circumstances, and section 48 provides that, in cases governed by section 47(c), “malice is not inferred from the communication.” (Italics added.) The statutes contain no specific language directed to the burden-of-proof issue, and, in particular, to the question whether, for purposes of section 47(c), the plaintiff bears the burden of proving that a statement, made upon a privileged occasion, was made with malice, or the defendant instead bears the burden of proving the absence of malice.

In the absence of an express statutory directive as to the burden-of-proof issue, we turn to the legislative history of section 47(c). As we shall explain, unlike the language of the statute, the provision’s legislative history sheds considerable light upon the legislative intent regarding the burden-of-proof issue.

We recently had occasion to examine the legislative history of section 47(c) in Brown, supra, 48 Cal.3d 711, 726-729 , in addressing the question whether section 47(c) establishes a broad “public-interest” privilege extending considerably beyond the traditional “common-interest” privilege. Because much of the discussion in Brown is pertinent to the question presented in the case now before us, we quote from that opinion at some length.

“[Section 47(3)] was enacted as part of the Civil Code in 1872. At that time, in the common law of England and the United States, defamation was subject to strict liability, that is, liability without fault as to truth or falsity. (Eldredge, [The Law of Defamation (1978)], § 5, pp. 14-25; Prosser & Keeton, The Law of Torts (5th ed. 1984) § 113, p. 804.) The standard of liability was succinctly phrased in Lord Mansfield’s often quoted statement that, ‘Whenever a man publishes he publishes at his peril.’ (The King v. Woodfall (1774) Loftt 776, 781 [98 Eng.Rep. 914, 916].) Justice Holmes subsequently stated the rule in equally clear fashion: ‘If the publication was [libelous] the defendant took the risk.’ (Peck v. Tribune Co. (1909) 214 U.S. 185, 189 [53 L.Ed 960, 962, 29 S.Ct. 554 ] . . . .)

“To ameliorate the harshness of the strict-liability standard, certain privileges and defenses developed in the common law. The one that is most relevant... is the common-interest privilege, which protected communications made in good faith on a subject in which the speaker and hearer shared an interest or duty.[ ] This privilege applied to a narrow range of private interests. The interest protected was private or pecuniary; the relationship between the parties was close, e.g., a family, business, or organizational interest; and the request for information must have been in the course of the relationship. . . .

“The legislative history of section 47(3) indicates the Legislature intended to codify the narrow common law privilege of common interest, not to create any broad news-media privilege. . . . [f] . . . We also find significance in the drafters’ comments to section 47(3). The wording of section 47(3) was identical to that of section 31 of the original New York Civil Code published in 1865 as a codification of the common law (N.Y. Civ. Code, § 31 (Field 1865)). In the comments following section 47(3), its drafters cited two New York cases dealing with the common-interest privilege [Lewis and Herrick v. Chapman (1857) 16 N.Y. 369 and Thorn v. Moser (N.Y. Sup.Ct. 1845) 1 Denio 488 ] . . . [f] . . . [and] also cited a treatise on the common law, Hilliard, The Law of Torts or Private Wrongs (1859) chapter XIV, page 317.” ( Brown, supra, 48 Cal.3d at pp. 726-728, fns. omitted, citing code comrs. note foll. 1 Ann. Civ. Code, § 47 (1st ed. 1872, Haymond & Burch, comrs.-annotators) p. 25.)

The authorities cited by the drafters of section 47(c), and noted by us in Brown, supra, 48 Cal.3d at pages 728-729, speak directly to the burden-of-proof issue presented in the case now before us. The treatise authored by Hilliard, The Law of Torts or Private Wrongs (1859) (hereafter Hilliard), contains the traditional statement of the common law common-interest privilege set forth in an English case, Harrison v. Bush (1855) 119 Eng.Rep. 509, as follows: “A communication made bona fide upon any subject matter in which the party communicating has an interest, or in reference to which he has a duty, is privileged, if made to a person having a corresponding interest or duty, although it contain criminatory matter which, without this privilege, would be slanderous and actionable.” (See Hilliard, supra, ch. XIV at pp. 317-318.) The Hilliard treatise explains that, when the facts establish the statement was made upon a privileged occasion, the burden rests upon the plaintiff to establish malice: “The rule is well settled, that, if the words were spoken or the publication made upon a just occasion, the communication is privileged, and express malice must be shown in order to maintain an action.” (Hilliard, supra, ch. XIV, at p. 317, initial italics in original, latter italics added, fn. omitted.) In an accompanying footnote, the treatise cites two English cases addressing this issue. In Taylor v. Hawkins (1851) 117 Eng.Rep. 897, 902, Lord Campbell stated: “The rule is, that, if the occasion be such as repels the presumption of malice, the communication is privileged, and the plaintiff must then, if he can, give evidence of malice: if he gives no such evidence, it is the office of the Judge to say that there is no question for the jury, and to direct a nonsuit or a verdict for the defendant.” Similarly, in Wright v. Woodgate (1835) 150 Eng.Rep. 244, 246, Baron Parke stated: “[T]he occasion on which the communication was made rebuts the inference prima facie arising from a statement prejudicial to the character of the plaintiff, and puts it upon him to prove that there was malice in fact—that the defendant was actuated by motives of personal spite or ill-will, independent of the occasion on which the communication was made.” (See Hilliard, supra, ch. XIV at p. 317, fn. 1(b).)

The Lewis and Herrick and Thorn cases, also cited by the drafters of section 47(c), are equally explicit on this point. (See Lewis and Herrick v. Chapman (1857) 16 N.Y. 369, 373 [“The term ‘privileged,’ as applied to a communication alleged to be libelous, means simply that the circumstances under which it was made were such as to repel the legal inference of malice, and to throw upon the plaintiff the burden of offering some evidence of its existence beyond the mere falsity of the charge.” (Italics added.)]; Thorn v. Moser (N.Y. Sup.Ct. 1845) 1 Denio 488, 492-493 [“In general the mere utterance of slanderous words constitutes a ground of action, and the plaintiff need only prove their speaking to authorize a recovery. But to this there are well known exceptions growing out of the occasion of speaking the words and the attendant circumstances. These may be such as to repel the legal presumption of malice, which usually attaches to the speaker of defamatory words, and may require the plaintiff, before he can recover, to prove the existence of malice in fact on the part of the defendant. . . . Indeed, whenever a person speaks in the performance of any duty, legal or moral, public or private, or in the assertion of his own rights, or to vindicate and protect his interest, no action will lie against him without proving express malice, however untrue what is said may be.” (Italics added.)].)

In sum, as we set forth in Brown, supra, 48 Cal.3d at pages 726-729, it is clear that, in enacting section 47(c), the Legislature intended to codify without change the common law common-interest privilege. At common law, that privilege embodied a two-step analysis, under which the defendant bore the initial burden of demonstrating that the allegedly defamatory communication was made upon a privileged occasion, and the plaintiff then bore the burden of proving that defendant had made the statement with malice. Accordingly, the legislative history of section 47(c) supports the Court of Appeal’s determination in the present case that the trial court erred in instructing the jury that, under the common-interest privilege, defendants bore the burden of proof upon the question of malice.

C.

In challenging the Court of Appeal’s conclusion as to the burden-of-proof issue, plaintiff relies heavily upon a line of decisions, commencing with this court’s opinion in Snively v. Record Publishing Co. (1921) 185 Cal. 565 [ 198 P. 1 ] (hereafter Snively), which state that a defendant who relies upon the common-interest privilege of section 47(c) as a defense to a defamation action has the burden of proving that the statement was made upon a privileged occasion and that the statement was made without malice. (See Snively, supra, 185 Cal. at p. 578 ; see also Siemon v. Finkle (1923) 190 Cal. 611, 617 [ 213 P. 954 ] [following Snively]; Stevens v. Snow (1923) 191 Cal. 58, 64-65 [ 214 P. 968 ] [same]; Maher v. Devlin (1928) 203 Cal. 270, 281-284 [ 263 P. 812 ] [same]; Longsworth v. Curson (1922) 56 Cal.App. 489, 497 [ 206 P. 779 ], hg. den. Apr. 17, 1922; Miles v. Rosenthal (1928) 90 Cal.App. 390, 408 [ 266 P. 320 ]; Morcom v. San Francisco Shopping News (1935) 4 Cal.App.2d 284, 288-289 [ 40 P.2d 940 ]; Fairfield v. Hagan (1967) 248 Cal.App.2d 194, 204-205 [ 56 Cal.Rptr. 402 ] [following Morcom]; Swope v. Moskovitz (1967) 253 Cal.App.2d 514, 517 [ 61 Cal.Rptr. 277 ] [citing Snively].) Although plaintiff acknowledges that the Snively decision was disapproved in our recent decision in Brown , insofar as Snively suggested that the common-interest privilege codified in section 47(c) applies broadly to all news reports of matters of public interest (see Brown, supra, 48 Cal.3d at pp. 732-733, fn. 18), plaintiff maintains that Snively's analysis of the burden-of-proof issue was not disapproved in Brown and remains viable today. Plaintiff is correct that our decision in Brown did not address the burden-of-proof issue discussed in Snively , but, for a number of reasons, we cannot agree with plaintiff that the Court of Appeal erred in failing to adhere to the Snively line of decisions.

To begin with, in stating that a defendant, under the common-interest privilege, has the burden of proving that the statement in question was made without malice, the Snively court cited no authority related to the common-interest privilege, and, in particular, failed to consider or analyze the common law background or legislative history of section 47(c) that we have discussed above. As we have explained, at the time section 47(c) was enacted, the common law rule was well established that, once the defendant proved that an allegedly defamatory statement was made upon a privileged occasion, the plaintiff bore the burden of proving that the statement was made with malice. Section 47(c) codified that rule, but the court in Snively apparently was unaware of that circumstance.

Second, although Snively's statement regarding the burden-of-proof issue was followed in several cases (cited above) in the years immediately after the decision in Snively , the rationale of the conclusion reached in that case largely was undermined by this court’s subsequent decision in Locke v. Mitchell (1936) 7 Cal.2d 599, 602-604 [ 61 P.2d 922 ]. As noted above (see fn. 13, ante), the court in Snively had reasoned that, because the “absence of actual malice was an essential part of [the] defense [provided by the common-interest privilege], it was necessary for the defendants to prove such absence in order to sustain this defense.” (Snively, supra, 185 Cal. at p. 578 .) In Locke v. Mitchell , however, this court, in clarifying the operation of the common-interest privilege, explained that, "while in the case of a false and unprivileged publication, libelous [per se], malice is implied, and lack of it is a matter of defense, which need not be pleaded [citing, inter alia, Snively], this is not true of a qualified privilege, for section 48 of the Civil Code expressly declares that ‘malice is not inferred from the communication or publication’ in the cases provided for in subdivisions 3, 4 and 5 of section 47. Hence, where the complaint discloses a case of qualified privilege, no malice is presumed and in order to state a cause of action the pleading must contain affirmative allegations of malice in fact. [Citations.]” (Locke, supra, 7 Cal.2d at p. 602 , italics added.)

We recognize that the Locke decision, by its terms, dealt only with the question of which party bore the burden of pleading malice, and did not address specifically the question of which party bore the burden of proof with regard to malice. Nonetheless, the common law decisions establishing the common-interest privilege, as we already have seen, had established that, when the evidence demonstrated a statement had been made upon a privileged occasion, the plaintiff in an action for defamation could prevail only if he or she proved that the statement was made with malice. Following the Locke decision, subsequent decisions of this court applying the common-interest privilege—while not addressing explicitly the burden-of-proof issue —implicitly recognized that, once it is established that an allegedly defamatory statement was made upon an occasion that gave rise to a qualified privilege under section 47(c), a plaintiff may recover damages for defamation only if he or she presents evidence sufficient to establish that the statement was made with malice. (See, e.g., Sanborn v. Chronicle Pub. Co. (1976) 18 Cal.3d 406, 413-414 [ 134 Cal.Rptr. 402 , 556 P.2d 764 ]; Brewer v. Second Baptist Church (1948) 32 Cal.2d 791, 796-800 [ 197 P.2d 713 ]; Emde v. San Joaquin County etc. Council (1943) 23 Cal.2d 146, 154-155, 161 [ 143 P.2d 20 , 150 A.L.R. 916 ].)

Although this court has not had occasion, in the 58 years that have elapsed since the Locke decision, to address explicitly the question of which party bears the burden of proof under section 47(c) as to the issue of malice, the overwhelming majority of Court of Appeal decisions that have addressed this point have held explicitly that, where the defendant establishes that the defamatory communication was made upon a privileged occasion, the plaintiff bears the burden of proving that the defendant acted with malice. (See, e.g., Manguso v. Oceanside Unified School Dist., supra, 153 Cal.App.3d 574, 580-581 ; Williams v. Taylor, supra, 129 Cal.App.3d 745, 752 ; Gantry Constr. Co. v. American Pipe & Constr. Co., supra, 49 Cal.App.3d 186, 198 ; Deaile v. General Telephone Co. of California (1974) 40 Cal.App.3d 841, 846-849 [ 115 Cal.Rptr. 582 ]; Warfield v. McGraw-Hill, Inc. (1973) 32 Cal.App.3d 1041, 1047 [ 108 Cal.Rptr. 652 ]; McCunn v. California Teachers Assn. (1970) 3 Cal.App.3d 956, 962 [ 83 Cal.Rptr. 846 ]; Freeman v. Mills (1950) 97 Cal.App.2d 161, 166-169 [ 217 P.2d 687 ]; Shoemaker v. Friedberg (1947) 80 Cal.App.2d 911, 919 [ 183 P.2d 318 ]; contra, Swope v. Moskovitz, supra, 253 Cal.App.2d 514, 517 [relying upon Snively]; Fairfield v. Hagan, supra, 248 Cal.App.2d 194, 204-205 [relying upon Morcom v. San Francisco Shopping News, supra, 4 Cal.App.2d 284 , one of the decisions that chose to follow Snively].)

Thus, although plaintiff faults the Court of Appeal in the present case for failing to follow Snively , the passage in Snively upon which plaintiff relies has not been followed by the overwhelming majority of California decisions that have addressed this issue in the course of the last several decades. (See also 5 Witkin, Summary of Cal. Law, supra, Torts, § 520, p. 610 [“Ordinarily, privilege must be specially pleaded by the defendant, and the burden of proving it is on him. [Citations.] But where the complaint shows that the communication or publication is one within the classes qualifiedly privileged, it is necessary for the plaintiff to go further and plead and prove that the privilege is not available as a defense in the particular case, e.g., because of malice. [Citations.]” (Italics added.)].)

Furthermore, the Court of Appeal’s conclusion as to the burden-of-proof issue not only is supported by the common law background and legislative history of section 47(c), and by most of the recent California decisions discussing this issue, but also conforms to the application of the common-interest privilege in the great majority of jurisdictions throughout the United States. (See generally, Prosser & Keeton, supra, Defamation, § 115, pp. 835-836 [citing cases]; Harper et al., supra, Defamation, § 5.29, pp. 248-249, fn. 25 [citing cases].)

In addition to relying upon Snively and its progeny, plaintiff contends that the Court of Appeal’s decision in the present case—and, by implication, all of the other Court of Appeal decisions, cited above, which held the burden of proof as to malice is upon the plaintiff—are inconsistent with the provisions of Evidence Code section 500, which sets forth a general rule with regard to allocation of the burden of proof. That statute, enacted in 1965 as part of California’s initial codification of the rules of evidence, provides in full: “Except as otherwise provided by law, a party has the burden of proof as to each fact the existence or nonexistence of which is essential to the claim for relief or defense that he is asserting.” (Ibid.)

Plaintiff s contention is unavailing for several reasons. First, the introductory clause of Evidence Code section 500—“[e]xcept as otherwise provided by law”—makes it clear that the section was not intended to supersede rules (allocating the burden of proof) embodied in other statutory provisions or court decisions. (Evid. Code, § 160.) Plaintiff has not pointed to anything in the legislative history of Evidence Code section 500 suggesting that the statute was intended to alter the operation of the common law rules, relevant to the burden of proof as to the common-interest privilege, which were codified by the adoption of sections 47(c) and 48. Second, even without regard to its introductory clause, Evidence Code section 500 does not purport to determine which facts are “essential” to the plaintiffs “claim for relief’ and which facts are “essential” to a claimed defense, but rather leaves those substantive determinations to be resolved in light of the particular cause of action or defense at issue. As we have seen, from its origins at early common law the common-interest privilege has been understood to require that the plaintiff prove the existence of “malice” as an essential requirement of a cause of action for defamation when the evidence establishes that the statement was made upon a privileged occasion. Accordingly, the rule placing upon the plaintiff the burden of proving malice, for purposes of section 47(c), does not conflict with Evidence Code section 500.

Finally, plaintiff contends that our recent decision in Brown, supra, 48 Cal.3d 711 , contains several statements “from which it could be inferred” that a defendant bears the burden of proving an absence of malice. As we shall explain, plaintiff is mistaken.

In Brown, supra, 48 Cal.3d 711 , the sole issue presented was whether former section 47, subdivision 3, afforded a broad “public-interest privilege” to the news media to make false statements regarding a private individual. ( 48 Cal.3d 711,719 , italics added.) Defendants, members of the news media, argued that when news organizations publish and broadcast matters of public interest, they share a “common interest with their audiences,” warranting protection under the statute. (Ibid.) In rejecting the defendants argument, we addressed two issues pertinent to our present discussion.

First, we held: “Section 47(3) provides a privilege to specified communications made ‘without malice.’ .... If section 47(3) applies to the occasion on which a communication is made and if it was made without malice, it is privileged and cannot constitute a defamation under California law.” (Brown, supra, 48 Cal.3d at p. 723 , original italics.) Plaintiff contends that, properly interpreted, the quoted statement requires that, in order to prevail upon the defense of the common-interest privilege, a defendant must prove the occasion was privileged and the statement was made without malice. Yet Brown does not so hold. Indeed, Brown does not even address the burden-of-proof issue in the context of the common-interest, privilege, and therefore the expansive interpretation of that decision urged by plaintiff is erroneous.

Second, in the context of examining the public-interest privilege in Brown , we held: “Interpreting section 47(3) to require a private person to prove malice might have another curious result. The distinction between public and private persons reflects a special solicitude for private reputation. Under a section 47(3) privilege, however, a private person might have to show a greater degree of culpability by the defendant than would a public person under the common law fair-comment defense and its constitutional-malice standard. . . . If so, a private person subject to section 47(3) would have a greater burden in obtaining redress for a damaged reputation than the high court has seen fit to impose on public persons.” (Brown, supra, 48 Cal.3d at p. 745 , fa. omitted.) Significantly, this discussion in Brown was directed toward explaining why section 47(3) did not embrace the public-interest privilege urged by the defendants in that case. The discussion was unrelated to the distinct question presented here, which involves a nonmedia defendant who properly has asserted the statutorily endorsed common-interest privilege as an affirmative defense. Accordingly, we reject plaintiff’s contention that Brown is supportive of her position.

HI.

Although, as we have explained, the trial court’s instructions to the jury in the present case erroneously allocated to defendants the burden of proving an absence of malice for purposes of the common-interest privilege, for the reasons set forth below we conclude that this error does not warrant reversal of the trial court’s judgment in favor of plaintiff.

In order to persuade an appellate court to overturn a jury verdict because of instructional error, an appellant must demonstrate that “the error was prejudicial (Code Civ. Proc., § 475) and resulted in a ‘miscarriage of justice.’ ” (Pool v. City of Oakland (1986) 42 Cal.3d 1051, 1069 [ 232 Cal.Rptr. 528 , 728 P.2d 1163 ], quoting Cal. Const., art. VI, § 13; People v. Watson (1956) 46 Cal.2d 818, 834 [ 299 P.2d 243 ].) Instructional error ordinarily is considered prejudicial only when it appears probable that the improper instruction misled the jury and affected the verdict. (Krouse v. Graham (1977) 19 Cal.3d 59, 72 [ 137 Cal.Rptr. 863 , 562 P.2d 1022 ].) In the present case, as to the issue of malice, the trial court instructed the jury (pursuant to a modified version of BAJI No. 7.05) that the qualified privilege “is lost ... if the person making the statement was: [f] 1. Motivated by hatred or ill-will toward the plaintiff which induced the publication; or [f] 2. Was without a good-faith belief in the truth of the statement.” This instruction correctly stated the law. (See Frommoethelydo v. Fire Ins. Exchange (1986) 42 Cal.3d 208, 217 [ 228 Cal.Rptr. 160 , 721 P.2d 41 ]; Sanborn v. Chronicle Pub. Co., supra, 18 Cal.3d at p. 413 , quoting Roemer v. Retail Credit Co. (1975) 44 Cal.App.3d 926, 936 [ 119 Cal.Rptr. 82 ] [“ ‘The malice necessary to defeat a qualified privilege is “actual malice” which is established by a showing that the publication was motivated by hatred or ill will towards the plaintiff or by a showing that the defendant lacked reasonable grounds for belief in the truth of the publication and therefore acted in reckless disregard of the plaintiff’s rights (citations).’ ” (Italics by the court in Roemer.)].) As to the burden-of-proof issue, the trial court erroneously instructed the jury that defendants bore the burden of proving that their statements to seminar attendees were made without malice. The trial court’s error in instructing the jury as to the burden of proof clearly was harmless, however, because the jury also was instructed as to punitive damages (pursuant to a modified version of BAJI No. 14.71). In this regard, the jury was informed that, if it found “by clear and convincing evidence” that defendants’ conduct was intended “to cause injury to the plaintiff or [constituted] despicable conduct. . . carried on by the defendants] with a willful and conscious disregard for the rights of others,” malice (or “oppression,” a term defined in the instruction as “despicable conduct that subjects a person to cruel and unjust hardship in conscious disregard of that person’s rights”) thereby was established. (Italics added.) The jury’s affirmative finding as to malice (or oppression) demonstrates that the jury found plaintiff had proved not only that defendants acted with the requisite reprehensible motivation constituting an abuse of the privileged occasion, thereby defeating the qualified privilege, but that defendants’ conduct also was intentionally injurious to, or in conscious disregard of, plaintiff’s rights, thereby meeting the heightened requirements of malice (or oppression) necessary to support an award of punitive damages. (See § 3294, subd. (c)(1), (2).) Thus, it is clear the jury’s verdict would have been the same had it been instructed correctly that, in order to establish that defendants abused the privileged occasion, plaintiff bore the burden of proving that defendants’ defamatory remarks were made with malice. Under these circumstances, reversal of the judgment rendered by the trial court is not warranted. (See People v. Watson, supra, 46 Cal.2d at p. 836 .)

IV.

We reverse the judgment of the Court of Appeal insofar as it overturns the trial court’s judgment in favor of plaintiff.

Lucas, C. J„ Mosk, J„ Kennard, J„ Arabian, J„ Baxter, J„ and Werdegar, J., concurred.

All further statutory references are to the Civil Code unless otherwise indicated. Prior to reconfiguration of the statute in 1990 into lettered subdivisions, section 47(c) was designated section 47, subdivision (3). Where in prior decisions the statute is cited under its former designation as section 47, subdivision (3) (or “section 47(3)”), we have retained that designation. Heinz Reusser’s tape-recorded remarks were as follows: “I’m Heinz Reusser. The reason I like to be on the panel here is my concern about how we influence our horses with drugs and surgery and how that affects, really affects, the future quality and future potential of our breed and to me they are very much the same thing as a breeder. We know that people breed to winning horses and if these winning horses have only been winning because they had the pharmacist at their side, then we are leading people that are breeding these horses down the wrong path. Especially if they don’t know. And the same is true if they are breeding or buying for breeding horses that are surgically altered, taking genetic problems out of the horse and later we breed to them. And, of course, they come back and I think that these two things have the same effect on what we achieve in breeding with many of our breeders being led down the wrong path that may cost them tens or hundreds of thousands of dollars until they find out what happened. “In my specific case, and I just want to mention this to you, I have a mare, a multiple champion mare, had a surgically altered neck-ewe and it’s very visible and I think that it established that was done to the mare. And now I am looking at the first foal and the first foal has exactly the same cosmetic conformational problem as the mother had except you didn’t know that. That is going to come up three years and fifty thousand dollars later. So, I think drugs and the methods of surgically altering a horse, if for the purpose of winning in the show ring, are some of the worst tools to serve our breed with. This is what I would like you to think about and see that we can come back from these ills. I think they are terrible things. Sylvia, do you have the pictures here? She took some pictures of that mare so you can look at it. We are not telling you some wild story, ok?” A question was raised at trial regarding the term “neck-ewe” that appears in the above passage. Although “ewe-neck” is a recognized conformational defect found in some Peruvian Paso horses, “neck-ewe” apparently is not a recognized term of art. At trial, the parties contested whether Heinz Reusser had transposed the words, “ewe,” and “neck,” or had completed his sentence with the word, “neck.” The dispute is insignificant in the resolution of the present case. At trial, Sylvia Reusser testified that, in reply to John Burges’s inquiry, she stated, “You might know the horse.” Burges testified that her reply was, “Yeah, you know the mare.” Heinz Reusser testified he did not seek an explanation from Lundquist as to how Perla de Oro had obtained the scar, because he believed Lundquist initially had been dishonest with him regarding the breeding of Perla de Oro to La Briego, and because Lundquist had not disclosed the existence of the scar to him prior to the sale. Plaintiff’s instruction No. 20, a modified version of BAJI No. 2.60, in pertinent part instructed the jury regarding the burden of proof, as follows: “In plaintiff’s first cause of action for defamation and slander, the plaintiff has the burden of establishing by a preponderance of the evidence the following: “1. That the statements made by defendants on August 14, 1988, were untrue and intentionally published by the defendants; “2. That the natural and probable effect on the average listener was to defame plaintiff; “3. That the publication was the proximate cause of damage to plaintiff. U “In response to plaintiff’s first cause of action for defamation and slander, defendant has asserted as a defense that the statements were true or that he was privileged. As to these defenses, the defendant has the burden of proving by a preponderance of the evidence all of the facts necessary to establish: “1. That the statements made by defendant on August 14, 1988 were true; or “2. That when defendant made the statements on August 14, 1988: “a. He made the statements without malice towards the plaintiff; and “b. He made them to person[s] who were interested in the subject matter of the statement; and “[c.] He was interested in the subject matter of the statements or stood in such relation to the person[s] interested as to afford a reasonable ground for supposing the motive for the communication innocent, or was requested by the person[s] interested to give the information. . . .” With respect to plaintiff’s prayer for punitive damages, the trial court instructed the jury, pursuant to modified versions of BAJI Nos. 2.62 and 14.71, that, in seeking an award of punitive damages, plaintiff bore the burden of proving, by clear and convincing evidence, all the facts necessary to establish that defendants acted with malice or oppression toward plaintiff. The Court of Appeal also held that the trial court’s instruction, assigning to the Reussers the burden of establishing an absence of malice, prejudicially affected the jury’s verdict as to Vivienne Lundquist’s cause of action for intentional infliction of emotional distress. Additionally, the Court of Appeal held that the jury’s award of punitive damages did not “cure[] the instructional error,” because, had the jury been instructed properly as to which party bore the burden of proof relating to malice, it might never have reached the question of punitive damages. For these reasons, the Court of Appeal reversed the entire judgment. Two weeks before oral argument, plaintiff’s counsel informed this court that the parties had settled their dispute. Although this development suggested that the case had become moot, and, as a general rule, this court will not render an advisory opinion (Coleman v. Department of Personnel Administration (1991) 52 Cal.3d 1102, 1126 [ 278 Cal.Rptr. 346 , 805 P.2d 300 ]), we instead follow the well-established line of judicial authority recognizing an exception to the mootness doctrine, and permitting the court to decline to dismiss a case rendered moot by stipulation of the parties where the appeal raises issues of continuing public importance. (See, e.g., Burch v. George (1994) 7 Cal.4th 246, 253, fn. 4 [ 27 Cal.Rptr.2d 165 , 866 P.2d 92 ]; In re Kieshia E. (1993) 6 Cal.4th 68, 74, fn. 5 [ 23 Cal.Rptr.2d 775 , 859 P.2d 1290 ]; Abbott Ford, Inc. v. Superior Court (1987) 43 Cal.3d 858, 868-869, fn. 8 [ 239 Cal.Rptr. 626 , 741 P.2d 124 ]; Daly v. Superior Court (1977) 19 Cal.3d 132, 141 [ 137 Cal.Rptr. 14 , 560 P.2d 1193 ]; Liberty Mut. Ins. Co. v. Fales (1973) 8 Cal.3d 712, 715-716 [ 106 Cal.Rptr. 21 , 505 P.2d 213 ]; D.I. Chadbourne, Inc. v. Superior Court (1964) 60 Cal.2d 723, 731-732, fn. 5 [ 36 Cal.Rptr. 468 , 388 P.2d 700 ]; County of Madera v. Gendron (1963) 59 Cal.2d 798, 804 [ 31 Cal.Rptr. 302 , 382 P.2d 342 , 6 A.L.R.3d 555 ]; see also Cal. Rules of Court, rule 19(b) [once the record is filed in the reviewing court, dismissal following abandonment of the appeal is discretionary, not mandatory]; 9 Witkin, Cal. Procedure (3d ed. 1985) Appeal, §§ 505-506, pp. 491- 493.) Because the present case provides this court with an opportunity to reconcile disparate lines of judicial authority and thereby clarify the law with respect to the common-interest privilege (see post, pp. 1202-1203, 1208-1210), we refrain from dismissing the action, instead retaining jurisdiction for the purpose of resolving the legal issues presented. The cited passage of the treatise authored by Prosser and Keeton states: “The burden is upon the defendant in the first instance to establish the existence of a privileged occasion for the publication, by proof of a recognized public or private interest which would justify the utterance of the words. . . . Once the existence of the privilege is established, the burden is upon the plaintiff to prove that it has been abused by excessive publication, by use of the occasion for an improper purpose, or by lack of belief or grounds for belief in the truth of what is said.” (Prosser & Keeton, supra, Defamation, § 115, p. 835, italics added, fns. omitted.) Similarly, the treatise authored by Harper et al. states: “As to ‘malice,’ i.e., actual ill will or improper purpose that constitutes an abuse of the occasion and therefore defeats a qualified privilege, the jury makes the findings, subject to the general and usual control of the court, the burden of proof being on the plaintiff(2 Harper et al., supra, Defamation, § 5.29, p. 248, italics added, fns. omitted.) Section 613 of the Restatement Second of Torts provides in relevant part that “(1) In an action for defamation the plaintiff has the burden ofproving, when the issue is properly raised, ... [5] ... (h) the abuse of a conditional privilege” (italics added), and comment g accompanying section 613 (at p. 309) explains that “[i]f [the defendant] relies upon the defense that the publication of the communication was conditionally privileged, [the defendant] has the burden of proving it. If he sustains this burden by evidence of the requisite quantity and quality, he will prevail unless the plaintiff takes up and sustains the burden of proving that the privilege was abused.” (Italics added.) “Section 47(c) provides that a privileged publication is one made: “In a communication, without malice, to a person interested therein, (1) by one who is also interested, or (2) by one who stands in such a relation to the person interested as to afford a reasonable ground for supposing the motive for the communication to be innocent, or (3) who is requested by the person interested to give the information.” We observe that the common-interest privilege has proved to be a source of vexation and bafflement to courts and commentators alike. (See, e.g., Institute of Athletic Motivation v. University of Illinois (1980) 114 Cal.App.3d 1, 11 [ 170 Cal.Rptr. 411 ] [“[T]he scope of the privilege ... is not capable of precise or categorical definition . . . .”]; see also Brown, supra, 48 Cal.3d at pp. 723-733 & fn. 18 [resolving various points of confusion related to the privilege]; 4 Levy et al., Cal. Torts (1993) Defamation Defenses and Privileges, § 45.12[4], pp. 45-66-45-67 [contrasting various judicial interpretations of the privilege within California]; Prosser & Keeton, supra, Defamation, § 115 pp. 825, 835 [noting the “complexity” of the qualified privilege structure and “the very vagueness of the rules” pertaining to the burden-of-proof issue].) Initially enacted concurrently with section 47 in 1872, section 48 prescribes: “In the case provided fo/c in [section 47(c)], malice is not inferred from the communication.” The “good faith” requirement distinguishes a conditional (or “qualified”) privilege from an absolute one. (See, e.g., Lee & Lindahl, 3 Modern Tort Law (rev. ed. 1988) Defamation, § 36.20, p. 250 [“The distinction between absolute and qualified privileges is essentially that an absolute privilege confers immunity regardless of motive while a qualified privilege can be lost if the defendant acted out of malice.” (Fn. omitted.)]; Eldredge, The Law of Defamation (1978) § 93, p. 510 [“Conditionally privileged occasions are created in order to permit the publisher of the defamation to protect the interest which is entitled to protection. If the publication is made primarily for the purpose of protecting or advancing that interest, the publisher has a proper purpose and, with respect to the purpose element, does not abuse the occasion. On the other hand, where the primary purpose is another purpose, e.g., a desire to injure the defamed person, this is an abuse of the occasion and no privilege comes into being.” (Fns. omitted.)]; 2 Harper et al., supra, Defamation, § 5.21, pp. 179-180 [“The difference between absolute and qualified privilege lies in the effect of the motive and purpose of the defamer. The malice or malevolent purpose of the defamer is of no consequence if the occasion is absolutely privileged, but if the privilege is conditional only, proof of actual malice or ill will makes the defendant liable. In other words, malice is irrelevant if the words are spoken on an occasion absolutely privileged, as the immunity is absolute and indefeasible. But the insulation of an occasion only qualifiedly privileged is overcome by proof of malice.” (Fn. omitted.)].) The court in Snively held: “[B]y averring in their answer that the publication was on a privileged occasion and was made without malice, the defendants directly put in issue the question of actual malice. As absence of actual malice was an essential part of that defense, it was necessary for the defendants to prove such absence in order to sustain this defense. The withdrawal by the plaintiff of his claim for punitive damages and the striking out of the averments of actual malice from the complaint did not establish the fact that defendants published the cartoon without actual malice, nor excuse them from the necessity of proving absence of such malice, if they desired to establish that defense.” ( 185 Cal. at p. 578 .)