OPINION

MOSK, J.

In Bertero v. National General Corp. (1974) 13 Cal.3d 43 [ 118 Cal. Rptr. 184 , 529 P.2d 608 , 65 A.L.R.3d 878 ] ( Bertero ), we held that a suit for malicious prosecution lies for bringing an action charging multiple grounds of liability when some but not all of those grounds were asserted with malice and without probable cause. In the case at bar we are called on to reconsider the question. After doing so, we conclude that we should adhere to the prevailing Bertero rule and therefore affirm the judgment of the Court of Appeal.

Plaintiff Arthur J. Crowley appeals from a judgment of dismissal following the sustaining of a general demurrer without leave to amend to his complaint for malicious prosecution against defendant Carole Katleman and her attorneys, the law firm of Hufstedler, Miller, Kaus & Ettinger, [1] and individual attorneys Warren L. Ettinger and Dan Marmalefsky (hereafter collectively the defendant attorneys).

(1) Our task in reviewing a judgment of dismissal following the sustaining of such a demurrer is to determine whether the complaint states, or can be amended to state, a cause of action. For that purpose we accept as true the properly pleaded material factual allegations of the complaint, together with facts that may properly be judicially noticed. ( Blatty v. New York Times Co. (1986) 42 Cal.3d 1033, 1040-1041 [ 232 Cal. Rptr. 542 , 728 P.2d 1177 ]; Blank v. Kirwan (1985) 39 Cal.3d 311, 318 [ 216 Cal. Rptr. 718 , 703 P.2d 58 ].) [2]

Arthur Crowley was Beldon Katleman's best friend, next-door neighbor, and attorney. In 1973 Beldon Katleman married Carole Katleman, a woman some 30 years his junior. After a brief marriage characterized by the probate court as "stormy," he divorced her in 1975. Crowley represented Beldon Katleman in the divorce proceedings, and as a result of that representation Carole Katleman became extremely hostile towards Crowley.

On January 2, 1976, Beldon Katleman executed a will, naming Crowley as executor. Crowley did not draft the will, nor did he participate in its drafting or its formal execution. The will recited that Katleman was not married; that he had an adult daughter by a prior marriage and two grandchildren by that daughter; and that he had no siblings, but that his mother was still living. In the will Katleman expressly declined to provide for his adult daughter and her issue. Instead, he made a specific bequest to his mother's longtime servant, and disposed of the residue as follows: if his mother survived him the residue would be held in trust for her benefit during her lifetime, and after her death would be distributed to Crowley; if she did not survive him, he gave the residue directly to Crowley. Crowley was named trustee. The will included a standard no contest clause disinheriting any beneficiary or heir who contested it.

In 1980 Beldon married Carole Katleman for the second time; and although, according to the probate court, "the second marriage also had its stormy moments when he threatened to again divorce Mrs. Katleman," they were still married when he died on September 28, 1988. Beldon Katleman never revoked his 1976 will, nor did he execute a subsequent will. Because Beldon Katleman's mother had died in 1982, Crowley became the principal beneficiary.

Shortly after Beldon Katleman's death, Crowley offered Carole Katleman one-half of her deceased husband's estate. [3] She refused his offer, and instead told third parties she would have Crowley disbarred and would "spend every penny or dime" to make sure he received nothing from the estate. She also told Crowley she was not aware of any will or codicil executed by Beldon Katleman other than his 1976 will. A search for such a document turned up none.

On October 4, 1988, Crowley filed a petition to probate Beldon Katleman's will. The court appointed Crowley special administrator of the estate.

On October 28, 1988, Carole Katleman, represented by the defendant attorneys, filed a will contest. As amended, the contest alleged in six separate counts six grounds for invalidating the will, to wit, that (1) Crowley exerted undue influence over Beldon Katleman; (2) Beldon Katleman revoked the will by destroying it; (3) the will was not in fact his last will; (4) he lacked testamentary capacity when he executed the will; (5) the will was not duly executed; and (6) Crowley defrauded Beldon Katleman to induce him to make the will. Carole Katleman then successfully petitioned the probate court to remove Crowley as special administrator of the estate because of the pendency of her will contest.

On December 6, 1989, the probate court granted Crowley's motion for summary adjudication of issues as to the ground of the will contest alleging lack of due execution, declaring that the will had been properly executed and witnessed. The court denied the motion as to the remaining grounds, ruling there were triable issues of material fact as to each.

Shortly before trial of the will contest Crowley again offered Carole Katleman one-half of the estate, but she again refused his offer.

After substantial discovery, the will contest was litigated in a trial lasting almost three weeks. On August 3, 1990, the probate court ruled that none of the six grounds alleged by Carole Katleman for invalidating the will was meritorious. Rather, the court adjudged that the will was not the product of either undue influence or fraud by Crowley, Beldon Katleman did not revoke the will by destroying it, the will was his last will, he had testamentary capacity when he executed the will, and the will was duly executed. The court therefore ordered the will admitted to probate and appointed Crowley its executor.

Carole Katleman took an appeal from the judgment. On May 22, 1991, however, she filed a voluntary dismissal of the appeal with prejudice. The judgment thereby became a final decision on the merits in Crowley's favor.

While the will contest was pending Carole Katleman also filed a claim for a share of the estate as an omitted spouse. (Prob. Code, § 6560.) Crowley opposed the claim on the ground, inter alia, that by filing the will contest Carole Katleman triggered the no contest clause of the will and thus gave up her omission rights. On August 12, 1991, the probate court ruled to the contrary as a matter of law, concluded that Carole Katleman was an omitted spouse, and awarded her the share prescribed by statute, i.e., all the community property and one-half of Beldon Katleman's separate property. ( Ibid. ) In so ruling, however, the court observed that "Carole's will contest does indeed seem to be vindictive.... But even if her attack was pure vengeance, and no matter whether Mr. Crowley's righteous outrage is justified, the enforcement of the no contest clause is not a proper substitute for a malicious prosecution action for whatever damages Mr. Crowley can prove."

Some six weeks later Crowley filed the present action for malicious prosecution against Carole Katleman and the attorney defendants. The first two causes of action are against Carole Katleman. They allege that the will contest terminated in Crowley's favor and that Carole Katleman acted maliciously and without probable cause in contesting the will on the grounds that (1) it was not duly executed, (2) it was void for fraud, (3) Beldon Katleman lacked testamentary capacity, (4) it was not his last will, and (5) Beldon Katleman revoked the will by destroying it. It is further alleged that the will contest "was not premised on an honest or good faith belief by [Carole Katleman] of the merits of such claims, but was instead based upon her malicious, vindictive hatred of [Crowley], to cause [him] to suffer emotional distress, to injure his reputation, and her desire to assert as many claims as possible against him, out of spite." We observe that the first two causes of action allege that only five of the six grounds of the will contest lacked probable cause; they are silent as to the undue influence ground.

The third cause of action is against the attorney defendants. It alleges generally that they instigated and continued the will contest maliciously and without probable cause. Specifically, it alleges that the attorney defendants knew or should have known there was no probable cause for contesting the will on the grounds that (1) it was not duly executed, (2) it was not Beldon Katleman's last will, (3) it was void for fraud, and (4) Beldon Katleman lacked testamentary capacity; it further asserts that no reasonable attorney would have believed these grounds of the contest were legally tenable. This cause of action thus alleges that only four of the six grounds of the will contest lacked probable cause, and is silent as to the undue influence and revocation grounds.

Carole Katleman and the attorney defendants (hereafter collectively defendants) filed a general demurrer to the malicious prosecution complaint, asking the court to take judicial notice of the probate proceedings. In their points and authorities defendants contended that (1) Crowley "tacitly acknowledged" that the undue influence ground of the will contest was legally tenable by omitting it from the list of grounds alleged to lack probable cause; [4] (2) by denying Crowley's motion for summary adjudication of issues as to all grounds of the will contest except lack of due execution, the probate court "necessarily determined" there was probable cause for the remaining grounds; (3) under Friedberg v. Cox (1987) 197 Cal. App.3d 381 [ 242 Cal. Rptr. 851 ] ( Friedberg ), the absence of probable cause for one ground of the will contest (i.e., lack of due execution) will not support a malicious prosecution action when, as here, there was probable cause for the remaining grounds; and (4) in any event, probable cause to claim lack of due execution was shown by the testimony of the two surviving witnesses to the will, who testified in the will contest that they could not recall Beldon Katleman's signing the will in their presence and asking them to witness it. [5]

Perhaps aware of the inadequacy of the arguments made in their points and authorities, defendants shifted their ground at the hearing on the demurrer: instead of contending that the rule of Friedberg applied because the court ruling on the motion for summary adjudication of issues had necessarily determined there was probable cause for all grounds of the will contest except lack of due execution, they argued first and foremost that the Friedberg rule applied because there was probable cause for one ground of the contest, i.e., undue influence. They characterized this ground as the "primary theory" of the will contest, and urged there was probable cause for this ground because (1) the complaint does not challenge it and (2) the court trying the will contest ruled that the confidential relationship between Crowley and Beldon Katleman had given rise to a presumption of undue influence (although the presumption was, admittedly, rebutted at trial).

On January 28, 1992, the court sustained defendants' demurrer without leave to amend, but failed to clearly state its reasons. Although the code requires that "the court shall include in its decision or order a statement of the specific ground or grounds upon which the decision or order is based" (Code Civ. Proc., § 472d), here the court recited only that it took judicial notice of the probate proceedings and that it "bases its decision on Sheldon Appel Co. v. Albert & Oliker, 47 Cal.3d 863 [ 254 Cal. Rptr. 336 , 765 P.2d 498 ] (1989) and Friedberg v. Cox, 197 Cal. App.3d 381 [ 242 Cal. Rptr. 851 ] (1987)." The court thereafter dismissed the malicious prosecution action in its entirety, and Crowley took this appeal.

The Court of Appeal reversed the judgment "under compulsion" of Bertero, supra, 13 Cal.3d 43, 55-57 . The court and defendants strongly criticized the Bertero rule, however, and we granted review to consider their points.

(2) "To establish a cause of action for the malicious prosecution of a civil proceeding, a plaintiff must plead and prove that the prior action (1) was commenced by or at the direction of the defendant and was pursued to a legal termination in his, plaintiff's, favor [citations]; (2) was brought without probable cause [citations]; and (3) was initiated with malice [citations]." ( Bertero, supra, 13 Cal.3d at p. 50 .)

In the case at bar it is undisputed that the will contest was initiated by defendants and that it terminated in a decision on the merits in Crowley's favor as to each ground of the contest. At this stage of the proceedings defendants do not contest the allegation that they acted with malice. The dispute relates to the third element of the cause of action, i.e., lack of probable cause to bring the contest.

(3a) Because the case is before us on a demurrer, the issue is whether the complaint properly pleads the element of probable cause. Specifically, the issue is whether a malicious prosecution action for bringing a will contest on multiple grounds may be maintained when the plaintiff does not allege that all the grounds asserted in the contest lacked probable cause. In the case at bar, as noted above, none of the three causes of action alleges that defendants lacked probable cause for the undue influence ground.

As the Court of Appeal correctly observed, "This case is virtually identical to Bertero." In Bertero, supra, 13 Cal.3d 43 , the employee plaintiff (Bertero) sued the defendant employers for breach of an employment contract. By way of affirmative defenses the answer attacked the validity of the contract on three grounds, alleging that Bertero (1) obtained the contract by duress, (2) obtained the contract by undue influence, and (3) gave no consideration for the contract. The defendants then filed a cross-complaint against Bertero to recover salary already paid to him under the contract, alleging the same three grounds of invalidity as the answer. The matter was tried and Bertero prevailed in all respects: the judgment declared the employment contract valid, awarded Bertero damages for its breach, and dismissed the cross-complaint with prejudice.

After the judgment was affirmed on appeal, Bertero filed another action against the same defendants for malicious prosecution of their failed cross-complaint, charging that all three grounds of the cross-complaint were malicious and lacked probable cause. Again Bertero prevailed, and was awarded additional damages. On appeal from that judgment the defendants challenged, inter alia, an instruction that allowed the jury to find for Bertero even if only one of the three theories of liability in the cross-complaint lacked probable cause. [6] Affirming the judgment with a minor modification, this court held the instruction correct. (13 Cal.3d at pp. 55-57.)

(4) We began by reviewing the dual harms to society and to the individual that the cause of action for malicious prosecution is designed to redress: "The malicious commencement of a civil proceeding is actionable because it harms the individual against whom the claim is made, and also because it threatens the efficient administration of justice." ( Bertero, supra, 13 Cal.3d at p. 50 .)

Turning to the challenged instruction, we found support for it in Singleton v. Perry (1955) 45 Cal.2d 489 [ 289 P.2d 794 ] ( Singleton ) and Albertson v. Raboff (1956) 46 Cal.2d 375 [ 295 P.2d 405 ] ( Albertson ). In Singleton the defendant signed two criminal complaints against the plaintiff. After the charges were dismissed the plaintiff filed two malicious prosecution actions against the defendant, consolidated for trial. The plaintiff prevailed in only one of the actions, and the defendant appealed. In affirming the judgment we did not question that malicious prosecution would lie for the groundless charge even though there may have been probable cause for the other. We quoted with approval the view of an earlier decision of a sister state: "`The authorities show ... that, in order to maintain [a malicious prosecution action], "it is not necessary that the whole proceeding be utterly groundless, for, if groundless charges are maliciously and without probable cause, coupled with others which are well founded, they are not on that account less injurious, and, therefore, constitute a valid cause of action." [Citations.]'" ( 45 Cal.2d at p. 497 .)

In Albertson, supra, 46 Cal.2d 375 , the defendant brought an action against the plaintiff to recover (1) the balance due on a promissory note and (2) title to real property on a theory of fraudulent conveyance. The plaintiff prevailed only on the latter claim, and sued the defendant for maliciously prosecuting that portion of the case. We cited Singleton, supra, 45 Cal.2d at page 497 , as holding that "a defendant cannot escape liability for the malicious prosecution of an unjustified charge by joining with it a justified charge," and we concluded in an alternate holding that malicious prosecution would lie for the claim of fraudulent conveyance despite probable cause for the claim on the promissory note. ( Albertson, supra, 46 Cal.2d at p. 385 .)

The defendants in Bertero, supra, 13 Cal.3d 43 , sought to distinguish the foregoing two cases on the ground that in Bertero "only one remedy the recovery of monies already paid Bertero under the contract was sought with three theories for the recovery suggested." ( Id. at p. 56, italics in original.) To apply the above discussed reasoning to "cases in which the plaintiff in the former action asserted separate theories of recovery," the defendants argued, "will result in potential plaintiffs being unable to rely on multiple theories of liability stemming from a single occurrence." ( Ibid. ) We rejected this attempted distinction, explaining that "A plaintiff acting in good faith may safely sue on alternative theories after full disclosure to counsel when he possesses a reasonable belief in the validity of each of those theories. If his original pleading (or cross-pleading) advances a theory which subsequent research or discovery proves to be untenable the pleading may be amended. We see no reason for permitting plaintiffs and cross-complainants to pursue shotgun tactics by proceeding on counts and theories which they know or should know to be groundless." ( Id. at p. 57, fn. omitted, italics in original.)

(3b) As noted above, defendants in the case at bar also contend there was probable cause for the undue influence ground of the will contest because the court trying the contest found that the confidential relationship between Crowley and Beldon Katleman had given rise to a presumption of such influence. Again we rejected an identical contention in Bertero : "Our conclusion that an action for malicious prosecution lies when but one of alternate theories of recovery is maliciously asserted disposes of a further contention of [the defendants in Bertero ]. They argue a statutory presumption of a lack of consideration and undue influence [citation] in the creation of the employment contract in favor of Bertero. Their theories of recovery in their cross-pleading, however, encompassed duress as well. Whatever the merits of the claimed presumptions, and we do not reach that issue, the cause of action for malicious prosecution could be predicated on maliciously asserted charges that Bertero exercised duress in obtaining the employment contract." ( 13 Cal.3d at p. 57, fn. 5 .)

Nor could defendants herein contend they were compelled to assert all the statutory grounds for a will contest under pain of being deemed to have waived them: as we said in rejecting a similar argument in Bertero , "A litigant is never compelled to file a malicious and fabricated action. It is not the assertion of a claim that is actionable but rather the malicious character of the assertion." ( 13 Cal.3d at p. 52 .)

For all these reasons the Court of Appeal was correct in concluding that "The holding in Bertero is controlling." Under the rule of that decision, the complaint in the case at bar states a cause of action for malicious prosecution even though it does not allege that every one of the grounds asserted in the will contest lacked probable cause. And under the rule of Auto Equity Sales, Inc. v. Superior Court (1962) 57 Cal.2d 450, 455 [ 20 Cal. Rptr. 321 , 369 P.2d 937 ], the Court of Appeal was also correct in concluding that it was bound by Bertero to reverse the judgment dismissing the action.

Unable to distinguish Bertero, defendants ask us to overrule it.

They rely first on Sheldon Appel Co. v. Albert & Oliker (1989) 47 Cal.3d 863 [ 254 Cal. Rptr. 336 , 765 P.2d 498 ] ( Sheldon Appel ), but they overstate the relevance of that decision to the particular issue in the case at bar. To begin with, there can be no claim that Sheldon Appel is in point either on its facts or on its law. We there reviewed a judgment finding a law firm liable in malicious prosecution for filing a complaint to impose a vendor's lien on real property on behalf of a client. The prior action was brought on a single theory of liability, rather than, as here, on multiple theories. In these circumstances it is not surprising that although our opinion in Sheldon Appel answered four questions of law relating to proof of probable cause, not one of those questions had anything to do with the multiple-theory issue in Bertero and the case at bar. [7] Indeed, we cited Bertero no less than five times in the Sheldon Appel opinion, with no hint of disapproving its holding on the present issue.

Unable to rely on either the facts or the law of Sheldon Appel, defendants quote from a preliminary policy statement with which we prefaced the body of the opinion. (5) In that policy statement we reiterated the traditional view that malicious prosecution is a "disfavored cause of action" because of its potentially chilling effect on the public's willingness to resort to the courts for settlement of disputes. ( 47 Cal.3d at p. 872 .) We adhere to that view. (See, e.g., Pacific Gas & Electric Co. v. Bear Stearns & Co. (1990) 50 Cal.3d 1118, 1131 [ 270 Cal. Rptr. 1 , 791 P.2d 587 ].) But we were also fully cognizant of the same view in Bertero , where the defendants expressly reminded us that "malicious prosecution is not a tort `favored by the law'" ( 13 Cal.3d at p. 53 ). Nevertheless we warned, "This convenient phrase should not be employed to defeat a legitimate cause of action. We responded to an argument similar to defendants' over 30 years ago, reasoning, `... we should not be led so astray by the notion of a "disfavored" action as to defeat the established rights of the plaintiff by indirection; for example, by inventing new limitations on the substantive right, which are without support in principle or authority....'" ( Ibid., quoting Jaffe v. Stone (1941) 18 Cal.2d 146, 159 [ 114 P.2d 335 , 135 A.L.R. 775 ]; accord, Leonardini v. Shell Oil Co. (1989) 216 Cal. App.3d 547, 566-567 [ 264 Cal. Rptr. 883 ].) [8]

Nor, for the same reason, should we expand the substantive right. Defendants further quote the portion of our policy statement in Sheldon Appel, supra, 47 Cal.3d at pages 873-874, in which we recommended the use of statutory sanctions against frivolous claims or delaying tactics. (E.g., Code Civ. Proc., § 128.5.) Yet we urged that step not because we believed the tort of malicious prosecution should be abolished or further restricted, but because we believed it should not be expanded: our statement was in reply to the suggestion by some commentators that to combat recent increases in groundless litigation "a reassessment of the traditional `disfavored' status of the malicious prosecution tort, and a relaxation of some of the traditional elements of the tort, may be in order." ( 47 Cal.3d at p. 872 .) It was in that context that we expressed support for statutory sanctions against frivolous claims or delaying tactics, concluding: "Because these avenues appear to provide the most promising remedies for the general problem of frivolous litigation, we do not believe it advisable to abandon or relax the traditional limitations on malicious prosecution recovery." ( Id. at p. 874, italics added.) The implication, of course, was that within those limitations an action for the tort may properly be maintained. [9] As counsel for defendants acknowledged at oral argument, to follow the rule of Bertero today is not to "expand" the tort of malicious prosecution, because it has been the law of this state for 20 years.

(3c) Defendants next contend that we should abandon the rule of Bertero, supra, 13 Cal.3d 43 , and replace it with a new rule based on the "primary right" theory. To understand this contention it will be helpful to briefly review the main points of the primary right theory.

(6) The primary right theory is a theory of code pleading that has long been followed in California. It provides that a "cause of action" is comprised of a "primary right" of the plaintiff, a corresponding "primary duty" of the defendant, and a wrongful act by the defendant constituting a breach of that duty. ( McKee v. Doud (1908) 152 Cal. 637, 641 [ 93 P. 854 ].) The most salient characteristic of a primary right is that it is indivisible: the violation of a single primary right gives rise to but a single cause of action. ( Slater v. Blackwood (1975) 15 Cal.3d 791, 795 [ 126 Cal. Rptr. 225 , 543 P.2d 593 ].) A pleading that states the violation of one primary right in two causes of action contravenes the rule against "splitting" a cause of action. ( Wulfjen v. Dolton (1944) 24 Cal.2d 891, 894-895 [ 151 P.2d 846 ].)

As far as its content is concerned, the primary right is simply the plaintiff's right to be free from the particular injury suffered. ( Slater v. Blackwood, supra, 15 Cal.3d 791, 795 .) It must therefore be distinguished from the legal theory on which liability for that injury is premised: "Even where there are multiple legal theories upon which recovery might be predicated, one injury gives rise to only one claim for relief." ( Ibid. ) The primary right must also be distinguished from the remedy sought: "The violation of one primary right constitutes a single cause of action, though it may entitle the injured party to many forms of relief, and the relief is not to be confounded with the cause of action, one not being determinative of the other." ( Wulfjen v. Dolton, supra, 24 Cal.2d 891, 895-896 , italics deleted.)

(7) The primary right theory has a fairly narrow field of application. It is invoked most often when a plaintiff attempts to divide a primary right and enforce it in two suits. The theory prevents this result by either of two means: (1) if the first suit is still pending when the second is filed, the defendant in the second suit may plead that fact in abatement (Code Civ. Proc., § 430.10, subd. (c); Wulfjen v. Dolton, supra, 24 Cal.2d 891, 894-895 ); or (2) if the first suit has terminated in a judgment on the merits adverse to the plaintiff, the defendant in the second suit may set up that judgment as a bar under the principles of res judicata ( Panos v. Great Western Packing Co. (1943) 21 Cal.2d 636, 638-640 [ 134 P.2d 242 ]). The latter application of the primary right theory appears to be most common: numerous cases hold that when there is only one primary right an adverse judgment in the first suit is a bar even though the second suit is based on a different theory (e.g., Johnson v. American Airlines, Inc. (1984) 157 Cal. App.3d 427, 432 [ 203 Cal. Rptr. 638 ]) or seeks a different remedy (e.g., Stafford v. Yerge (1954) 129 Cal. App.2d 165, 171 [ 276 P.2d 649 ]). [10]

(3d) One may well ask what this theory of pleading has to do with the case at bar. It would obviously be relevant if, for example, Crowley had filed a second malicious prosecution action alleging that defendants lacked probable cause for the ground of their will contest that he omitted from his present complaint, i.e., undue influence. In that event defendants could have invoked the primary right theory to support a plea in abatement or, if the present action fails, the bar of res judicata. (Cf. Dryer v. Dryer (1964) 231 Cal. App.2d 441, 446-449 [ 41 Cal. Rptr. 839 ] [judgment refusing to set aside transfer of property to surviving spouse on ground of undue influence, held a bar to a subsequent action for same relief on ground of fraud].)

But this case presents no such scenario. Rather, defendants contend we should use the primary right theory for a wholly different purpose, i.e., to determine when the probable cause element of a malicious prosecution suit is satisfied in cases in which, as here, the prior action alleged multiple theories of liability. It is difficult to discuss this contention because defendants fail to articulate their reasoning clearly and consistently. They appear to have shifted their ground again and to have abandoned the argument they made at the hearing on their demurrer, i.e., that the undue influence ground was the "primary theory" of their will contest. The defects in that argument were twofold. First, under both the statute (Prob. Code, § 8252, subd. (a)) and the allegations of the complaint in the will contest, the several grounds of such a contest are of equal dignity none is "primary." Second, the primary right theory that defendants invoke does not concern itself with theories of liability such as undue influence but with the plaintiff's underlying right to be free from the injury itself: defendants' argument would have converted the primary right theory into a "primary theory theory."

Instead, in their opening brief in this court defendants now argue that (1) although the cause of action in the prior proceeding stated multiple grounds or theories of liability, it must nevertheless have been premised on the violation of a single primary right, and therefore, (2) if there was probable cause to assert the violation of that primary right on any one theory of liability, such probable cause is sufficient to defeat a malicious prosecution claim even if the other theories of liability lacked probable cause.

The reasoning is flawed by a non sequitur. It is true that under the primary right theory a properly pleaded cause of action must be premised on a single primary right even though it states multiple grounds of liability. But it does not follow from the primary right theory that probable cause to assert that cause of action on one ground of liability defeats a malicious prosecution claim when the other grounds lacked probable cause. Whether such "partial probable cause" is sufficient for this purpose, as we shall see, is a question of policy under the substantive law of malicious prosecution; the primary right theory of pleading simply does not address the matter. [11]

Lacking a logical basis for their contention, defendants rely on a judicial analogy: they cite Friedberg, supra, 197 Cal. App.3d 381 , and Paramount General Hospital v. Jay (1989) 213 Cal. App.3d 360 [ 261 Cal. Rptr. 723 ]. We are not averse to drawing analogies from the primary right theory when appropriate. (See, e.g., Bay Cities Paving & Grading, Inc. v. Lawyers' Mutual Ins. Co., supra, 5 Cal.4th 854, 860 [in deciding the meaning of the word "claim" in an insurance policy, the fact that the claimant had only one cause of action under the primary right theory, "though not controlling, is illustrative"].) Defendants' reliance on Friedberg , however, is not persuasive for several reasons.

First, the portion of the Friedberg opinion cited by defendants (197 Cal. App.3d at pp. 387-388) appears to have been largely unnecessary to the decision. In that case one attorney (Ingraham) sued another (Friedberg) for his share of $86,000 in fees generated by litigation in which they had both been counsel. Ingraham alleged three theories of liability: joint venture, tortious interference with contract, and quantum meruit. The court granted a nonsuit as to the theories of joint venture and interference with contract, but the jury found for Ingraham on the quantum meruit theory and assessed damages against Friedberg in the amount of $12,900. Friedberg nevertheless filed an action for malicious prosecution against Ingraham and the attorneys who represented him in the fee action (hereafter collectively Ingraham), alleging that in that action Ingraham had prosecuted his unsuccessful theories of joint venture and interference with contract maliciously and without probable cause. The court granted summary judgment for Ingraham, and the Court of Appeal affirmed.

The sole issue on appeal was whether the fee action had terminated favorably to Friedberg, as required to support a malicious prosecution claim. ( Bertero, supra, 13 Cal.3d at p. 50 .) The issue was not difficult: on its face the action had evidently not terminated favorably to Friedberg, because the judgment assessed substantial damages against him. Friedberg therefore urged the Court of Appeal to go behind the judgment and hold that the favorable termination requirement was met by the interlocutory ruling granting his motion for nonsuit as to two of Ingraham's three theories of liability, i.e., the theories of joint venture and interference with contract.

The Friedberg court rejected that contention on the sufficient ground that "`[T]he criterion by which to determine which party was successful in the former action is the decree itself in that action. The court in the action for malicious prosecution will not make a separate investigation and retry each separate allegation without reference to the result of the previous suit as a whole....'" ( 197 Cal. App.3d at p. 385 .) The court relied on Murdock v. Gerth (1944) 65 Cal. App.2d 170, 177 [ 150 P.2d 489 ] ( Murdock ), for the proposition that "in determining whether a proceeding alleged to have been maliciously prosecuted has been terminated in favor of the party injured by such proceeding, consideration should be given to the judgment as a whole. To hold otherwise would defeat the purpose of the rule which seeks to prevent collateral attack upon judgments of duly constituted courts. Hence, the decree or judgment itself in the former action is the criterion by which to determine who was the successful party in such proceeding." (Italics added.)

The Murdock court did not mention the primary right theory of pleading, and the Friedberg court likewise did not need to invoke that theory. It relied on the theory in order to distinguish Albertson, supra, 46 Cal.2d 375 , but Albertson was distinguishable on other grounds. As noted above, in the prior action in Albertson the defendant sued the plaintiff to recover (1) the balance due on a promissory note and (2) title to real property on a theory of fraudulent conveyance; the plaintiff lost on the first claim and took an appeal, but won on the second claim and sued the defendant for maliciously prosecuting it. The defendant contended in this court that the malicious prosecution complaint was premature because the plaintiff's appeal from the adverse portion of the judgment in the prior action was still pending. In rejecting the point, we relied exclusively on the settled rule that an appeal may be taken from only a portion of a judgment when that portion is "severable" in the sense that the issues raised in the appeal can be resolved without regard to the issues determined by the portion of the judgment that was not appealed. ( Id. at p. 378, citing, inter alia, American Enterprise, Inc. v. Van Winkle (1952) 39 Cal.2d 210, 216-217 [ 246 P.2d 935 ].) Because this rule does not turn on whether or not the appealable portion of the judgment adjudicated a "cause of action," the opinion in Albertson did not mention the primary right theory. Rather, in holding that the malicious prosecution complaint met the favorable termination requirement, we said simply that "As indicated above, that part of the judgment in the former action that determined that defendant had no interest in or a right to a lien upon plaintiff's real property is now final and constitutes a termination of that separable part of the proceeding favorable to plaintiff." ( 46 Cal.2d at p. 382 .) Because there was no such partial appeal from the judgment in Friedberg , the Friedberg court could have distinguished Albertson on that ground alone.

In addition, the reliance of the Friedberg court on the primary right theory to define the prior action for purposes of the favorable termination requirement of the law of malicious prosecution appears to suffer from the same non sequitur as we identified above. The Friedberg court reasoned that under the primary right theory, "if Ingraham had sued to recover attorney's fees solely on the theory of, e.g., quantum meruit[,] and the matter had gone to final judgment, a later action to recover attorney's fees based on joint venture or tortious interference with contract would be foreclosed by the doctrine of res judicata." ( 197 Cal. App.3d at p. 388 .) This is true; but it does not follow from the primary right theory that when, as in Friedberg , the three theories of liability are litigated in a single action and the latter two are stricken on a motion for nonsuit, that ruling does not constitute at least a "partial favorable termination" for malicious prosecution purposes. Whether such a termination is sufficient to support a malicious prosecution action is, again, a question of policy under the substantive law of that tort.

(8) In any event, Friedberg was not a probable cause case but a favorable termination case. Prior opinions have stressed that the two elements of the tort serve different purposes: "[Plaintiff] confuses the elements of probable cause and favorable termination. Whether a prior action was legally tenable goes to the issue of probable cause, that is, did the defendant have an honest and reasonable belief in the truth of the allegations. [Citation.] Whether a prior action was terminated favorably tends to show the innocence of the defendant in the prior action [citations], and is not affected by the objective tenability of the claim. In short, these two elements of the malicious prosecution tort serve different purposes, and the legal tenability of the underlying action is not the standard by which to judge whether the action was terminated in [plaintiff's] favor." ( Warren v. Wasserman, Comden & Casselman (1990) 220 Cal. App.3d 1297, 1303 [ 271 Cal. Rptr. 579 ].)

For these reasons, even the Friedberg court recognized that Bertero did not affect the favorable termination requirement: as summarized in a recent decision, "the Supreme Court's holding that a malicious prosecution suit may be maintained where only one of several claims in the prior action lacked probable cause ( Bertero v. National General Corp., supra, 13 Cal.3d at pp. 55-57) does not alter the rule there must first be a favorable termination of the entire action. ( Friedberg v. Cox, supra, 197 Cal. App.3d at pp. 386-387.) In Bertero , the question whether all or only part of the prior action had to be without probable cause arose only after judgment had been reached in the plaintiff's favor in the prior action as a whole." ( Jenkins v. Pope (1990) 217 Cal. App.3d 1292, 1300 [ 266 Cal. Rptr. 557 ].)

(3e) Finally, defendants criticize the rule of Bertero, supra, 13 Cal.3d 43 , on various grounds. They make essentially five arguments, adopting in large part the views of the Court of Appeal herein, but none is persuasive.

1. Defendants' main objection is that the Bertero rule is assertedly incompatible with "the fundamental interest which the malicious prosecution tort is designed to protect `the interest in freedom from unjustifiable and unreasonable litigation'" ( Sheldon Appel, supra, 47 Cal.3d at p. 882 ). Defendants argue that when there is probable cause for at least one of the theories of liability asserted in the prior action, the defendant has to defend against that theory in any event and hence the "litigation" is not unjustifiable and unreasonable. Of course, the defendant also has to defend against the theories of liability asserted in the prior action without probable cause, but defendants dismiss that fact with the explanation that in so doing the defendant "has simply been required to respond to additional allegations and arguments directed at a single prior right."

The explanation, however, begs the question. By defining the "litigation" in issue as the prior action per se rather than each theory of liability litigated, defendants assume the point to be proved. It is true that such a defendant must in any event defend against the one valid theory of liability; but the defendant's obligation also to defend against the invalid theories of liability may well be so burdensome as the complaint alleges in the case at bar that it amounts to an impairment of the defendant's interest in freedom from unjustifiable and unreasonable litigation. Whether this is so is a question to be answered, again, not by the primary right theory but by the substantive law of malicious prosecution.

2. (9), (3f) Next, defendants contend in effect that the Bertero rule is no longer necessary because the trial court now has statutory powers to deal with frivolous or delaying conduct that it lacked in 1974 when Bertero was decided and that provide a remedy superior to the cause of action for malicious prosecution in these circumstances. For this proposition defendants rely on a passage in Sheldon Appel in which we took note of legislative measures designed "to facilitate the early weeding out of patently meritless claims and to permit the imposition of sanctions in the initial lawsuit against both litigants and attorneys for frivolous or delaying conduct." (47 Cal.3d at pp. 873-874.) Among the statutes cited in Sheldon Appel defendants particularly stress Code of Civil Procedure section 128.5, enacted in 1981 (hereafter section 128.5). That statute authorizes trial courts to award certain expenses, including attorney fees, resulting from "bad-faith actions or tactics that are frivolous or solely intended to cause unnecessary delay." ( Id., subd. (a).) It is true that "actions or tactics" are defined to include, as an example, "the filing and service of a complaint" ( id., subd. (b)(1)), but it does not follow that the Legislature intended to substitute this remedy for the cause of action for malicious prosecution, less still to overrule Bertero .

The legislative history shows that the Legislature's intent was far more modest. Three years before section 128.5 was enacted, this court noted in Bauguess v. Paine (1978) 22 Cal.3d 626, 635 [ 150 Cal. Rptr. 461 , 586 P.2d 942], that although the Legislature had long recognized (in Code Civ. Proc., § 128) the inherent power of trial courts to ensure orderly proceedings and compel obedience to their orders, it had not authorized the award of attorney fees as a sanction for attorney misconduct violating that power. We then declined to expand the foregoing inherent power to include that sanction, stressing that it should be circumscribed by legislative safeguards and guidelines. (22 Cal.3d at pp. 636-639.)

The Legislature soon filled that gap by enacting section 128.5, expressly declaring that "It is the intent of this legislation to broaden the powers of trial courts to manage their calendars and provide for the expeditious processing of civil actions by authorizing monetary sanctions now not presently authorized by the interpretation of the law in Baug[u]ess v. Paine (1978), 22 Cal.3d 626 [ 150 Cal. Rptr. 461 , 586 P.2d 942 ]." (Stats. 1981, ch. 762, § 2, p. 2968.) "Managing calendars" and "processing civil actions" are not, of course, the primary problems addressed by the tort of malicious prosecution.

As enacted, moreover, the statute mentioned "making or opposing motions without good faith" as its only example of frivolous actions or delaying tactics; it was silent as to the filing of a complaint. The latter example was added in the 1985 revision of the statute. (Stats. 1985, ch. 296, § 1, p. 1335.) But in the same revision the Legislature also added an express declaration (now section 128.5, subdivision (e)) that "The liability imposed by this section is in addition to any other liability imposed by law for acts or omissions within the purview of this section." (Italics added.) This declaration has been correctly cited for the proposition that "The purpose of the section was to broaden the authority of the courts to manage their calendars expeditiously; the section was not intended as a substitute for substantive causes of action arising out of the underlying facts." ( Brewster v. Southern Pacific Transportation Co. (1991) 235 Cal. App.3d 701, 711 [ 1 Cal. Rptr.2d 89 ], fn. omitted.) Among those substantive causes of action is, properly circumscribed, the action for malicious prosecution.

There is still another reason to conclude that the Legislature did not intend either to substitute section 128.5 for the cause of action for malicious prosecution or to overrule Bertero : the remedies are not coextensive. Section 128.5 allows compensation only for out-of-pocket litigation costs, including attorney fees, that directly result from the objectionable conduct; the relief cannot include consequential damages. ( Brewster v. Southern Pacific Transportation Co., supra, 235 Cal. App.3d at pp. 710-712.) By contrast, Bertero stressed that a plaintiff who pleads and proves a case of malicious prosecution may recover not only litigation costs and attorney fees but also "compensation for injury to his reputation or impairment of his social and business standing in the community [citations], and for mental or emotional distress [citation]." ( 13 Cal.3d at p. 51 , fn. omitted.) In appropriate cases such compensation can be well justified and significant in amount. [12]

For all these reasons, section 128.5 and the cause of action for malicious prosecution provide distinct remedies that are at most alternatives to each other: as explained in the companion case of Estate of Katleman, supra, 13 Cal. App.4th 51, 67 , echoing the view of the trial court herein, "an adequate remedy for a frivolous or vindictive will contest is available in the form of an action for malicious prosecution." In a footnote at this point the Court of Appeal continued: "Alternatively, the trial court might have made an award of sanctions against Carole [Katleman] if it found her action to be in bad faith. (Code Civ. Proc., § 128.5; Sheldon Appel Co. v. Albert & Oliker (1989) 47 Cal.3d 863, 873-874 ....) Additional remedies are unnecessary and inappropriate." ( 13 Cal. App.4th at p. 67, fn. 8 , italics added.) [13]

3. Defendants also charge that under the Bertero rule the apportionment of damages between the theories of liability that are and are not supported by probable cause is difficult and "highly speculative." There is no showing, however, that juries cannot perform that task fairly and consistently if they are properly instructed they draw more subtle distinctions every day. Moreover, any difficulty in this regard is chargeable to the tortfeasor: in language quoted in part in Bertero, supra, 13 Cal.3d at page 56 , we rejected an identical contention in Singleton, supra, 45 Cal.2d 489 . There the issue was "Must plaintiff, having shown that she was damaged by the prosecution, go further and show specifically that her damage was attributable to the prosecution on the unjustified charge rather than that on the justified charge?" ( Id. at p. 497.) In answer, we quoted with approval from an earlier decision of a sister state: "`[I]t is manifest that whatever difficulty, or impossibility even, there may be, in discriminating between the injuries, resulting from the good and bad counts, thus improperly blended, is chargeable to the wrongful act of the defendants themselves, and, upon principle, it would seem that they should not now be permitted to plead their own wrong in their own justification.... [¶] `Indeed, it would seem almost a mockery to hold that, by uniting groundless accusations with those for which probable cause might exist, the defendants could thereby escape liability, because of the injured party's inability to divide his damages between the two with delicate nicety. Such, we think, is not the law.'" ( Id. at pp. 497-498.)

4. Defendants next adopt the complaint of the Court of Appeal herein to the effect that the Bertero rule is assertedly incompatible with the modern practice of pleading "inconsistent counts" i.e., alternative factual or legal theories when the pleader is in doubt as to which theory most accurately reflects the events and can be established by the evidence. (See, e.g., Grudt v. City of Los Angeles (1970) 2 Cal.3d 575, 586 [ 86 Cal. Rptr. 465 , 468 P.2d 825 ], and cases cited.) The concern is unfounded. Bertero was not meant to discourage that practice, and we are not aware that it has had any such effect during the past 20 years. Certainly defendants make no such showing. As noted above, the Bertero rule requires only that the plaintiff refrain from asserting or pursuing such theories if they lack probable cause and are motivated by malice. ( 13 Cal.3d at p. 57 .) The plaintiff remains free to allege any and all "inconsistent counts" that a reasonable attorney would find legally tenable on the basis of the facts known to the plaintiff at the time. (See Sheidon Appel, supra, 47 Cal.3d at pp. 878, 885-886.)

5. (10), (3g) Lastly, defendants contend the Bertero rule is "particularly inappropriate" in a will contest. Their arguments, however, are primarily directed at showing that the tort of malicious prosecution itself is a "particularly inappropriate" remedy for will contests brought with malice and without probable cause. Defendants' main premise is that such a contest is not a new and separate action but the continuation of an existing proceeding, i.e., the petition to probate the will; they so contend apparently in order to rely on Coleman v. Gulf Ins. Group (1986) 41 Cal.3d 782, 793-794 [ 226 Cal. Rptr. 90 , 718 P.2d 77 , 62 A.L.R.4th 1083 ](no cause of action for malicious prosecution of an appeal). They also stress that such a contest does not seek affirmative relief from the proponent of the will e.g., money damages but simply seeks to prevent the will from being admitted to probate. They point out that the contest is initiated by filing an "objection" to the petition (Prob. Code, § 8250, subd. (a)), and they conclude that the grounds of the contest are therefore "analogous to affirmative defenses." [14] Crowley disputes this characterization of a will contest, stressing that the pleading that initiates it, however denominated, is in the form of a civil complaint, that a summons issues on that complaint, directing the proponent of the will to file a responsive pleading ( id., § 8250, subd. (a)), and that the latter may then answer or demur ( id., § 8251, subd. (a)). Defendants reply that a responsive pleading is not mandatory and failure to file such a pleading does not result in default but in a hearing on the merits. ( Id., § 8251, subd. (c)(1).)

Defendants' premise is mistaken: "When a will is contested before probate there are two separate and distinct proceedings pending before the court. One is the petition for the probate of the will; the other is the contest of the probate of the will." ( Estate of Relph (1923) 192 Cal. 451, 458 [ 221 P. 361 ]; accord, Estate of Stone (1943) 59 Cal. App.2d 263, 268 [ 138 P.2d 710 ] ["We must not confuse the will contest with the proceeding for the probate of the will, for they are separate proceedings."]; Estate of Raymond (1940) 38 Cal. App.2d 305, 307 [ 100 P.2d 1085 ].)

To be sure, a will contest cannot be initiated until the petition to probate the will has been filed, and to that extent it is dependent on the latter. The answer to the defendants' point, however, is not to be found in the technical niceties of will contest pleading, but in the reality of the matter. Whether they are called affirmative relief or affirmative defenses, the formal assertion of grounds to contest a will has the effect of injecting new factual and legal issues into the probate process, and of placing on the will's proponent the burden of mounting a defense to those issues. It is true the contest does not pray for money damages; but it ordinarily seeks to deny the beneficiaries their inheritances and to enrich the contestants pro tanto, and the case at bar illustrates that the dollar amount at stake can be substantial. (See fn. 3, ante. ) As the case at bar also illustrates, the filing of such a contest can dramatically transform the probate of a will from a routine ex parte procedure often conducted by simple affidavit (Prob. Code, § 8220, subds. (a) & (b)), into sharply adversarial and hotly contested litigation, necessitating lengthy and expensive discovery and trial.

Accordingly, whether or not an abusive will contest is technically a separate and distinct action for some purposes, its effect is nevertheless to infringe on the interest of the will's proponent in freedom from unjustifiable and unreasonable litigation. For purposes of the law of malicious prosecution, therefore, the contest satisfies the requirement of a "prior action ... commenced by or at the direction of the defendant...." ( Bertero, supra, 13 Cal.3d at p. 50 .) No less than the abusive cross-complaint asserted in Bertero , "When such action is prompted by malice and is not based on probable cause, it is actionable as in the case of other affirmative, malicious prosecutions." ( Id. at p. 53.) The case at bar is by no means the first instance of a malpractice action premised on such a will contest. (See, e.g., MacDonald v. Joslyn (1969) 275 Cal. App.2d 282 [ 79 Cal. Rptr. 707 , 35 A.L.R.3d 641 ]; Fairchild v. Adams (1959) 170 Cal. App.2d 10 [ 338 P.2d 191 ].) [15]

"The elements of the common law malicious-prosecution cause of action have evolved over time as an appropriate accommodation between the freedom of an individual to seek redress in the courts and the interest of a potential defendant in being free from unjustified litigation." ( Oren Royal Oaks Venture v. Greenberg, Bernhard, Weiss & Karma, Inc. (1986) 42 Cal.3d 1157, 1169 [ 232 Cal. Rptr. 567 , 728 P.2d 1202 ].) When the prior action charged multiple grounds of liability and there was probable cause for some grounds but not for others, the question arises whether the malicious prosecution plaintiff has satisfied the requirement of showing that the prior action was brought without probable cause. We have seen that neither the statutes enacted nor the cases decided since Bertero, supra, 13 Cal.3d 43 , satisfactorily answer this question. Instead the solution lies in identifying the competing policies at work and in determining which preponderates when applied in the circumstances of the case at bar.

Defendants identify the two main policies that are served by the remedy of imposing sanctions for frivolous or delaying conduct in the original action (hereafter the sanctions remedy): (1) it encourages free access to the courts for the settlement of disputes, and (2) it avoids burdening the judicial system by additional litigation. (See Sheldon Appel, supra, 47 Cal.3d at pp. 872-873.)

As explained in Bertero , the cause of action for malicious prosecution (hereafter the tort remedy) is likewise intended to redress two kinds of harm. (1) The individual who is sued "is harmed because he is compelled to defend against a fabricated claim which not only subjects him to the panoply of psychological pressures most civil defendants suffer, but also to the additional stress of attempting to resist a suit commenced out of spite or ill will, often magnified by slanderous allegations in the pleadings." (13 Cal.3d at pp. 50-51.) And (2) "The judicial process is adversely affected by a maliciously prosecuted cause not only by the clogging of already crowded dockets, but by the unscrupulous use of the courts by individuals `... as instruments with which to maliciously injure their fellow men.'" ( Id. at p. 51, italics added.)

We begin with the second pair of policies served by these remedies: each seeks to avoid burdening the judicial system by additional litigation. Does one outweigh the other? Defendants assert that the judicial burden caused by a subsequent malicious prosecution action is greater than the burden of defending against groundless theories of recovery in the original action. In some cases this may be true; but in other cases and we must accept the complaint's allegations that this is such a case to mount a defense against multiple baseless and malicious grounds may well be no less onerous than to mount a defense against separate causes of action. This is because the burden of litigating such grounds depends on such diverse factors as how many grounds are asserted, how different are the facts that must be proved to support and defeat them, how extensive is the discovery necessary to develop those facts, how many witnesses and documents are required to make that proof even, indeed, how thoroughly the parties prepare their case and how vigorously they present and oppose it at trial. In light of these variables the most we can fairly say is that the two "judicial burden" policies tend to equalize in the balance, i.e., that the judicial cost of a subsequent malicious prosecution action is not necessarily greater than that of defending against multiple baseless and malicious grounds in the original action.

We are therefore remitted to comparing the first pair of policies stated above: we must decide which weighs more in this context (1) the policy of encouraging free access to the courts, served by the sanctions remedy, or (2) the policy of redressing the harm suffered by individuals compelled to defend against unjustifiable litigation, served by the tort remedy.

It is true that untrammeled access to the courts promotes social peace by providing the citizenry with an alternative to potentially dangerous self-help methods of redressing private grievances. But it is not an unmixed blessing: many of our courts are burdened by overcrowded dockets and long delays, and all litigation exacts both public and private costs. We are willing as a society to incur those burdens and costs when the litigation is well founded or, even when ultimately unsuccessful, was at least initiated with probable cause and without malice. In those circumstances the balance tips in favor of the policy of encouraging judicial access. That policy becomes counterproductive, however, when it operates to promote litigation that is groundless and motivated by malice; such litigation has no place in our judicial system, and we are therefore unwilling to bear its costs. [16] After careful consideration, we see no reason to reach a different result when the litigation in question is the assertion of baseless and malicious grounds of liability in a single lawsuit: in both instances the balance tips in favor of the policy of making whole the individuals harmed by such abuse of our courts.

We may draw a persuasive analogy from the cases involving the "litigation privilege" of Civil Code section 47, subdivision (b), an important statutory privilege that protects any "publication or broadcast" made in the course of a "judicial proceeding." (11) The policy served by that privilege is essentially identical to the policy of judicial access at issue here: as we reiterated in our recent decision in Silberg v. Anderson (1990) 50 Cal.3d 205, 213 [ 266 Cal. Rptr. 638 , 786 P.2d 365 ] ( Silberg ), "The principal purpose [of the litigation privilege] is to afford litigants and witnesses [citation] the utmost freedom of access to the courts without fear of being harassed subsequently by derivative tort actions." (Accord, Rubin, supra, 4 Cal.4th at p. 1194.) It is well settled that this privilege is absolute ( Silberg, supra, 50 Cal.3d at p. 215), and that it applies to any communication whether or not it amounts to a publication ( id. at p. 212), that it applies not only to judicial proceedings but to all truth-seeking inquiries, including legislative and other official proceedings ( id. at p. 213), and although originally limited to defamation actions, that it now applies to virtually all other torts ( id. at pp. 215-216, citing many cases).

It is equally well settled, however, that the litigation privilege does not apply to the tort of malicious prosecution. ( Rubin, supra, 4 Cal.4th at p. 1194; Silberg, supra, 50 Cal.3d at pp. 212, 216; Ribas v. Clark (1985) 38 Cal.3d 355, 364 [ 212 Cal. Rptr. 143 , 696 P.2d 637 , 49 A.L.R.4th 417 ]; Kilgore v. Younger (1982) 30 Cal.3d 770, 778 [ 180 Cal. Rptr. 657 , 640 P.2d 793 ]; Albertson, supra, 46 Cal.2d 375, 382 .) And the reason for this rule is also settled: we reiterated in Silberg that "Malicious prosecution actions are permitted because `[t]he policy of encouraging free access to the courts [that underlies the litigation privilege] ... is outweighed by the policy of affording redress for individual wrongs when the requirements of favorable termination, lack of probable cause, and malice are satisfied.'" (50 Cal.3d at p. 216, quoting Albertson, supra, 46 Cal.2d at p. 382 , italics added.) The policy of encouraging judicial access served by the litigation privilege is strong, yet the cases uniformly hold that when the litigation is groundless and motivated by malice the balance tips in favor of the policy of redressing the individual harm inflicted by that litigation. Because the policy of judicial access in the case at bar is essentially identical to the policy underlying the litigation privilege, we find those cases persuasive.

(3h) For all these reasons we reaffirm the rule of Bertero, supra, 13 Cal.3d 43 , as the law governing the issue presented by this case. [17]

The judgment of the Court of Appeal is affirmed.

Lucas, C.J., Kennard, J., Baxter, J., George, J., and Werdegar, J., concurred.

ARABIAN, J., Dissenting.

A wealthy businessman, with a net worth estimated at $10 million, dies. Instead of leaving his estate or any part of it to his wife, as might be expected, his will directs that the entire fortune go to his lawyer, a man who happens to be his neighbor and, we are told, best friend. Can any sensible person doubt that, whatever the testator's intent, these facts are a compelling blueprint for a will contest? And if the circumstances suggest that the aggrieved widow is likely to file a lawsuit attacking the validity of the will, is it any surprise that her conscientious lawyers will feel bound to allege every alternative theory supporting the claim that there was wrongdoing behind the testamentary disposition of her husband's millions?

I take these propositions to be self-evident. Yet the majority would permit the beneficiary of this most peculiar will to sue the widow and her lawyers for tort damages if any one of the several theories alleged in her will contest petition is rejected as lacking probable cause. I cannot join in a holding that not only perpetuates dubious law for no better reason than that it exists, but is blind to realities of contemporary litigation practice that should impel the court to the opposite result. More fundamentally, I dissent from the majority's gratuitously restrictive construction of Code of Civil Procedure section 128.5 (hereafter, section 128.5), a construction that threatens to hobble an effective judicial response to the pressing burden of excessive litigation a response we explicitly and unanimously embraced a mere five years ago. In my view, the circumstances of this case, however much they merited the imposition of substantial monetary sanctions against petitioners in the will contest itself, do not justify yet another derivative tort action.

A malicious prosecution suit imposes substantial litigation costs on both the litigants and the courts not least because malice is such a highly factual issue that it often precludes summary disposition. It is for these reasons, among others, that we have made it clear that the policies controlling the availability of the malicious prosecution tort action transcend the interest in protecting the defendant in the prior lawsuit from having to defend against spurious claims and make it a "disfavored" cause of action that is "carefully circumscribed." ( Sheldon Appel Co. v. Albert & Oliker (1989) 47 Cal.3d 863, 872 [ 254 Cal. Rptr. 336 , 765 P.2d 498 ] ( Sheldon Appel ); Rubin v. Green (1993) 4 Cal.4th 1187, 1194 [ 17 Cal. Rptr.2d 828 , 847 P.2d 1044 ]; Pacific Gas & Electric Co. v. Bear Stearns & Co. (1990) 50 Cal.3d 1118, 1131 [270 Cal. Rptr. 1, 791 P.2d 587 ].) In light of these concerns, I would not only reconsider the correctness of our holding in Bertero v. National General Corp. (1974) 13 Cal.3d 43 [ 118 Cal. Rptr. 184 , 529 P.2d 608 , 65 A.L.R.3d 878 ] ( Bertero ), but would hold that where, as here, alternative theories of liability in support of a single unitary right are alleged by the plaintiff in the complaint in the first action and one of them is determined to have been supported by probable cause, the defendant in the first action may not pursue a derivative malicious prosecution claim. Instead, the defendant must rely on those sanctions for plaintiff's misconduct made available by statute in the original action. Such a result, of course, would require us to overrule our contrary holding in Bertero .

Although the majority savages defendants' arguments, its critique has little to do with the question before us. It is true that the "primary right" concept is in origin a creature of 19th century code pleading and a construct originally developed for different analytical purposes. It does not follow, however, either that the notion itself cannot usefully be applied by analogy to the resolution of problems presented in the derivative litigation context, or that its adoption here is unsound. The use of the "primary right" model as a means of explicating the differing interests that may be at stake in a malicious prosecution action criticized at such length by the majority is, after all, only a metaphor. [1] More importantly, the foundation supporting our holding in Bertero, supra, 13 Cal.3d 43 , has been eroded by intervening changes in legislation and in this court's own perspective on the balance to be struck between "the freedom of an individual to seek redress in the courts and the interest of a potential defendant in being free from unjustified litigation." ( Oren Royal Oaks Venture v. Greenberg, Bernhard, Weiss & Karma, Inc. (1986) 42 Cal.3d 1157, 1169 [ 232 Cal. Rptr. 567 , 728 P.2d 1202 ].)

In short, the majority's reasoning regarding the implications of recent decisions of this court on the vitality of Bertero, supra, 13 Cal.3d 43 , is not only flawed, but, in its lengthy wrangling over the alleged irrelevance of the notion of primary rights, the opinion "thrusts at lions of [its] own imagining." ( Board of Education of Kiryas Joel Village School Dist. v. Grumet (1994) 512 U.S. ___ , ___ [ 129 L.Ed.2d 546, 564 , 114 S.Ct. 2481, 2493 .) The real question before us is not the red herring of whether defendants or the Court of Appeal correctly parsed the nature of the "primary rights" doctrine. It is whether, on the merits of defendants' argument, we should adjust more closely the conflicting interests of judicial access versus retaliatory lawsuits by abandoning the "alternate theory" holding of Bertero, supra, 13 Cal.3d 43 , in favor of a rule that is more responsive to underlying policy concerns. I believe we should.

The Court of Appeal characterized the complaint in the original will contest action as one seeking redress of a single, "primary right" namely, whether Beldon Katleman's will should have been admitted to probate. From that premise, it reasoned that, in light of the policies shaping the availability of the malicious prosecution tort, a distinction ought to be drawn for purposes of applying the probable cause element one of the components of the tort required to be established by the plaintiff between complaints lacking probable cause with respect to one or more multiple primary rights and those in which some but less than all of several alternative theories supporting a single primary right lack probable cause.

The reasons why a "primary right" analysis or its modern equivalent should be applied in cases such as this one is obvious. As most practitioners are aware, in framing the complaint in a civil case, counsel often may be uncertain which theory of liability has the best chance of succeeding as the lawsuit unfolds; the plaintiff in, say, a wrongful termination of employment suit might allege multiple alternative theories of relief rescission, breach of a written contract, breach of an oral contract, breach of an implied-in-fact contract, and wrongful termination in violation of public policy as supporting the vindication of a single right. It is not only common to allege alternative theories of liability in connection with a single claim for relief arising out of a single transaction or event, but "`[e]ven where there are multiple legal theories upon which recovery might be predicated, one injury gives rise to only one claim for relief. [Citation.]'" ( Bay Cities Paving & Grading, Inc. v. Lawyers' Mutual Ins. Co., supra, 5 Cal.4th 854, 860 , italics in original.)

Given that widespread practice and its justification, it is a hard rule that permits the defendant in the first action to then sue the plaintiff in that action in a subsequent malicious prosecution tort suit on the ground that some of the theories supporting a single claim for relief lacked probable cause. This is especially true where one (or more, but less than all) of the alternative theories is supported by probable cause. Having to defend against, say, three (or even five) theories of recovery supporting a single claim for relief, only one of which is determined to be supported by probable cause, is not qualitatively so different from having to defend against a single theory. Certainly this holds true in the case of the wrongful termination lawsuit imagined above, and certainly the burden is markedly less than that of having to defend against several different causes of action, one of which is sound and the others fabricated.

Whether denominated a "primary right" or a single cause of action, such unitary claims for relief typically arise from a "transaction or series of connected transactions" (as section 24, subdivision 1 of the Restatement Second of Judgments has it). For that reason, a defendant required to defend against a complaint pleading the five alternative theories of recovery in the wrongful termination action imagined above is likely to be under less of a defensive burden than, say, the defendant in Singleton v. Perry (1955) 45 Cal.2d 489 [ 289 P.2d 794 ], where the plaintiff swore to two distinct criminal complaints, one for the theft of a Cadillac and another for the theft of jewelry and other personal property. ( Id. at p. 492.) To my mind, such pragmatic concerns justify treating the two types of cases differently.

In addition to such practical considerations, our holding in Bertero, supra, 13 Cal.3d 43 , falls short in several other respects. As the Court of Appeal summed it up, Bertero 's "alternate theory rule invites a multitude of unwarranted litigation, encourages excessive and repetitive litigation, discourages citizens from bringing meritorious civil disputes to the courts, and is inconsistent with modern pleading practice"; while what it termed the "primary right" theory "suffers from none of these deficiencies and adequately protects defendants from unmeritorious lawsuits." If these characterizations are accurate as I believe them to be then I cannot imagine why we would be unwilling to embrace the result urged upon us by defendants. My point is assuredly not that we should throw overboard the remedy of the malicious prosecution tort altogether. It is simply that, in assessing the conditions for its maintenance in particular circumstances, we should continue to do what we have pledged to do at least since our decision five years ago in Sheldon Appel, supra, 47 Cal.3d 863 : Take special care to weigh on which side of the competing interests the balance falls. Unlike the majority, applying that calculus here leads me to conclude that plaintiff's remedy lies in the sanctions that were available to him in the will contest proceeding, not in another lawsuit.

The majority attempts to counter this reasoning by asserting that, as a practical matter, there are too many variables affecting a given piece of litigation to ensure that implementing a "primary rights" -like limitation on the scope of malicious prosecution actions would serve its desired end of reducing the judicial burden of retaliatory claims. But that reasoning, while perhaps descriptively accurate, misses the point. Judges must often craft rules designed for the average run of cases precisely because they would be paralyzed if required to account for the atypical handful that lie at the margins. It is thus no answer to say that we are foreclosed from modifying Bertero because a new rule would not work with complete efficiency in all cases. Of course it would not; but neither need it do so in order to justify its adoption.

What is worse, every time this court upholds the availability of a malicious prosecution action, our ruling has two consequences, both problematical: We authorize yet another derivative or retaliatory lawsuit, a creature we have repeatedly looked on with a skeptical eye, and we limit incrementally the ideal of full and open access to the courts by the threat to litigants and their counsel of retaliatory suits. (See, e.g., Rubin v. Green , 4 Cal.4d 1187, 1196-1199; Sheldon Appel, supra, 47 Cal.3d 863 ; Silberg v. Anderson (1990) 50 Cal.3d 205 [ 266 Cal. Rptr. 638 , 786 P.2d 365 ]; Oren Royal Oaks Venture v. Greenberg, Bernhard, Weiss & Karma, Inc., supra, 42 Cal.3d 1157 ; Pacific Gas & Electric Co. v. Bear Stearns & Co., supra, 50 Cal.3d 1118 .)

There is another consideration supporting defendants' argument: The veritable sea change that has taken hold in social and judicial attitudes toward multiplying litigation in the 20 years since we decided Bertero, supra, 13 Cal.3d 43 . In Sheldon Appel, supra, 47 Cal.3d 863 , at page 872, this court observed that "in recent years ... the large volume of litigation filed in American courts had become a matter of increasing concern, and in some quarters it has been suggested that a reassessment of the traditional `disfavored' status of the malicious prosecution tort, and a relaxation of some of the traditional elements of the tort, may be in order." We noted, however, that "most of the academic commentators have concluded that expansion of the malicious prosecution tort is not a promising remedy for the problem," and that the courts of several other states "have recently addressed this same question and, in thoughtful opinions, have rejected attempts to broaden the application of the tort...." ( Id. at p. 873.)

After reviewing the competing policy choices, we concluded that the preferable course was to emphasize the availability of sanctions within the original action itself as a means of penalizing groundless claims, rather than expanding the scope of the malicious prosecution tort. We also took note of recent measures enacted by the Legislature "to facilitate the early weeding out of patently meritless claims and to permit the imposition of sanctions in the initial lawsuit against both litigants and attorneys for frivolous or delaying conduct." ( Sheldon Appel, supra, 47 Cal.3d at pp. 873-874.) Among the measures we cited as indicative of legislative attitudes were the remedies provided by Code of Civil Procedure sections 437c, 1038, 128.5 and 409.3 remedies that were not available to a defendant when Bertero, supra, 13 Cal.3d 43 , was decided. We concluded that "[b]ecause these avenues appear to provide the most promising remedies for the general problem of frivolous litigation, we do not believe it advisable to abandon or relax the traditional limitations on malicious prosecution recovery." ( Sheldon Appel, supra, 47 Cal.3d at p. 874 .)

Less than two years ago, we applied the teachings of Sheldon Appel, supra, 47 Cal.3d 863 , in Rubin v. Green, supra, 4 Cal.4th 1187 . There, we held that a lawsuit alleging the wrongful solicitation of clients filed by the defendant in a pending action against the attorneys for the opposing plaintiffs was not maintainable. Among other reasons, we relied on the fact that such a lawsuit was "inconsistent with the choice made in Sheldon Appel, supra, 47 Cal.3d 863 , where we specifically discounted another round of litigation as an antidote for the fevers of litigiousness, preferring instead the increased use of sanctions within the underlying lawsuit and legislative measures." ( 4 Cal.4th at p. 1199 .)

In assessing the significance of this development, it is important to underline the conditions that drove our reasoning in Sheldon Appel and the exact nature of our conclusion. The majority opinion rejects defendants' reliance on our endorsement in Sheldon Appel of measures such as section 128.5 on the ground that the Legislature did not intend "to substitute this remedy for the cause of action for malicious prosecution...." (Maj. opn., ante, at p. 687.) Our unanimous opinion in Sheldon Appel, supra, 47 Cal.3d 863 , was not, however, parsing the intent behind intervening legislation. Instead, we wrote as the state's highest court, responsible for shaping the scope and availability of common law tort remedies in light of the perceived wisdom of the day. As a part of that climate of opinion, we relied on statutory efforts to reduce frivolous litigation, including enactment of section 128.5, as emblematic of legislative attitudes toward both the "litigation explosion" and effective correctives for it. Taking our cue from the choice evident in such measures, we reasoned in effect that the courts might usefully supplement the solutions of positive law by declining to expand the malicious prosecution tort, relying instead on intrasuit sanctions. ( 47 Cal.3d at p. 874 .) [2]

It is a short step from embracing the proposition that statutory sanctions for frivolous claims are a superior substitute for an expanded malicious prosecution action as a means of reducing groundless lawsuits, to adopting the view that, under circumstances in which an adequate intrasuit remedy is available to redress certain types of groundless pleadings, the availability of the derivative tort remedy should be contracted. Indeed, the symmetry by which the lesser harm of a groundless alternate theory is redressed by the intramural remedy of sanctions is virtually exact. The reduced burden of defending against improbable (so to speak) theories of recovery allegedly supporting a single claim for relief is sufficiently compensated for by the imposition of monetary sanctions in the initial lawsuit, while the greater harm of defending against fabricated causes of action continues to justify the derivative malicious prosecution claim. The net result, of course, is an incremental reduction in the quantum of litigation and the associated expenditure of resources, a more closely calibrated deterrent to judicial access, and a sanction more commensurate with the offense.

The majority rejects this result by the expedient of failing to discriminate between groundless claims and groundless theories : "We are willing," the majority writes, "... to incur [the] burdens and costs [of litigation] when [it] is well founded or, even when ultimately unsuccessful, was at least initiated with probable cause and without malice.... That policy becomes counterproductive, however, when it operates to promote litigation that is groundless and motivated by malice; such litigation has no place in our judicial system...." (Maj. opn., ante, at p. 694, italics added.) The simple fact one which the majority persists in ignoring is that by definition a claim seeking to vindicate a single primary right and falling within the alternate theory rule is not entirely groundless. At least one of the several theories alleged in support of the underlying primary right is supported by probable cause, again by definition. As a result, the underlying claim itself is incapable of supporting a subsequent malicious prosecution action. That conclusion, in my view, is sufficient to place the alternate theory cases on a different footing from those alleging groundless causes of action.

Likewise, the majority's argument that we should not overrule Bertero because the remedies available to litigants in the form of intrasuit sanctions and retaliatory malicious prosecution actions are not coextensive again misses the point. To be sure, they are not coextensive. That is the point. The defendant subjected to the single claim/multiple theory lawsuit is remitted to an intrasuit remedy precisely because the burden of defense is assessed as having been less than that facing a party forced to defend against several groundless causes of action. After all, a lesser invasion of the defendant's interest in being free from unjustified litigation merits a commensurably lesser sanction.

Under circumstances in which it is possible to vindicate the interests in promoting judicial access and restraining additional litigation, while at the same time responding to the interests of those who are forced to defend against groundless allegations, this court should not hesitate to do so. Given the hindsight furnished by a generation of experience since our decision in Bertero, supra, 13 Cal.3d 43 , this is, to my mind, a case in which the policies promoting judicial access and curbing lawsuits are not outweighed by those that encourage derivative litigation. I would overrule our alternate theory holding in Bertero and direct the Court of Appeal to affirm the judgment of the trial court sustaining defendants' demurrer to the complaint.

On November 30, 1994, the opinion was modified to read as printed above.

[1] The firm has since changed its name.

[2] The summary that follows is based on the complaint and a probated will attached thereto as an exhibit, supplemented by certain court records in the underlying probate proceeding that the parties request us to judicially notice pursuant to Evidence Code section 459. We also incorporate certain facts recited in the published decision in an appeal from a portion of the probate proceeding ( Estate of Katleman (1993) 13 Cal. App.4th 51 [ 16 Cal. Rptr.2d 468 ]), again, facts that we may judicially notice. ( Weiner v. Mitchell, Silberberg & Knupp (1980) 114 Cal. App.3d 39, 45-46 [ 170 Cal. Rptr. 533 ].)

[3] At the time of Beldon Katleman's death his estate was valued in excess of $10 million.

[4] Defendants make no such claim with regard to the omission of the revocation ground from the third cause of action. We therefore take no further notice of that omission.

[5] We pause to observe that although defendants' first point may be conceded at this stage of the proceedings, their second point was without merit for the reasons stated in Lucchesi v. Giannini & Uniack (1984) 158 Cal. App.3d 777, 785-788 [ 205 Cal. Rptr. 62 ]; their third point was premised on their second point; and their fourth point failed because, as the Court of Appeal herein concluded, even if judicial notice is taken of the cited testimony of the two witnesses it does not establish probable cause as a matter of law.

[6] The instruction stated in part: "you are instructed that a defendant in a malicious prosecution action cannot escape liability for the malicious prosecution of a claim for which he did not have probable cause by joining it with a claim for which he did have probable cause...." ( Bertero, supra, 13 Cal.3d at p. 55, fn. 4 .)

[7] The four questions we addressed in Sheldon Appel were: "(1) the respective roles of the court and the jury in the determination of [the probable cause] element; (2) whether probable cause is to be tested by an objective or a subjective standard, or some combination of the two; (3) whether the adequacy of a defendant attorney's investigation or legal research is relevant to the probable cause determination; and (4) whether expert testimony is admissible on the issue." ( 47 Cal.3d at p. 874 .)

[8] In any event, because the law favors the prompt settlement of estates (e.g., Estate of Horn (1990) 219 Cal. App.3d 67 , 71 [ 268 Cal. Rptr. 41 ], and cases and authorities cited), the will contest brought by defendants herein was itself a "disfavored" cause of action. Little is gained by the use of such epithets.

[9] We recently made that implication express in Rubin v. Green (1993) 4 Cal.4th 1187 [ 17 Cal. Rptr.2d 828 , 847 P.2d 1044 ] ( Rubin ). There we held that one who is the target of a threatened lawsuit cannot maintain a retaliatory action charging the attorneys for the opposing party with "soliciting" the suit. In so holding we again emphasized such a plaintiff's right to invoke, in the underlying suit, a variety of statutory sanctions against frivolous claims or delaying tactics ( id. at p. 1199); but we concluded by acknowledging that "Ultimately, of course, plaintiff is free to prosecute a malicious prosecution action, provided the requisite conditions are pleaded and proven." ( Id. at p. 1200, fn. omitted.)

[10] The primary right theory is also occasionally invoked to resolve a question of venue. (E.g., Peiser v. Mettler (1958) 50 Cal.2d 594, 603-606 [ 328 P.2d 953 , 74 A.L.R.2d 1 ].) And before the adoption of the modern rule that a complaint may be amended after the statute of limitations has run provided recovery is sought "on the same general set of facts" ( Austin v. Massachusetts Bonding & Insurance Co. (1961) 56 Cal.2d 596, 600 [ 15 Cal. Rptr. 817 , 364 P.2d 681 ]), the primary right theory was invoked to determine when such an amendment was permissible because it did not state "a different cause of action." (E.g., Big Boy Drilling Corp. v. Rankin (1931) 213 Cal. 646, 648-649 [ 3 P.2d 13 ].)

[11] In their reply brief defendants once more shift their argument. They urge that regardless of the assertion of invalid theories of liability, a malicious prosecution claim cannot be maintained if there was probable cause to assert the violation of the primary right itself or, as the Court of Appeal put it, when the cause of action is brought with probable cause. The argument is no more persuasive than its predecessors. When a complaint alleges multiple theories of liability or "counts," the counts "are merely ways of stating the same cause of action differently." ( Bay Cities Paving & Grading, Inc. v. Lawyers' Mutual Ins. Co. (1993) 5 Cal.4th 854, 860, fn. 1 [ 21 Cal. Rptr.2d 691 , 855 P.2d 1263 ].) Accordingly, the only way that a litigant can show probable cause for the cause of action as a whole or for the "primary right" is to show probable cause for each of the counts or theories alleged. In this event the whole is indeed the sum of the parts.

[12] Finally, we understand that trial courts may be more reluctant to charge litigants or attorneys appearing before them with bad faith than juries to whom such persons are total strangers. The case at bar presents a striking example of this phenomenon. It will be remembered that in August 1991 the probate court awarded Carole Katleman her statutory share of the estate as an omitted spouse, i.e., one-half of Beldon Katleman's separate property and all the community property. At the time of Beldon Katleman's death, however, his estate real property, stocks and other investments, bank accounts, etc. stood in his sole name as his separate property. Accordingly, in October 1992 Carole Katleman filed a "property characterization petition" seeking to have the entire estate declared community property. Her claim was that he had "transmuted" his separate property into community property in 1980; at the time, such transmutations could be accomplished orally. (Now see Fam. Code, § 852 [transmutations after Jan. 1, 1985, must be "made in writing by an express declaration"].) To that end Carole Katleman testified that at the time of their second marriage Beldon Katleman made certain remarks to her concerning his separate property that led her to believe he was giving her a community interest in all such property and had the legal effect of making such a gift.

The probate court rejected her petition, ruling that except for a car and a ring stipulated to have been interspousal gifts, Beldon Katleman's entire estate was his separate property. In its statement of decision the court declared that it found Carole Katleman's testimony to be wholly unbelievable and false, a "manufactured script" designed to defeat Crowley's rights under the will. The court based this finding on stipulated acts and omissions of the parties during their two marriages, Carole Katleman's sworn statements in prior documents filed in the probate proceeding, her answers to interrogatories, her testimony in the will contest, her unexplained delay in making the subject claim, and her demeanor on the stand. The court concluded: "The claim of Mrs. Katleman now before the Court is absolutely without merit. Her willingness to commit perjury for the pecuniary gain of obtaining Mr. Katleman's estate is overwhelming."

Nevertheless, the same court thereafter denied Crowley's motion that Carole Katleman be sanctioned under section 128.5 for pressing her perjurious claim of oral transmutation, on the grounds that "The great lack of merit of her claim did not amount to bad faith" and "the Court cannot say the case was one totally devoid of merit or solely intended to harass Mr. Crowley or cause unnecessary delay." The court also declined to sanction the defendant attorneys, extolling them as persons "well known to this Court to be attorneys of great excellence and integrity."

[13] After this case was argued in this court legislation was enacted that will suspend the operation of section 128.5 until January 1, 1999, substituting in its place, for a four-year trial period, a statute modeled on recently revised rule 11 of the Federal Rules of Civil Procedure (28 U.S.C.). (Stats. 1994, ch. 1062, §§ 1-3, eff. Jan. 1, 1995.) The effect, if any, of this legislation on the question before us is obviously not presented in this appeal.

Another post- Bertero statute cited by defendants is Code of Civil Procedure section 1038, but it too offers an extremely limited remedy. It provides that if a trial court, on motion of the defendant prior to judgment, determines in certain cases that the action was not brought in good faith and with reasonable cause, it shall award the defendant reasonable and necessary attorney fees and expert witness costs. But by its terms the statute applies only to (1) actions brought against public entities and public employees and (2) actions for indemnity or contribution. ( Id., subd. (a).) And even in those limited cases, the statute applies only when the action is terminated by a successful defense motion for summary judgment, nonsuit or directed verdict ( id., subd. (d)); an ordinary judgment for the defendant, no matter how favorable, is insufficient to support the remedy. Finally, although parties who actually invoke this remedy waive their remedy of malicious prosecution, the two remedies are in fact alternatives: the statute expressly declares that "Failure to make the motion shall not be deemed a waiver of the right to pursue a malicious prosecution action." ( Id., subd. (c).)

[14] Defendants note that one of those grounds lack of due execution is analogous to a denial, because the proponent of the will has the burden of proving such execution in any event. (Prob. Code, § 8252, subd. (a).)

[15] There is no merit in defendants' additional contention that Bertero, supra, 13 Cal.3d 43 , is contrary to the rule that an interim adverse judgment on the merits, even though subsequently set aside on motion or on appeal, conclusively establishes probable cause for the prior action. (See, e.g., Bealmear v. So. Cal. Edison Co. (1943) 22 Cal.2d 337, 340 [ 139 P.2d 20 ]; Fairchild v. Adams, supra, 170 Cal. App.2d 10, 15 .) Neither in Bertero nor in the case at bar was there any such interim judgment; as the Court of Appeal herein correctly held, the denial of Crowley's motion for summary adjudication of issues as to all grounds except lack of due execution was not a judgment on the merits for that purpose. ( Lucchesi v. Giannini & Uniack, supra, 158 Cal. App.3d 777, 785-788 .)

[16] Indeed, at common law to promote such litigation constituted the offense of barratry. (See Rubin, supra, 4 Cal.4th at p. 1190 .)

[17] In view of this conclusion we need not reach defendants' secondary contention that the Court of Appeal erred in holding that judicial notice of the trial testimony of two witnesses to the will would not establish probable cause as a matter of law for the ground of lack of due execution.

[1] In point of fact, I am not at all persuaded that the use of a "primary right" analysis would fail to provide a workable means of segregating those malicious prosecution actions that should go forward and those that should be barred in favor of sanctions within the prior action. Applied under its more familiar label of "cause of action" (see, e.g., 4 Witkin, Cal. Procedure (3d ed. 1985) Pleading, § 23, pp. 66-67), and interpreted in light of modern transactional notions (cf. Rest.2d, Judgments, § 24, subd. (1)), the concept seems sufficiently stable to be applied effectively by the trial courts in passing on, say, a demurrer to a malicious prosecution complaint. (Cf. Bay Cities Paving & Grading, Inc. v. Lawyers' Mutual Ins. Co. (1993) 5 Cal.4th 854 [ 21 Cal. Rptr.2d 691 , 855 P.2d 1263 ].)

[2] Fortunately, the damage inflicted by the majority's constrictive gloss of section 128.5 is erased by the recent repeal of the statute and the enactment of new sanctions provisions incorporating the substance of rule 11 of the Federal Rules of Civil Procedure (28 U.S.C.), itself recently revised. (Assem. Bill No. 3594 (1993-1994 Reg. Sess.), signed by the Governor on Sept. 28, 1994, eff. Jan. 1, 1995.) Among other expansions in the trial court's powers to sanction misconduct by counsel or a party, the new statute suspends the operation of section 128.5 for four years and substitutes in its place the text of federal rule 11, modified in minor particulars. The new statute authorizes trial judges to sanction attorneys, their firms and clients for violating a certification that a complaint (as well as other filings) is not filed "primarily for an improper purpose," that the claims are warranted by existing law (with certain exceptions), and that allegations have factual support. (Assem. Bill No. 3594, supra, enacting Code Civ. Proc., § 128.7, subd. (b).) If these terms are not intended to reach misconduct analogous to that addressed by malicious prosecution actions, I am at a loss to fathom what would.