Opinion
KENNARD, J.
Plaintiff, a child injured during birth, alleges that defendant hospital intentionally destroyed evidence relevant to his malpractice action against the hospital. He seeks to bring a separate tort cause of action against defendant hospital for its alleged intentional spoliation—that is, intentional destruction or suppression—of evidence.
The intentional destruction of evidence is a grave affront to the cause of justice and deserves our unqualified condemnation. There are, however, existing and effective nontort remedies for this problem. Moreover, a tort remedy would impose a number of undesirable social costs, as well as running counter to important policies against creating tort remedies for litigation-related misconduct. As we shall explain below, we conclude that when the alleged intentional spoliation is committed by a party to the underlying cause of action to which the evidence is relevant and when the spoliation is or reasonably should have been discovered before the conclusion of the underlying litigation, it is preferable to rely on existing nontort remedies rather than creating a tort remedy.
I
Plaintiff Kristopher Schon Bowyer, through his guardian ad litem, brought a medical malpractice action against defendant Cedars-Sinai Medical Center (hereafter sometimes hospital) and others for injuries he allegedly sustained because of oxygen deprivation during birth. In the course of pretrial discovery, plaintiff’s attorney sought from defendant hospital copies of plaintiff’s [5]*5medical records; defendant hospital was unable to locate certain records, including fetal monitoring strips recording plaintiff’s heartbeat during labor.
Plaintiff’s attorney thereafter filed a second amended complaint, adding a cause of action for intentional spoliation of evidence and alleging that the hospital had intentionally destroyed the missing records to prevent plaintiff from prevailing in his malpractice action. The complaint sought punitive damages on plaintiff’s cause of action for intentional spoliation. Defendant hospital moved to strike plaintiff’s punitive damages claim on the ground that plaintiff had not complied with Code of Civil Procedure section 425.13, and the trial court granted the motion. Under section 425.13, a plaintiff may not file a complaint seeking punitive damages in an action arising out of the professional negligence of a health care provider unless the court grants an order permitting the complaint to be filed; the court may grant the order only if the plaintiff establishes through affidavits a substantial probability of prevailing on the punitive damages claim.
Plaintiff then moved under Code of Civil Procedure section 425.13 for leave to file a third amended complaint seeking punitive damages. The trial court granted plaintiff’s motion. Defendant hospital petitioned the Court of Appeal for a writ of mandate. After issuing the alternative writ, the Court of Appeal denied defendant’s petition in a written opinion holding that section 425.13 did not apply to plaintiff’s claim of intentional spoliation because the alleged spoliation did not occur while defendant hospital was rendering professional medical services to plaintiff. We granted review to decide whether this court should recognize a tort remedy for the intentional destruction or suppression of evidence by a party to the underlying litigation and, if so, whether section 425.13 applies to claims for punitive damages for acts of intentional spoliation by a health care provider.
II
At the threshold of this case stands the question of whether this court should recognize a tort remedy for the intentional spoliation of evidence by a party to the underlying cause of action to which the evidence is relevant (what we shall term a “first party” spoliator) when, as here, the spoliation is or reasonably should have been discovered before the trial or other decision on the merits of the underlying cause of action. This court has not previously addressed the question of whether tort remedies should exist for acts of spoliation.1
Plaintiff, however, raises a procedural objection to our consideration of this threshold issue. Plaintiff contends that because the existence of the [6]*6tort was not an issue raised in the courts below, we should not decide it. We disagree. Our power of decision, of course, extends to the entire case (Cal. Rules of Court, rule 29.2(a)), although as a matter of policy we ordinarily exercise that power only with respect to issues raised in the Court of Appeal (id., rule 29(b)). The petition for review that we granted squarely raised the issue of whether this court should recognize a tort cause of action for intentional first party spoliation, and the issue has been extensively briefed not only by the parties but also by numerous amici curiae. It is an issue of law that does not turn on the facts of this case, it is a significant issue of widespread importance, and it is in the public interest to decide the issue at this time. Given the prior recognition of the tort by the lower courts (see Willard v. Caterpillar, Inc. (1995) 40 Cal.App.4th 892 [48 Cal.Rptr.2d 607]; Smith v. Superior Court (1984) 151 Cal.App.3d 491 [198 Cal.Rptr. 829]), delaying until some future case an announcement of our conclusion that a tort remedy should not be recognized in the circumstances present here would be extremely wasteful of the resources of both courts and parties, for they would continue to litigate such cases on the assumption that the tort exists.
This is not the first occasion on which we have addressed a dispositive issue not raised by the parties below. In Fisher v. City of Berkeley, we decided a potentially dispositive threshold issue raised for the first time in this court by an amicus curiae (the validity under federal antitrust law of Berkeley’s rent control ordinance) because it was an issue of law not turning on disputed facts and because it was an important question of public policy. (Fisher v. City of Berkeley (1984) 37 Cal.3d 644, 654, fn. 3 [209 Cal.Rptr. 682, 693 P.2d 261] [“parties may advance new theories on appeal when the issue posed is purely a question of law based on undisputed facts, and involves important questions of public policy”]; see also Ford v. Gouin (1992) 3 Cal.4th 339, 346-350, fn. 2 [11 Cal.Rptr.2d 30, 834 P.2d 724, 34 A.L.R.Sth 769] (plur. opn. of Arabian, J.) [deciding case by applying Harbors and Navigation Code section 658, a ground never raised in the trial court, appellate court, or this court]; 3 Cal.4th at pp. 364-369 (cone, and dis. opn. of George, J., joined by Lucas, C. J.) [same]; id. at p. 369 (dis. opn. of Mosk, J.) [same].) Moreover, here the trial court was bound by prior appellate decisions recognizing the tort of spoliation (see Auto Equity Sales, Inc. v. Superior Court (1962) 57 Cal.2d 450, 455 [20 Cal.Rptr. 321, 369 P.2d 937]) and it would therefore have been pointless to raise the issue there (see [7]*7Moradi-Shalal v. Fireman’s Fund Ins. Companies (1988) 46 Cal.3d 287, 292, fn. 1 [250 Cal.Rptr. 116, 758 P.2d 58] [deciding issue raised for the first time in this court]; Fisher v. City of Berkeley, supra, 37 Cal.3d at p. 655, fn. 3 [same]). Finally, for this court to decide the application of Code of Civil Procedure section 425.13’s pleading rule to the tort of intentional first party spoliation without deciding whether the tort itself exists would risk rendering a purely academic and hypothetical decision.2
[8]*8In considering whether to create a tort remedy for intentional first party spoliation that is or reasonably should have been discovered before trial of the underlying action, we begin with certain general principles of tort law. “A tort, whether intentional or negligent, involves a violation of a legal duty, imposed by statute, contract or otherwise, owed by the defendant to the person injured.” (5 Witkin, Summary of Cal. Law (9th ed. 1988) Torts, § 6, p. 61.) At issue here is whether to impose on parties to a lawsuit a duty to avoid the intentional destruction of evidence relevant to the lawsuit. As we have stated, the concept of duty “ ‘is a shorthand statement of a conclusion, rather than an aid to analysis in itself.’ ” (Dillon v. Legg (1968) 68 Cal.2d 728, 734 [69 Cal.Rptr. 72, 441 P.2d 912, 29 A.L.R.3d 1316].) It is “ ‘only an expression of the sum total of those considerations of policy which lead the law to say that the particular plaintiff is entitled to protection.’ ” (Ibid.) Thus, we must examine and weigh the relevant “considerations of policy” that favor or oppose a tort remedy for intentional first party spoliation.
No one doubts that the intentional destruction of evidence should be condemned. Destroying evidence can destroy fairness and justice, for it increases the risk of an erroneous decision on the merits of the underlying cause of action. Destroying evidence can also increase the costs of litigation as parties attempt to reconstruct the destroyed evidence or to develop other evidence, which may be less accessible, less persuasive, or both.
That alone, however, is not enough to justify creating tort liability for such conduct. We must also determine whether a tort remedy for the intentional first party spoliation of evidence would ultimately create social benefits exceeding those created by existing remedies for such conduct, and outweighing any costs and burdens it would impose. Three concerns in particular stand out here: the conflict between a tort remedy for intentional first party spoliation and the policy against creating derivative tort remedies for litigation-related misconduct; the strength of existing nontort remedies for spoliation; and the uncertainty of the fact of harm in spoliation cases.
Our inquiry into whether to create a tort remedy for the intentional spoliation of evidence must begin with a recognition that using tort law to correct misconduct arising during litigation raises policy considerations not present in deciding whether to create tort remedies for harms arising in other contexts. In the past, we have favored remedying litigation-related misconduct by sanctions imposed within the underlying lawsuit rather than by [9]*9creating new derivative torts. In Sheldon Appel Co. v. Albert & Oliker (1989) 47 Cal.3d 863, 873 [254 Cal.Rptr. 336, 765 P.2d 498], we rejected a proposed expansion of the tort of malicious prosecution with the following observation: “While the filing of frivolous lawsuits is certainly improper and cannot in any way be condoned, in our view the better means of addressing the problem of unjustified litigation is through the adoption of measures facilitating the speedy resolution of the initial lawsuit and authorizing the imposition of sanctions for frivolous or delaying conduct within that first action itself, rather than through an expansion of the opportunities for initiating one or more additional rounds of malicious prosecution litigation after the first action has been concluded.”
On other occasions as well, we have warned of the dangers of creating new torts to remedy litigation-related misconduct. In Rubin v. Green (1993) 4 Cal.4th 1187 [17 Cal.Rptr.2d 828, 847 P.2d 1044], in the course of balancing the utility of a tort remedy for litigation-related misconduct (improper attorney solicitation of clients) against the burdens it would impose, we noted: “[I]t does not follow [from the existence of litigation-related misconduct] that we should adopt a remedy that itself encourages a spiral of lawsuits. fi[] . . . fi[] . . . [W]e [have] 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.” (Id. at p. 1199.) And, in Silberg v. Anderson (1990) 50 Cal.3d 205 [266 Cal.Rptr. 638, 786 P.2d 365], in the course of discussing the litigation communications privilege (Civ. Code, § 47, subd. 2), we observed: “[T]he law places upon litigants the burden of exposing during trial the bias of witnesses and the falsity of evidence, thereby enhancing the finality of judgments and avoiding an unending roundelay of litigation . . . . fi[] For our justice system to function, it is necessary that litigants assume responsibility for the complete litigation of their cause during the proceedings. To allow a litigant to attack the integrity of evidence after the proceedings have concluded, except in the most narrowly circumscribed situations, such as extrinsic fraud, would impermissibly burden, if not inundate, our justice system.” (50 Cal.3d at p. 214.)
Perjury, like spoliation, undermines the search for truth and fairness by creating a false picture of the evidence before the trier of fact. Perjury does so by creating false evidence; spoliation does so by destroying authentic evidence. Yet we have held that there is no civil remedy in damages against a witness who commits perjury when testifying. (Taylor v. Bidwell (1884) 65 Cal. 489, 490 [4 P. 491].) In reaching that conclusion, we relied on a New York case that concluded “it would be productive of endless litigation” to permit the victim of a judgment allegedly based on false testimony to bring [10]*10an action for damages. {Smith v. Lewis (N.Y. 1808) 3 Johns. 157, 168 (Kent, C. J.).)
In reliance on our decision in Taylor v. Bidwell, supra, 65 Cal. 489, one Court of Appeal later held that there can be no tort action for the concealment or withholding of evidence. {Agnew v. Parks (1959) 172 Cal.App.2d 756, 765-766 [343 P.2d 118].) Other Court of Appeal decisions have rejected other attempts, put forward under a variety of legal theories, to seek damages for the presentation of false evidence. {Mercury Casualty Co. v. Superior Court (1986) 179 Cal.App.3d 1027, 1034-35 [225 Cal.Rptr. 100] [rejecting fraud action based on allegations that insurer of opposing party in underlying action had presented false testimony in that action]; Rios v. Allstate Ins. Co. (1977) 68 Cal.App.3d 811, 817-819 [137 Cal.Rptr. 441] [rejecting insured’s action for bad faith alleging that, in arbitration between insurer and insured, insurer had presented false evidence and testimony]; Kachig v. Boothe (1971) 22 Cal.App.3d 626, 636, 640-641 [99 Cal.Rptr. 393] [rejecting action for fraud and intentional infliction of emotional distress arising out of prior judgment that rested on false testimony and false documentary evidence].)
These cases denying a tort remedy for the presentation of false evidence or the suppression of evidence rest on a concern for the finality of adjudication. This same concern underlies another line of cases that forbid direct or collateral attack on a judgment on the ground that evidence was falsified, concealed, or suppressed. After the time for seeking a new trial has expired and any appeals have been exhausted, a final judgment may not be directly attacked and set aside on the ground that evidence has been suppressed, concealed, or falsified; in the language of the cases, such fraud is “intrinsic” rather than “extrinsic.” {La Salle v. Peterson (1934) 220 Cal. 739, 740-742 [32 P.2d 612]; Pico v. Cohn (1891) 91 Cal. 129, 133-135 [25 P. 970]; Burch v. Hibernia Bank (1956) 146 Cal.App.2d 422, 432-433 [304 P.2d 212]; 8 Witkin, Cal. Procedure, supra, Attack on Judgment in Trial Court, § 242, p. 757 [“fraud occurring in the course of the proceeding is not a ground for equitable relief’].) Similarly, under the doctrines of res judicata and collateral estoppel, a judgment may not be collaterally attacked on the ground that evidence was falsified or destroyed. {Jorgensen v. Jorgensen (1948) 32 Cal.2d 13,18-19 [193 P.2d 728]; Adams v. Martin (1935) 3 Cal.2d 246, 248-249 [44 P.2d 572]; Eichman v. Fotomat Corp. (1983) 147 Cal.App.3d 1170, 1175-1176 [197 Cal.Rptr. 612].)
As we explained more than a century ago, the rule against vacating judgments on the ground of false evidence or other intrinsic fraud serves the important interest of finality in adjudication: “[W]e think it is settled beyond [11]*11controversy that a decree will not be vacated merely because it was obtained by forged documents or perjured testimony. The reason of this rule is, that there must be an end of litigation; and when parties have once submitted a matter ... for investigation and determination, and when they have exhausted every means for reviewing such determination in the same proceeding, it must be regarded as final and conclusive . . . . [f] . . . [W]hen [the aggrieved party] has a trial, he must be prepared to meet and expose perjury then and there. . . . The trial is his opportunity for making the truth appear. If, unfortunately, he fails, being overborne by perjured testimony, and if he likewise fails to show the injustice that has been done him on motion for a new trial, and the judgment is affirmed on appeal, he is without remedy. The wrong, in such case, is of course a most grievous one, and no doubt the legislature and the courts would be glad to redress it if a rule could be devised that would remedy the evil without producing mischiefs far worse than the evil to be remedied. Endless litigation, in which nothing was ever finally determined, would be worse than occasional miscarriages of justice . . . .” (Pico v. Cohn, supra, 91 Cal. 129, 133-134; accord, United States v. Throckmorton (1878) 98 U.S. 61, 68-69 [25 L.Ed. 93].)
Weighing against our recognition of a tort cause of action for spoliation in this case are both the strong policy favoring use of nontort remedies rather than derivative tort causes of action to punish and correct litigation misconduct and the prohibition against attacking adjudications on the ground that evidence was falsified or destroyed. In particular, there are a number of nontort remedies that seek to punish and deter the intentional spoliation of evidence.
Chief among these is the evidentiary inference that evidence which one party has destroyed or rendered unavailable was unfavorable to that party. This evidentiary inference, currently set forth in Evidence Code section 413 and in the standard civil jury instructions, has a long common law history. (See The Pizarra (1817) 15 U.S. (2 Wheat.) 227, 240 [4 L.Ed. 226] (per Story, J.); 2 McCormick on Evidence (4th ed. 1992) § 265, pp. 191-192; 2 Wigmore on Evidence (Chadboum rev. 1979) §§ 278, 291, pp. 133, 221; Maguire & Vincent, Admissions Implied From Spoliation or Related Conduct (1935) 45 Yale L.J. 226.) For example, in the case of Armory v. Delamirie (1722 K.B.) 93 Eng. Rep. 664, a chimney sweep sought to recover a jewel he had given to a jeweler for appraisal. When the jeweler failed to produce the jewel at trial, the court instructed the jury “that unless the [jeweler] did produce the jewel, and shew it not to be of the finest water, they should presume the strongest against him, and make the value of the best jewels the measure of their damages . . . .” (Ibid.) This court, too, has long recognized the appropriateness of this inference. (Fox v. Hale & Norcross S. M. Co. (1895) 108 Cal. 369, 415-417 [41 P. 308].)
[12]*12As presently set forth in Evidence Code section 413, this inference is as follows: “In determining what inferences to draw from the evidence or facts in the case against a party, the trier of fact may consider, among other things, the party’s . . . willful suppression of evidence relating thereto . . . The standard California jury instructions include an instruction on this inference as well: “If you find that a party willfully suppressed evidence in order to prevent its being presented in this trial, you may consider that fact in determining what inferences to draw from the evidence.” (BAJI No. 2.03 (8th ed. 1994).) Trial courts, of course, are not bound by the suggested language of the standard BAJI instruction and are free to adapt it to fit the circumstances of the case, including the egregiousness of the spoliation and the strength and nature of the inference arising from the spoliation.
In addition to the evidentiary inference, our discovery laws provide a broad range of sanctions for conduct that amounts to a “misuse of the discovery process.” (Code Civ. Proc., § 2023, subd. (b).) Section 2023 of the Code of Civil Procedure gives examples of misuses of discovery, including “[flailing to respond or to submit to an authorized method of discovery” (id., subd. (a)(4)) or “[mjaking an evasive response to discovery.” (Id., subd. (a)(6).) Destroying evidence in response to a discovery request after litigation has commenced would surely be a misuse of discovery within the meaning of section 2023, as would such destruction in anticipation of a discovery request.
The sanctions under Code of Civil Procedure section 2023 are potent. They include monetary sanctions, contempt sanctions, issue sanctions ordering that designated facts be taken as established or precluding the offending party from supporting or opposing designated claims or defenses, evidence sanctions prohibiting the offending party from introducing designated matters into evidence, and terminating sanctions that include striking part or all of the pleadings, dismissing part or all of the action, or granting a default judgment against the offending party. Plaintiff remains free to seek these remedies in this case.
Another important deterrent to spoliation is the customary involvement of lawyers in the preservation of their clients’ evidence and the State Bar of California disciplinary sanctions that can be imposed on attorneys who participate in the spoliation of evidence. As a practical matter, modem civil discovery statutes encourage a lawyer to marshal and take charge of the client’s evidence, most often at an early stage of the litigation. In doing so, a lawyer customarily instructs the client to preserve and maintain any potentially relevant evidence, not only because it is right for the client to do so but also because the lawyer recognizes that, even if the evidence is [13]*13unfavorable, the negative inferences that would flow from its intentional destruction are likely to harm the client as much as or more than the evidence itself.
In addition, the risk that a client’s act of spoliation may suggest that the lawyer was also somehow involved encourages lawyers to take steps to protect against the spoliation of evidence. Lawyers are subject to discipline, including suspension and disbarment, for participating in the suppression or destruction of evidence. (Bus. & Prof. Code, § 6106 [“The commission of any act involving moral turpitude, dishonesty or corruption . . . constitutes a cause for disbarment or suspension.”]; id., § 6077 [attorneys subject to discipline for breach of Rules of Professional Conduct]; Rules Prof. Conduct, rule 5-220 [“A member shall not suppress any evidence that the member or the member’s client has a legal obligation to reveal or to produce.”].) The purposeful destruction of evidence by a client while represented by a lawyer may raise suspicions that the lawyer participated as well. Even if these suspicions are incorrect, a prudent lawyer will wish to avoid them and the burden of disciplinary proceedings to which they may give rise and will take affirmative steps to preserve and safeguard relevant evidence.
Finally, Penal Code section 135 creates criminal penalties for spoliation. “Every person who, knowing that any book, paper, record, instrument in writing, or other matter or thing, is about to be produced in evidence upon any trial, inquiry, or investigation whatever, authorized by law, willfully destroys or conceals the same, with intent thereby to prevent it from being produced, is guilty of a misdemeanor.” (Ibid)
These nontort remedies for spoliation are both extensive and apparently effective for, although real, the problem of spoliation does not appear to be widespread. The committee of distinguished judges and attorneys in charge of preparing the standard California jury instructions describes as “relatively rare [the] case in which evidence has been willfully suppressed.” (Use Note to BAJI No. 2.03, supra, p. 22.) The reported California cases describing instances of intentional spoliation are not numerous. The infrequency of spoliation suggests that existing remedies are generally effective at deterring spoliation.
III
Another consideration weighing against recognition of a tort remedy for intentional first party spoliation is the uncertainty of the fact of harm in spoliation cases. It seems likely that in a substantial proportion of spoliation cases the fact of harm will be irreducibly uncertain. In such cases, even if the [14]*14jury infers from the act of spoliation that the spoliated evidence was somehow unfavorable to the spoliator, there will typically be no way of telling what precisely the evidence would have shown and how much it would have weighed in the spoliation victim’s favor. Without knowing the content and weight of the spoliated evidence, it would be impossible for the jury to meaningfully assess what role the missing evidence would have played in the determination of the underlying action. The jury could only speculate as to what the nature of the spoliated evidence was and what effect it might have had on the outcome of the underlying litigation.
One court considering the question has observed the following: “[I]t is impossible to know what the destroyed evidence would have shown. ... It would seem to be sheer guesswork, even presuming that the destroyed evidence went against the spoliator, to calculate what it would have contributed to the plaintiff’s success on the merits of the underlying lawsuit. . . . The lost evidence may have concerned a relevant, but relatively trivial matter. If evidence would not have helped to establish plaintiff’s case an award of damages for its destruction would work a windfall for the plaintiff.” (Petrik v. Monarch Printing Corp. (1986) 150 Ill.App.3d 248, 260-261 [103 Ill.Dec. 774, 501 N.E.2d 1312, 1320].)
In the many spoliation cases in which the fact of harm is uncertain, a tort remedy for first party spoliation would not accurately compensate for losses caused by spoliation or correct errors in the determination of the issues in the underlying litigation.3 In the past, we have considered the uncertainty of determining hypothetically whether a particular plaintiff would have prevailed on a legal claim as sufficient reason for refusing to recognize a tort remedy for other forms of wrongful conduct. (Taylor v. Hopper (1929) 207 Cal. 102, 103-105 [276 P. 990] [refusing to recognize a cause of action for fraudulent inducement of a settlement of a legal claim because, given the uncertainty of whether the plaintiff would have prevailed on the legal claim, [15]*15“there is no practicable measure of damages for the action sought to be maintained”]; see also Agnew v. Parks, supra, 172 Cal.App.2d 756, 768-769 [rejecting action for fraud that allegedly caused plaintiff to lose a prior lawsuit because of uncertainty as to whether plaintiff would have prevailed in the absence of the alleged fraud].)
IV
The costs that a tort remedy would impose also weigh against creation of a spoliation tort remedy. The uncertainty of the fact of harm, in addition to making a tort remedy a poor instrument for compensating spoliation victims, would create the risk of erroneous determinations of spoliation liability (that is, findings of liability in cases in which availability of the spoliated evidence would not have changed the outcome of the underlying litigation). An erroneous determination of spoliation liability would enable the spoliation victim to recover damages, or avoid liability, for the underlying cause of action when the spoliation victim would not have done so had the evidence been in existence. The availability of punitive damages would only magnify the cost of erroneous liability determinations. The risk of erroneous spoliation liability could also impose indirect costs by causing persons or entities to take extraordinary measures to preserve for an indefinite period documents and things of no apparent value solely to avoid the possibility of spoliation liability if years later those items turn out to have some potential relevance to future litigation.
There is also the cost to defendants and courts of litigating meritless spoliation actions. A separate tort remedy would be subject to abuse, for in many cases potentially relevant evidence will no longer exist at the time of trial, not because it was intentionally destroyed but simply because it has been discarded or misplaced in the ordinary course of events. (Comment, Spoliation of Evidence: A Troubling New Tort (1989) 37 U. Kan. L.Rev. 563, 592 [“A new cause of action could accrue each time a plaintiff loses a lawsuit, for in most cases there is likely to be some piece of potential evidence that is not available at the time of trial.”].) Many corporations and other entities, for example, have document retention policies under which they destroy at stated intervals documents for which they anticipate having no further need. (See Willard v. Caterpillar, Inc., supra, 40 Cal.App.4th 892, 919-924; Akiona v. U.S. (9th Cir. 1991) 938 F.2d 158, 161; Lewy v. Remington Arms Co., Inc. (8th Cir. 1988) 836 F.2d 1104, 1111-1112; Fedders & Guttenplan, Document Retention and Destruction: Practical, Legal and Ethical Considerations (1980) 56 Notre Dame L.Rev. 5, 7, 11-17, 53-55.) The mere fact of destruction, however, would permit a disappointed litigant to sue the prevailing party for spoliation, and in many cases the issue of the [16]*16defendant’s purpose in destroying the evidence, like many other issues turning on intent and state of mind, could only be resolved at trial. In this case, for example, plaintiff contends that “a trier of fact could easily find intentional spoliation of evidence” from the mere fact that defendant hospital no longer possesses the records in question.
Moreover, if, as plaintiff seeks to do here, a spoliation tort cause of action were tried jointly with the claims in the underlying action, a significant potential for jury confusion and inconsistency would arise. The jury in such a case logically would first consider the underlying claims; for if it awards the spoliation victim complete relief on the underlying claims, then the spoliation has caused no harm to the spoliation victim’s position in the underlying litigation. In doing so, the jury would consider any acts of spoliation by applying the evidentiary inference of Evidence Code section 413. If the jury rejects the spoliation victim’s position on the underlying claims, it has either rejected application of the evidentiary inference to the case before it (e.g., because the spoliation victim has not demonstrated that the spoliation was intentional), or has determined that, even applying the inference in favor of the spoliation victim, the other evidence in the case compels a different result. The jury would then consider and decide the spoliation tort claim; in doing so, however, it would necessarily be reconsidering its adjudication that either no intentional spoliation occurred or that the spoliated evidence would not have led to a different result. At the least, this would be confusing to the jury; at most, it would lead to inconsistent results.
On the other hand, pursuing a spoliation tort remedy in a proceeding separate from the underlying action would result in duplicative proceedings without avoiding the potential for inconsistent results. The spoliation action would require a “retrial within a trial,” for all of the evidence in the underlying action would have to be presented again so that the spoliation jury could determine what effect the spoliated evidence would have ha.d in light of all the other evidence. (Federated Mut. v. Litchfield Prec. Comp., supra, 456 N.W.2d 434, 438; 2 Dobbs, Law of Remedies (2d ed. 1993) § 6.6(2), p. 139 [“In some cases proof adequate to show prospective economic loss with reasonable certainty may entail a very elaborate presentation which is a burden to the courts. For instance, in a spoliation of evidence claim, the plaintiff can only prove the loss of his personal injury case by showing that the evidence the defendant lost or altered would have resulted in a favorable verdict at certain fixed damages; and this in turn can only be shown by presenting to the jury in the [spoliation] case all of the proof that would have gone to the jury in the personal injury trial.”].) This duplication of effort would be burdensome both to the parties and to the judicial system, [17]*17and it would also produce an even greater risk of inconsistent results than would deciding a spoliation tort cause of action as part of the underlying litigation.
Nor would an intentional first party spoliation tort cause of action offset these costs by significantly increasing deterrence of the destruction of evidence beyond that afforded by existing remedies. It seems dubious that a tort remedy would increase the frequency with which destruction comes to light and thereby increase deterrence. As discussed above, the motive and opportunity to discover instances of spoliation are at their greatest during discovery in the underlying action. There is no reason to conclude that instances of spoliation that remain hidden during discovery in the underlying action would come to light afterward solely by reason of the existence of a tort remedy. Nor, given the uncertainty of harm from acts of evidence destruction, would a tort remedy increase deterrence by more accurately compensating the spoliation victim and thus reducing the benefit to the spoliator.
Conclusion
As we noted at the outset, the intentional spoliation of evidence by a party to the litigation to which it is relevant is an unqualified wrong. We conclude, however, that it is the rare case in which a tort remedy for an intentionally caused harm is not appropriate. The remedies already available in first party spoliation cases to the spoliation victim, especially the evidentiary inference provided by Evidence Code section 413 and the discovery remedies of Code of Civil Procedure section 2023, provide a substantial deterrent to acts of spoliation, and substantial protection to the spoliation victim. Given that existing remedies will in most cases be effective at ensuring that the issues in the underlying litigation are fairly decided, whatever incremental additional benefits a tort remedy might create are outweighed by the policy considerations and costs described above. By opening up the decision on the merits of the underlying causes of action to speculative reconsideration regarding how the presence of the spoliated evidence might have changed the outcome, a tort remedy would not only create a significant risk of erroneous findings of spoliation liability but would impair the fundamental interest in the finality of adjudication and the stability of judgments.
Accordingly, we hold that there is no tort remedy for the intentional spoliation of evidence by a party to the cause of action to which the spoliated evidence is relevant, in cases in which, as here, the spoliation victim knows or should have known of the alleged spoliation before the trial or other [18]*18decision on the merits of the underlying action.4 We reverse the decision of the Court of Appeal denying the petition for a writ of mandate and remand the cause with directions to that court to issue a writ instructing the trial court to vacate its order granting plaintiff leave to file his third amended complaint.
George, C. J., Mosk, J., Werdegar, J., Chin, J., and Brown, J., concurred.