OPINION

GEORGE, C. J.

The mother of Katie Janeway, a developmentally disabled 14 year old, signed an application form releasing the City of Santa Barbara and its employees (hereafter the City or defendants) from liability for "any negligent act" related to Katie's participation in the City's summer camp for developmentally disabled children. Katie drowned while attending the camp, and her parents (plaintiffs, real parties in interest in the present proceedings) commenced this suit. The Court of Appeal below (1) held unanimously that the agreement embodied in the application form was effective and enforceable insofar as it concerned defendants' liability for future ordinary negligence, but (2) concluded, by a two-to-one vote, that a release of liability for future gross negligence generally is unenforceable, and that the agreement in this case did not release such liability.

In granting review, we limited the issue to be briefed and argued to the second issue — whether a release of liability relating to recreational activities generally is effective as to gross negligence. As explained below, we answer that question in the negative, and affirm the judgment rendered by the Court of Appeal. We conclude, consistent with dicta in California cases and with the vast majority of out-of-state cases and other authority, that an agreement made in the context of sports or recreational programs or services, purporting to release liability for future gross negligence, generally is unenforceable as a matter of public policy. Applying that general rule in the case now before us, we hold that the agreement, to the extent it purports to release liability for future gross negligence, violates public policy and is unenforceable.

I

The relevant facts were properly set forth by the Court of Appeal below, and we adopt that recitation with minor supplementation and stylistic changes.

The City has provided extensive summer recreational facilities and activities for children, including a camp for children with developmental disabilities — Adventure Camp. Katie Janeway, who suffered from cerebral palsy, epilepsy, and other similar developmental disabilities, participated in Adventure Camp in 1999, 2000, 2001, and 2002.

Adventure Camp was conducted from noon until 5:00 p.m. on weekdays for approximately three weeks in July and August. Camp activities included swimming, arts and crafts, group games, sports, and field trips. In 2002, as in prior years, swimming activities were held on two of five camp days each week in a City swimming pool.

In 2002, the application form for Adventure Camp included a release of all claims against the City and its employees from liability, including liability based upon negligence, arising from camp activities. Katie's mother, Maureen Janeway, signed the release. She had signed similar releases covering Katie's participation in the camp in prior years.

Maureen Janeway disclosed Katie's developmental disabilities and medical problems to the City, specifically informing the City that Katie was prone to epileptic seizures, often occurring in water, and that Katie needed supervision while swimming. In addition, the City was aware that Katie had suffered seizures while attending Adventure Camp events in 2001. She had a seizure when sitting on the pool deck and another seizure at the skating rink. Paramedics were called after her seizure on the pool deck. Nevertheless, Maureen Janeway indicated that Katie was a good swimmer, and she never sought to prevent or restrict Katie's participation in the swimming portion of Adventure Camp.

Based upon the information provided by Maureen Janeway and Katie's history of seizures, the City took special precautions during the Adventure Camp swimming activities in 2002. The City assigned Veronica Malong to act as a "counselor." Malong's responsibility was to keep Katie under close observation during the camp's swimming sessions. Previously, Malong, a college student, had worked for one year as a special education aide at the middle school attended by Katie. Malong had observed Katie experience seizures at the school, and she received instruction from the school nurse regarding the handling of those seizures. Malong also attended training sessions conducted by the City concerning how to respond to seizures and other first aid matters.

Katie participated in the first swimming day at the 2002 Adventure Camp without incident. On the second swimming day she drowned.

Approximately one hour before drowning, while waiting to enter the locker room at the pool, Katie suffered a mild seizure that lasted a few seconds. Malong observed the seizure and sent another counselor to report the incident to a supervisor. According to the pleadings, the supervisor stated that the report never was received. Malong watched Katie for approximately 45 minutes following the mild seizure. Then, receiving no word from her supervisor, Malong concluded that the seizure had run its course and that it was safe for Katie to swim.

Malong sat on the side of the pool near the lifeguard, watching the deep end of the pool. In addition to the Adventure Camp participants, there were as many as 300 other children in the pool area. Malong watched Katie jump off a diving board and swim back to the edge of the pool. At Malong's insistence, Katie got out of the pool and rested for a few minutes. Malong then asked Katie whether she wished to dive again, and Katie said she did. Katie dove into the water, bobbed to the surface, and began to swim toward the edge of the pool. As Katie did so, Malong momentarily turned her attention away from Katie. When Malong looked back no more than 15 seconds later, Katie had disappeared from her sight. After Malong and others looked for Katie somewhere between two and five minutes, an air horn blew and the pool was evacuated. Lifeguards pulled Katie from the bottom of the pool, and she died the next day.

Katie's parents, Terral and Maureen Janeway, filed a wrongful death action alleging the accident was caused by the negligence of the City and Malong. Relying upon the release, defendants moved unsuccessfully for summary judgment and summary adjudication. Defendants then sought relief in the Court of Appeal, filing a petition for writ of mandate. (Code Civ. Proc., § 437c, subd. (m)(1).) As noted earlier, the appellate court denied the petition, holding (1) the agreement was effective and enforceable insofar as it concerned defendants' liability for future ordinary negligence, but (2) concluding a release of liability for future gross negligence generally is unenforceable, and the agreement in this case did not validly release such liability. As observed above, we address only the second holding.

II

A

We begin by defining the terms that underlie the issue presented. "Ordinary negligence" — an unintentional tort — consists of a failure to exercise the degree of care in a given situation that a reasonable person under similar circumstances would employ to protect others from harm. (See, e.g., Donnelly v. Southern Pacific Co. (1941) 18 Cal.2d 863, 869 [ 118 P.2d 465] ( Donnelly).)

"Gross negligence" long has been defined in California and other jurisdictions as either a "`"want of even scant care"'" or "`"an extreme departure from the ordinary standard of conduct."'" ( Eastburn v. Regional Fire Protection Authority (2003) 31 Cal.4th 1175, 1185-1186 [ 7 Cal.Rptr.3d 552, 80 P.3d 656] ( Eastburn), and cases cited; accord, Colich Sons v. Pacific Bell (1988) 198 Cal.App.3d 1225, 1240 [ 244 Cal.Rptr. 714] (Colich); Kearl v. Board of Medical Quality Assurance (1986) 189 Cal.App.3d 1040, 1052-1053 [ 236 Cal.Rptr. 526]; see also, e.g., Prosser Keeton, The Law of Torts (5th ed. 1984) § 34, pp. 211-212 (Prosser and Keeton); 57A Am.Jur.2d (2004) Negligence, § 227, p. 296.)

B

The traditional skepticism concerning agreements designed to release liability for future torts, reflected in Gardner, supra, 180 Cal.App.3d 713, and many other cases, long has been expressed in Civil Code section 1668 (hereafter cited as section 1668), which (unchanged since its adoption in 1872) provides: "All contracts which have for their object, directly or indirectly, to exempt any one from responsibility for his [or her] own fraud, or willful injury to the person or property of another, or violation of law, whether willful or negligent, are against the policy of the law."

C

In Tunkl v. Regents of University of California (1963) 60 Cal.2d 92 [ 32 Cal.Rptr. 33, 383 P.2d 441] ( Tunkl), we applied section 1668 in the context of a release required by a nonprofit research hospital as a condition of providing medical treatment. In that case, the plaintiff had signed a contract releasing the operators of the hospital — the Regents of the University of California — "`from any and all liability'" for "`negligent . . . acts or omissions of its employees'" so long as the hospital used due care in selecting those employees. ( 60 Cal.2d at p. 94.) Thereafter, the plaintiff sued for ordinary negligence based on the treatment received from two of the hospital's doctors.

Turning to section 1668, Justice Tobriner's unanimous opinion for the court noted that past decisions had differed concerning the reach of that statute ( Tunkl, supra, 60 Cal.2d 92, 96-97), but that those decisions agreed in one significant respect: they consistently "held that [an agreement's] exculpatory provision may stand only if it does not involve [and impair] `the public interest.'" ( Id., at p. 96.) Exploring the meaning and characteristics of the concept of "public interest" as illuminated by the prior cases ( id., at pp. 96-98), we read those precedents as recognizing a general rule that an " exculpatory clause which affects the public interest cannot stand." (Id., at p. 98, italics added.)

Tunkl next addressed the "factors or characteristics" that underlie the concept of "public interest" in the context of an agreement releasing liability for future ordinary negligence. ( Tunkl, supra, 60 Cal.2d 92, 98.) In passages widely quoted and followed or adopted as a guide by numerous out-of-state decisions addressing the enforceability of such agreements, we wrote: "The social forces that have led to such characterization are volatile and dynamic. No definition of the concept of public interest can be contained within the four corners of a formula. The concept, always the subject of great debate, has ranged over the whole course of the common law; rather than attempt to prescribe its nature, we can only designate the situations in which it has been applied. We can determine whether the instant contract does or does not manifest the characteristics which have been held to stamp a contract as one affected with a public interest." ( Tunkl, supra, 60 Cal.2d at p. 98.)

We found in the prior cases a "rough outline" of the "type of transaction in which exculpatory provisions will be held invalid," explaining: "[T]he attempted but invalid exemption involves a transaction which exhibits some or all of the following characteristics. It concerns a business of a type generally thought suitable for public regulation. The party seeking exculpation is engaged in performing a service of great importance to the public, which is often a matter of practical necessity for some members of the public. The party holds himself out as willing to perform this service for any member of the public who seeks it, or at least for any member coming within certain established standards. As a result of the essential nature of the service, in the economic setting of the transaction, the party invoking exculpation possesses a decisive advantage of bargaining strength against any member of the public who seeks his services. In exercising a superior bargaining power the party confronts the public with a standardized adhesion contract of exculpation, and makes no provision whereby a purchaser may pay additional reasonable fees and obtain protection against negligence. Finally, as a result of the transaction, the person or property of the purchaser is placed under the control of the seller, subject to the risk of carelessness by the seller or his agents." ( Tunkl, supra, 60 Cal.2d 92, 98-101, fns. omitted.)

We continued our analysis in Tunkl by stressing that considerations of public policy did not bar all contracts releasing future liability for negligence, and by drawing a distinction between such permissible releases and those that implicate at least some of the circumstances described above. ( Tunkl, supra, 60 Cal.2d 92, 101.) We commented that when certain of these characteristics are present, the transaction is such that "the releasing party does not really acquiesce voluntarily in the contractual shifting of the risk," and further that when the "service is one which each member of the public, presently or potentially, may find essential to him," the releasor "faces, despite his economic inability to do so, the prospect of a compulsory assumption of the risk of another's negligence." ( Id., at p. 101.) Applying the public interest characteristics articulated above to the facts of the transaction then before us in Tunkl, we concluded that the release exhibited not only some of those characteristics, but all of them, and that the contract of exculpation for negligence committed by the hospital's employee doctors "affect[ing] the public interest" was invalid. ( Id., at pp. 101-102.)

D

E

As the parties observe, no published California case has upheld, or voided, an agreement purporting to release liability for future gross negligence. Some decisions have stated, in dictum, that such a release is unenforceable. ( Farnham v. Superior Court (1997) 60 Cal.App.4th 69, 74 [ 70 Cal.Rptr.2d 85] ["exemptions from all liability for . . . gross negligence . . . have been consistently invalidated"]; Health Net, supra, 113 Cal.App.4th 224, 234 [liability for future gross negligence cannot be released].) Others carefully have specified that liability for "ordinary" or "simple" negligence generally may be released (that is, so long as doing so is consistent with Tunkl, supra, 60 Cal.2d 92) — thereby implicitly differentiating gross negligence from the class of conduct as to which liability generally may be released. Indeed, for more than three decades, Witkin has asserted that California law categorically bars the prior release of liability for future gross negligence: "The present view is that a contract exempting from liability for ordinary negligence is valid where no public interest is involved. . . . [¶] But there can be no exemption from liability for intentional wrong [or] gross negligence. . . ." (1 Witkin, supra, Contracts, § 660, pp. 737-738, italics added; see also 1 Witkin, Summary of Cal. Law (9th ed. 1987) Contracts, § 631, p. 569 [same]; 1 Witkin, Summary of Cal. Law (8th ed. 1973) Contracts, § 485, pp. 411-412 [essentially identical]; 1 Witkin, Summary of Cal. Law (7th ed. 1960) Contracts, § 200, p. 226 ["The Contracts Restatement declares that a person can contract to exempt himself from liability for ordinary negligence, but not for gross negligence"].) As defendants observe, however, Witkin does not cite any relevant California decision in support of that proposition.

On the other hand, as defendants and their amici curiae also observe, a number of cases have upheld agreements insofar as they release liability for future ordinary negligence in the context of sports and recreation programs, on the basis that such agreements do not concern necessary services, and hence do not transcend the realm of purely private matters and implicate the "public interest" under Tunkl, supra, 60 Cal.2d 92. Our lower courts have upheld releases of liability concerning ordinary negligence related to gymnasiums and fitness clubs, auto and motorcycle racing events, ski resorts and ski equipment, bicycle races, skydiving or flying in "ultra light" aircraft, and various other recreational activities and programs such as horseback riding, white-water rafting, hypnotism, and scuba diving. Most, but not all, other jurisdictions have held similarly. In light of these decisions, some more recent appellate decisions have concluded categorically that private agreements made "in the recreational sports context" releasing liability for future ordinary negligence "do not implicate the public interest and therefore are not void as against public policy." ( Benedek, supra, 104 Cal.App.4th at pp. 1356-1357.)

III

In the absence of an authoritative discussion in any California opinion concerning the enforceability of an agreement releasing liability for future gross negligence, we consider the law of other jurisdictions. We find that the vast majority of decisions state or hold that such agreements generally are void on the ground that public policy precludes enforcement of a release that would shelter aggravated misconduct. (See, e.g., Xu v. Gay (2003) 257 Mich.App. 263 [ 668 N.W.2d 166, 170] (Xu); Zavras v. Capeway Rovers Motorcycle Club (1997) 44 Mass.App.Ct. 17 [ 687 N.E.2d 1263, 1265] (Zavras); Wolf v. Ford (1994) 335 Md. 525 [ 644 A.2d 522, 525]; New Light Co. v. Wells Fargo Alarm Servs. (1994) 247 Neb. 57 [ 525 N.W.2d 25, 29-31] (New Light); Wheelock v. Sport Kites, Inc. (D. Hawaii 1993) 839 F.Supp. 730, 736 ( Wheelock) [applying Hawaii law]; Boyce v. West (1993) 71 Wn.App. 657 [ 862 P.2d 592, 597] (Boyce); Sommer v. Federal Signal Corp. (1992) 79 N.Y.2d 540 [ 593 N.E.2d 1365, 1370-1371, 583 N.Y.S.2d 957]; Buckner v. Varner (Tenn.Ct.App. 1990) 793 S.W.2d 939, 941; Wade v. Watson (N.D.Ga. 1981) 527 F.Supp. 1049, 1051-1052 [applying Ga. law]; Shelby Mut. Ins. v. Grand Rapids (1967) 6 Mich.App. 95 [ 148 N.W.2d 260, 262].)

A

The text writers reflect this majority rule. For example, in Champion, Fundamentals of Sports Law (1990), the author observes: "[I]t is universally held that a release will not bar a claim for gross negligence. That is true even though the same exculpatory clause would bar an [action] for simple negligence." (Id., § 11.2, p. 209, italics added; see also id., § 11.6, p. 215.) Leading treatises are in accord; indeed, some of them state categorically that any attempt to release liability for future gross negligence is void as against public policy. Yet other treatise writers and law review authors have offered similar, albeit slightly moderated characterizations of the law, reflecting the circumstance that there are at least a handful of cases from a few jurisdictions that, without discussing the general rule or authorities set forth above, enforce contracts releasing liability for future gross negligence in the context of agreements signed by motor vehicle racing participants.

B

IV

Defendants and their supporting amici curiae argue that we should not be guided by these out-of-state cases and authorities, for three reasons. They assert that (1) enforcement of agreements releasing liability for future gross negligence is mandated by section 1668, and a contrary rule would violate both that statute and the holding in Tunkl, supra, 60 Cal.2d 92; (2) many out-of-state decisions supporting the proposition that future gross negligence cannot be released are distinguishable and hence inapt; and (3) considerations of public policy, properly understood, mandate not the majority rule — generally voiding releases of liability for future gross negligence — but the opposite, that is, a rule enforcing releases of liability for future gross negligence.

A

1

In this respect we agree with the Court of Appeal below, which observed that section 1668 "has not been . . . interpreted to authorize any and all releases that are not expressly invalidated." As the lower court also noted, " Tunkl itself went beyond the language of Civil Code section 1668 to invalidate releases of liability for negligence under certain circumstances. . . ." To be more explicit: our unanimous decision in Tunkl, supra, 60 Cal.2d 92, did precisely what defendants and their supporting amici curiae assert is precluded by section 1668 — our decision found a release of liability for future ordinary negligence void on public policy grounds other than those set forth in section 1668. Indeed, Tunkl sets forth a categorical rule: Any exculpatory clause (even one releasing liability for future ordinary negligence) is unenforceable if it relates to a transaction that adequately exhibits at least some of the six characteristics set forth in that case, and thereby "affects the public interest." We could not accept the statutory argument advanced by defendants and their supporting amici curiae, without at the same time fundamentally undermining and effectively overruling Tunkl — and we decline any implied invitation to do so. Accordingly, we reject defendants' argument that, by enacting section 1668 more than 130 years ago, our Legislature established a policy generally allowing releases of liability for future gross negligence, and hence a policy precluding this court from adopting, on public policy grounds, the opposite — and clearly majority — rule.

2

We also reject the similar argument, advanced by defendants and their amici curiae, that we may not recognize or employ, as a basis for invalidating a release, any public policy rationale different from that set out in our decision in Tunkl, supra, 60 Cal.2d 92.

As we have observed ante, in part III.B, the out-of-state decisions and other authority holding agreements releasing liability for future gross negligence to be unenforceable are based, not on Tunkl's public interest, "transaction-focused" analysis, but instead upon a separate and different public policy rationale focusing upon the degree or extent of the misconduct at issue, in order to discourage (or at least not facilitate) aggravated wrongs. Defendants and their amici curiae, however, assert that if a particular agreement releasing liability for "negligence" is, as the Court of Appeal found in the present case, enforceable under the Tunkl public interest analysis (an issue that, as observed ante, fn. 1, we do not address), then, also pursuant to Tunkl, such an agreement "can and should be enforced for all negligence" — that is, ordinary and gross negligence. Justice Baxter, in his dissenting opinion in this matter, post, embraces the same view.

We did not address in Tunkl whether an agreement purporting to release liability for future gross negligence could be enforced; we considered only the circumstances in which a release of liability for the type of negligence at issue in that case — future ordinary negligence — might be unenforceable. Our recognition in Tunkl that the concept of "public interest" is dynamic, not static; our refusal to rigidly "prescribe its nature"; and our explication of only a "rough outline" of the type of transaction as to which a release of liability for ordinary negligence would be unenforceable ( Tunkl, supra, 60 Cal.2d at p. 98), all belie the suggestion that we now should read Tunkl as implicitly foreclosing a different public policy analysis in the context of an agreement purporting to release liability for future gross negligence. Certainly, nothing in Tunkl is inconsistent with the public-policy-based majority rule described above. Nor can Tunkl reasonably be read to stand for the proposition that, assuming Tunkl's public interest factors do not preclude enforcement of an agreement releasing liability for future ordinary negligence, this same agreement also should, or even may, be construed and enforced to release liability for future gross negligence.

B

Defendants contend that many out-of-state decisions supporting the proposition that liability for future gross negligence cannot be released arise in jurisdictions that define this form of negligence not as California does (as either (1) a failure to exercise even slight care, or (2) an extreme departure from the ordinary standard of conduct — see ante, pt. II.A), but instead define that term as conduct tantamount to wanton, reckless, or willful misconduct. Even if some decisions arguably are distinguishable on that basis, however, significant other out-of-state authority is not so readily distinguishable.

For example, the State of Washington, which views gross negligence consistent with the California definition, has long held void and unenforceable any attempted release of liability for a negligent act that "falls greatly below the standard established by law for protection of others." ( Vodopest, supra, 913 P.2d 779, 783, italics added.) The same approach appears to apply in Massachusetts, which also long has viewed gross negligence consistent with the California definition. (See Zavras, supra, 687 N.E.2d 1263, 1265-1266 fn. 4 [noting general rule that liability for "ordinary" negligence may be released, but that liability for "gross" negligence — defined as the "`absence of slight diligence, or the want of even scant care'" — may not]; see also Sharon v. City of Newton (2002) 437 Mass. 99 [ 769 N.E.2d 738, 748, fn. 12] ( Sharon) [citing Zavras with approval].) Similarly, Nebraska, which also long has viewed gross negligence consistent with the California definition, has refused to permit the release of liability for such future conduct. ( New Light, supra, 525 N.W.2d 25, 30-31 [defendant barred from insulating itself for damages caused by its own gross negligence, defined as failure to employ even "slight care" in the performance of its duty].) In other words, it appears that these states — and Washington in particular, for many decades — have enforced what is effectively the same rule that defendants and their amici curiae assert should be rejected as unwarranted and unworkable in California.

C

Ultimately, defendants and their amici curiae argue that rejection of the majority rule described above, and adoption of the opposite rule proposed by them, is mandated by public policy, as they perceive it. They stress the asserted uncertainty of the gross negligence standard and argue that unless providers of recreational services and related programs can be assured that agreements purporting to release liability for future gross negligence will be enforced, (1) subsequent suits against recreational service providers — private, public, for-profit, or nonprofit — will not be readily resolvable in favor of defendants on summary judgment, with the result that unwarranted liability will be threatened or imposed, and (2) service providers will react by greatly restricting, or simply declining to afford, such services or programs in California.

1

We do not agree that adoption of the foregoing majority rule in the setting of the definition of "gross negligence" employed in this state (failure to exercise even slight care, or an extreme departure from the ordinary standard of conduct) would prove unworkable, or that application of such a standard would frustrate the proper termination of suits on summary judgment or foster untoward liability. As the parties acknowledge, the same definition long has been employed in cases applying numerous California statutes that confer limited immunity for negligence while expressly exempting immunity for gross negligence. Despite the concerns of defendants and their amici curiae, in light of the experience under these statutes it does not appear that the application of a gross negligence standard, as defined in California, has a tendency to impair the summary judgment process or confuse juries and lead to judgments erroneously imposing liability. To the contrary: "These statutes reflect the sound legislative judgment that, under a gross negligence standard, meritless suits will typically be disposed of by summary judgment; that when a case goes to trial the jury, instructed on this standard, will be less likely to confuse injury with fault; and that verdicts reflecting such confusion will be more readily reversed, whether by the trial or appellate court, than under an ordinary negligence standard." ( Kahn v. East Side Union High School Dist. (2003) 31 Cal.4th 990, 1020 [4 Cal.Rptr.3d 103, 75 P.3d 30] (cone. opn. of Werdegar, J.).) In this respect, we emphasize the importance of maintaining a distinction between ordinary and gross negligence, and of granting summary judgment on the basis of that distinction in appropriate circumstances. (See Decker v. City of Imperial Beach (1989) 209 Cal.App.3d 349, 358 [ 257 Cal.Rptr. 356] ["Generally it is a triable issue of fact whether there has been such a lack of care as to constitute gross negligence" — "but not always"]; see also, e.g., Eastburn, supra, 31 Cal.4th 1175, 1185-1186 [trial court properly precluded amendment of a complaint to allege gross negligence]; DeVito v. State of California (1988) 202 Cal.App.3d 264, 272 [ 248 Cal.Rptr. 330] [summarily concluding that a complaint "alleges no facts showing `an extreme departure from the ordinary standard of care'"].)

2

As defendants observe, some cases and other authorities assert, albeit without citing any empirical evidence, that upholding agreements releasing liability for future negligence is necessary in order to ensure the continued availability of sports recreation and related programs. (E.g., Hohe, supra, 224 Cal.App.3d 1559, 1564; YMCA, supra, 55 Cal.App.4th 22, 27-28.) Defendants and their amici curiae embrace this broad premise and argue by analogy that the same principle applies with respect to agreements releasing liability for future gross negligence.

Defendants assert that unless recreation service providers dependably can enforce agreements to release liability for both future ordinary negligence and future gross negligence, "the inevitable result will be fewer — and more expensive — programs," and that (quoting Allabach, supra, 46 Cal.App.4th 1007, 1016) ultimately, "`"many popular and lawful recreational activities are destined for extinction."'"

The various amici curiae in support of defendants echo and amplify these predictions. For example, amici curiae NASCAR and the California Speedway Association assert that limiting agreements releasing liability for future ordinary negligence, while not permitting the release of liability for future gross negligence, ultimately will "deprive [the public] of the . . . opportunity to participate and recreate in many . . . cherished [pastimes]," including being spectators at NASCAR and similar motor vehicle racing events. Likewise, amici curiae Bally Total Fitness Corporation and 24 Hour Fitness USA, Inc., claim the appellate decision below, enforcing the release as to negligence but not as to gross negligence, "[wreaks] havoc on recreational providers," leading them to a "precipice from which there will be no return." Similarly, the brief of the International Health, Racquet, and Sportsclub Association and California Clubs of Distinction twice declares that "the effect of [enforcing a release as to negligence but not as to gross negligence] cannot be overstated" — and suggests that unless releases of liability for future gross negligence are enforced, there will be "far reaching and devastating consequences," rendering commercial health and racquet clubs "a thing of the past." Amici curiae Sierra Club, League of California Cities, and California State Association of Counties make similar, albeit slightly less strident, assertions.

We are sensitive to the policy arguments advanced by defendants and their amici curiae that caution against rules triggering wholesale elimination of beneficial recreational programs and services — and we are especially sensitive to the concerns relating to the continued availability of programs such as the one here at issue, serving the recreational needs of developmentally disabled children. But we find no support for such broad predictions in the present setting.

a

Although, as noted, some cases and authorities assert that upholding releases of liability for ordinary negligence may help ensure the continuation of sports recreation and related programs (see ante, at fn. 27), we do not discern in those cases any discussion of an asserted corresponding need to recognize and enforce agreements releasing liability for future gross negligence, and indeed we find little supporting that position even in the law review literature upon which defendants rely. We also find it significant that, as observed ante, part IV.B, the States of Washington, Massachusetts, and Nebraska all effectively bar release of liability for gross negligence, as that term is defined in California. We would expect that if, based upon the experience of these sister states, there existed substantial evidence supporting the ominous forecasts of defendants and their amici curiae concerning the future of recreational services in California under the same system, defendants and their amici curiae would highlight that information. And yet, no such information has been provided to us.

b

Indeed, if the premise of defendants and their amici curiae were correct — that is, if failing to enforce agreements releasing liability for future gross negligence would imperil the very existence of sports and recreational industries — we at least would expect to see some analogous evidence in the experience of those states that prohibit even agreements releasing liability for future ordinary negligence. Ordinary negligence, after all, occurs much more commonly than gross negligence, and hence judicial decisions holding unenforceable any release of liability for ordinary negligence would, under the theory of defendants and their amici curiae, pose a much greater threat to the continued availability of recreational sports programs than would a rule holding unenforceable releases of liability for gross negligence generally. And yet, as explained below, in numerous contexts concerning recreational sports and related programs, courts categorically have voided agreements releasing liability for future ordinary negligence without (so far as we can discern) triggering in any substantial degree the dramatically negative effects predicted by defendants and their amici curiae.

Many thousands of contracts that have been entered into, releasing liability for future ordinary negligence in the context of recreational sports and related programs, are unenforceable in most states. This is so because, although courts in California and a few other states have enforced agreements, signed by parents, releasing liability for future ordinary negligence committed against minor children in recreational and related settings, that position apparently represents a minority view. "A clear majority of courts . . . have held that a parent may not release a minor's prospective claim for negligence." ( Hawkins ex rel. Hawkins v. Peart (2001) 2001 UT 94 [ 37 P.3d 1062, 1065-1066] [voiding agreement signed on behalf of minor releasing liability for future negligence concerning horseback riding], and cases and other authorities cited.)

In addition, we observe that Vermont has voided agreements releasing liability for future ordinary negligence in the context of recreational skiing and racing; Connecticut has acted similarly concerning "snow tubing" and horseback riding lessons; West Virginia has voided a release of liability for ordinary negligence executed by a university student who was injured while playing "club" rugby; and Washington has voided agreements releasing public school districts from liability for future ordinary negligence related to interscholastic athletics. Virginia long has categorically and broadly voided all preinjury releases, even in the recreational sports context. Perhaps most significantly, the New York Legislature, for three decades, has barred enforcement of agreements between operators of "gymnasium[s]" and places of "amusement or recreation, or similar establishment[s]," and their paying members or customers, purporting to release liability for future negligence by the operator. (N.Y. Gen. Oblig. Law, § 5-326.) Pursuant to this statute, New York courts have found releases to be void and unenforceable in the context of suits for personal injuries caused by ordinary negligence related to automobile racing at commercial racetracks; skiing and ski lessons at resorts; horseback riding organized and operated by a business firm or riding stable business; recreational parachuting or skydiving lessons; flag football played in a league run by a corporation; tennis played at a country club at which the plaintiff was a member; and riding a "mechanical bull" in a bar.

We brought the cases from these six states (Connecticut, Utah, Vermont, Virginia, Washington, and West Virginia) and the New York statute to the parties' attention and solicited supplemental briefing concerning defendants' policy argument that enforcing releases of liability for future ordinary negligence, but not for future gross negligence, would lead to the demise or substantially diminished availability of recreational services and programs. Thereafter, pursuant to a request by defendants, we allowed additional supplemental briefing. The ensuing briefing, however, disclosed no empirical study suggesting that holdings such as those described above, precluding the release of liability for future ordinary negligence (or for that matter, similar holdings under Tunkl, supra, 60 Cal.2d 92), have triggered the predicted elimination or even widespread substantial reduction of the affected services or programs. Indeed, defendants forthrightly concede in their supplemental briefs that they found no empirical support for such assertions.

Defendants caution, however, against any attempt to assess "`the societal effects of judicial holdings'" (quoting Choper, Consequences of Supreme Court Decisions Upholding Individual Constitutional Rights (1984) 83 Mich. L.Rev. 1, 7), and they suggest that because of legal, economic, social and other differences between the seven jurisdictions discussed above and California, the experiences of those states "probably" are not predictive of what might occur in California if we were to decline to enforce releases of liability for future gross negligence. Nevertheless, and seemingly in conflict with their own admonition about attempting to assess the societal effects of judicial holdings, defendants speculate that the rules employed in the seven jurisdictions described above, declining to enforce releases of liability for future ordinary negligence, " may have led or may lead to the diminished availability or even the demise of recreational services and programs" in those states. Furthermore, defendants suggest that, even without empirical evidence of any negative effects in those states, but in light of some law review articles generally predicting such effects if releases of liability for future ordinary negligence are not enforced (see ante, fn. 27), we should assume such effects have occurred and will occur in those jurisdictions, and that such effects also would occur in California, were we to adopt a rule posing even a comparatively lesser threat to the continued availability of recreational sports and sports programs — that is, a rule generally enforcing releases of liability for future ordinary negligence, but generally declining to enforce releases of liability for future gross negligence.

We find defendants' arguments unpersuasive. Of course legal, economic, social, and other differences can make interjurisdictional comparisons inexact. But that does not mean we should ignore what might be gleaned from the legal laboratory that is the product of our federal system, under which states may, and do, undertake different solutions to common problems. The circumstance that neither defendants nor their supporting amici curiae have found from the experience of our sister states any substantial empirical evidence supporting their dire predictions is, we believe, both relevant and telling.

Indeed, it appears that the experience of our sister states has not borne out the predictions of defendants and their supporting amici curiae. In Virginia and New York, for example — where, as noted above, agreements to release future liability for ordinary negligence causing personal injury long have been categorically barred by case law or generally precluded by statute, as construed by case law — service providers have been subjected to the potential of liability substantially greater than that facing their counterparts in California and most other jurisdictions, which (as observed ante, part II.E) generally uphold such releases. And yet, our research suggests that the predicted demise of recreational opportunities apparently has not come to pass in Virginia or New York.

For example, amicus curiae NASCAR's brief predicts the downfall of spectator auto racing unless agreements releasing liability for future gross negligence regularly are enforced. According to NASCAR's official Web site, however, of the 31 NASCAR-affiliated major speedways located in the United States and Mexico, two are, and long have been, located in Virginia, and one is, and long has been, located in New York. In other words, despite Virginia's and New York's strict "no release of liability for ordinary negligence" rules, which subject NASCAR to greater potential liability than the mere "no release of liability for gross negligence" rule at issue in the present case, NASCAR-sponsored racing appears not to have disappeared in those states.

Likewise, amicus curiae Bally Total Fitness Corporation's prediction of calamity in the health club industry if releases of liability for future gross negligence are not enforced appears difficult to reconcile with the prevalence of that corporation's business in those two states. Bally's official Web site discloses that it operates seven clubs in Virginia, and 36 in New York. Amici curiae International Health, Racquet, and Sportsclub Association and California Clubs of Distinction similarly assert that commercial recreational services are in danger of extinction if releases of liability for future gross negligence are not enforced. According to the 2002 United States Economic Census (Aug. 2005), which reports on, among other things, each state's "fitness and recreational sports center[s]" (including health, fitness, swimming, racquet, and handball clubs, as well as roller skating and ice-skating rinks), in 2002 there were more than 750 such business locations in Virginia, and more than 1,800 in New York. Again, despite the strict Virginia and New York rules, which subject recreational service providers to far greater potential liability than the mere "no release of liability for gross negligence" rule at issue in the present case, it does not appear that commercial and organized recreational clubs have become "a thing of the past" in those states.

Nor are we aware of any empirical evidence to suggest, as defendants postulate, that a holding declining to enforce an agreement purporting to release liability for future gross negligence would jeopardize programs, such as the one here at issue, that provide recreational opportunities for developmentally disabled children — and indeed, initial research casts doubt upon such predictions.

We reject the arguments of defendants and their amici curiae that considerations of public policy mandate the adoption of a rule under which agreements releasing liability for future gross negligence always, or even generally, would be enforced.

V

As then-Justice Traynor observed in Donnelly, supra, 18 Cal.2d 863, the distinction between "ordinary and gross negligence" reflects "a rule of policy" that harsher legal consequences should flow when negligence is aggravated instead of merely ordinary. ( Id., at p. 871; accord, e.g., Colich, supra, 198 Cal.App.3d 1225, 1240.)

For the reasons discussed above — that is, adherence to the "public policy to discourage," or at least not facilitate, "aggravated wrongs" (Prosser Keeton, supra, § 68, p. 484) — and consistent with Donnelly, supra, 18 Cal.2d 863, and the Court of Appeal below, as well as the vast majority of other jurisdictions, we conclude that public policy generally precludes enforcement of an agreement that would remove an obligation to adhere to even a minimal standard of care. Applying that general rule here, we hold that an agreement purporting to release liability for future gross negligence committed against a developmentally disabled child who participates in a recreational camp designed for the needs of such children violates public policy and is unenforceable.

The Legislature, which already has enacted numerous statutes designed to protect from unfair liability various participants in and sponsors of socially useful enterprises (see ante, fn. 25), is of course free to enact additional legislation limiting, as necessary, the liability of specific recreational service providers. If those who provide such programs or other recreational sports services believe the viability of their particular industry rests upon the ability to secure valid releases of liability for future gross negligence — that is, exoneration for the providers' failure to employ even "slight care," or for an "extreme departure" from the ordinary standard of conduct — the proper forum in which to present that policy argument, and to seek that broad protection, is the Legislature.

VI

Defendants and some of their supporting amici curiae assert that by declining to enforce the release at issue in this case against a possible claim for gross negligence, we would be (1) recognizing a legal distinction between ordinary negligence and more aggravated misconduct, and thus (2) in essence recognizing, in these circumstances, the possibility of a "cause of action" for gross negligence. They assert we may not properly do either. As explained, we reject defendants' objections.

A

Defendants claim our courts "may not distinguish ordinary from gross negligence absent express legislative authorization." In support of this proposition, they cite the Legislature's 1874 repeal of statutes recognizing and defining "slight," "ordinary," and "gross" negligence. (See Walther v. Southern Pacific Co. (1911) 159 Cal. 769, 775 [ 116 P. 51].) Amicus curiae NASCAR echoes this view, asserting that, with respect to gross negligence and "other grades" of misconduct such as recklessness and willful misconduct, California "courts have uniformly agreed that none of [those classifications], in the absence of specific statutory creation, are to be treated differently [from] `ordinary' negligence" and that "there is no legal distinction" between the concepts of ordinary negligence, gross negligence, and recklessness "in the absence of a statute."

This assertion inaccurately characterizes the law. For example — and despite the absence of any statutory authorization for the distinction — we long have adhered to the common law rule that a contract may be reformed due to mutual mistake based upon "ordinary negligence," but not when the mistake is based upon "gross negligence." ( Hess v. Ford Motor Co. (2002) 27 Cal.4th 516, 529 [ 117 Cal.Rptr.2d 220, 41 P.3d 46]; see also Van Meter v. Bent Construction Co. (1956) 46 Cal.2d 588, 594-595 [ 297 P.2d 644] [allowing reformation upon a showing of gross negligence].)

Similarly, prior to abandonment of the common law doctrine of contributory negligence in favor of comparative fault in Li v. Yellow Cab Co. (1975) 13 Cal.3d 804 [ 119 Cal.Rptr. 858, 532 P.2d 1226] ( Li) — and despite the absence of any statutory authorization for making the distinction — recklessness by a tortfeasor long was recognized by California courts in order to ameliorate the harsh effects to a plaintiff of the contributory negligence bar. (See, e.g., 4 Witkin, Summary of Cal. Law (8th ed. 1974) Torts, § 668, pp. 2974-2976, and cases cited.) Now, in the post-Li context, the common law doctrine of assumption of risk continues to disprove the thesis that absent statutory authority, the courts are precluded from drawing legal distinctions between ordinary negligence and more aggravated categories of misconduct. The primary-assumption-of-risk doctrine involves injury-causing conduct by a defendant who, because of the setting and the relationship of the parties, owes no legal duty to protect a plaintiff against ordinary negligence. ( Knight v. Jewett (1992) 3 Cal.4th 296 [ 11 Cal.Rptr.2d 2, 834 P.2d 696] ( Knight).) In the context of active sports coparticipants, for example, this means that a defendant generally has no duty to eliminate, or protect a plaintiff against, risks inherent in a sport — that is, against ordinary careless conduct considered to be part of the sport. ( Id., at pp. 315-316.) And yet, Knight holds, such a defendant nevertheless may be liable for conduct "so reckless as to be totally outside the range of the ordinary activity." ( Id., at pp. 320-321, italics added.)

As shown, pursuant to our common law contract-reformation case law and the assumption-of-risk doctrine, and despite the absence of statutory authorization, California case law clearly distinguishes between the concepts of ordinary negligence and other, aggravated forms of misconduct such as gross negligence and recklessness.

B

Defendants and various supporting amici curiae also assert that California does not recognize any cause of action for "gross negligence" unless such an action is directly, or at least implicitly, authorized by one of the numerous statutes that employ gross negligence as the applicable standard. (See, e.g., statutes cited ante, fn. 25.) Defendants and their amici curiae rely upon Continental Ins. Co. v. American Protection Industries (1987) 197 Cal.App.3d 322 [ 242 Cal.Rptr. 784] ( Continental).

We do not view our holding — that an agreement purporting to release liability for future gross negligence committed against a developmentally disabled child who participates in a recreational camp designed for the needs of such children violates public policy and is unenforceable — as recognizing a cause of action for gross negligence. In any event, as explained below, the decision in Continental does not assist defendants.

Continental, supra, 197 Cal.App.3d 322, did not concern a release of future tort liability, but instead a liquidated damages provision of a contract for burglar alarm services. The provision limited damages for "`negligence'" to $250 ( id., at p. 328, fn. 4), and the plaintiff, an insurer, sought unsuccessfully to avoid that clause by amending its complaint to allege not ordinary negligence, but gross negligence. In affirming the trial court's order refusing to recognize the plaintiff's cause of action, the appellate court noted that numerous California decisions had discussed and applied the doctrine of gross negligence in the context of various statutory provisions establishing that specific level of negligence as the operative standard in particular situations ( id., at p. 329, fn. 5). The appellate court in Continental also quoted Prosser and Keeton's comments concerning the "`difficulty of drawing satisfactory lines of demarcation'" relating to degrees of negligence, and the ensuing elimination of the distinction between ordinary and gross negligence "` in most situations.'" ( Id., at p. 330, fn. 7, quoting Prosser Keeton, supra, § 34, p. 211, italics added.) The court then observed that after the decision in Li, supra, 13 Cal.3d 804, which as noted abandoned the all-or-nothing common law doctrine of contributory negligence in favor of comparative fault, "the need for categorization of misconduct into degrees has been radically diminished." (Continental, supra, 197 Cal.App.3d at p. 330, italics added.) From this, the court in Continental jumped to the broad conclusion that " any attempt to categorize gross negligence separately from ordinary negligence is unnecessary" (ibid., italics added) — and it determined that the trial court properly had declined to allow the plaintiff to amend its complaint to allege gross negligence. ( Ibid.) Subsequently, two decisions have, in offhand dicta, cited Continental as standing for the general proposition that "California does not recognize a distinct cause of action for `gross negligence' independent of a statutory basis." ( Saenz, supra, 226 Cal.App.3d 758, 766, fn. 9.)

We need not address here the question whether the court in Continental reached the correct decision in the context of the liquidated damages provision before it. We conclude, however, that the decision in Continental is distinguishable in the context of the release at issue in the present case. It is true that, after Li, in the context of comparative fault analysis, there typically is no need to distinguish gross negligence from ordinary negligence, because we now permit fact finders to compare the respective fault of the parties, regardless of the degree of negligence of each. ( Sorensen v. Allred (1980) 112 Cal.App.3d 717, 725-726 [ 169 Cal.Rptr. 441] [allowing comparison of negligent and "willful and wanton" (reckless) conduct].) It also is generally true that, with the advent of comparative fault, the need to categorize misconduct into degrees has been "diminished." ( Continental, supra, 197 Cal.App.3d 322, 330.) But as acknowledged by the court in Continental, Li's adoption of comparative fault obviated the need for the distinction only in "most" situations — not in all. ( 197 Cal.App.3d at p. 330, fn. 7; accord, Bielski v. Schulze (1962) 16 Wis.2d 1 [ 114 N.W.2d 105, 114] [observing that the adoption of comparative fault, and the abrogation of gross negligence as a general matter, nevertheless may require that the law continue to recognize gross negligence in the context of "anticipatory releases and exculpatory clauses"].)

Again, reference to Knight, supra, 3 Cal.4th 296, is illustrative. As noted above, in the context of primary assumption of risk (that is, liability of active sports coparticipants for injuries arising from the normal conduct of the sport), the absence of a duty to protect against ordinary negligence does not absolve a defendant from liability based upon reckless conduct. Similarly, in the present situation, it cannot be said that a legal distinction between ordinary negligence and gross negligence is "unnecessary" — indeed, a theory of gross negligence, if supported by evidence showing the existence of a triable issue, is the only negligence-based theory that is potentially open to plaintiffs.

VII

Werdegar, J., Chin, J., and Corrigan, J., concurred.