WHITEHEAD v. CITY OF OAKLAND (2025)

Civil Code section 1668 renders unlawful any contract that seeks, “directly or indirectly, to exempt any one from responsibility for his own fraud, or willful injury to the person or property of another, or violation of law, whether willful or negligent.” In this case, plaintiff Ty Whitehead alleges he suffered a serious head injury during a bicycle training ride for a charity fundraiser because defendant City of Oakland (the City) breached its statutory duty under Government Code section 835 to maintain a safe roadway for public use. The trial court granted summary judgment to the City, and the Court of Appeal affirmed, on the basis of a release and waiver of liability that Whitehead signed on the morning of the training ride. The release and waiver included a provision discharging the ride organizers, as well as any public entities providing facilities for the ride, from any liability for negligence. We conclude that such a release is “against the policy of the law” under Civil Code section 1668 (section 1668) to the extent it purports to relieve the City of liability for negligently violating a statutory duty relating to public safety. We therefore reverse the judgment of the Court of Appeal.

I. Background

Plaintiff Ty Whitehead suffered a traumatic brain injury in March 2017 while participating in a group training ride in preparation for AIDS/LifeCycle, a weeklong fundraiser bike ride from San Francisco to Los Angeles. At the time of the injury, Whitehead was riding downhill on Skyline Boulevard in Oakland with no other riders in the immediate vicinity. According to evidence offered by Whitehead, cyclists, even those not participating in the training ride, were “essentially required to ride in the center of the lane” when traversing the segment where the injury occurred. As his front tire went down sharply into a large, deep pothole near the center of the lane and came to a stop, Whitehead flipped forward over the front of the bike and hit the rear of his head on the pavement. Whitehead later explained that “it's amazing how just up on the hill a short distance you can look down the road, and the holes are very hard to see. The road looks complete just being a little bit up the hill.”

Earlier that day, prior to the training ride, Whitehead and other participants signed a release form. The document was entitled “AIDS/LifeCycle® Training Ride GENERAL INFORMATION AND RELEASE AND WAIVER OF LIABILITY, ASSUMPTION OF THE RISK, AND INDEMNITY AGREEMENT.” It included a section entitled “RISKS; ASSUMPTION OF RISK,” which provided: “I understand that the Event [defined as the 7-day AIDS/LifeCycle bike ride “and/or preseason training rides and activities leading up to the 7-day event”] is potentially a hazardous activity, and that accidents during the Event could lead to serious injury, death and/or property damage, both to me and to others. Risks associated with the Event may include, but are not limited to: [¶] using public streets and facilities where hazards such as broken pavement and road debris may exist; [¶] being struck by, or colliding with, other cyclists, spectators, automobiles, and road debris; [¶] ․ ; [¶] negligence or carelessness of ․ owners/lessors of the course or facility owners (which may include state and local governmental entities); [¶] negligence or carelessness in the implementation or enforcement of any rules, regulations or guidelines related to the Events and/or in the selection, use, or maintenance of any equipment, course, competition, facility or service related to the Events. [¶] I understand that the Event may expose me to risks other than those listed above and that the risks may not be reasonably foreseeable to me, [or the organizers]. [¶] In consideration for being allowed to participate in the Event, I hereby assume all risks associated with the Event, even those risks which are not reasonably foreseeable at this time.”

The “WAIVER AND RELEASE” clause provided: “To the maximum extent permitted by law, I hereby release, waive, forever discharge and covenant not to sue the Releasees (as defined in the next sentence) from all liabilities, claims, costs, expenses, damages, losses and obligations, of any kind or nature (whether in law or equity) (collectively the ‘Released Liabilities’), which may arise or result (either directly or indirectly) from any participation in the Event. ‘Releasees’ means ․ (B) the owners/lessors of the course or facilities used in the Event ․ and (D) the directors, officers, officials, employees and agents of the entities listed in (A)–(C). [¶] For the avoidance of doubt, the Released Liabilities include all bodily injury, death and/or property damage I may suffer which arises or results (either directly or indirectly) from my participation in the Event, including through any negligence of the Releasees.” (Italics added.)

Finally, the “INDEMNIFICATION” clause provided: “I hereby agree to indemnify, defend and hold harmless the Releasees from all liabilities, claims, costs, expenses, damages, losses and obligations, of any kind or nature (whether in law or in equity) (collectively, the ‘Claims’), which may arise or result (either directly or indirectly) from my participation in the Event.”

One year after the accident, Whitehead sued the City under Government Code section 835 et seq., alleging that the public roadway was in a dangerous condition. In December 2021, the trial court granted the City's motion for summary judgment on the ground that the release was valid and enforceable and barred Whitehead's claim against the City for liability arising from an allegedly dangerous condition of public property. That same day, the court denied Whitehead's motion for summary adjudication of the City's release defense. The court reasoned that “ ‘releases that do not involve transactions affecting “the public interest” may stand’ ” and concluded that “Plaintiff has not demonstrated that the subject of the waiver and release he signed affects the public interest.” In assessing whether the release implicated the public interest, the trial court relied exclusively on the multifactor test we announced in Tunkl v. Regents of University of California (1963) 60 Cal.2d 92, 32 Cal.Rptr. 33, 383 P.2d 441 (Tunkl), which evaluated the validity of a release of common law negligence claims.

The Court of Appeal affirmed. (Whitehead v. City of Oakland (2024) 99 Cal.App.5th 775, 318 Cal.Rptr.3d 168 (Whitehead).) Like the trial court, it relied exclusively on Tunkl and Court of Appeal decisions applying the Tunkl framework to other common law negligence claims. We granted review to decide whether the release relieved the City of liability for harm allegedly caused by maintaining a public roadway in a dangerous condition in violation of Government Code section 835.

II. Discussion

Courts have traditionally looked with disfavor at contractual agreements that purport to exculpate a party for future violations of the law. Such agreements can “pose a conflict between contract and tort law. On the one hand is the freedom of individuals to agree to limit their future liability; balanced against that are public policies underlying our tort system: as a general matter, we seek to maintain or reinforce a reasonable standard of care in community life and require wrongdoers — not the community at large — to provide appropriate recompense to injured parties.” (City of Santa Barbara v. Superior Court (2007) 41 Cal.4th 747, 754, 62 Cal.Rptr.3d 527, 161 P.3d 1095 (City of Santa Barbara).)

The Legislature struck the balance between contract and tort in section 1668 (see City of Santa Barbara, supra, 41 Cal.4th at pp. 754–755, 62 Cal.Rptr.3d 527, 161 P.3d 1095), which has since 1872 provided: “All contracts which have for their object, directly or indirectly, to exempt any one from responsibility for his 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.” Admittedly, we have had only limited opportunities to consider the contours of section 1668. (See Tunkl, supra, 60 Cal.2d at p. 95, 32 Cal.Rptr. 33, 383 P.2d 441.)

In Tunkl, we considered whether a hospital could secure a release from an entering patient for liability arising from common law ordinary negligence. (Tunkl, supra, 60 Cal.2d at p. 94, 32 Cal.Rptr. 33, 383 P.2d 441.) The patient alleged the negligence of two hospital physicians caused him injury, and the hospital defended on the ground that the patient had, prior to admission, agreed to release the hospital “ ‘from any and all liability for the negligent or wrongful acts or omissions of its employees.’ ” (Ibid.) In assessing whether the release violated public policy (id. at pp. 96–97, 101, 32 Cal.Rptr. 33, 383 P.2d 441), we examined whether it exhibited “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.” (Id. at pp. 98–101, 32 Cal.Rptr. 33, 383 P.2d 441, fns. omitted.) Although the agreement in Tunkl satisfied “all” of these characteristics and therefore was invalid, we cautioned that an invalid agreement to exculpate a party for ordinary common law negligence “need only fulfill some of the characteristics above outlined.” (Id. at p. 101, 32 Cal.Rptr. 33, 383 P.2d 441.)

In City of Santa Barbara, supra, 41 Cal.4th 747, 62 Cal.Rptr.3d 527, 161 P.3d 1095, which involved the drowning of a child at a summer camp for developmentally disabled children, we declined to apply the Tunkl framework to the parents’ cause of action against the camp organizer for gross negligence; we instead employed “a separate and different public policy rationale.” (City of Santa Barbara, at p. 764, 62 Cal.Rptr.3d 527, 161 P.3d 1095.) The Court of Appeal in that case had already upheld an agreement releasing the City of Santa Barbara and its employees “insofar as it concerned [their] liability for future ordinary negligence” by analyzing the Tunkl factors, and we did not grant review of that determination. (City of Santa Barbara, at p. 750, 62 Cal.Rptr.3d 527, 161 P.3d 1095.) Our review was limited to the question of “the enforceability of an agreement releasing liability for future gross negligence,” an issue in which there was an “absence of an authoritative discussion in any California opinion.” (Id. at p. 760, 62 Cal.Rptr.3d 527, 161 P.3d 1095.) To analyze this latter question, we relied “upon a public policy analysis that is different from the ‘public interest’ factors considered under Tunkl, supra, 60 Cal.2d 92 [32 Cal.Rptr. 33, 383 P.2d 441].” (City of Santa Barbara, at p. 762, 62 Cal.Rptr.3d 527, 161 P.3d 1095.) We also examined “the law of other jurisdictions,” the vast majority of which “state or hold” that agreements to relieve a party of liability for future gross negligence are void as against public policy. (Id. at p. 760, 62 Cal.Rptr.3d 527, 161 P.3d 1095.)

The case before us does not involve a claim for future ordinary or gross negligence under the common law. The complaint instead asserts a negligent violation of the City's statutory duty (see Gov. Code, § 835 et seq.) to maintain its streets in a reasonably safe condition for travel by the public. Consistent with our precedent (and with the law of other jurisdictions), we conclude that an agreement to exculpate a party for future violations of a statutory duty designed to protect public safety is against the policy of the law under Civil Code section 1668 and is not enforceable.

We first addressed the validity of an anticipatory release in the face of a claimed violation of a statutory duty over a century ago in Union Const. Co. v. Western Union Tel. Co. (1912) 163 Cal. 298, 125 P. 242. The case involved a claim that the defendant telegraph company had failed to deliver certain telegrams. The defendant company's failure was not only negligent, we observed, but also violated a specific statute requiring the company to use “ ‘great care and diligence in the transmission and delivery of messages.’ ” (Id. at p. 314, 125 P. 242.) But each of the messages had been “written upon a blank provided by the defendant containing a contract purporting to exempt it from liability for damages in excess of the charges for transmission.” (Id. at p. 309, 125 P. 242.) Under the circumstances, we found “much force” in the proposition that “ ‘[i]t would be against reason and public policy to hold that it is permissible for such a company to stipulate for immunity from liability for a failure to exercise the care and diligence that the statute under which it operates declares it shall exercise.’ ” (Id. at p. 315, 125 P. 242.) We found it unnecessary to rest our decision reversing the nonsuit exclusively on this proposition, however, because we reasoned that, in light of the above proposition, the release should not be construed to include a violation of the company's statutory duty to deliver a correctly transmitted message “unless no other meaning can reasonably be deduced.” (Id. at p. 316, 125 P. 242.)

Subsequent cases have followed and applied the proposition we identified in Western Union. In Hanna v. Lederman (1963) 223 Cal.App.2d 786, 36 Cal.Rptr. 150, the tenants’ property suffered water damage when the fire sprinkler system flooded the building. The tenants alleged that the sprinkler system, in violation of a municipal code ordinance, was not equipped with an audible alarm and that the lack of an audible alarm was a proximate cause of their losses. (Id. at pp. 787–788, 792, 36 Cal.Rptr. 150.) The defendant lessors sought to invoke an exculpatory clause in the commercial lease, in which the tenants agreed to waive all claims against the lessor for damages to their goods and merchandise. (Id. at pp. 788–789, 36 Cal.Rptr. 150.) The Court of Appeal declined to enforce the exculpatory clause. It held that “[s]ince the claim for damages because of negligence embodied in the first cause of action of each tenant was predicated upon the alleged violation of ․ the Municipal Code, the exculpatory provision could not be a defense to that cause of action if the evidence showed such violation to be a proximate cause of the tenant's loss.” (Id. at p. 792, 36 Cal.Rptr. 150.) Similarly, in Halliday v. Greene (1966) 244 Cal.App.2d 482, 53 Cal.Rptr. 267, the Court of Appeal declined to enforce an exculpatory clause in a residential apartment lease when a tenant was injured during an escape from a building fire. The plaintiffs alleged that the building had only one exit stairway, in violation of a general industry safety order. (Id. at p. 488, 53 Cal.Rptr. 267.) The court reasoned that “[i]f the safety order is applicable, the exculpatory clause is ineffective”: “Public policy, as expressed in section 1668, prohibits an agreement to relieve one of the consequences of his violation of the law, and this whether the violation be wilful or negligent.” (Halliday, at p. 488, 53 Cal.Rptr. 267.)

More recently, the Court of Appeal prevented the State Department of Health Care Services from relying on a release included in its agreement with a managed care health plan on the ground that the release was violative of public policy under section 1668. (Health Net of California, Inc. v. Department of Health Services (2003) 113 Cal.App.4th 224, 226–227, 6 Cal.Rptr.3d 235 (Health Net).) The release prohibited “recovery of damages (but not equitable relief) for any violation of statutory or regulatory law not made part of the parties’ contractual obligations.” (Id. at pp. 226–227, 6 Cal.Rptr.3d 235.) The health plan alleged that the Department had unlawfully assigned thousands of default enrollees to a competitor in violation of the Department's own regulations, and sought damages for the loss of its equitable share of new customers. (Id. at p. 230, 6 Cal.Rptr.3d 235.) In invalidating the release, the court declared that “California courts have construed the statute [section 1668] for more than 85 years to at least invalidate contract clauses that relieve a party from responsibility for future statutory and regulatory violations.” (Health Net, at p. 235, 6 Cal.Rptr.3d 235.)

And in Capri v. L.A. Fitness International, LLC (2006) 136 Cal.App.4th 1078, 39 Cal.Rptr.3d 425, the Court of Appeal declined to enforce a release to bar the plaintiff's claim that he slipped and fell on the fitness club's pool deck. The plaintiff alleged the club had allowed algae to grow there, in violation of state and local health and safety laws, and that the statutory and code violations proximately caused him to fall. (Id. at pp. 1082, 1085, 39 Cal.Rptr.3d 425.) Relying on Hanna, Halliday, and Health Net, the court concluded that “the plain language of section 1668 invalidates contract clauses seeking to relieve a party from responsibility for future statutory and regulatory violations.” (Capri, at p. 1087, 39 Cal.Rptr.3d 425; see also Epochal Enterprises, Inc. v. LF Encinitas Properties (2024) 99 Cal.App.5th 44, 62, 317 Cal.Rptr.3d 573 [“ ‘a party [cannot] contract away liability for his fraudulent or intentional acts or for his negligent violations of statutory law’ ”].)

The case law above is consistent with the prevailing rule in other jurisdictions. (See, e.g., Miller v. Crested Butte, LLC (Colo. 2024) 549 P.3d 228, 235–236 [“settled precedent from this court has established that a party cannot discharge its obligation to perform a statutory duty by way of an exculpatory agreement”]; JM Family Enterprises, Inc. v. Winter Park Imports, Inc. (Fla.Dist.Ct.App. 2009) 10 So.3d 1133, 1133 (per curiam) [“a release or exculpatory clause that attempts to prospectively insulate a party from liability for violating a statute or ordinance enacted to protect the public is generally unenforceable as against public policy”]; La Frenz v. Lake County Fair Board (1977) 172 Ind.App. 389, 360 N.E.2d 605, 609 [“where a safety statute enacted for the protection of the public is violated,” “the obligation and the right created by the statute are public ones which are not within the power of any private individual to waive”]; Lee v. Sun Valley Co. (1984) 107 Idaho 976, 695 P.2d 361, 364 [“while the agreement between Sun Valley and plaintiff does absolve Sun Valley from common law liabilities, it does not absolve Sun Valley from liability for possible violation of the public duty imposed by I.C. § 6-1204”]; Henry v. Mansfield Beauty Academy, Inc. (1968) 353 Mass. 507, 233 N.E.2d 22, 24 [“a contract cannot serve to shield the defendant from responsibility for violation of a statutory duty”]; James Vault & Precast Co. v. B&B Hot Oil Serv., Inc. (N.D. 2019) 927 N.W.2d 452, 466 [“a contractual provision purporting to exempt anyone from responsibility for a willful or negligent violation of statutory or regulatory law is against the policy of law and not enforceable”]; Boyd v. Smith (1953) 372 Pa. 306, 94 A.2d 44, 46 [“when the legislation in question is, as here, a police measure obviously intended for the protection of human life ․ public policy does not permit an individual to waive the protection which the statute is designed to afford him”]; Finch v. Inspectech, LLC (2012) 229 W.Va. 147, 727 S.E.2d 823, 832 [“a limitation of liability contractual provision may be invalidated as contrary to public policy if it absolves a party of liability for failure to conform to a statutorily imposed standard of conduct”]; accord, Street v. Darwin Ranch, Inc. (D.Wyo. 1999) 75 F.Supp.2d 1296, 1300 [“the Court finds no fault with Plaintiff's premise that a release agreement is invalid as against public policy if it conflicts with a standard of care imposed by a safety statute”]; see generally Rest.2d Contracts, § 195, com. a [“If, for example, a statute imposes a standard of conduct, a court may decide on the basis of an analysis of the statute, that a term exempting a party from liability for failure to conform to that standard is unenforceable”].)

Whitehead's claim fits squarely in this category. He alleges the City had a statutory duty to maintain its roadways in a safe condition for the public, including for those members of the public on bicycles. (See Gov. Code, § 835 et seq.; Williams v. County of Sonoma (2020) 55 Cal.App.5th 125, 132, 269 Cal.Rptr.3d 273 [“The County owes a duty to maintain safe roads for all foreseeable uses, including ․ bicycling as a means of transportation”]; Campbell v. Palm Springs (1963) 218 Cal.App.2d 12, 22, 32 Cal.Rptr. 164 [“It is a municipality's duty to keep its streets and sidewalks in a reasonably safe condition”].) He further alleges that the pothole created a dangerous condition in the roadway; that the City negligently breached its duty by failing to warn of, prevent, or correct the road's dangerous condition or designate a bicycle lane; and that he suffered economic and noneconomic injuries as a result of the City's violation of its statutory duty.

We have not been asked to decide whether Whitehead has adequately supported his claim that the City had a statutory duty to him, breached it, and thereby caused damages. Those questions are not before us, and we express no view on them. (See City of Santa Barbara, supra, 41 Cal.4th at p. 778, fn. 56, 62 Cal.Rptr.3d 527, 161 P.3d 1095.) The narrow issue before us is whether Whitehead's claim against the City, assuming it is otherwise valid, is barred by the release. (See Health Net, supra, 113 Cal.App.4th at p. 235, 6 Cal.Rptr.3d 235.) We hold that it is not.

The Court of Appeal came to the opposite conclusion. It correctly recognized that the validity of the release depended on section 1668. (Whitehead, supra, 99 Cal.App.5th at p. 781, 318 Cal.Rptr.3d 168.) And it acknowledged, albeit only in the background section of its opinion, that Whitehead's claim asserted the City had breached its statutory duty to maintain the road in a safe condition for the public. (Id. at p. 780, 318 Cal.Rptr.3d 168.) But the Court of Appeal, like the parties, proceeded on the assumption that Tunkl — which, as stated above, involved the validity of a release concerning a claim that the defendant breached a common law duty — governed the validity of a release concerning a claim that the City breached a statutory duty. This was error. Tunkl set forth an analytical model for the particular circumstance where a plaintiff seeks to invalidate an anticipatory release “on public policy grounds other than those set forth in section 1668.” (City of Santa Barbara, supra, 41 Cal.4th at p. 763, 62 Cal.Rptr.3d 527, 161 P.3d 1095.) Here, by contrast, Whitehead's justification for invalidating the release falls neatly within section 1668’s bar on attempts to relieve a party for a “violation of law, whether willful or negligent.” (See Gavin W. v. YMCA of Metropolitan Los Angeles (2003) 106 Cal.App.4th 662, 670, 131 Cal.Rptr.2d 168 [“ ‘This section made it clear a party could not contract away liability for his fraudulent or intentional acts or for his negligent violations of statutory law’ ”].) Notably, the Court of Appeal did not cite a single case that has applied Tunkl to sustain a release against an asserted breach of a statutory duty.

The City disagrees with the reasoning of the cases that failed to apply Tunkl to releases involving statutory violations. In its view, the phrase “violation of law” in section 1668 is sufficiently “expansive” to “encompass[ ] common-law claims as well as statutory ones.” From this premise, the City argues that the Tunkl test should apply equally to releases of both types of violations. The City's premise, however, is incorrect. If “violation of law” encompassed the common law as well as positive law, there would have been no need for the Legislature to separately prohibit agreements purporting to exempt a party from responsibility “for his own fraud” or for “willful injury to the person or property of another.” (§ 1668; see Health Net, supra, 113 Cal.App.4th at p. 233, 6 Cal.Rptr.3d 235.) Excluding the common law from the “law” referenced in section 1668 is the only construction that gives effect to all the words in the statute. 1 Moreover, this construction is consistent with our observation in City of Santa Barbara, supra, 41 Cal.4th at page 763, 62 Cal.Rptr.3d 527, 161 P.3d 1095, that Tunkl, which involved a claim of common law negligence, set forth grounds for invalidating an anticipatory release “on public policy grounds other than those set forth in section 1668.” There would have been no need for us to reach beyond the statute to assess a release involving ordinary negligence if ordinary negligence qualified as a “violation of law” within the terms of section 1668.

The City, like the Court of Appeal, relies heavily on Okura v. United States Cycling Federation (1986) 186 Cal.App.3d 1462, 231 Cal.Rptr. 429 (Okura). The City claims Okura “is directly on point”; the Court of Appeal deemed it “materially indistinguishable” from this case. (Whitehead, supra, 99 Cal.App.5th at p. 784, 318 Cal.Rptr.3d 168.) Although we find there are some similarities between this case and Okura, there are also some key differences.

The plaintiff in Okura suffered injuries when his bicycle hit loose debris as he was crossing railroad tracks during a bicycle race in Hermosa Beach organized by an affiliate of the United States Cycling Federation. (Okura, supra, 186 Cal.App.3d at pp. 1464–1465, 231 Cal.Rptr. 429.) Prior to the race, the plaintiff had executed a release that provided in relevant part: “ ‘In consideration of the acceptance of my application for entry in the above event, I hereby waive, release and discharge any and all claims for damages for death, personal injury or property damage which I may have, or which may hereafter accrue to me, as a result of my participation in said event. This release is intended to discharge in advance the promoters, sponsors, the U.S.C.F., the S.C.C.F., the promoting clubs, the officials, and any involved municipalities or other public entities (and their respective agents and employees), from and against any and all liability arising out of or connected in any way with my participation in said event, even though that liability may arise out of negligence or carelessness on the part of the persons or entities mentioned above.’ ” (Id. at p. 1465, 231 Cal.Rptr. 429.) The trial court granted summary judgment to the United States Cycling Federation, its local affiliate, and the City of Hermosa Beach. The Court of Appeal, relying on the Tunkl factors, affirmed. (Okura, at pp. 1466–1469, 231 Cal.Rptr. 429.) We do not find Okura illuminating on the claim presented here, however.

Okura articulated grounds for enforcing the release against the race organizers but did not purport to offer any separate analysis of the release with respect to the City of Hermosa Beach. Indeed, the court's discussion of the first Tunkl factor (i.e., whether the transaction “concerns a business of a type generally thought suitable for public regulation” (Tunkl, supra, 60 Cal.2d at p. 98, 32 Cal.Rptr. 33, 383 P.2d 441)) remarked simply that “[n]either the South Bay Wheelmen nor the United States Cycling Federation are subject to public regulation”; it made no mention at all of the municipality's duty to maintain safe public roadways. (Okura, supra, 186 Cal.App.3d at p. 1466, 231 Cal.Rptr. 429.) Furthermore, Okura involved a competitive race on a closed course (id. at p. 1464, 231 Cal.Rptr. 429; cf. Gov. Code, § 831.7, subd. (b)(3) [classifying “bicycle racing,” but not “riding a bicycle on paved pathways [or] roadways,” as a “ ‘Hazardous recreational activity’ ”]); here, by contrast, Whitehead has presented evidence that he was riding his bike on an open road, using the lane — as a member of the general public might — for its intended purpose. Finally, the only cause of action mentioned in the Okura opinion was one for common law negligence. (See Okura, supra, 186 Cal.App.3d at p. 1464, 231 Cal.Rptr. 429.) The court never considered how to analyze a release in the context of a statutory violation. (Cf. Capri v. L.A. Fitness International, LLC, supra, 136 Cal.App.4th at pp. 1084–1085, 39 Cal.Rptr.3d 425 [applying the Tunkl factors and upholding the release as to a common law negligence cause of action, but deeming the release unenforceable as to a cause of action based on negligent violations of the Health & Saf. Code].) We express no view on whether Okura was correctly decided or whether it might have been decided differently if the plaintiff had argued that Hermosa Beach violated a statutory duty. It is sufficient for our purposes to conclude that Okura’s reasoning does not apply here.

The City argues next that Government Code section 835 does not represent the kind of “law” contemplated by the bar on anticipatory releases because section 835 “does not prescribe any rules or standards public entities must comply with for maintaining their roads or streets” and “[i]n fact, does not require them to do anything at all.” Section 835, in the City's view, “simply describes the limited circumstances under which a public entity may lose its immunity to liability.”

The City cites no authority for its characterization of Government Code section 835. Its characterization, moreover, is contrary to our precedent. As we recognized in Zelig v. County of Los Angeles (2002) 27 Cal.4th 1112, 1133, 119 Cal.Rptr.2d 709, 45 P.3d 1171, under section 835 public entities owe members of the public a “duty not to maintain public premises in a dangerous condition.” (See also Williams v. County of Sonoma, supra, 55 Cal.App.5th at p. 132, 269 Cal.Rptr.3d 273; Campbell v. Palm Springs, supra, 218 Cal.App.2d at p. 22, 32 Cal.Rptr. 164.) Our case law thus demonstrates that section 835 does require municipalities to do something about public roadways in appropriate circumstances. Whether such circumstances are present here, of course, is beyond the scope of our opinion. (See ante, ––– Cal.Rptr.3d at p. ––––, ––– P.3d at p. ––––, .)

The City's interpretation of Government Code section 835 does not make logical sense, either. It is true, as the City asserts, that section 835 represents a limited waiver of immunity. But a mere waiver of immunity would not itself create a cause of action, especially since (as the City repeatedly observes) no common law claim can lie against a public entity. (See Gov. Code, § 815, subd. (a).) Accordingly, section 835 represents both a limited waiver of sovereign immunity and an articulation of a cognizable cause of action. (See Pfleger v. Superior Court (1985) 172 Cal.App.3d 421, 432, 218 Cal.Rptr. 371 [§ 835 “defines a form of governmental liability, not an immunity”].) The City's assumption that section 835 performs only the former function finds no support in the statute's text or in our precedent.

The City contends next that it should be entitled to enforce the release because of Government Code section 815, subdivision (b), which in pertinent part provides that “[t]he liability of a public entity established by this part ․ is subject to any defenses that would be available to the public entity if it were a private person.” But Civil Code section 1668 bars contracts purporting to exempt “anyone” — public or private — from responsibility for a “violation of law.” Accordingly, AIDS/LifeCycle would be equally unable, by virtue of Civil Code section 1668, to enforce the release against any claim based on a statutory violation. Government Code section 815 therefore does not aid the City in this case.

The City advises us that invalidating the release in this case will have “especially dire consequences for public entities, inevitably increasing their liability.” It offers hypotheticals in which, say, a group of senior citizens rents space from the public library to play bridge, during which an attendee trips over loose computer cables on the floor; or a member of a private karate club that rents space from an elementary school is injured at the school. If the releases secured by these organizations cannot be enforced by the public entities that provided the space where the injury occurred, the City ominously predicts, “public entities will have every incentive to restrict the use of their premises to avoid potentially exorbitant liabilities.” Alternatively, the City warns, public entities will “likely have to consider requiring permits for large training rides like the one in which Whitehead was injured,” as well as “demanding insurance and indemnification from the sponsors of recreational activities,” thereby making “recreational events harder to come by and more expensive.”

We note here, as we did in City of Santa Barbara, supra, 41 Cal.4th 747, 62 Cal.Rptr.3d 527, 161 P.3d 1095, that no empirical evidence has been offered to support these assertions. (Id. at pp. 768, 773–774, 62 Cal.Rptr.3d 527, 161 P.3d 1095.) Indeed, in light of the fact that many other jurisdictions similarly decline to enforce anticipatory releases of liability for injuries arising from statutory violations related to public safety — and even violations that are not so cabined — the absence of such evidence is “both relevant and telling.” (Id. at p. 774, 62 Cal.Rptr.3d 527, 161 P.3d 1095.) In addition, the City's warning that the increased risk of liability will cause public entities to prohibit, or make prohibitively expensive, the recreational use of their facilities seems exaggerated. The City already owes a duty to the public to maintain its public roadways in a safe condition. Any cyclist traveling this part of Skyline who suffered an injury but was not part of the training ride would presumably have been entitled to file an action against the City for violating its statutory duty under Government Code section 835. The City does not explain how its burdens would intolerably increase if Whitehead, traveling the same road in the same manner, had the same opportunity.

We are not presented with — and this case thus does not require us to decide — whether section 1668 necessarily invalidates anticipatory releases of a claim arising from every statutory violation. 2 Statutes, regulations, and ordinances have proliferated since the enactment of section 1668 in 1872. Whatever “violation of law” is encompassed by section 1668, though, the statute surely prohibits any effort to release a party from its specific statutory duty to ensure that public roadways are safe for foreseeable uses. (See Capri v. L.A. Fitness International, LLC, supra, 136 Cal.App.4th at p. 1087, 39 Cal.Rptr.3d 425 [“this portion of the agreement still seeks to exculpate respondent for its violation of the swimming pool safety statutes, and hence is invalid under section 1668”]; accord, Street v. Darwin Ranch, Inc., supra, 75 F.Supp.2d at p. 1300 [release is invalid “if it conflicts with a standard of care imposed by a safety statute”]; JM Family Enterprises, Inc. v. Winter Park Imports, Inc., supra, 10 So.3d at p. 1133 [release is invalid if it “attempts to prospectively insulate a party from liability for violating a statute or ordinance enacted to protect the public”]; La Frenz v. Lake County Fair Board, supra, 360 N.E.2d at p. 609 [obligations and rights created by “a safety statute enacted for the protection of the public ․ are public ones which are not within the power of any private individual to waive”]; Boyd v. Smith, supra, 94 A.2d at p. 46 [an individual may not waive “a police measure obviously intended for the protection of human life”].) To hold otherwise would substantially undermine the Legislature's ability to protect the public.

III. Disposition

Case law in this state and in other states shows that agreements to exculpate a party for future violations of statutes designed to protect public safety are unenforceable. In this case, the City sought to enforce a release to preclude an action that allegedly arose from a violation of its statutory duty to maintain safe roadways for the public. We have determined that such a release violates section 1668. Because the Court of Appeal affirmed the grant of summary judgment based on the release — a decision we have now found to be error — it did not consider “whether the doctrine of primary assumption of risk forecloses plaintiff's claim.” (Whitehead, supra, 99 Cal.App.5th at p. 790, 318 Cal.Rptr.3d 168.) That argument remains available to the City on remand. We reverse the judgment of the Court of Appeal and remand the cause to that court for further proceedings consistent with this opinion.

I agree with the majority that the City of Oakland may not enforce an agreement purporting to release it from liability for failing to make public roads safe for ordinary, foreseeable uses. I write separately to add a point in response to the City's arguments about the governing legal framework, as well as to offer a few observations about the scope of today's decision.

I'll start with the framework. As this case came to us, the parties’ debate revolved exclusively around the proper application of our seminal decision in Tunkl v. Regents of University of California (1963) 60 Cal.2d 92, 32 Cal.Rptr. 33, 383 P.2d 441 (Tunkl), concerning the enforceability under Civil Code section 1668 (section 1668) of contractual agreements to anticipatorily release a party from legal liability. We asked the parties to brief the question whether Tunkl governs in a case involving statutory violations. The City responded that Tunkl must govern, because Tunkl neither draws nor admits of any distinction between statutory and common law violations.

Well, yes and no. The Tunkl opinion proceeds essentially in two parts. At the outset, seeking to bring clarity and uniformity to a century's worth of disparate interpretations of section 1668, Tunkl identified and endorsed the common doctrinal thread between them: Voluntarily adopted exculpatory clauses in private contracts, though not categorically prohibited, are not enforceable if they “affect[ ] the public interest.” (Tunkl, supra, 60 Cal.2d at p. 98, 32 Cal.Rptr. 33, 383 P.2d 441.) Then, to this core general proposition, Tunkl added more specific guidance to assist in “placing particular contracts within or without the category of those affected with a public interest.” (Ibid.) Without venturing any comprehensive definition of the public interest, Tunkl identified six factors frequently cited as relevant to determining whether a contractual release of liability will be held invalid as affecting the public interest, including whether the parties’ transaction “concerns a business of a type generally thought suitable for public regulation”; whether “[t]he 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”; and whether, “[a]s a result of the essential nature of the service ․ the party invoking exculpation possesses a decisive advantage of bargaining strength against any member of the public who seeks his services.” (Id. at pp. 98–100, 32 Cal.Rptr. 33, 383 P.2d 441, fn.omitted.)

As to the first, broad proposition about the general enforceability of releases of liability, I think the City is correct that Tunkl drew no express distinction between releases of statutory claims and common law claims, appearing to regard both types of claims as implicating section 1668’s dictates about releases of liability for “ ‘[n]egligent ․ violation[s] of law.’ ” (Tunkl, supra, 60 Cal.2d at p. 95, 32 Cal.Rptr. 33, 383 P.2d 441.) We accepted the proposition that a “strict[ ]” reading of the language of section 1668 would encompass “any contract for exemption from liability for negligence” — presumably because a failure to comply with the legal duty of ordinary care is, strictly speaking, a “ ‘[n]egligent’ ” “ ‘violation of law.’ ” (Tunkl, at p. 95, 32 Cal.Rptr. 33, 383 P.2d 441.) We also cited, with seeming approval, a leading treatise's criticism of the view that a “ ‘ “violation of law” ’ ” could be read as “ ‘limited strictly to violation of statutes.’ ” (Id. at pp. 95–96, fn. 3, 32 Cal.Rptr. 33, 383 P.2d 441.) Indeed, we went out of our way to quote that criticism at some length: “ ‘Apart from the debatable interpretation of “violation of law” as limited strictly to violation of statutes, the explanation appears to make an unsatisfactory distinction between (1) valid exemptions from liability for injury or death resulting from types of ordinary or gross negligence not expressed in statutes, and (2) invalid exemptions where the negligence consists of violation of one of the many hundreds of statutory provisions setting forth standards of care.’ ” (Ibid.)

But the City's main argument for enforcing the liability release here depends not just on Tunkl’s general view of the enforceability of exculpatory clauses, but on application of the more specific guidance Tunkl supplied for determining whether a contract is one that affects the public interest. Relying on the six Tunkl public interest factors, the City argues that the release is enforceable because the subject of the contract — participation in a recreational bicycle ride — does not concern the provision of a necessary service. That is to say, there was no practical necessity for Whitehead to participate in a recreational long-distance bicycle ride; he could easily have avoided the risk by staying home.

This is where the City goes wrong. When Tunkl set out six factors to guide the inquiry into whether a particular contract is one that affects the public interest, Tunkl was at pains to note that it was not setting out a one-size-fits-all test for determining when a release of liability affects the public interest or goes “against the policy of the law.” (§ 1668; see Tunkl, supra, 60 Cal.2d at pp. 97–98, 32 Cal.Rptr. 33, 383 P.2d 441.) The six public interest factors set out in the opinion are, unsurprisingly, geared to the type of situation we confronted in that case: That is, they are designed to help identify the public policy interests in a private contractual release of liability for ordinary common law negligence where the transaction at issue is one in which the purveyor of an essential good or service uses its relative bargaining power to shift the risk of its negligence to the less powerful bargainer. (See Tunkl, at p. 101, 32 Cal.Rptr. 33, 383 P.2d 441.) Tunkl’s nonexhaustive list of six factors concerning the public's interest in particular private transactions does not take into account — nor was it meant to take into account — how the enactment of legislation imposing specific obligations or duties might properly guide a court's evaluation of whether it is “contrary to public policy” to allow contracts that prospectively limit liability for future violations of those duties. (Id. at p. 97, 32 Cal.Rptr. 33, 383 P.2d 441.)

This brings me to the scope of today's decision. The majority opinion today does not hold that section 1668 “necessarily invalidates anticipatory releases of a claim arising from every statutory violation”; it instead holds only that, whatever the scope of section 1668, “the statute surely prohibits any effort to release a party from its specific statutory duty to ensure that public roadways are safe for foreseeable uses.” (Maj. opn., ante, ––– Cal.Rptr.3d at pp. –––– – ––––, ––– P.3d at pp. –––– – ––––.) The reservation is indisputably appropriate; section 1668 clearly does not invalidate every anticipatory release of liability where liability is based in some part on statutory law. At the same time that it enacted section 1668 in 1872, the Legislature enacted other liability-related statutes that overlapped to some degree with common law causes of action, including, but not limited to, Civil Code section 1714. (The fraud statute, Civil Code section 1709, is another example.) As the majority notes, the Tunkl opinion would have been written very differently if the enactment of section 1714 — the statute that codifies the duty of ordinary care on which every claim of common law negligence is founded — were alone sufficient to invalidate a contractual release of liability. (See maj. opn., ante, at p. ––––, fn. 2, ––– P.3d at p. ––––, fn. 2; see also Tunkl, supra, 60 Cal.2d at p. 96, fn. 3, 32 Cal.Rptr. 33, 383 P.2d 441 [approvingly quoting commentator's skepticism that § 1668 was meant to invalidate all contractual releases for liability for negligence “ ‘where the negligence consists of violation of one of the many hundreds of statutory provisions setting forth standards of care’ ”].) And at the same time that the Legislature enacted section 1668, the Legislature also enacted Civil Code section 3513, which provides: “Any one may waive the advantage of a law intended solely for their benefit. But a law established for a public reason cannot be contravened by a private agreement.” (See, e.g., Bickel v. City of Piedmont (1997) 16 Cal.4th 1040, 1050, 68 Cal.Rptr.2d 758, 946 P.2d 427 [waiver of statutory rights regarding permit applications did not violate Civ. Code, § 3513 when the provisions were for the benefit of individual applicants and did not “seriously compromise any public purpose”].) If some statutory rights not affecting the public interest can be waived, it is not clear why section 1668 would categorically bar parties from reallocating the risks of negligent violation of such statutory rights by contract.

As to the specific statutory claim here, the majority opinion grounds its holding in Government Code section 835, a statute that sets forth the conditions for suing a public entity for dangerous conditions of public property. But it does not hold that every contractual release of claims that might be brought under section 835, concerning conditions of public property, is necessarily barred by section 1668; it deals only with the safety of the public roads — i.e., publicly owned property that is dedicated to use by the public. The public policy regarding the obligation to maintain the safety of such property is clear and, for present purposes, controlling. Moreover, today's opinion does not hold that section 1668 bars release of liability on every possible claim related to the condition of the roads. It is by now well established that the statutory duty reflected in section 835 concerns public entities’ duty to address conditions related to the safety of the roadway for typical, foreseeable uses, rather than the unique risks that might be related to other types of roadway use. (Legis. Com. coms., 32 pt. 2 West's Ann. Gov. Code (2012 ed.) foll. § 835, p. 99.) The conditions that make a road safe for ordinary use may not be the same as the conditions that make the road safe for landing a plane (ibid.) or, say, holding the cycling leg of the Olympic triathlon.

But this case does not involve a plane landing or, for that matter, a mass racing event. And as the majority says, the question before us is not whether Whitehead has presented an otherwise valid claim under Government Code section 835, based on the condition of the roads for ordinary, foreseeable uses by the public; it is only whether the law countenances an agreement to release a public entity from liability for failing to keep the roads safe for such uses. A city's duty to keep the public roads safe for ordinary public uses is a matter of undeniable public concern, and the policy of the law does not permit enforcement of a private contract purporting to sign that duty away.

FOOTNOTES

1.   At oral argument, the City stated that section 1668 “was not drafted as well as we might all like.” But we perceive no drafting anomaly. The canon against surplusage is a venerable one that predates section 1668’s enactment (see, e.g., People v Waterman (1866) 31 Cal. 412, 415) and enables a construction that gives effect to each part of the statute.

2.   The City warns that if this release is allowed to stand, every ordinary negligence claim could be recast as a violation of a statute: namely, Civil Code section 1714, subdivision (a), which in pertinent part provides, “Everyone is responsible, not only for the result of his or her willful acts, but also for an injury occasioned to another by his or her want of ordinary care or skill in the management of his or her property or person ․” We disagree. The release here purported to release the City from a specific statutory duty, not the generalized duty of care under section 1714. We note further that section 1714, like section 1668, was enacted as part of the original Civil Code in 1872. Yet notwithstanding section 1714, courts have recognized that parties have the freedom to contract a release of ordinary negligence in certain circumstances (see, e.g., Lewis Operating Corp. v. Superior Court (2011) 200 Cal.App.4th 940, 946, 132 Cal.Rptr.3d 849), and indeed the entire Tunkl framework would have been unnecessary if section 1714 had the sweeping effect the City fears.

Opinion of the Court by Evans, J.

We Concur: GUERRERO, C. J. CORRIGAN, J. LIU, J. KRUGER, J. GROBAN, J. JENKINS, J.