Concept
Releases and Covenants
Effect of judgment, release, and satisfaction on other tortfeasors
Releases and Covenants
A plaintiff injured by several tortfeasors often settles with one and sues on against the rest, so the law must decide whether the settlement ends the case against everyone. Joint and several liability sets up the question. As Cox v. Pearl Investment Co. states the doctrine, joint tortfeasors answer for the plaintiff’s entire injury, and the plaintiff may pursue all, some, or one of them for the full amount of the damages.1 The plaintiff is entitled to only one satisfaction, yet an unsatisfied judgment against one tortfeasor does not bar an action against another.2 A settlement falls between those two principles, and this unit asks whether it extinguishes the claim against the tortfeasors who did not sign. The credit sections that follow ask how far the payment shrinks a claim that survives.
Restatement (Second) of Torts § 885 answers in two subsections. A valid release of one tortfeasor from liability for a harm, given by the injured person, does not discharge others liable for the same harm unless it is agreed that it will discharge them.3 A covenant not to sue one tortfeasor, or not to proceed further against that tortfeasor, does not discharge any other tortfeasor liable for the same harm.4 A release is the injured person’s surrender of a claim, ordinarily in exchange for payment; a covenant not to sue is a promise to forgo pursuing one.
Today’s answer is recent, and it grew out of three older ones. The strict common law rule treated a release of one tortfeasor as a release of all. The covenant not to sue evaded that rule. The modern intent rule replaced both by asking what the parties agreed, and an agency exception keeps the older result alive for parties whose liability is purely vicarious. Each stage asks the same question: should a settlement that the plaintiff makes with one tortfeasor, for her own reasons, cost her the claim against the rest?
The Common Law Release Rule
The early cases held that a release of one tortfeasor released the plaintiff’s claims against all joint tortfeasors. If Fermi gave a release to Joule, he was deemed to have released Edison, Watt, and every other possible defendant, regardless of the amount Joule paid for it.5 The plaintiff’s intent did not matter. The contract rule for joint obligors worked the same way, discharging the others irrespective of the parties’ intention, and it survived into the twentieth century.6 A plaintiff who settled with one tortfeasor therefore met a consequence that nobody at the table had to mention: signing away the claim against one ended the lawsuit against all.
The rule rested on counting claims. The claims against Joule and Edison were treated as a single, indivisible cause of action, so a release that surrendered the claim as to one surrendered it as to all.7 The companion idea was that the tortfeasor who paid had satisfied the whole injury. In Abb v. Northern Pacific R. Co., the plaintiff took $300 and a one-year pass from a street railway company in return for a written release. The trial court read the paper as a mere covenant not to sue, told the jury it was no full bar to the action against the other defendant, and directed the jury to deduct the $300 from whatever it found the whole damage to be.8 The pleadings showed one wrong, the joint wrong of the street railway company and the appellant, and the opinion quotes an older Maine decision to explain why that mattered:
The difficulty in maintaining the suit against the others is that the law considers that the one who has paid for the injury occasioned by him, and has been discharged, committed the whole trespass and occasioned the whole injury, and that he has therefore satisfied the plaintiff for the whole injury which he received.9
Each step follows from the one before: one wrong, one injury, one satisfaction, and a paying tortfeasor deemed to have supplied it.
Had Pringle given Whiting a covenant not to sue instead, her claim against Shattuck would have survived, as the next subsection shows.10 The money and the loss are identical in the two versions, and only the paper differs. That reliance on form is the formalism later courts rejected.
The strict rule made sensible partial settlements self-defeating. A plaintiff may have good reasons to accept less than full value from one tortfeasor: she may doubt she can prove that party liable, the party’s insurance may cap what it can pay, or she may dread trying the case against a sympathetic or aggressive defendant.11 A risk-averse plaintiff would take $13,000 in hand over a 25 percent chance of collecting $27,000.12 Such a deal makes particularly good sense when the plaintiff keeps her right to sue another tortfeasor and can collect the balance there.13 Under the strict rule, the same deal forfeited that right, whatever the amount paid.14
The rule also trapped plaintiffs who thought they had protected themselves. Documents called releases sometimes purported to reserve rights against other tortfeasors, courts disagreed about their effect, and a plaintiff who believed she had reserved her claims could find them barred by an earlier settlement with someone else involved in the events.15 Contract law felt the same pressure. Because the joint-obligor rule was unfair, some courts confined it to formal releases under seal, and covenants not to sue soon circumvented it.16
Courts eventually abandoned the claim-counting. Even if Fermi’s damages are indivisible, he has an independent claim for them against each tortfeasor, so it is no longer logical to infer that he abandons claims against one by settling with another.17 The strict rule thus matters today chiefly as the background to the covenant not to sue and the intent rule. It still decides cases in any jurisdiction that has not abandoned it, so identify the governing rule before analyzing any settlement with one of several tortfeasors.
Covenants Not to Sue
A covenant not to sue is the plaintiff’s promise to refrain from pursuing a claim against one named tortfeasor, usually for payment, and it does not discharge any other tortfeasor liable for the same harm.20 The Restatement’s words “not to proceed further” reach a plaintiff who has already sued and agrees to go no further. Unlike the release rule, the covenant rule carries no “unless agreed” qualifier, because a covenant leaves the others undischarged by its nature.
A covenant surrenders nothing. At common law a release extinguished the plaintiff’s rights against all joint tortfeasors, but a covenant was enforceable only by the person to whom it was given and did not affect her rights against anyone else.21 Contract law draws the same line. A release is an executed transaction, a covenant not to sue is executory, and the covenant is no defense either to the covenantee or to the other joint obligors. A covenantee who is sued anyway is protected by a court order barring the creditor from levying against the covenantee’s property.22 That distinction gave lawyers their way around the strict rule: the common law was soon circumvented by covenants not to sue.23
The plaintiff does not recover twice, because the covenantee’s payment is credited against the surviving claim. In a jurisdiction that holds a covenant not to sue waives no rights against other tortfeasors, Pringle may still sue Shattuck. If she recovers $50,000, the $40,000 Whiting already paid is credited against the judgment, and Shattuck owes only $10,000.24 She is made whole by the two recoveries together, and no more than whole, since a payment made in compensation of the claim diminishes the claim against the tortfeasors at least to the extent of the payment.25 How the credit is measured when the rules differ is taken up in the sections on credit for payments.
The caption on a document does not decide which kind it is. Under the old regime, a paper denominated a release that purported to reserve rights against other tortfeasors left everyone unsure of its effect.26 In the parallel law of joint obligors, a release or discharge that manifests a contrary intention is treated as a covenant not to sue.27 The modern intent rule, taken up next, removes the need for such label games by asking directly what the parties agreed.
The Modern Intent Rule
Section 885(1) moves the question from the label on the instrument to the agreement behind it. A valid release of one tortfeasor does not discharge others liable for the same harm unless it is agreed that it will.30 The default follows from the reasoning that overthrew the strict rule: the plaintiff has an independent claim against each tortfeasor, so settling with one no longer implies abandoning the others.31 The exception preserves freedom of contract. A settling tortfeasor who wants complete peace can bargain for a release that says it discharges the others, and the plaintiff, who is free to agree, bears the consequence of that choice.
Fermi’s release to Joule shows the default at work. If the release states that it does not release Edison, Fermi keeps his claim against Edison.32 The result is the same if the release is silent, because silence is not an agreement to discharge, and § 885(1) demands an agreement. Pringle’s settlement shows how much the form of the paper once mattered. Under the strict rule, a release to Whiting ended her claim against Shattuck while a covenant would have preserved it. Under § 885(1) both instruments yield one answer: Pringle may pursue Shattuck, and a $50,000 judgment is reduced by the $40,000 already paid, so Shattuck owes $10,000.33
Because § 885(1) asks what was agreed, the inquiry can reach beyond the instrument’s words. Under the Restatement, the plaintiff’s intent to reserve rights against other tortfeasors need not be written and can be proved by parol evidence, meaning evidence from outside the document.34 The old demand for a written, concurrent reservation came from the parol evidence rule.35 Joint-obligor law made the same shift, converting a rule that defeated the parties’ intention into a rule of presumptive intention, so that releasing one obligor releases no others unless the terms so provide.36
Statutes write the same idea into law. Under the Uniform Contribution Act, a release or covenant does not discharge any of the other tortfeasors from liability unless its terms so provide.37 California’s Code of Civil Procedure § 877 likewise provides, among other things, that a release does not discharge a third party unless its terms so provide.38 Both statutes point to the instrument’s terms where the Restatement points to the parties’ agreement, and both treat releases and covenants alike. In a statutory state, therefore, the place to look for a discharge is the document itself.
Some states add a stricter protection. In McMillen v. Klingensmith, the Texas Supreme Court held that a settlement document releases from liability only those tortfeasors named or otherwise specifically identified in it, and no others.39 The Oklahoma Supreme Court reached the same rule in Moss v. City of Oklahoma City, holding that a release discharges other potential tortfeasors only if they are named or otherwise specifically identified in it as persons to be discharged, and it extended the rule to satisfaction of an agreed or consent judgment.40 The requirement gives the plaintiff a concrete check before she signs, since the question is whether the instrument points at that tortfeasor. In such a state, the tenant’s release in the hypothetical discharges the landlord only if it names or specifically identifies him, and a general reference to every other person liable may not be enough.
A release that leaves the others undischarged still lowers what they owe. Under the Uniform Contribution Act, it reduces the claim against the others by the amount stipulated in the release or the covenant, or by the consideration paid for it, whichever is greater, and it discharges the settling tortfeasor from all liability for contribution to any other tortfeasor.41 Restatement (Second) of Torts § 885(3) sets a floor of the same kind: a payment by any person in compensation of the claim diminishes the claim against the tortfeasors at least to the extent of the payment.42 Yost v. American Overseas Marine Corp. puts the two halves together: a release or covenant not to sue given to one tortfeasor does not release the other unless it so provides, but payments made by one of the tortfeasors reduce the amount payable by others.43 The credit follows from the single recovery the plaintiff is allowed, since a plaintiff who took part of her loss from one tortfeasor and the whole of it from another would be paid twice for the same dollars. The sections on measuring the credit and on settlement and contribution develop both points.
Vicariously Liable Parties
Employer and employee show where the older release rule survives. The employee is directly liable for personal tortious conduct and the employer is vicariously liable for it, which makes them joint and several tortfeasors, and in most jurisdictions a release of one joint tortfeasor in that setting releases all of them, unlike a covenant not to sue.47 Vicarious liability is liability imposed on one person for another’s tort because of the relationship between them, as with an employer’s respondeat superior liability for an employee’s torts. The exception thus reaches a party whose liability is solely vicarious, and it needs a reason of its own, since § 885(1) would otherwise leave that party in the case unless the release said otherwise.48
The reason lies in what such a party is asked to pay for. An independent concurrent tortfeasor answers for its own fault, so a settlement with someone else leaves the claim against it untouched. A party liable only by relationship has committed no wrong of its own, and its liability is derivative of the actor’s tort. A release surrenders the claim against the actor, and the derivative liability goes with it.49
Agency law shows a related concern in its judgment rules. Damages in a later judgment against a principal can be no greater than those obtained against the agent, so that the principal’s right of indemnity against the agent is not compromised.50 A release rule that left the employer exposed to the full claim after the employee settled would sit uneasily beside that protection.
The exception stops at the edge of derivative liability. Whiting’s leaving the power on and Shattuck’s swing of the ladder were separate acts of negligence, so Pringle’s release of Whiting did not discharge Shattuck, who answers for his own fault. Whiting’s payment still counts against the loss, and Shattuck owes only the balance, but he owes it because his own negligence caused the fire.51 The same line runs through a single defendant whose liability has two sources.
Schubert v. August Schubert Wagon Co. marks the boundary from the other side. An employee driving on his employer’s business injured his wife and escaped liability because of intra-family tort immunity, yet the court held that the employer was not exonerated when the employee had escaped liability “upon grounds not inconsistent with the commission of a wrong unreleased and unrequited.”52 Immunity shields one defendant from one plaintiff and leaves the wrong standing, so the employer’s liability still has a wrong to rest on. A release is the plaintiff’s own act of surrendering the claim, and it removes the very condition the court relied on.
The rule varies, so confirm which rule your jurisdiction applies. Where the agency exception does not apply, the employer is treated like any other tortfeasor who was not discharged by the release. A covenant not to sue the employee, unlike a release, does not release the employer, because it leaves the employee’s underlying wrong standing.53