Defining the Same Claim

Defining the Same Claim

Every other element of claim preclusion can be undisputed and the case can still turn on one question: does case two assert a new claim, or the old claim in a new form? Once the first judgment is valid, final, and on the merits, and the parties line up the same way, the fight is over how much of the dispute that judgment extinguished. A broad definition of a claim sweeps in every ground that grows out of the same events, so the first judgment leaves little behind. A narrow definition treats more of the second suit as new, so the judgment reaches less and the second suit more often goes forward.

The choice among definitions is a choice about repose. Roman law called the idea res judicata, “a matter already judged”: once a plaintiff has had a fair chance to litigate against a defendant, the defendant is entitled to a sense of repose, and a final judgment spares defendants endless suits over one dispute, saves resources, and prevents inconsistent results.¹ A broad definition serves those aims, but it costs the claimant who failed to foresee a theory or a harm, since that ground is lost with the rest. A narrow definition protects a claimant who wants to sue in stages, but it exposes the defendant to serial suits over one episode. Each test below draws that balance at a different point.

Three definitions are in use. A growing number of state and federal courts define a claim by the transaction from which it arises, following § 24 of the Restatement (Second) of Judgments. Some states keep the narrower same-evidence test, and a much smaller number use the primary-rights test, which treats each violated right as a separate claim.² When case one ended in a state-court judgment, the court that hears case two does not choose among them. Under 28 U.S.C. § 1738, a state court’s judicial proceedings receive in every court in the United States the same full faith and credit they have by law or usage in the courts of the state from which they are taken, so the rendering state’s definition of a claim measures how much the judgment decided.³ The majority in Frier called that the generally accepted rule: to decide a judgment’s preclusive effect, look to the jurisdiction that rendered it.⁴

Two cautions keep this step in its place. Defining the claim presupposes a judgment that qualifies, and a pending appeal does not by itself defeat preclusion: the court hearing case two may stay the action until the appeal ends, or it may apply res judicata and dismiss on the strength of the trial court’s judgment.⁵ And the definition of a claim is not the only point on which jurisdictions differ. Rule 41(b)’s statement that most dismissals operate as adjudications on the merits does not itself settle whether a judgment is “on the merits” for claim preclusion, and Semtek shows that jurisdictions may treat a dismissal for the running of the statute of limitations differently.⁶

The subsections that follow take the transactional test first, then the narrower definitions, then the rule against splitting one transaction by theory, damages, or remedy, and last the separate transactions and later-accruing claims that stay outside the first claim.

The Transactional Test

Federal courts increasingly define a claim by § 24 of the Restatement (Second) of Judgments, which says that a valid final judgment extinguishes all causes of action arising from the same “transaction, or series of connected transactions.”⁷ A person who suffers several injuries in one accident has causes of action that all arise from one transaction. A person who has one accident with the same defendant in January and, coincidentally, another in March has causes of action from two transactions.⁸

The definition borrows the joinder rules’ unit. A party who asserts a right to relief arising out of a transaction or occurrence must join every claim she has from it, or res judicata bars the omitted claims.⁹ Preclusion therefore turns on the right to join a claim in case one, not on whether the plaintiff asserted it. A claim need not have been litigated to be barred; it need only have been available.¹⁰ Schulansky’s suit against Ronan over the addition to her house shows how far that reaches. After she recovered $7,000 from Ronan on a contract claim, she could not sue him again on a breach of contract theory, and she could not sue him for fraud connected with the addition work, although she had never pleaded fraud, tried the case on it, or thought of it.¹¹ The lesson is that her lawyer had better think of it sooner rather than later.

The transactional test displaced older definitions because those definitions sorted claims by something other than the events. The same-evidence test ties a claim to the theory the plaintiff asserts, so two grounds from one transaction can be separate causes of action when the proof for each differs, while the transactional approach is more pragmatic.¹² The evidence comparison also resists application, since some of the evidence will be identical in almost any pair of related suits and the court must decide how much overlap is enough.¹³ Under the Restatement, by contrast, it is dispositive that two causes of action arising out of one transaction could be joined as one convenient trial unit.¹⁴ The Supreme Court has endorsed the transactional definition for federal preclusion in Lucky Brand Dungarees, Inc. v. Marcel Fashions Group, Inc., and since 1982 a growing number of state and federal courts have adopted it.¹⁵

River Park, Inc. v. City of Highland Park applies the test to a change of theory. The plaintiffs alleged that the city had been slow to process their permit and had then bought the property for itself. After their federal due process suit under 42 U.S.C. § 1983 failed, they sued in state court on three state-law theories, among them breach of implied contract and abuse of governmental power.¹⁶ They conceded that both actions arose out of the processing of their applications and sought damages for the same wrongful acts, but they argued that the suits differed in their theories of relief.¹⁷ The Illinois Supreme Court called that argument contrary to transactional analysis, under which suits on different theories may be one cause of action, and it held that the state claims and the section 1983 claim were the same cause of action because they arose from the same core of operative facts.¹⁸ Each theory would have called for somewhat different proof, and that difference no longer controlled.¹⁹

Frier reaches the same result through Illinois’s version of the test. A car owner lost state replevin actions over cars the city had towed, then sued the city in federal court under section 1983 for denying him a prompt hearing.²⁰ Section 1738 required the federal court to give the Illinois judgments the effect Illinois would give them.²¹ The replevin theory required a showing that the cars were taken without lawful process, and the due process theory rested on the same absence of a hearing, so the court found a common core of operative facts.²² Nor could the owner say the due process claim was unavailable in the first round, because he could have joined a count for declaratory relief and punitive damages with his replevin count.²³

Same Evidence and Primary Rights

Not every jurisdiction defines a claim by transaction, and the narrower definitions matter even to a court that would use the transactional test for its own judgments. A state judgment carries its home state’s definition with it, and the court hearing case two applies the preclusion rules of the rendering court.²⁵ Suppose East Dakota defines a claim narrowly, and a West Dakota court whose own rules would bar the second action hears it. Because East Dakota would not bar it, West Dakota allows it.²⁶ The rule has a practical payoff: the parties know the reach of a judgment the day it is rendered, and no one gains by filing case two in a state with looser rules of res judicata.²⁷

Under the same-evidence test, two suits involve the same claim only if the evidence needed to sustain the second suit would have sustained the first, or if the same facts were essential to maintain both actions.²⁸ The test asks what each theory requires the plaintiff to prove, so it follows the plaintiff’s theories, and two grounds that grow from one transaction can be separate causes of action because the evidence needed to support them differs.²⁹

River Park shows the two tests pulling apart. The federal due process claim required the plaintiffs to prove a federally protected property interest and a due process violation, which might have called for evidence of which city officials acted, whether they were state actors, and how the city’s processing procedures violated the plaintiffs’ rights. The state theories had elements of their own, so a court applying the same-evidence test might have held that claim preclusion did not apply.³⁰ Under the transactional test, the federal claim and all three state claims arose from one group of operative facts and were one claim.³¹ The Illinois Supreme Court rejected the narrower test that some Illinois courts had used. It had already described transactional analysis in earlier decisions such as Torcasso, and the test it displaced has the flaw that some of the evidence is identical in almost any pair of related suits.³²

The primary-rights test looks to the right that was violated and finds a separate claim for each violation.³³ Harm to a person and harm to property from one collision violate different rights, so a judgment for one leaves the other open.

Narrow definitions fit a system that restricts joinder. In the early days of English common law, the right to join claims was extremely limited, and claims were treated as distinct categories that different courts often handled.³⁴ Many states later allowed plaintiffs to split claims by type of damages, on the theory that these involved discrete rights protected by distinct causes of action.³⁵ Modern federal and state joinder rules let a plaintiff put every theory into one suit, so plaintiffs seldom can say that they could not have asserted an omitted claim in the first action.³⁶ A definition tied to distinct rights makes sense when the system keeps them apart, and a definition tied to the transaction makes sense when the system lets them travel together.

The separate opinion in Frier shows the narrower approach inside one state’s law. The majority found a common core of operative facts. The separate opinion read Illinois law differently: Illinois defines the common set of facts for claim preclusion as those necessary to sustain the cause of action, not those that could be conveniently litigated in one lawsuit. That focus on the elements of the causes of action and the proofs at trial, instead of the policy advantages of trying both actions together, would have defeated preclusion there.³⁷ That two wholly different causes of action arising out of one transaction could be joined as a convenient trial unit was irrelevant under that Illinois approach, although it would be dispositive under the Restatement.³⁸ The same two suits are therefore barred under one definition and open under the other, and because a state’s definition can change over time, as River Park shows, the answer must come from the rule the rendering court actually applies.³⁹

The first move in a problem is therefore to identify the court that rendered case one and the definition its law applies. A student who applies the transactional test by reflex will bar suits that a narrow rule leaves open, and one who assumes a narrow rule without checking will miss a bar that a transactional rule imposes. The same inquiry applies to a federal judgment, but the source of its preclusion law is a separate question taken up later.

Theories, Damages, and Remedies

Section 24 of the Restatement (Second) of Judgments extinguishes the plaintiff’s rights to remedies against the defendant with respect to all or any part of the transaction, or series of connected transactions, out of which the action arose.⁴¹ The language speaks of the transaction and its remedies. The legal theory, the kind of damages, and the form of relief a plaintiff chose all sit inside one claim, and none of them marks where one claim ends and another begins. A plaintiff tries to divide one transaction in three ways: by theory, by damages, and by remedy. Each attempt fails for the same reason, because the plaintiff framed the first suit, could have included the omitted piece, and cannot hold it back for later.

A plaintiff who loses on one theory cannot try another on the same facts. Rizzuto, who sued over a vicious grounder, could not sue again on a different theory of relief based on the same grounder, since res judicata bars the claims asserted and also the others arising out of that occurrence that could have been asserted.⁴² The joinder rules make that duty fair. Rule 18(a) allows Elbers, who sues Doe Chemical Company for negligently manufacturing a drug, to add a strict liability claim on the same events, and Rule 8(d)(3) permits alternative or inconsistent pleading, but if he leaves the strict liability claim out, res judicata bars it later, because it is only another theory of recovery for the same injuries.⁴³ An employee who loses on an alleged oral promise of two years of employment and then sues on the same facts for the reasonable value of her services, a theory called quantum meruit, meets the same result, because a modern court would probably treat both suits as one transaction.⁴⁴

The bar works as an incentive. Claim preclusion gives plaintiffs a powerful reason to allege all of their claims, and their best theories, in a single case, so they cannot try one approach in the first case and a different one in the second to see which works better.⁴⁵ That is why complaints often contain several counts on different legal theories: it is better to be comprehensive in the first action, subject to Rule 11, than to fight res judicata in a second.⁴⁶

The same reasoning governs the kind of damages. Rizzuto’s later claim for punitive damages arose from the same incident as his earlier suit, so it could have been, and therefore should have been, asserted there, and the court will not hear it however clear his right to it.⁴⁷ Schulansky’s second suit for a different type of damage caused by the same construction work failed for the same reason.⁴⁸ Some jurisdictions once let plaintiffs divide damages by type: automobile plaintiffs could sue for personal injuries and for vehicle damage in separate actions, on the theory that these involved discrete rights protected by distinct causes of action, as in Vasu v. Kohlers, Inc. That rule has largely been abandoned, and Rush v. City of Maple Heights effectively overruled Vasu in Ohio.⁴⁹ A state that still treats the injuries as separate claims nonetheless sends its rule along with its judgment. A court in another state, asked whether a property-damage suit follows a personal-injury judgment from that state, must ask how the rendering state would treat the second suit, even when its own law would bar it.⁵⁰ Installments follow the one-suit logic too: a landlord whose tenant is six months behind when suit is filed must sue for all six months at once, and any month left out is waived.⁵¹

Form of relief is the third variable. A demand for legal relief, generally money damages, and a demand for equitable relief such as an injunction belong to one claim when they rest on the same facts, so both must be requested in the same action. A copyright owner who believes a defendant is infringing cannot sue for an injunction and then sue separately for money damages.⁵² Claims are counted the same way in other settings: a complaint that asserts one legal right states one claim for relief even when it seeks several remedies.⁵³ History explains why the rule had to be learned. Under early English common law, claims were conceptualized in distinct categories and often handled by different courts, so the system itself could force a plaintiff to split a case.⁵⁴ Federal courts and many state systems chose the other course and allowed many claims to be heard in a single action, so a plaintiff can seldom say that a remedy was out of reach in case one.⁵⁵

A plaintiff who wins cannot return for more either. When Schulansky recovered on her claim, the claim merged into the judgment, and she could enforce the judgment through an execution, but no further suit could be brought on the extinguished claim.⁵⁶ If the defendant’s assets sit in another state, she brings an enforcement action there. That action is a collateral proceeding that does not reallege the claim, so claim preclusion does not apply to it.⁵⁷

The hardest splitting problem arises when harm grows after trial. A plaintiff who sues for an eye injury, seeks all known damages, and omits headaches that develop later might argue that the headaches were not part of the first suit. Courts generally reject that argument, and the usual rule requires a plaintiff to recover in the original action all damages, including future damages reasonably likely to ensue, as Restatement (Second) of Judgments § 25 comment c states.⁵⁸ The injury was before the first court, and the plaintiff should have offered evidence that such injuries commonly lead to headaches so the jury could award damages for that probable consequence. If plaintiffs could return whenever their harm exceeded expectations, judgments would never be final.

Latent disease is the recognized pressure point. A worker who recovers for asbestosis may develop cancer decades later, and strict claim-preclusion doctrine bars a second suit for the cancer. Some courts allow it anyway: New Jersey recognized the cancer claim after an earlier asbestosis recovery in Mauro v. Raymark Industries, Inc., while the Fifth Circuit, applying Texas law in Gideon v. Johns-Manville Sales Corp., required a single action, although Texas has since adopted a two-disease approach in Pustejovsky v. Rapid-American Corp.⁵⁹ These courts bend the traditional rule case by case, so a problem about later-developing harm calls for identifying the jurisdiction’s own approach.

A question that describes the second suit as seeking something different invites a wrong answer. A new theory, a new category of damages, and a new form of relief each tempt a student to call the claim new, yet the question is always whether the second suit arises from the transaction the first judgment already reached.⁶⁰ Only a showing that the first court could not have entertained the theory or granted the remedy takes the second suit outside the bar. Allen could not have included his patent theory in the first state-court action because the federal courts have exclusive jurisdiction over patent claims, so he will usually not be barred from asserting it later, a point developed in a later section.⁶¹

Separate Transactions and Later Claims

The preceding subsections widened the claim: a new theory, another type of damages, or a different remedy resting on the same facts is the same claim and belongs in case one. This subsection supplies the limit. Section 24 defines a claim by the single transaction out of which the first action arose, so a claim arising from a separate transaction lies outside it.⁶³ So does a claim on a breach that had not yet occurred when case one was filed.⁶⁴

Rule 18(a) tempts students to miss the limit. It lets a plaintiff join every claim against an opposing party, including claims that have nothing to do with one another, so it seems to follow that whatever the rule allowed in case one must be included there or lost. For grounds arising from the same transaction, preclusion does turn permission into a duty.⁶⁵ For grounds that do not satisfy the definition of a single claim, the “may” of the joinder rule does not mean “must.” Roosevelt, for example, is free to bring a separate action on a contract claim even though she could have joined it in the first suit.⁶⁶

The line sits at the transaction because that is where the doctrine’s reasons run out. A defendant’s interest in repose is an interest in not enduring endless lawsuits arising from the same dispute.⁶⁷ A second suit about a different job, a different accident, or a different year repeats nothing, because the events differ and no earlier judgment is at risk of being contradicted. A rule forcing joinder across unrelated transactions would also load every lawsuit between two people with the whole inventory of their grievances.

The Schulansky problem draws the line inside a single relationship between the same two parties. Her second action against Ronan on a different legal theory for the New Hampshire addition is barred, and so is one that merely asserts a different type of damage caused by that work. Her claim for Ronan’s work on her Massachusetts house is different. It arises from a separate transaction, so she was not required to assert it in the first action even though it had already accrued.⁶⁸ Accidents follow the same pattern. Merkle’s two accident claims against Rico arose out of different occurrences at different times, so suing Rico over one accident does not bar a separate action against him for the other.⁶⁹ Likewise, Rizzuto’s negligence suit leaves him free to sue on a contract to promote Allston gloves on television, because that claim rests on a separate transaction, even though Rule 18(a) would have let him join the two. The contract claim had accrued at the time of the negligence suit, and he was still not required to assert it there.⁷⁰

Successive payments under one contract divide on a different fact, the date the claim accrued. A plaintiff who sues on a contract for one year’s default may sue again for a default in a later year, because no one can be expected to predict future defaults and sue on them before they occur, and separate breaches in successive years are different occurrences even under a single contract.⁷¹ A landlord who sues a tenant for back rent must seek all the rent due but not future defaults, since the claim for future nonpayment could not have been raised until it accrued.⁷² Rent already due at filing sits on the other side of the line, and a landlord who leaves out any month waives it.⁷³

The Massachusetts job and the later-year default reach the same result for different reasons. The Massachusetts claim had accrued and could have been joined, but it was no part of the New Hampshire transaction, so joinder was optional. The later default belonged to the same contract, but joinder was impossible because the claim had not accrued. A month of rent already due at filing is joinable, accrued, and part of the same relationship, so the rule requires it to travel with the rest.

Even where claim preclusion does not apply, issue preclusion may. If Stamski sues Jane a second time over facts that had not taken place when he sued her the first time, claim preclusion does not bar the suit, but issue preclusion will probably bar him from relitigating the meaning of the lease provision decided in the first case, since it would waste the parties’ and the court’s resources to decide it twice.⁷⁴

Tax litigation supplies a compact version of the same line, because each taxable year generates its own liability and its own cause of action.⁷⁵ A taxpayer who litigates an income tax deficiency for 2020 in the Tax Court cannot later file a refund suit for 2020, even if the refund suit raises issues the Tax Court never considered, since the year is the claim and the new issues are only new grounds within it. The same taxpayer may raise those issues for 2021 without meeting claim preclusion, although collateral estoppel keeps the taxpayer from relitigating an issue actually litigated and necessarily determined in the earlier case.⁷⁶

A question on this point gives the two suits a great deal in common: the same parties, the same kind of contract or accident, and an argument that the plaintiff could have joined everything in case one. Those similarities are the distraction. The questions that decide the outcome are whether the second suit rests on events that formed part of the transaction case one adjudicated and, if it rests on a later breach, whether that breach had occurred when case one was filed. When case one was decided in a state court, the court hearing case two applies the rendering state’s definition of a claim, and a narrower definition leaves even more open, so a claim from a separate transaction survives under any of the definitions in this section.⁷⁷