Gibbs and Constitutional Power

Gibbs and Constitutional Power

A state claim that presents no federal question and lacks diversity has no jurisdictional footing of its own. United Mine Workers of America v. Gibbs explains when a federal court may hear such a claim anyway. The Court held that a federal court may do so when the state claim and a federal claim derive from a common nucleus of operative fact and are such that a plaintiff would ordinarily be expected to try them all in one judicial proceeding. That test displaced the narrower cause-of-action inquiry of Hurn with a question about how the underlying facts relate to one another.

The shift rests on the unit that Article III uses. The Court concluded that Article III grants jurisdiction over entire “cases,” not just over particular claims or issues in a case, so a court holding one proper federal claim has constitutional power to hear the whole dispute between the parties.¹ A federal claim and a state claim that grow out of the same events therefore make up one constitutional case. A narrower rule would split that case between two court systems, and when all the claims arise out of the same transaction or occurrence, it is far more efficient to have one court hear the entire matter.²

Gibbs does not turn every related state claim into a federal matter, because it divides the inquiry into two steps. The first step asks about power: does Article III permit the federal court to hear the state claim at all? The answer is yes when the federal claim is substantial and the two claims share a common nucleus of operative fact. The second step asks about discretion. Even when power exists, it need not be exercised in every case, because supplemental jurisdiction is a doctrine of discretion and not of the plaintiff’s right.³ Power is a constitutional boundary that no court can enlarge for its own convenience; discretion is a practical judgment about whether this court should use that power in this case. Section 1367(a) later wrote the boundary into its text by extending supplemental jurisdiction to claims so related to the claims within the court’s original jurisdiction that they form part of the same case or controversy under Article III,⁴ so what Gibbs says about power remains the interpretive background for the statute.

The subsections below follow that structure: the case itself, then the three conditions of power, then the discretion that sits on top of them.

Gibbs: Facts and Holding

The dispute behind Gibbs had one plaintiff, one defendant, a federal claim, a state claim, and no diversity to support the state claim on its own. Grundy Company hired Paul Gibbs to superintend a new mine and gave him a contract to haul its coal, but the plan to open the mine collided with a rivalry between two unions over who would represent the region’s miners. Armed members of a United Mine Workers local kept the mine from opening, and Gibbs lost both the job and the haulage contract. He blamed a concerted union plan and sued only the international union, in the federal district court for the Eastern District of Tennessee.⁵

Gibbs pleaded two claims out of those events. The federal claim alleged a secondary boycott under § 303 of the Labor Management Relations Act, that is, boycotting one party (Gibbs) to affect a labor dispute with a different party, the mine owner.⁶ The state claim alleged, under Tennessee law, an unlawful conspiracy and boycott aimed at maliciously interfering with his employment contract and his haulage contract.⁷ The federal claim supplied original jurisdiction, and such a jurisdictionally sufficient claim is called an anchor claim or freestanding claim. The state claim had no basis of its own. It presented no federal question, and diversity was absent because an unincorporated association such as the union is regarded as a citizen of every state in which it has members.⁸ Its only footing was the doctrine of pendent jurisdiction, the older name for what is now called supplemental jurisdiction.⁹

The Court began with a threshold question: could the district court properly hear the Tennessee claim at all?¹⁰ Hurn had allowed a state claim to ride along when the two claims were “little more than the equivalent of different epithets to characterize the same group of circumstances,” and the Court found the tendency to confine Hurn to such cases unnecessarily grudging.¹¹ In its place the Court stated a broader rule. Pendent jurisdiction exists whenever there is a claim arising under federal law and the relationship between that claim and the state claim permits the conclusion that the entire action comprises one constitutional “case.” Three conditions fill in that relationship:

1. The federal claim must have substance sufficient to confer subject-matter jurisdiction.
2. The federal and state claims must derive from a common nucleus of operative fact.
3. Considered without regard to their federal or state character, the claims must be such that the plaintiff would ordinarily be expected to try them all in one judicial proceeding.

When all three hold, federal courts have power to hear the whole.¹²

Nothing in Article III’s text mentions related state claims, yet Gibbs held that the provision implicitly authorizes jurisdiction over them.¹³ The case itself fits the rule. Both claims grew out of a single labor dispute at a Tennessee coal mine, and under modern pleading rules the parties would ordinarily expect to try such related claims in one proceeding.¹⁴ The claims arose from the same nucleus of operative fact and reflected alternative remedies; indeed, the verdict sheet authorized only one award of damages, so recovery could not be given separately on the federal and state claims.¹⁵

Finding power completed only half of the analysis. The Court added that the power need not be exercised in every case in which it exists, because pendent jurisdiction is a doctrine of discretion, not of the plaintiff’s right.¹⁶ The Court reversed the Sixth Circuit’s affirmance,¹⁷ but its jurisdictional ruling became the framework that still governs related state claims.¹⁸ What puts a state claim beyond that power is the factual relationship between the claims, as the next example shows.

The songwriter and Gibbs each hold a substantial federal claim, so that feature cannot explain the different results. What differs is the factual relationship between the claims: Gibbs’s two claims drew on a single set of events, and the songwriter’s drew on two. A student who reasons that any federal claim draws every state claim along with it skips the common-nucleus requirement, and that requirement is what filters out the libel claim.

A Substantial Federal Claim

The first condition of power concerns the claim that anchors the case. The federal claim “must have substance sufficient to confer subject matter jurisdiction on the court.”²⁰ The requirement follows from how supplemental jurisdiction works. A related state claim holds no ticket of its own and gets into federal court only by attaching to a claim that does, so the court must have original jurisdiction over at least one claim in the action.²¹ If the supposed anchor gives the court no genuine jurisdiction, the state claim has nothing to attach to, which is why the statute reaches only civil actions of which the district courts have original jurisdiction.²²

A substantial claim need not be a winning one. In Hurn v. Oursler the federal infringement question was “not plainly unsubstantial,” so the federal court’s jurisdiction was rightly upheld and the infringement claim was properly disposed of on the merits.²³ The line falls at “plainly unsubstantial” for two reasons. If jurisdiction depended on winning, a court would have to decide the merits to learn whether it had power to decide them, and every loss would retroactively erase the authority to enter it. If any federal label sufficed, a plaintiff could turn a state dispute into a federal case by attaching an empty federal theory to it. Because the question of power is ordinarily resolved on the pleadings, substantiality is measured by what the complaint alleges and the law that governs it when the case begins, and the eventual verdict plays no part.²⁴

Gibbs itself shows the distinction at work. The district court dismissed as unproved the § 303 claims that the union tried to induce coal operators other than Grundy to stop doing business with Gibbs, but it submitted the § 303 claims relating to Grundy to the jury. The jury returned verdicts against the union on those claims, and the verdicts were set aside only on the union’s motion for a directed verdict and judgment n.o.v.²⁵ A federal claim that survives to a jury verdict and is lost afterward was a real anchor all along.

A later loss can still matter, but for a different question. If the federal claims are dismissed before trial, even though they were not insubstantial in a jurisdictional sense, Gibbs says the state claims should be dismissed as well.²⁶ That statement concerns discretion, which a later subsection develops, while the present requirement concerns whether power exists at all.

Suppose the homeowner had instead alleged a colorable federal statutory violation that the court later rejected for want of proof on one element. The same warranty claim would then sit within the court’s power, and the loss of the federal claim would bear only on whether the court should keep the state claim. The plumbing facts are identical in both versions. What differs is whether the federal allegation was ever a genuine basis for jurisdiction, and the usual exam error is to treat a federal claim that loses as though it had been insubstantial from the start.

Common Nucleus of Operative Fact

The second condition asks how the added claim relates to the anchor claim, and the Court answered in one sentence: the state and federal claims “must derive from a common nucleus of operative fact.”²⁸ Operative facts are the events that matter to a claim’s outcome, the things a party must prove to win, as distinct from background that merely sets the scene. A nucleus is common when the same events sit at the center of both claims. Section 1367(a) carries the requirement into the statute by reaching only claims so related to the anchor claim that they form part of the same case or controversy under Article III,²⁹ and a claim derived from a common nucleus of operative fact satisfies that statutory language.³⁰

The test compares facts and ignores labels, which is its break from Hurn. A federal claim and a state claim can carry different names and rest on different bodies of law and still share a nucleus. An employee fired because of his religion who sues under Title VII and adds a state wrongful discharge claim has two claims that both address the termination of employment on the basis of religion, so the firing is the shared nucleus and the state claim is within the court’s power.³¹ Measuring relatedness by events follows from what a joint proceeding saves. When two claims draw on the same events, the same witnesses and documents prove both, so one proceeding costs less than two.³² When the only connection is a remote background link, a combined trial saves little, and calling the second claim part of the same case would stretch federal jurisdiction past what one case can hold. A comparison of facts is also a task a judge can actually perform, because the judge need only compare the stories the two claims tell.

The nucleus test sits close to the same-transaction-or-occurrence standard that governs claim preclusion and the joinder of claims, so the rules on joining claims and the rules on jurisdiction over them work in coordination. The tests use different words but lead to the same results, so a state claim arising from the same transaction or occurrence as the federal claim safely satisfies the nucleus test.³³ An exam answer should therefore treat a same-transaction claim as qualifying and a looser link as arguable, with the argument resting on how much proof the two claims share.

An exam stem rarely announces whether two claims share a nucleus. It supplies two stories and expects a comparison of what each claim requires the plaintiff to prove. An arrestee who sues the arresting officer under a federal civil-rights statute for excessive force and joins a state-law battery claim from the same scuffle has two claims that both turn on what happened during the arrest, so the battery claim shares the nucleus. Another arrestee who adds a claim that the officer never paid her $500 for work done on the officer’s house presents a different picture. Bad feeling over the debt may have helped bring on the arrest, but proving the debt requires proving the work and the nonpayment, neither of which occurred during the encounter, and a causal link that thin is too attenuated to put the debt claim in the same nucleus.³⁴

The same line separates cases outside the arrest setting. A federal antitrust claim and a state unfair-competition claim arising from the same alleged conduct usually share a nucleus, because many facts that prove the unfair competition also help prove the price-fixing or monopoly. A slander claim, by contrast, often requires proof of different facts than a personal-injury claim, even when the defendant allegedly slandered the plaintiff immediately after the accident and the events were part of a series.³⁵ Closeness in time does not make the facts the same: the injury case turns on how the collision happened and what it did to the plaintiff’s body, while the slander case turns on what was said afterward and what it did to the plaintiff’s reputation.

The last step matters because failing the nucleus test closes only the supplemental route. A claim with its own jurisdictional basis needs no anchor: a plaintiff who pleads a patent claim, for example, invokes the district court’s own subject-matter jurisdiction under 28 U.S.C. § 1338, and must plead and establish it.³⁶

Claims Tried Together

The third formulation in Gibbs asks a practical question after the factual one. If the plaintiff’s claims, considered without regard to whether each is federal or state, are such that the plaintiff would ordinarily be expected to try them all in one judicial proceeding, then, assuming the federal issues are substantial, the federal court has power to hear the whole.³⁸ The instruction to disregard each claim’s federal or state character asks the court to picture the dispute as an ordinary lawsuit with no line between court systems. In that picture a person wronged by one course of events decides which claims to put before a single judge and jury, and for claims that grow out of the same events the usual answer is all of them. Section 1367(a) carries the idea into the statute by reaching claims so related to the anchor claim that they form part of the same case or controversy under Article III.³⁹

The test measures a plaintiff’s expectations because the procedural system builds those expectations. Under the Federal Rules the impulse is toward the broadest scope of action consistent with fairness to the parties, and joinder of claims, parties, and remedies is strongly encouraged.⁴⁰ A jurisdictional rule that split apart claims the Rules invite a plaintiff to join would force two lawsuits where the Rules contemplate one. Claim preclusion applies firmer pressure. A plaintiff who omits a claim arising out of the same transaction or occurrence as the claim sued on must join it now or lose the ability to bring it later.⁴¹

That pressure explains why the jurisdictional test and the preclusion test have to line up. A dismissal for lack of subject-matter jurisdiction should not operate preclusively to bar the plaintiff from bringing the case again in the proper court, because such a dismissal is not “on the merits.”⁴² The connection also appears in Gibbs itself. The Court observed that Hurn’s citation of a res judicata decision shows that the weighty policies of judicial economy and fairness to parties reflected in res judicata were themselves strong counsel for a rule that would let federal courts dispose of the state claims as well as the federal claims.⁴³ The working result is a rule of thumb: when claim preclusion would force a plaintiff to join an additional claim, the nucleus-of-common-fact test of § 1367(a) is met as well.⁴⁴

The religious-discrimination example shows the alignment at work. The employee’s federal claim and the state wrongful discharge claim both address the termination of employment on the basis of religion, so they arise out of the same transaction or occurrence. Rule 18 permits the employee to join the wrongful discharge claim, and claim preclusion would compel the joinder, so the employee would ordinarily be expected to try both claims together and the nucleus test is satisfied.⁴⁵

Counterclaims show the same line from the defendant’s side, because the Rules sort them by the same same-transaction-or-occurrence standard: a counterclaim is compulsory under Rule 13(a) when it arises out of the transaction that is the subject of the plaintiff’s claim, and permissive under Rule 13(b) when it does not. Suppose a Tennessee corporation sues its former CEO, a Florida citizen, for $100,000 on a non-compete clause, with diversity supplying the anchor claim. If the CEO counterclaims for $50,000, alleging that the clause was fraudulently inserted into the contract, the counterclaim concerns the same contractual terms and their validity. It is compulsory, and although it has no basis of its own as a state-law claim worth less than $75,000, it satisfies § 1367(a) because it shares a common nucleus with the original claim. A counterclaim for bonuses the company failed to pay three years earlier concerns different events altogether. That claim is permissive, it does not arise from a common nucleus of facts with the non-compete claim, and supplemental jurisdiction is unavailable for it.⁴⁶

The line falls there for reasons that follow from what a joint trial accomplishes. The fraud counterclaim and the non-compete claim ask one factfinder to examine the same contract, the same negotiations, and the same terms, so a single trial costs less than two and cannot produce conflicting findings about the document. A claim for bonuses requires proof of a bonus arrangement and a payment history that play no part in the non-compete dispute, so a combined trial saves nothing and only attaches an unrelated state claim to a federal case. If such claims could ride along, a party could put any state-law grievance against an opponent into federal court simply because the opponent chose to sue there, and the limits on federal jurisdiction would give way.

Identity of the parties therefore decides nothing, and the loan claim fails for the same reason the bonus counterclaim failed. The inference runs in one direction: a claim that preclusion or compulsory joinder would force into the suit qualifies without further argument, while a claim that is merely related can still qualify but needs an argument built on the proof the claims share.

Power Versus Discretion

Showing that a federal court has power over a related state claim does not settle whether it must hear the claim. After stating the conditions of power, Gibbs added that the power “need not be exercised in every case in which it is found to exist,” because pendent jurisdiction “is a doctrine of discretion, not of plaintiff’s right.”⁴⁸ The inquiry therefore has a second step. The first step asks whether Article III reaches the state claim, a question answered by the substantial federal claim, the common nucleus of operative fact, and the expectation of a single trial. The second asks whether the court should use that power in the case before it, and a different set of considerations answers it.⁴⁹

Those considerations rest on what justifies hearing the state claim at all. Gibbs located the justification in judicial economy, convenience, and fairness to litigants, and it said that when those are missing a federal court “should hesitate to exercise jurisdiction over state claims, even though bound to apply state law to them.”⁵⁰ Economy explains the first part of the hesitation: a combined proceeding is worth having only if it saves the parties and the court some work, so a state claim that adds nothing to a federal case has lost the reason the power exists. Comity explains the second part. State courts are the authoritative readers of state law, so needless decisions of state law should be avoided both as a matter of comity and to give the parties a surer-footed reading of the applicable law.⁵¹

Gibbs names several circumstances that tip the balance, and each points in a recognizable direction:⁵²

• When the federal claims are dismissed before trial, even though they were not insubstantial in a jurisdictional sense, the state claims should be dismissed as well.
• When the state issues substantially predominate, whether in terms of proof, the scope of the issues raised, or the comprehensiveness of the remedy sought, the state claims may be dismissed without prejudice and left to state tribunals.
• When the state claim is closely tied to questions of federal policy, the argument for keeping it is particularly strong.
• When trying divergent legal theories together creates a likelihood of jury confusion, that risk may justify separating the state and federal claims for trial.

The first two factors push toward sending the state claim away, and the third pushes toward keeping it. The fourth depends on whether the confusion can be managed, and Gibbs itself shows that it sometimes can. None of these is a formula that dictates a result. They guide a judgment that belongs to the trial judge, who sees the case as it develops.

Because the judgment depends on how the case develops, the discretionary question stays open for the whole litigation. The question of power will ordinarily be resolved on the pleadings, but pretrial procedures or the trial itself may reveal “a substantial hegemony of state law claims,” meaning a dominance of state law over the case, or a likelihood of jury confusion that no one could have anticipated when the complaint was filed.⁵³ A plaintiff may well know the nature of the proofs and the relative importance of the claims from the start, and a federal court’s wide latitude over state-law questions does not oblige it to tolerate a litigant’s effort to impose on it what is in effect only a state-law case. Once a state claim “constitutes the real body of a case, to which the federal claim is only an appendage,” the state claim may fairly be dismissed.⁵⁴ Power over the claim exists the whole time, so this is the point at which discretion polices a thin federal claim attached to a state dispute.

Gibbs itself shows discretion exercised in favor of keeping the claim. The trial court submitted the § 303 claims relating to Grundy to the jury, which returned verdicts against the union that were set aside only on post-verdict motions, and the district judge considered the haulage-contract claim proved as to liability and failing only for lack of proof of damages.⁵⁵ Some risk of jury confusion existed, since differing standards of proof of union involvement applied, but a special verdict form lessened it, and the question whether preemption limited the state claim gave a special reason to keep the case, because federal courts are particularly appropriate bodies for applying preemption principles.⁵⁶ The Court concluded that the district court might, in its sound discretion, have dismissed the state claim, but that the circumstances showed no error in refusing to do so.⁵⁷ The ruling approved a choice that could have gone either way.

What would send a case like Gibbs the other way? An exam stem usually supplies the answer by letting the federal claim fall away early or by letting the state claim swallow the case.

The tenant’s case differs from Gibbs in how far the federal claim shaped the proceeding. In Gibbs the federal claims were tried to a jury and the preemption question tied the state claim to federal policy, so keeping the state claim served economy and the federal interest together. Here nothing federal gets decided, so no work is saved by keeping the state claims, while the open state-law question presents the comity concern in its strongest form. The dismissal is without prejudice, which leaves the tenant free to pursue the claims in state court.⁵⁸ The loss of a federal claim before trial does not erase the court’s power over the state claim, as the earlier discussion of a substantial federal claim anticipated. It changes the discretionary balance and puts retention of the state claim to the judge’s judgment.

Congress later wrote several of these considerations into the statute. Section 1367(c) allows a court to decline supplemental jurisdiction when the claim raises a novel or complex issue of state law, when it substantially predominates over the claims within original jurisdiction, or when the court has dismissed all claims over which it has original jurisdiction, among other grounds that later subsections take up.⁵⁹ Each ground compresses a consideration from Gibbs: novel state law echoes the concern for a surer-footed reading of state law, predominance echoes the factor of the same name, and dismissal of all original-jurisdiction claims echoes early dismissal of the federal claims.