Introduction

Subject-matter jurisdiction tells a plaintiff which court system may hear her claim, and personal jurisdiction tells her which states' courts may bind the defendant. Even after both are satisfied, one question remains open: within the chosen system, which particular court may actually try the case? In the federal system that is the question of venue, and Congress, rather than the Constitution, supplies the answer. Venue rules take the set of courts that jurisdiction leaves available and narrow it further, so that the place of trial bears some sensible relationship either to the parties or to the events in suit. A lawsuit can be jurisdictionally flawless and still be filed in a district with no real connection to anyone involved, and venue doctrine exists to prevent exactly that.

Because venue is a creature of statute built around convenience, it behaves differently from the jurisdictional doctrines that precede it. It is a protection the defendant holds and can give up, which means the objection must be raised early and is easily lost, and a venue defect never threatens the validity of a judgment the way a defect in subject-matter jurisdiction does. The consequences nonetheless run in both directions: a case filed in the wrong district is exposed to dismissal or transfer, while a case filed in a technically proper but badly located district can be moved to a better one, sometimes with consequences for the law that will govern the merits.

The chapter follows the order in which a court would take these questions up. It first identifies the districts where venue is proper under the general venue statute, a task that turns on where each kind of defendant resides. It then turns to contested venue: how the objection is raised and how it is lost, what a court may do with a case filed in the wrong district, how a court sitting in a proper district may nonetheless send the case elsewhere for convenience, and which state's law applies once the case arrives.

What Venue Does

Venue is a statutory requirement that does two jobs at once: it regulates the flow of judicial business within a court system, spreading cases among districts rather than letting them pool wherever plaintiffs prefer, and it identifies a forum with a logical connection to the dispute or the parties. In the federal system, venue points to a particular judicial district within a state, which makes it the finest-grained of the three forum requirements: subject-matter jurisdiction selects a court system, personal jurisdiction selects the states in which the defendant may be sued, and venue selects the district within those states where the suit may proceed. The general venue statute, 28 U.S.C. § 1391, supplies the governing rules for most civil actions, and it operates "[e]xcept as otherwise provided by law," so a specialized venue statute — the patent venue statute is the classic example — displaces it wherever one applies¹.

The district-level focus matters because personal jurisdiction is measured against states as wholes. A defendant with statewide contacts in Texas is amenable to suit everywhere in Texas, so if jurisdiction were the only constraint, a plaintiff could sue him in a Texas district hundreds of miles from every witness and every scrap of evidence. Venue is the doctrine that asks whether that particular district is a proper place for trial even though the state as a whole plainly is. Without it, the breadth of personal jurisdiction inside a large state would leave defendants and witnesses exposed to litigation in districts that have nothing to do with them, and the venue statute answers that exposure by demanding a connection — the defendants' home or the events themselves — between the chosen district and the case.

Because venue serves convenience rather than power, the law treats it as a personal privilege of the defendant, and a privilege can be surrendered. A case filed in a district the statute does not authorize is laid in improper venue, but the defect disappears if the defendant consents to it or fails to object in time, and the court then proceeds as though venue were proper. That waivability is the sharpest line between venue and subject-matter jurisdiction: a challenge to whether the federal courts may hear the claim at all attacks the court's power, belongs to no party, and can never be waived, while a venue challenge protects only the defendant's convenience and evaporates the moment the defendant shows he is willing to litigate where the plaintiff filed. Keeping those two challenges separate is essential, because the waiver rules studied later in this chapter apply to one and never to the other.

Proper Venue Under Section 1391(b)

Section 1391(b) lists the three bases of proper venue in most federal civil actions: a district where any defendant resides, if all defendants reside in the same state; a district where a substantial part of the events or omissions giving rise to the claim occurred, or where a substantial part of the property in dispute is situated; and, only when neither of those identifies any district, a fallback keyed to personal jurisdiction². Two features of this design should be fixed before the details. First, more than one district is often proper at the same time, and the plaintiff needs only one basis for the district she chooses; it does not matter that other bases fail for that district or that other districts would also qualify. Second, the statute pairs a party-based rule with an event-based rule because each captures a different kind of legitimate connection: a district where the defendants live is convenient for the defense, while a district where the events occurred is convenient for the evidence, and either connection makes the forum a sensible place for trial.

Residential Venue

Under § 1391(b)(1), venue lies in "a judicial district in which any defendant resides, if all defendants are residents of the State in which the district is located"³. The provision works in two steps, and each step operates at a different level of geography. The first is a state-level condition: all defendants must be residents of a single state, or the provision supplies nothing at all. The second is a district-level choice: once the condition is met, any one defendant's home district is a proper venue for the whole case. So if two defendants both reside in New York, one in the Southern District and one in the Eastern District, venue is proper in either district. Move one defendant to Colorado and leave the other in Ohio, and (b)(1) identifies no proper district anywhere in the country, because the defendants are no longer residents of a single state; the plaintiff must then build venue on the transactional basis instead.

The all-defendants condition is what keeps this rule fair. Without it, a plaintiff suing defendants scattered across several states could force all of them to a district that is home to only one, and a residence-based venue would then be no convenience at all for the others. Once every defendant shares a state, by contrast, the worst that any one defendant's home district imposes on the rest is an intrastate burden, which the statute treats as tolerable. One caution before moving on: the word "resides" in (b)(1) is a defined term with a different content for individuals, entities, and corporations in multi-district states⁴, and the whole of the next section is devoted to those definitions.

Transactional Venue

Section 1391(b)(2) makes venue proper in a district in which a "substantial part of the events or omissions giving rise to the claim occurred, or a substantial part of property that is the subject of the action is situated"⁵. The operative word is "substantial," and it deliberately tolerates more than one answer: the test asks whether the district's connection to the claim is genuine and significant, and it does not require anyone to crown the single district where the most important event happened. A contract negotiated and to be performed in one district supports transactional venue there even though the defendant resides somewhere else entirely, and in a products case both the district where the product was manufactured and the district where it caused injury will ordinarily qualify. The rule runs out, though, where the connection is derivative rather than event-based: a district whose only link to the case is that the plaintiff felt economic effects there, with no events or omissions giving rise to the claim actually occurring in the district, does not satisfy (b)(2)⁶.

The generous phrasing is a deliberate correction of an older formula. Before 1990 the statute laid venue in the district "where the claim arose," and courts read that language to point to one district, which forced litigation — Leroy v. Great Western United Corp. is the well-known example — over which single district in a multi-event dispute deserved the title. The 1990 amendment substituted the substantial-part language precisely because several places usually have a genuine evidentiary connection to a dispute, and forcing courts to pick one wasted effort on a question with no right answer. The event-based connection also does independent work that the residence basis cannot: it guarantees that the forum has local access to the witnesses and proof, wherever the defendants happen to live.

Fallback Venue

Section 1391(b)(3) provides that "if there is no district in which an action may otherwise be brought as provided in this section," the plaintiff may sue in any district in which any defendant is subject to the court's personal jurisdiction with respect to the action⁷. The trigger deserves emphasis, because it is the point most often misread: the fallback opens only when no district anywhere in the United States satisfies (b)(1) or (b)(2), and the absence of a proper district in the plaintiff's preferred state changes nothing. Where the ordinary bases identify several proper districts, a plaintiff may not invoke (b)(3) to reach a district she finds more convenient; the fallback exists to fill a gap, and where there is no gap it has no work to do.

The gap it fills is real. Suppose a plaintiff sues one defendant from Colorado and one from Ohio over an accident that occurred entirely in Canada. The defendants share no state, so (b)(1) supplies no district; the events occurred in no American judicial district, so (b)(2) supplies none either; and without a fallback the plaintiff could satisfy jurisdiction somewhere yet have proper venue nowhere. In that situation (b)(3) opens any district in which a defendant is subject to personal jurisdiction — the Ohio defendant's amenability at home, for example, makes an Ohio district a proper venue⁸. The fallback supplies only venue, however; personal jurisdiction over each defendant remains a separate requirement that the plaintiff must still satisfy in the chosen district. Confining the provision to the true-gap case is what keeps it from swallowing the connection requirements the first two clauses impose: the loosest connection the statute knows is tolerated only because, in these rare cases, nothing better exists.

Where Defendants Reside

Residential venue is only as clear as the word "resides," and § 1391(c) supplies its content "[f]or all venue purposes" with a separate definition for each kind of defendant: natural persons, entities, and defendants not resident in the United States⁹. Section 1391(d) then adds a special rule for corporations sued in states with more than one district¹⁰. The definitions differ because the connection that makes a district a fair home forum differs by kind of defendant: a person has one true home, an entity has none in that sense, and a foreign defendant has no American home at all. Before working through them, note the confusion this vocabulary invites. Venue residence for a corporation is a different concept from citizenship under the diversity statute: a corporation is a citizen of at most two states for diversity purposes, but it may reside, for venue, in every district in which it is subject to personal jurisdiction for the action. Treating the two as interchangeable is among the most common analytical mistakes in this area.

Natural Persons

Under § 1391(c)(1), a natural person — including an alien lawfully admitted for permanent residence — is deemed to reside in the judicial district in which that person is domiciled¹¹. Domicile carries the same meaning it has in the diversity context: the person's true, fixed home, to which he intends to return, so a defendant carries exactly one venue residence at a time. In Thompson v. Greyhound Lines, Inc. the court located an individual defendant's venue residence in precisely this way, finding his domicile and treating the district containing it as his residence for the (b)(1) analysis¹². An individual domiciled in the Southern District of Alabama therefore resides there for venue even if he owns a vacation house in another district, and a defendant who works weekdays in one district but keeps his home and his intent to remain in another resides only in the latter.

The single-district answer is the point of the rule. Anything looser — physical presence, property ownership, frequent visits — would make an individual "reside" in many districts at once, and a plaintiff could then manufacture residence-based venue out of incidental contacts, which would defeat the purpose of anchoring venue to a defendant's genuine home¹³. Domicile picks out one stable district per person and gives the (b)(1) analysis a fixed input rather than a moving target.

Entities

Section 1391(c)(2) governs any entity with the capacity to sue and be sued in its common name, incorporated or not, and it draws a sharp line between the entity's two litigating postures: as a defendant, the entity resides in any judicial district in which it is subject to personal jurisdiction with respect to the civil action in question, but as a plaintiff, it resides only in the district of its principal place of business¹⁴. Each half has its own logic. The defendant-side rule borrows personal jurisdiction as the measure of residence because amenability is itself the constitutional test of fairness: an entity that can lawfully be sued in a district for this very claim cannot complain that trial there is unfairly inconvenient. In Zippo Manufacturing Co. v. Zippo Dot Com, Inc. a California company's paid subscription contracts with Pennsylvania residents supported specific jurisdiction in the Western District of Pennsylvania for the trademark claims arising from them, and because the company was subject to jurisdiction there for that action, it resided there and venue was proper¹⁵. The plaintiff-side rule does subtler work: an entity with nationwide operations could otherwise parlay its own widespread activities into claimed residences, and venue advantages, in dozens of districts, so when the entity chooses to sue, the statute gives it only its true home.

The same borrowing means that entity residence rises and falls with personal jurisdiction doctrine. A corporation that shipped an injurious product into a district and is subject to specific jurisdiction there for the resulting claim resides in that district as a defendant, making venue proper under (b)(1) if the all-defendants condition is met; the same corporation suing as a plaintiff resides only where its principal place of business sits¹⁶. And because the Supreme Court has narrowed general jurisdiction to the states where a corporation is essentially at home, the districts in which a corporation resides for claims arising elsewhere have narrowed correspondingly — a large factory or sales presence in a district no longer makes the corporation a resident there for unrelated claims.

Defendants Not Resident in the United States

Section 1391(c)(3) states two rules for a defendant not resident in the United States: such a defendant may be sued in any judicial district, and the joinder of such a defendant is disregarded in determining where the action may be brought with respect to the other defendants¹⁷. The first rule reflects that venue has no convenience work to do for this defendant. A person with no American residence has no home district that a venue rule could protect, so the fairness of the forum is policed entirely by personal jurisdiction, and once jurisdiction exists any district will do. The second rule protects the domestic co-defendants' venue analysis from distortion: a foreign individual's presence in the suit is ignored when testing whether all defendants reside in one state, so joining him cannot destroy an otherwise proper residence-based venue for the Americans¹⁸.

The one real trap here is the boundary between (c)(2) and (c)(3). A foreign corporation that is subject to personal jurisdiction in a district is deemed a resident of that district under (c)(2), which means it never enters (c)(3) at all: it counts fully in the venue analysis, its residence is computed like any other entity's, and its joinder is not disregarded. Only a defendant with no United States residence — a foreign individual with no American domicile is the clean case — gets the sue-anywhere treatment and the disregard rule.

Corporations in Multi-District States

Section 1391(d) answers a problem the entity definition creates in states with more than one district. Personal jurisdiction is measured against the state as a whole, so a corporation amenable to suit anywhere in Texas would, under (c)(2) alone, "reside" in every Texas district, including remote ones where it has no presence whatever, and plaintiffs could then choose the emptiest district in the state. Section 1391(d) closes that opening: in a multi-district state, a corporate defendant is deemed to reside in any district within which its contacts would be sufficient to subject it to personal jurisdiction if that district were a separate state, and if no district qualifies, in the district with which it has the most significant contacts — the most-significant-contacts test¹⁹. The district-as-state device confines residence to districts where the corporation's own contacts would justify suit, which restores at the district level the connection requirement that statewide amenability erases.

A matched pair shows the test in operation. Suppose a Delaware corporation's only Virginia connection is a contract negotiated and to be performed in Richmond, which sits in the Eastern District of Virginia. Testing the Eastern District as though it were a separate state, the corporation's Richmond contacts would support specific jurisdiction for a suit on that contract, so the corporation resides in the Eastern District under (d) and venue is proper there under (b)(1). The Western District of Virginia comes out the other way on the same facts: the corporation has no contacts there, so if that district were a separate state it could not be sued in it, and it does not reside there even though it is amenable to suit throughout Virginia. The final clause — the most-significant-contacts fallback — guarantees that a corporation resident in the state under (c)(2) always has at least one home district, so the two provisions never combine to leave a corporation resident in a state but resident in none of its districts²⁰.

Challenging Improper Venue

A defendant objects to venue through Rule 12(b)(3), which lists improper venue among the defenses that may be asserted by motion, or by pleading the defense in the answer; if the defendant chooses the motion route, the motion must be made before the responsive pleading²¹. Timing then does the real work. Rule 12(g)(2) forbids a party who has made a pre-answer Rule 12 motion from making a second one raising a defense that was available but omitted, and Rule 12(h)(1) declares the venue defense waived if it is left out of that first motion or out of the first responsive pleading²². So a defendant who files a pre-answer motion to dismiss for failure to state a claim without including the venue objection has lost it for good, and the court will hear the case in the improper district. A defendant who raises improper venue in a timely 12(b)(3) motion, or in the answer when no pre-answer motion was made, preserves the objection and puts the court to a choice of remedies.

The strictness of that waiver regime follows directly from venue's character as a convenience privilege. A defendant willing to litigate in the chosen district loses nothing that venue was designed to protect, so the system demands the objection at the threshold, before either side has spent resources, and treats silence as acquiescence. Subject-matter jurisdiction sits at the opposite pole because it guards the court's power rather than any party's comfort: a court must dismiss for lack of subject-matter jurisdiction whenever the defect appears, however late, and no party conduct can cure it. The classification therefore controls everything about timing — the same silence that forfeits a venue objection has no effect at all on a jurisdictional one.

When the objection is preserved and the district is in fact wrong, § 1406(a) gives the court two options: it may dismiss the case, or, in the interest of justice, transfer it to any district in which the case could have been brought²³. Transfer is usually the sounder choice, and the reason is what dismissal costs a plaintiff who made an honest venue mistake. A dismissed plaintiff must refile, pay a new filing fee, and serve process again, and if the limitations period ran between the original filing and the dismissal, the claim is gone; a transferred case, by contrast, keeps its original filing date and simply continues in the proper district. Courts therefore ordinarily find transfer in the interest of justice whenever a proper federal district exists.

The transfer power reaches even a court that could not itself decide the case. In Goldlawr, Inc. v. Heiman the transferor court lacked personal jurisdiction over two corporate defendants, and the Supreme Court held that the absence of jurisdiction did not defeat the court's power to transfer under § 1406(a) rather than dismiss²⁴. The holding fits the remedy's purpose: the plaintiff who filed in a district lacking venue has often filed in one lacking personal jurisdiction as well, and by the time the defect is established the limitations period may have run, so conditioning the transfer power on jurisdiction over the defendant would disable it in exactly the cases where dismissal would be most unjust.

Transfer From a Proper Venue

The venue statute deliberately tolerates multiple proper districts, and a district can satisfy § 1391(b) while sitting far from every witness and every document in the case. Section 1404(a), the federal venue transfer statute, addresses that situation: for the convenience of parties and witnesses, and in the interest of justice, a district court may transfer any civil action to any other district where it might have been brought, or to any district to which all parties have consented²⁵. Keep this device separate from the one just studied. Section 1406(a) cures a filing in the wrong district; § 1404(a) presupposes that venue in the original court is proper and moves the case anyway because another proper district is a better place to try it. Congress built the provision as an intra-system replacement for the common-law forum non conveniens dismissal: rather than dismissing and forcing the plaintiff to refile elsewhere, with all the limitations risk that entails, the court sends the living case to the better federal district, and it may do so on a lesser showing of inconvenience than the common-law doctrine required.

The statutory precondition is that the transferee district be one where the action "might have been brought," which means the transferee court must have had subject-matter jurisdiction, personal jurisdiction over the defendants, and proper venue at the time the suit was filed²⁶. In Hoffman v. Blaski the Supreme Court held that this is measured as of the original filing and without regard to the defendant's later conduct, so a defendant cannot obtain transfer to a district where venue and jurisdiction were lacking by offering to waive its objections there. The limit keeps transfer from becoming a device by which defendants acquire a forum the plaintiff never had a right to choose. Congress later added one measured exception: the statute now permits transfer to a district to which all parties have consented, so the plaintiff's agreement — never the defendant's unilateral waiver — can open an otherwise unavailable district²⁷.

Within those bounds the decision is discretionary, and either party may move — the plaintiff included — or the court may raise the question itself. Courts weigh private-interest factors such as access to sources of proof, the availability of compulsory process for unwilling witnesses, and the cost of attendance for willing ones, alongside public-interest factors such as court congestion, the local interest in deciding local controversies at home, and the forum's familiarity with the governing law, and the plaintiff's choice of forum carries substantial weight, so the movant bears the burden of showing that the balance favors the move. A case properly filed in a district far from the accident scene may accordingly be transferred to the district where the collision occurred and the witnesses live, since the plaintiff could have filed there originally; the Fifth Circuit, sitting en banc in In re Volkswagen of America, Inc., went so far as to police that discretion by mandamus where the chosen Texas district had no meaningful connection to the collision and the sources of proof all lay elsewhere²⁸. The section's remaining subsections handle smaller movements within a single district: on motion, consent, or stipulation of all parties a case may be transferred between divisions of the same district, and the court may order trial at any place within the division where the case is pending²⁹.

Law Applied After Transfer

A § 1404(a) transfer is designed to change the courtroom and nothing else, and the choice-of-law rule enforces that design. In Van Dusen v. Barrack the Supreme Court held that after a transfer from a proper venue, the transferee court must apply the state law, including the choice-of-law rules, that the transferor court would have applied had the case stayed put. The mechanism behind the rule starts from Erie's framework: a federal court sitting in diversity applies the choice-of-law rules of the state in which it sits, so if a transfer switched the applicable rules, moving the case could change its outcome. Defendants could then invoke a convenience statute to strip plaintiffs of the substantive law their proper filing secured, and every transfer motion would become a fight over legal advantage rather than a comparison of courtrooms. Fixing the law at the transferor's keeps § 1404(a) about convenience.

In Ferens v. John Deere Co. the Court extended the transferor-law rule to transfers the plaintiff itself requests³⁰. There a Pennsylvania plaintiff whose tort claim was time-barred at home filed it in a Mississippi federal court to capture Mississippi's six-year limitations period, then moved under § 1404(a) to transfer the case to Pennsylvania for convenience; the Pennsylvania transferee court was required to apply the Mississippi limitations rule the transferor would have applied³¹. The Court accepted even this pointed forum shopping for systemic reasons. A rule that switched depending on who moved would be unadministrable when both parties, or the court itself, initiate the transfer, and the plaintiff's power to capture favorable law by filing in a proper but inconvenient forum already exists without any transfer at all — denying transferor law would only push the plaintiff to stay in the inconvenient forum to keep it, defeating the convenience purpose the statute exists to serve.

The rule applies only to transfers from a proper venue, and the § 1406 situation comes out the other way. When a case filed in an improper district is transferred under § 1406(a), the transferee court applies the law of its own state, choice-of-law rules included. The contrast follows from the same logic run in reverse. The transferor-law rule preserves an advantage the plaintiff legitimately secured by filing where the statute entitled her to file; a plaintiff who filed in a district the venue statute never authorized captured nothing, so there is nothing for the transfer to preserve, and carrying the improper forum's law along would reward filing in obviously wrong courts precisely to shop for favorable law. A single question therefore sorts every transfer problem: was the original district proper? If yes, the law travels with the case; if no, the transferee's own law governs.

Forum Non Conveniens