A Summons That Has to Cross an Ocean and a Legal System

Picture the moment a breach-of-contract complaint is signed and ready to file in a U.S. district court. The plaintiff is a Los Angeles importer; the defendant is a manufacturer in Shenzhen that shipped two containers of defective goods and stopped answering emails. The lawyer drafts the summons, attaches the complaint, and then hits the wall that defines U.S.–China litigation: there is no process server who can walk into a factory in Guangdong, hand over the papers, and sign a return. Between the plaintiff and valid service stand an ocean, a sovereign that guards its courts jealously, and a 1965 treaty that China has implemented about as restrictively as any party to it.

That treaty—the Hague Service Convention—turns what is a clerical afterthought in a domestic case into a strategic problem that can determine whether the lawsuit goes anywhere at all. Service of process is not a formality. In American litigation it is the predicate for personal jurisdiction: a court generally cannot bind a defendant it has not properly served, and the Supreme Court has held that "before a federal court may exercise personal jurisdiction over a defendant, the procedural requirement of service of summons must be satisfied" (Omni Capital Int'l, Ltd. v. Rudolf Wolff & Co., 484 U.S. 97, 104 (1987)). Botch service on a Chinese defendant and the consequences cascade: the court may quash service or dismiss, a default judgment may later be vacated as void, and—the quiet killer—even a judgment that survives in the United States may be worthless because a Chinese court will refuse to enforce a judgment obtained through service that violated Chinese law. The method of service therefore reaches both the front end of the case (jurisdiction) and the back end (collection). It is worth getting right.

This guide walks through the whole problem: when the Convention applies, how the Central Authority route actually works in China, why the familiar shortcuts are unavailable, and—crucially, because this is where the law is moving fastest—when a U.S. court will let you skip the Central Authority altogether and serve a Chinese defendant by email. We then turn to the planning moves, the FRCP 4(d) waiver and the SinoType contractual workaround, that can make the entire treaty apparatus unnecessary. It is written for lawyers and non-lawyers alike, and it takes a practitioner's neutral view. Because treaty membership, designated authorities, fees, and processing times shift over time, confirm the specifics against current primary sources before relying on them in a live matter.

For the cousins of this article, see our companion pieces on methods and strategy for serving defendants in China and the IP-specific playbook in serving legal documents to Chinese defendants in IP litigation. This piece is the strategic overview; those two go deeper on tactics and on the trademark and counterfeiting context, respectively.

The Hague Service Convention in Brief

The full title is a mouthful: the Convention on the Service Abroad of Judicial and Extrajudicial Documents in Civil or Commercial Matters, concluded at The Hague on November 15, 1965 (20 U.S.T. 361, T.I.A.S. No. 6638). Everyone calls it the Hague Service Convention. Its purpose is refreshingly practical for an international instrument: create a reliable, standardized channel for moving legal papers across borders; make sure defendants sued abroad actually receive timely notice; and make proof of service clean and predictable. More than 80 contracting states have joined—the number keeps climbing—including, importantly for our purposes, both the United States and China.

The Convention's scope is fixed by Article 1, which provides that it applies "in all cases, in civil or commercial matters, where there is occasion to transmit a judicial or extrajudicial document for service abroad." Hold onto that phrase—"occasion to transmit . . . for service abroad"—because it is the hinge on which the entire subject turns. When a U.S. company sues a Shenzhen manufacturer and must serve it in China, the machinery is ordinarily triggered.

The Convention's signature innovation is the Central Authority. Each contracting state designates an office to receive incoming service requests from other states, arrange for service under local law, and return proof. In the United States, the Central Authority is the Department of Justice's Office of International Judicial Assistance, which executes incoming requests through a private contractor. China has designated its Ministry of Justice—operating through its International Legal Cooperation Center in Beijing—as its Central Authority. Service into China runs through that office under Article 5. As the leading practice guidance puts it, litigants overwhelmingly prefer the Central Authority channel precisely because it "eliminate[s] doubts as to the validity of the service method used," which in turn helps ensure that a U.S. judgment will be recognized where the defendant actually has assets.

One scope point that trips people up: the Convention governs only the initial service of process—the summons and complaint—not later filings in the case. Once a Chinese defendant has appeared, ordinary domestic service rules govern the motions and discovery that follow; the treaty is a front-door requirement, not a hall pass you have to renew for every document (see, e.g., Trade Well Int'l v. United Cent. Bank, 2014 WL 4546022 (W.D. Wis. Sept. 12, 2014)).

When Does the Convention Even Apply? Schlunk and the Forum-Law Question

Before asking how to serve under the Convention, a litigant must answer a logically prior question: does the Convention apply at all? Everything flows from the answer, and the answer comes from the Supreme Court's foundational decision in Volkswagenwerk Aktiengesellschaft v. Schlunk, 486 U.S. 694 (1988).

Schlunk lays down two principles that pull in opposite directions, and the tension between them is the whole game. First, where the Convention applies, compliance is mandatory and exclusive. By operation of the Supremacy Clause (U.S. Const. art. VI, cl. 2), the Convention "pre-empts inconsistent methods of service prescribed by state law in all cases to which it applies," and a plaintiff cannot simply ignore it and serve some other way. Second—and here is the limitation that makes the entire planning strategy possible—whether the Convention applies in the first place is decided not by the treaty but by the internal law of the forum. The Convention is triggered only when the forum's own service law creates an "occasion to transmit a document abroad for service." If the forum's law lets you complete valid service without sending anything across a border, there is no occasion to transmit abroad, and the Convention "does not apply, even though the [defendant] is abroad."

Schlunk itself proves the point. The plaintiff served a German automaker by serving its wholly owned Illinois subsidiary as the parent's involuntary agent, under Illinois law. Because that service was completed inside the United States, nothing was transmitted abroad, so the Convention had no role and the German parent's treaty argument failed. The Court was candid that this approach allows forum law to shrink the Convention's reach, but it found that result baked into Article 1's text.

The practical upshot for a China case is a two-step inquiry. Step one: does the forum's service law require transmitting documents to China to accomplish valid service? If the Chinese company can be validly served domestically—through a U.S. agent, a domestic subsidiary acting as involuntary agent, or a contractual waiver of formal service—the Convention never attaches. Step two: if the forum's law does require reaching into China to serve, then the Convention governs and must be followed to the letter. Most of the time, when a Chinese company genuinely has to be served in China, the answer is that the Convention applies, which is why understanding China's restrictive implementation is unavoidable. But the existence of step one is the reason the smartest litigants spend their energy trying to stay out of the Convention's grip rather than wrestling with it.

What Counts as a "Civil or Commercial Matter"?

The Convention reaches only "civil or commercial matters," and that deceptively simple phrase has generated real uncertainty because the treaty never defines it and the drafting history is thin. Contracting states have construed it inconsistently, and the divergence tracks the deeper split between common-law and civil-law traditions.

The United States and the United Kingdom read "civil or commercial" broadly, treating the Convention as reaching essentially everything that is not criminal—including tax, administrative, and other public-law proceedings. Two Special Commissions of the Hague Conference (in 2003 and again in 2009) endorsed exactly this generous reading, urging a "broad" and "liberal" interpretation of the phrase. Several civil-law states read it narrowly, importing their domestic distinction between private law (civil and commercial) and public law (administrative, fiscal, and the like) to push public-law matters outside the Convention. The friction is not hypothetical: in one well-known episode, a German court refused to treat a U.S. qui tam action under the False Claims Act as a "civil or commercial matter," reasoning that the suit reached beyond a private dispute, and a U.S. court deferred to that characterization (see United States ex rel. Bunk v. Birkart Globistics GmbH & Co., 2010 WL 423247 (E.D. Va. Feb. 4, 2010)). The one near-universal point of agreement is that the Convention does not reach criminal matters.

Where does China fall? China has not announced a clear position in widely reported decisions, but as a civil-law jurisdiction that guards its judicial sovereignty closely, China is generally expected to read "civil or commercial" narrowly, likely excluding public-law and administrative proceedings as the Convention operates within its borders. For the garden-variety U.S.–China commercial contract dispute this rarely bites: a breach-of-contract or commercial-tort claim between two businesses sits comfortably inside even a stingy reading. But a U.S. litigant whose claim has a regulatory, tax, or public-law flavor should not assume the Central Authority channel is open, and the 2009 Special Commission's advice is sound—where there is doubt, describe the nature of the cause of action in the request itself so the receiving authority can see that the matter qualifies. The scope question is usually easy for commercial disputes and genuinely contestable at the margins.

Serving Through China's Central Authority: The Standard Route

When the Convention applies and a Chinese defendant must be served in China, the standard—and, in practice, the only reliably effective—route is service through China's Central Authority. The mechanics are exacting, and small errors cause rejection and months of delay.

A request goes to China's Ministry of Justice (through the International Legal Cooperation Center) with a specific package: a completed Form USM-94—the standardized Request for Service Abroad of Judicial or Extrajudicial Documents annexed to the Convention—signed by an attorney; the original English-language documents to be served, with the summons bearing the issuing court's seal; and a Chinese translation of everything to be served. As an officer of the court, a U.S. attorney is a competent "forwarding authority" under the Convention and may sign and submit the USM-94 directly; you do not need the U.S. Marshal to do it (Charleston Aluminum, LLC v. Ulbrinox S. de R.L. de C.V., 2013 WL 152895 (D.S.C. Jan. 15, 2013)).

Two features of the package deserve emphasis. Translation is mandatory and is the single biggest line item. Article 5 lets a Central Authority require that served documents be in the official language, and China requires Chinese. The request form itself—being a transmittal document rather than something served on the defendant—generally need not be translated under Article 7, though translating it in full is a defensible best practice. The summons must carry the issuing court's seal, because an unsealed summons is a classic ground for rejection. China has in recent years stood up an online portal for judicial-assistance requests, allowing electronic submission rather than mailing hard copies across the Pacific—a genuine modernization that has eased the mechanics, though counsel should confirm current portal procedures before relying on them.

China charges a fee for executing service, assessed reciprocally at the equivalent of what the United States charges to process an incoming request. Because the current U.S. fee runs on the order of $95, a U.S. litigant should budget the equivalent—and China requires payment by wire transfer, not check, so getting the payment mechanism wrong stalls everything. These are trivial sums against litigation costs. The procedural particularity is the real cost: under Article 4, a Central Authority may bounce any request that does not conform to the Convention's requirements, and China does, returning non-conforming packages with an explanation and a delay measured in months.

And then there is time, the hardest part of all. The Hague Conference's own materials say most Central Authorities can serve within two months; in practice, even ordinary destinations run closer to six months to serve and return proof. China is far slower. Backlogs have routinely stretched Chinese processing to a year or more, and during some periods requests have sat pending for roughly two years before a certificate comes back. A plaintiff serving through the Central Authority must build the case around that delay—calendaring the request early, moving for extensions of the service deadline, and setting honest client expectations from day one. Note a quirk that cuts in the plaintiff's favor here: Federal Rule of Civil Procedure 4(m)'s 90-day service clock does not apply to service on a foreign corporation in a foreign country (it expressly excludes service under Rule 4(f) and 4(h)(2)), so a plaintiff diligently working the Central Authority channel is not on the domestic stopwatch—though courts still expect reasonable diligence and will not let a case sit indefinitely.

The Central Authority route has one great virtue that justifies all of this friction: it is the method China itself recognizes, so service this way is secure and does not jeopardize later enforcement in China. It is slow, cumbersome, and translation-heavy—but it is the gold standard for a reason.

Getting the Request Right the First Time

Because a non-conforming request can sit for months only to bounce, the difference between a request that lands and one that comes back is usually a handful of avoidable details. A few practical points materially improve the odds.

Get the translation right and complete. China requires a Chinese translation of the documents to be served, and quality and completeness both matter; a partial or sloppy translation invites rejection. Use translators experienced with legal documents and Chinese judicial-assistance practice, and confirm whether a translator's certification of accuracy is expected. Confirm the summons bears the issuing court's seal. Verify the defendant's exact registered legal name and address in Chinese—a mismatch between the served name and the registered entity can derail both service and, later, enforcement against that entity's assets. Mind the payment mechanics: wire, not check, and in the right amount. Use the online portal where it is available to skip transoceanic mail. Complete the USM-94 carefully, remembering the appearance date typically corresponds to the answer deadline and the "place" is the U.S. court where the case sits. And above all, build the timeline into the case from the start—treat Central Authority service as a long-lead-time item, not a last-minute chore. None of these steps is hard; overlooking any one of them can cost a year.

Why the Usual Shortcuts Do Not Work for China

U.S. litigators serving foreign defendants instinctively reach for something faster and cheaper than the Central Authority. For China, almost every familiar shortcut is foreclosed, and understanding precisely why prevents expensive mistakes.

Service by mail. The most common alternative elsewhere is Article 10(a), which provides that the Convention "shall not interfere with . . . the freedom to send judicial documents, by postal channels, directly to persons abroad"—but only where the destination state has not objected, and only where the forum's own law authorizes mail service. For decades U.S. courts split on whether "send" in Article 10(a) even covered service, until the Supreme Court settled it in Water Splash, Inc. v. Menon, 581 U.S. 271 (2017), holding that the Convention does not prohibit service by mail in countries that have not objected. Here is the catch that makes Water Splash a near-irrelevance for China: China expressly objected to Article 10 when it acceded. That objection is a binding treaty reservation. The result is categorical—mail service into China is not effective, full stop—and the same objection forecloses Article 10's other limbs (service through judicial officers and other competent persons under 10(b)–(c)). The Practical Law guidance lists China by name among the states objecting to service through local agents under Articles 10(b) and (c). A plaintiff who mails a summons to a Chinese defendant has not served it.

Diplomatic and consular channels. Article 8 lets a state effect service abroad through its own diplomatic or consular agents, but China permits service through U.S. diplomatic or consular agents only on a person who is a U.S. national—which a Chinese company is not. And independently of China's reservation, U.S. consular officers are generally prohibited from serving legal process on behalf of private litigants. The route is closed from both directions.

Local agents and publication. Service through a local agent (Articles 10(b)–(c)) is foreclosed by China's objection, as just noted. Service by publication is a domestic improvisation the Convention does not authorize at all where it applies. Chinese domestic procedure recognizes some of these methods for its own cases, but a U.S. plaintiff who borrows them to serve a Chinese defendant where the Convention governs risks both invalidating jurisdiction in the U.S. court and creating a fatal enforcement problem in China.

That leaves one genuinely live alternative—court-ordered alternative service under FRCP 4(f)(3)—which is important enough, and contested enough, to get its own section.

The Exclusivity Principle, Two-Sided

A theme runs through all of the above and deserves its own statement: where the Convention applies, most U.S. courts treat it as exclusive. The reasoning is Schlunk's—the Supremacy Clause plus Article 1's mandatory "shall apply" language—and the consequence is that a litigant cannot, where the treaty governs, simply pick a faster domestic-style method and disregard it.

What makes this especially binding in the China context is that the exclusivity runs two ways. From the U.S. side, a non-Convention method risks invalidation under Schlunk's preemption holding. From the Chinese side, China's own procedural law directs that, in matters of judicial assistance with a treaty partner, the treaty's procedures must be followed, with diplomatic channels reserved for situations where no treaty governs. The implication is that China expects service from the United States to come through the Convention and is unlikely to recognize a non-Convention method—mail, agent, publication—when a treaty relationship exists. The two legal systems, approaching from opposite directions, converge on the same conclusion: where the Convention applies, the Central Authority route is the safe one. That convergence is exactly why the strategies that avoid triggering the Convention in the first place are so valuable—and why creative alternatives, however tempting, carry enforcement risk that pure U.S.-law analysis can hide.

The Live Frontier: FRCP 4(f)(3), Email, and the Schedule A Cases

Here is where the doctrine is moving, and where a careful litigant can find real leverage that did not exist a generation ago.

Start with the rule. Federal Rule of Civil Procedure 4(f) governs service on individuals in a foreign country, and Rule 4(h)(2) extends most of it to foreign corporations and associations. Rule 4(f)(1) authorizes service "by any internationally agreed means of service that is reasonably calculated to give notice, such as those authorized by the Hague Convention." And Rule 4(f)(3) authorizes service "by other means not prohibited by international agreement, as the court orders." That last clause is the opening. The Ninth Circuit's influential decision in Rio Properties, Inc. v. Rio International Interlink, 284 F.3d 1007 (9th Cir. 2002), held that 4(f)(3) is not a method of last resort to be tried only after the Convention fails; it stands on equal footing with the other subsections, and a district court may authorize alternative service—including, in that case, service by email—on an elusive foreign defendant whenever the court finds it appropriate and not prohibited by international agreement. Rio itself blessed email service on an internet-based Costa Rican defendant that had gone to some trouble to be hard to find.

The phrase that decides whether Rio's logic reaches a Chinese defendant is "not prohibited by international agreement." And this is where courts have genuinely split.

The skeptical view. Where the Convention applies, it is mandatory and exclusive under Schlunk; China has objected to alternative postal and agent service under Article 10; and email service—so the argument runs—is functionally a transmittal abroad that China's objections were meant to foreclose. On this view, ordering email service on a defendant at a known address in China would authorize a "means prohibited by international agreement" and should not be done. Several courts have refused to authorize email service into China precisely because the Convention provides the exclusive method when the defendant's Chinese address is known.

The permissive view. A large and growing body of district-court authority—concentrated in, but not limited to, intellectual-property and counterfeiting cases—takes the opposite tack. The reasoning has two moves. First, the Convention applies only when there is "occasion to transmit a document abroad for service"; if the defendant's physical address in China is genuinely unknown (as is routinely true of anonymous online sellers operating behind marketplace storefronts and pseudonyms), then under Article 1 the Convention does not apply at all, and 4(f)(3) is unconstrained by it. Second, even where an address is known, these courts reason that email and electronic transmittal are not the "postal channels" China objected to under Article 10, so an email order is not a means "prohibited by" the Convention—it is simply a means the Convention does not address. The Supreme Court's observation in Schlunk that the Convention does not apply when service can be made without transmitting documents abroad does heavy lifting here.

This second view has become the workhorse of the so-called Schedule A cases—mass actions, common in the Northern District of Illinois and elsewhere, in which a brand owner sues dozens or hundreds of anonymous online sellers (typically operating through Chinese-based e-commerce storefronts) listed only on a sealed "Schedule A." Because the defendants are deliberately anonymous and reachable only through marketplace messaging systems and the email addresses tied to their seller accounts, courts have routinely authorized service by email and electronic publication under Rule 4(f)(3), reasoning that traditional Central Authority service is impossible against defendants whose real names and addresses no one knows, and that electronic service to the very accounts the sellers use to do business is "reasonably calculated . . . to apprise interested parties" of the action under the due-process standard of Mullane v. Central Hanover Bank & Trust Co., 339 U.S. 306, 314 (1950). For a deeper treatment of the IP-specific tactics, including the use of marketplace platforms and asset freezes, see serving legal documents to Chinese defendants in IP litigation.

How should a practitioner navigate this split? Honestly, and with the enforcement endgame in view. If the defendant is a named manufacturer at a known Shenzhen address and you intend to enforce in China, the conservative and correct course is the Central Authority, full stop—an email-service order in that posture invites both a motion to quash and a refusal to enforce in China. If the defendant is an anonymous online seller whose physical address is unknown after diligent investigation, 4(f)(3) email service is not only defensible but often the only realistic path, and the case law is increasingly with you. The decisive, fact-intensive question is almost always whether the defendant's address is truly "unknown"—which, under cases like Compass Bank v. Katz, 287 F.R.D. 392 (S.D. Tex. 2012), requires a genuine, documented diligence effort, not a shrug. A litigant who skips the diligence and asks for email service on a defendant whose address a five-minute corporate-registry search would have revealed should expect to lose the motion.

A final caution that the permissive cases sometimes underplay: even where a U.S. court authorizes and accepts email service, that does nothing to guarantee a Chinese court will later recognize a judgment obtained that way. For an anonymous seller with no Chinese assets to reach, that may not matter—the leverage is the marketplace freeze and the U.S. judgment. But for a defendant whose assets sit in China, an end-run around the Convention can win the battle and lose the war. The discovery and evidence-gathering side of these cross-border cases raises parallel headaches; our practical discovery refresher covers the federal toolkit and the pitfalls that attend reaching information held abroad.

The Risk of Serving in Violation of Foreign Law

A subtle but important question lurks beneath all of this: what happens when service is defective under foreign law but might satisfy U.S. requirements? The two systems can diverge, and the divergence carries risk.

Some U.S. courts have held that service defective under foreign law does not necessarily invalidate service for purposes of U.S. jurisdiction, reasoning that the only question for the U.S. court is whether service satisfied U.S. law and constitutional due process. Under that view, a U.S. court might find it has personal jurisdiction even though the service would not be honored abroad. That is cold comfort if the plaintiff needs to enforce its judgment in China, because a judgment obtained through service that violated Chinese law is very likely unenforceable there—Chinese courts treat proper service as a precondition to recognition, and a service defect hands them an easy ground to refuse. A plaintiff who cuts corners therefore takes a real and asymmetric risk: it may win a U.S. judgment it cannot collect against the very assets that motivated the suit. Where the defendant's assets are in China—as they so often are—valid service under the Convention is not merely a U.S. procedural box to tick; it is a precondition to ever realizing on the judgment. Litigants violate foreign service law at their peril.

This is also the reason a Central Authority's refusal to serve on sovereignty or security grounds (permitted under Article 13) should set off alarm bells. As the practice guidance notes, such a refusal may signal that the destination state will not enforce any resulting U.S. judgment either—so if the defendant's assets are all in that state, counsel should reassess whether the case is worth pursuing at all.

The Waiver Route Under FRCP 4(d)

Before reaching for the Convention's machinery, consider whether the defendant will simply waive formal service. Rule 4(d) lets a plaintiff request that a defendant waive service of a summons, and—critically—the Convention's requirements are not implicated when a defendant voluntarily waives, because a voluntary waiver means there is no occasion to transmit documents abroad "for service." The advisory committee notes to Rule 4 expressly contemplate using the waiver mechanism to avoid the expense of formal service abroad.

The waiver route benefits both sides, which is why it is worth offering even to an adverse foreign defendant. The plaintiff escapes the cost and delay of translation and Central Authority service. The defendant who waives buys time: under Rule 4(d)(3), a defendant that timely returns a waiver gets a longer period to respond—90 days from the date the request was sent, rather than the standard 21 days after service. And waiving service does not waive objections to personal jurisdiction or venue (Rule 4(d)(1), (d)(5))—the defendant gives up nothing on those fronts. There is even a cost-shifting stick: a U.S.-domiciled defendant that refuses a waiver without good cause can be ordered to pay the expenses of formal service, though that lever has little practical bite against a defendant located in China. The obvious limitation is that waiver requires cooperation, which an evasive foreign defendant may have no reason to give. But it costs almost nothing to request, and where the defendant has its own reasons to engage—an ongoing business relationship, a counterclaim it wants to bring, a reputational interest—waiver can make the entire Convention problem evaporate.

The SinoType Workaround: Contracting Around the Convention

The most consequential recent development—and the one that hands U.S. companies real leverage at the drafting stage—is the California Supreme Court's decision in Rockefeller Technology Investments (Asia) VII v. Changzhou SinoType Technology Co., Ltd., 9 Cal. 5th 125, 460 P.3d 764 (2020), cert. denied, 141 S. Ct. 374 (2020). It holds that parties can, by contract, waive formal service of process in favor of an agreed notification method—and that when they do, the Hague Service Convention does not apply.

The facts make the principle concrete. A U.S. investment firm and a Chinese company signed a memorandum of understanding that did three relevant things: it submitted the parties to the jurisdiction of California courts; it provided that the parties would give notice to each other by Federal Express or similar courier; and it stated that the parties consented to service of process in accordance with those notice provisions. A dispute arose; the U.S. party pursued arbitration as the MOU allowed; the Chinese party did not appear; and a very large arbitration award—north of $400 million—was confirmed into a judgment after the petition and summons were transmitted by courier exactly as the MOU specified. The Chinese company later moved to set the judgment aside, arguing that the failure to serve through the Hague Service Convention rendered it void.

The California Supreme Court disagreed and upheld the judgment, reasoning straight from Schlunk. Whether there is "service abroad" triggering the Convention is determined by the law of the forum, and under California law parties may agree to waive formal service and substitute an alternative method of notification. Because these parties had done precisely that—replacing formal service with an agreed courier-notification method and consenting to it—there was no occasion to transmit documents abroad "for service," so the Convention simply did not apply. The court added a pointed policy observation: letting a party accept a contract's benefits and then evade its dispute-resolution machinery by retreating behind treaty formalities would invite exactly the gamesmanship the parties bargained to avoid.

SinoType is powerful, but its limits must be understood honestly, because boilerplate will not get you there. First, it is a California decision interpreting California law (including the California Arbitration Act's tolerance for informal service arrangements). Courts in other states, and federal courts, are not bound by it and may analyze the question differently—though its reasoning rests on the Supreme Court's own Schlunk framework and is widely cited and influential. Second, it turned on clear, specific contractual language: an express submission to jurisdiction, an express agreed notification method, and an express consent to service by that method. Vague or partial language may not suffice. Third, it arose in the posture of confirming an arbitration award, a setting in which courts are especially protective of the parties' bargain. A company cannot assume that any garden-variety notice clause will replicate the result. The lesson is not "rely on SinoType"—it is "draft deliberately, with SinoType in mind, so that your clause squarely fits its holding." For the architecture of the surrounding dispute-resolution clause, see drafting dispute-resolution clauses for international contracts.

A Caveat on the Back End: Enforcement in China

Even a U.S. judgment obtained through impeccable service faces a further hurdle: enforcement in China. Chinese courts have historically been cautious about recognizing and enforcing foreign court judgments, traditionally demanding either a treaty basis or a showing of reciprocity—though China's approach to reciprocity has been evolving toward greater openness in recent years. The practical consequences are twofold. First, this reinforces why service matters so much: a defect in service gives a Chinese court the easiest possible reason to refuse recognition. Second, it is a major reason sophisticated parties so often prefer arbitration for China-related deals. An arbitral award is far more readily enforceable across borders under the New York Convention—to which both China and the United States are parties—than a foreign court judgment is. The service problem and the enforcement problem are two ends of the same rope, and both pull toward planning the dispute-resolution structure at the contracting stage rather than improvising after a dispute erupts. We compare these forum choices in detail in arbitration, mediation, and choosing a dispute-resolution forum.

What U.S. Companies Should Do Up Front

The recurring theme of this entire subject is that the time to solve the service problem is before any dispute—at the contracting stage—not after a complaint is drafted. A U.S. company doing business with a Chinese counterparty can dramatically reduce its future litigation friction with a few deliberate provisions.

First, choose the forum and governing law intentionally—ideally a U.S. court or, often better for enforceability, a U.S.-seated or neutral arbitral forum—and obtain the Chinese counterparty's express submission to that jurisdiction. Second, include a precise notice-and-service clause that specifies an agreed method of notification (for example, delivery by a named international courier to designated addresses) together with an express waiver of formal service of process and an express consent to service by the agreed method—the SinoType structure, drafted specifically rather than vaguely. Third, consider requiring the counterparty to designate an agent for service of process in the United States. Because service on a domestic agent is completed domestically, nothing is transmitted abroad, and the Convention drops out under Schlunk—the same logic that defeated the Convention argument in Schlunk itself. Fourth, for many China deals, seriously weigh arbitration over litigation, for the combined service and enforcement advantages.

Drafted carefully, these provisions convert a future dispute from a multi-year, translation-laden service ordeal into a routine matter resolved in a U.S. or arbitral forum. Drafted poorly—or omitted—they leave the U.S. company to the slow mercy of China's Central Authority. The investment of a few sentences at contracting buys enormous leverage later, which is among the best returns in all of transactional drafting.

A Worked Example

The following is a hypothetical, offered to illustrate the doctrine; it does not describe any real party or dispute.

A U.S. distributor signs a supply agreement with a Chinese manufacturer. The agreement is silent on jurisdiction and service. Two years later quality problems erupt, negotiations collapse, and the distributor sues in U.S. federal court. To serve the manufacturer—a named entity at a known factory address in Dongguan—the distributor must use the Hague Service Convention. It prepares the summons (with the court's seal) and complaint, obtains certified Chinese translations at meaningful expense, completes Form USM-94, wires the reciprocal fee, and submits the package to China's Ministry of Justice through the online portal. Then it waits. Months pass; the request sits in a backlog. The distributor monitors the docket, follows up, and eventually moves to confirm that Rule 4(m)'s clock does not run against it. Roughly a year later a certificate of service comes back, and the litigation can finally proceed. Every shortcut the distributor's lawyer considered was unavailable: mail was foreclosed by China's Article 10 objection; an email-service motion under 4(f)(3) was a non-starter because the defendant was a named company at a known address, so the Convention plainly applied and an end-run would have risked a quash motion and unenforceability in China; and there was no U.S. agent and no contractual waiver because the contract created neither. The case was delayed and made more expensive purely by the absence of planning.

Now rewind to the contract. Suppose the same supply agreement had included a clause submitting both parties to a specified U.S. court (or a U.S.-seated arbitration), providing for notice by a named international courier to designated addresses, and expressly waiving formal service with consent to service by that courier method—the SinoType structure—and perhaps requiring the manufacturer to designate a U.S. agent for service. When the dispute arose, the distributor would simply deliver the summons and complaint by the agreed courier method (or serve the U.S. agent), the Convention would not apply because nothing had to be transmitted abroad "for service," and the case would proceed without the year-long detour. Same dispute, same parties—an entirely different procedural experience, decided by a few sentences drafted years before anyone imagined a lawsuit.

Practical Takeaways

For the U.S. plaintiff already in a dispute with a Chinese defendant, triage the situation first. If the defendant is a named entity at a known address and you may need to enforce in China, the conservative path is the Central Authority: prepare the USM-94, obtain Chinese translations, submit through the Ministry of Justice (using the online portal where available), wire the reciprocal fee, and—critically—plan for delay, leaning on Rule 4(m)'s foreign-service carve-out and moving for extensions as needed. If the defendant is an anonymous online seller whose address is genuinely unknown after documented diligence, FRCP 4(f)(3) alternative service—email and electronic means—is often the only realistic route and is increasingly well supported in the Schedule A case law, provided you do the diligence the cases require. In either posture, first explore whether the defendant will waive under Rule 4(d). And resist the temptation to improvise mail, agents, or publication: China's Article 10 objection makes them ineffective, and shortcuts that fail under Chinese law imperil both U.S. jurisdiction and any later enforcement in China.

For the transactional lawyer or business drafting a contract with a Chinese counterparty, solve the problem before it exists. Specify a U.S. or arbitral forum and governing law; obtain the counterparty's submission to jurisdiction; include a precise notice-and-service clause with an express waiver of formal service and consent to an agreed notification method, following the SinoType model; consider requiring a U.S. agent for service; and weigh arbitration for its superior cross-border enforceability. A few well-drafted sentences can render the entire Hague Service Convention apparatus unnecessary.

For both audiences, the unifying lesson is this: service on a Chinese defendant is governed by a treaty China has implemented restrictively; the Convention is mandatory and exclusive where it applies, but it applies only when the forum's law requires transmitting documents abroad "for service"; and careful planning—a Rule 4(d) waiver, a SinoType-style service waiver, a U.S. agent, or an arbitral forum—can avoid its burdens entirely. Understand when the Convention applies, follow it scrupulously when it does, exploit the genuine 4(f)(3) opening when the defendant is truly elusive, and structure relationships up front so that, ideally, none of it is ever needed.

Frequently Asked Questions

Can I just FedEx the summons to a Chinese company? No. Private courier is treated as a postal channel under Article 10, and China has objected to Article 10. Courier service into China is not effective to confer jurisdiction, and a Chinese court will not recognize a judgment built on it. (Note the irony that SinoType upheld courier notification—but only because the parties had contractually waived formal service, which took the matter outside the Convention entirely.)

How long does Central Authority service in China really take? Plan on a year, and do not be shocked by longer. The Hague Conference says most authorities serve within two months, and ordinary destinations run around six; China's backlogs have stretched processing to a year or more, with some requests pending close to two years. Calendar the request early and rely on the fact that Rule 4(m)'s 90-day clock does not apply to foreign service under Rule 4(f)/(h)(2).

Does the 90-day service deadline in Rule 4(m) apply to my Chinese defendant? Not to service on the foreign corporation in China. Rule 4(m) expressly excludes service under Rule 4(f) and Rule 4(h)(2). You still must act diligently, but you are not on the domestic stopwatch.

Can a court let me serve a Chinese defendant by email? Sometimes. Under Rio Properties and Rule 4(f)(3), courts can order email service when it is "not prohibited by international agreement." Courts split on whether that includes a Chinese defendant at a known address (many say no, because the Convention applies and is exclusive). But where the defendant's address is genuinely unknown—the typical anonymous online seller in a Schedule A case—many courts hold the Convention does not apply at all and authorize email service. The pivotal question is whether you have done real diligence to find an address.

What is a "Schedule A" case and why does email service work there? It is a mass IP enforcement action against dozens or hundreds of anonymous online sellers (often Chinese-based) listed on a sealed schedule. Because the sellers are deliberately anonymous and reachable only through marketplace messaging and account emails, courts routinely authorize service by email and electronic publication under Rule 4(f)(3), reasoning that the defendants' physical addresses are unknown and that electronic service to their own business accounts satisfies due process under Mullane.

Do I have to translate everything into Chinese? The documents to be served, yes—China requires a Chinese translation under Article 5, and translation is the largest single cost of the process. The USM-94 request form itself generally need not be translated under Article 7, though some practitioners translate it as a belt-and-suspenders measure.

What happens if I serve in a way that's valid in the U.S. but invalid in China? You may keep U.S. personal jurisdiction—some courts say U.S. law is the only yardstick for that—but you will likely be unable to enforce the judgment in China, where proper service is a precondition to recognition. If the assets are in China, that can make the whole exercise pointless.

Is there any way to avoid the Convention entirely? Yes—plan ahead. Get the counterparty's contractual submission to jurisdiction plus an express waiver of formal service and consent to an agreed notification method (the SinoType structure), or require a U.S. agent for service so that service is completed domestically and nothing is transmitted abroad under Schlunk. A Rule 4(d) waiver after the fact also works if the defendant cooperates.

Related Articles

This article is provided for general informational purposes and does not constitute legal advice. Treaty membership, designated authorities, fees, and processing times change over time and vary by forum and jurisdiction; current primary sources should be consulted, and qualified counsel engaged, before any reliance in a particular matter.