The Territorial Independence of Intellectual Property Rights

I. Introduction

Among the intellectual property (IP) law questions that have most confounded U.S. courts is the extent to which an IP owner has the right to prevent imports of protected goods that were first sold outside of the country. For many years, imports of copyrighted, patented, and trademarked goods were treated differently, and there was much uncertainty about whether the place of manufacture—inside or outside the U.S.—changed the outcome of the analysis. Equally muddled has been the jurisprudence relating to the extraterritorial reach of IP laws, particularly with respect to infringement. In case after case, federal courts have struggled to reconcile ambiguous or conflicting statutory provisions and competing public policy priorities. The struggle has brought them recurrently to interpretations of IP statutes starkly at odds with basic principles of both IP law and international law. The result is an increasingly incoherent U.S. IP jurisprudence when goods embodying IP cross national borders.

A careful analysis of the jurisprudence addressing IP disputes involving international activity reveals the root cause of the confusion. That cause is a misunderstanding of basic principles of international law relating to the nature of territorial jurisdiction. With some exceptions, these principles give states a predominant, though nonexclusive, interest in legislating for persons, property, and the status of both within their own territories. There are, of course, other bases for state jurisdiction, and these can be relevant to IP disputes. But the primacy of the state’s territorial jurisdiction in international law has important implications for the geographic scope of each state’s IP laws and how its courts resolve disputes involving foreign trade and alleged cross-border infringement.

Of course, the jurisdictional doctrines of public international law are not necessarily decisive in a dualistic country like the U.S. In the U.S. constitutional system, Congress may legislate contrary to international law, and such legislation would override prior, conflicting international law.1 Congress could adopt IP statutes that extend the proscriptions of U.S. IP laws to foreign conduct, property, or persons. Congress has, in fact, sometimes explicitly cast the net of patent, copyright, and trademark law beyond U.S. borders.2 However, when Congress has not clearly sought to apply U.S. IP statutes extraterritorially, courts are sometimes confronted with a conflict between ambiguous statutory language and U.S. obligations under public international law.

In theory, there are established jurisprudential methods to reconcile the two, such as The Charming Betsy doctrine3 and the EEOC v. Aramco presumption against extraterritorial interpretations of U.S. statutes.4 But courts have applied these doctrines inconsistently, and in no U.S. Supreme Court case has a core aspect of IP law resulting from these doctrines been recognized and properly applied. The result has been IP doctrines lacking coherence both under international law and U.S. IP legislation.

The purpose of this article is to reassert the primacy of each state’s territorial jurisdiction as a fundamental basis for resolving international IP disputes. It identifies the principle that I have elsewhere termed the “territorial independence of IP laws” as specially relevant to the problems of parallel imports and cross-border IP infringement,5 and it explains how the proper application of the territorial independence principle resolves IP disputes in a manner that avoids running afoul of international law, maintains the integrity of basic U.S. principles of statutory construction, and remains consistent with the various federal statutes protecting IP rights. The territorial independence principle arises from the basic doctrine of international law that states have primary prescriptive jurisdiction with regard to their own territories, and this has important implications for how IP laws should be interpreted in multinational IP disputes.

Part II.A begins by analyzing the conflict between how U.S. courts approach questions of cross-border IP infringement and the general jurisprudence relating to statutory construction. It starts with the Supreme Court’s strong presumption against extraterritorial interpretation of statutes and explains the presumption’s incompatibility with a leading case of alleged international IP infringement, Steele v. Bulova.6 In a recent case, Abitron Austria v. Hetronic International,7 the Court reaffirmed the relevance of the presumption even while failing to overturn Steele, thereby keeping alive the conflicting jurisprudence it had begun in that case. Part II.B examines the history of parallel import jurisprudence and exposes the tension between the presumption against extraterritoriality and the Court’s application of an international IP exhaustion doctrine. This portion of the discussion is necessarily somewhat complex, because the Court developed distinct copyright, patent, and trademark jurisprudence on the subject and only later undertook a partial reconciliation of these disparate doctrines.

Part III explains the principle of the territorial independence of IP rights and applies it to resolve the problems within U.S. IP jurisprudence identified in Parts II.A and II.B. Part III.A explains how the rules of international law governing state jurisdiction apply to IP legislation. Part III.B analyzes how these principles are integrated into federal IP statutes and demonstrates how the principles accord with other doctrines of jurisprudence, including international comity and the act of state doctrine. Part III.C concludes by explaining how regional IP protection systems, such as those used in Europe, create exceptions to the territorial independence principle while remaining consistent with its general application to states lacking such regional arrangements.

Part IV concludes the analysis by applying the territorial independence principle to resolve the major problems of extraterritorial IP infringement and parallel importation. It also makes clear that the doctrine solves not only a set of important jurisprudential conflicts but also certain public policy concerns—while acknowledging that it is no panacea. International IP conflicts sometimes raise public policy problems that territorial independence cannot solve and may even exacerbate. The Supreme Court has invoked such policy problems to justify judicial decisions at odds with the territorial independence principle. However, this Part argues that courts should not disregard fundamental principles of international law and statutory interpretation to solve these policy problems. Rather, the policy problems are best solved by the political branches through international treaties and appropriate legislation. Part IV will conclude with a general description of the optimal solution to these problems.

II. The International IP Problem

A. Confusion About the Extraterritorial Reach of IP Laws

1. The presumption against extraterritorial legislation and the Charming Betsy Doctrine

Since its 1991 decision in EEOC v. Aramco, the U.S. Supreme Court has increasingly applied a presumption against extraterritoriality as a canon of statutory interpretation.8 In Aramco, a U.S. citizen employed by a Delaware corporation sought to enforce Title VII of the Civil Rights Act of 1964 against his employer, claiming discrimination on the basis of race, religion, and national origin. Unlike most employment discrimination cases, the plaintiff’s duties were performed entirely in an office outside the U.S., specifically in Saudi Arabia. The Court’s analysis began with a presumption that the “‘legislation of Congress, unless a contrary intent appears, is meant to apply only within the territorial jurisdiction of the United States.’”9 Congress is assumed to legislate “against the backdrop of the presumption against extraterritoriality” and, therefore, overcoming the presumption requires a clear expression of congressional intent.10

Title VII’s jurisdictional language was11 very broad. It applied to any industry “affecting commerce” and defined commerce as “trade, traffic, commerce, transportation, transmission, or communication among the several States; or between a State and any place outside thereof . . . .”12 Because the employer was engaged in commerce between the U.S. and Saudi Arabia, Title VII by its plain language forbade the defendant from discriminating against its U.S. employees regardless of the physical location of their workplace. Yet, the Court found the statutory language ambiguous, requiring deduction “by inference from boilerplate language which can be found in any number of congressional Acts, none of which have ever been held to apply overseas.”13 Insisting that Title VII lacked a sufficiently explicit reference to foreign employment of U.S. citizens, the Court held that the plaintiff had not overcome the strong presumption against extraterritoriality.14

Whatever the deficiencies of the Court’s interpretation of the Civil Rights Act in Aramco, the general presumption against extraterritoriality advances two significant public policies. The first is the conservation of governmental resources. If federal statutes are routinely interpreted to apply beyond U.S. borders, then governmental agencies and courts must regulate and adjudicate conduct that may have no significant impact on the interests of the U.S. government or U.S. citizens and residents. Given the finite resources of any country, it makes little sense to divert enforcement resources for foreign conduct that does not significantly affect U.S. markets, territory, or persons. Although the Court did not invoke this rationale when announcing the presumption in EEOC v. Aramco, it nonetheless gives the doctrine some support.

The justification upon which the Court explicitly relied was that the presumption “serves to protect against unintended clashes between our laws and those of other nations which could result in international discord.”15 An interpretation of a statute that regulates conduct in foreign states could offend foreign sovereigns if the policy advanced by the legislation conflicts with their interests in regulating conduct within their own territory. This could have been merely a prudential consideration, but it finds support in another canon of statutory construction.

It is a longstanding principle of U.S. jurisprudence that “an act of Congress ought never to be construed to violate the law of nations if any other possible construction remains . . . .”16 The presumption against interpreting federal legislation to violate international law, commonly called the Charming Betsy doctrine, has been applied many times since and has become an established canon of statutory construction.17 The presumption serves important public policy interests, the most prominent of which is that, like the presumption against extraterritoriality, it is intended to diminish the chances of inadvertently putting the U.S. in conflict with its obligations to other states under both treaties and customary international law binding on the U.S. Through the Charming Betsy canon, the Supreme Court has sought to avoid embarrassing the political branches with jurisdictionally aggressive interpretations of legislation that Congress did not clearly intend.18 The doctrine denotes that, to the extent that U.S. IP statutes can reasonably be interpreted not to intrude on IP legislation in foreign states and not to presume to legislate in foreign territory, such restrictive interpretations should prevail.

Yet, if the primary justification for the presumption against extraterritoriality really is to avoid breaches of international law likely to create conflict with other states, the canon proposed by the Court is not the optimal solution, as John Knox has pointed out in the American Journal of International Law.19 Although states have primacy in their jurisdiction over persons, property, and conduct within their respective territories, public international law is fairly permissive in regulating state exercises of prescription jurisdiction.20 International law allows states to regulate the foreign conduct of its nationals and permanent residents, foreign acts having a substantial effect within the state’s territory, and several other classes of conduct outside that territory.21 To the extent the presumption against extraterritoriality is intended to foreclose unnecessary offense to other states, it sweeps too broadly. As Professor Knox has observed, it should be possible to tailor the relevant canon of statutory construction to harmonize with these limits by shifting the relevant paradigm from extraterritoriality to international law’s more nuanced conception of extrajurisdictionality.22 The relevant inquiry should be whether the statutory construction comports with international law, not merely whether the construction would apply the state’s law beyond the state’s borders. This approach, Knox argues, comports with the original basis for conservatism in construing the reach of U.S. statutes before EEOC v. Aramco.23 Knox also makes a strong case that a presumption against extrajurisdictionality better accords with U.S. interests as Congress tends to conceive them in its domestic legislation.24

Regardless of its wisdom or consistency with precedent, EEOC v. Aramco created a new rule strongly unfavorable to the extraterritorial application of U.S. legislation without a clear manifestation of congressional intent. In subsequent cases, the Court has repeatedly applied this presumption to a wide variety of statutes having jurisdictional language deemed insufficiently lucid. These include cases relating to securities regulation,25 the Alien Tort Statute,26 federal prohibitions on money laundering and fraud,27 and most recently, as will be discussed, the Lanham Act of 1946.28 Yet, one of its cases stands in the way of the application of that doctrine to federal trademark and unfair competition law.

2. Steele v. Bulova and the global U.S. trademark

Steele v. Bulova29  is a mysterious case; the Supreme Court’s ratio decidendi is laconic and poorly articulated, leaving much of the decision’s significance to speculation. It is speculation within a sphere circumscribed by the facts of the case, however, and that at least offers guidance about the range of possibilities behind the Court’s reasoning and intentions. Before Steele, the Court clearly signaled that U.S. trademark laws are confined to U.S. territory. It had observed in 1918, for example, that a foreign trademark “did not entitle its owner to claim exclusive trade-mark rights in the U.S. as against one who in good faith had adopted a like trade-mark here prior to the entry of the foreigner into this market.”30 The natural corollary of this observation is that U.S. trademark owners can claim no rights in foreign markets.

Steele involved an infringement complaint by a U.S. trademark owner, Bulova Watch Co., against a U.S. citizen relating to activities almost entirely outside the U.S. Bulova had registered its trademark for watches in the U.S., but it had no Mexican trademark registration.31 Petitioner Steele registered the BULOVA mark in Mexico and began manufacturing and selling watches there bearing his mark. Bulova sued Steele, complaining of trademark infringement under the Lanham Act of 1946, which provides for a civil action by a trademark owner against “any person who shall, in commerce, infringe [a U.S.] trade-mark.”32 Commerce, in turn, is defined generically as “all commerce which may lawfully be regulated by Congress.”33

Steele had never manufactured or sold watches in the U.S. nor imported them into the U.S. However, the Court observed that Steele’s conduct had some indirect connection to the U.S. in three ways. First, Steele was a U.S. citizen and resident, although all relevant commercial transactions occurred in Mexico. Second, Steele had purchased some watch components for his Mexican manufacturing operation from suppliers in the U.S. Third, some persons who had purchased Steele’s watches in Mexico brought them to the U.S. and then complained about their quality to one of Bulova’s sales representatives.34 There was no other U.S. connection. Of course, none of these three facts by itself could constitute trademark infringement. Nonetheless, the Court majority concluded that Steele’s activities, “when viewed as a whole,” fell within the scope of the Lanham Act’s prohibition on infringement of a U.S. trademark.35

The route by which the Court arrived at this conclusion is cryptic. It first observed that international law does not prohibit Congress to legislate with respect to the activities of its citizens abroad.36 While undoubtedly true, the point is only relevant if Congress intended the Lanham Act to have global effect with regard to U.S. citizens and residents. This is plainly not the case. The Lanham Act does not prohibit U.S. citizens or residents from reproducing in foreign states trademarks registered in the U.S. Never before had the Act been authoritatively interpreted to such effect. It is, in fact, common for the same trademark to be registered in the U.S. as well as multiple other countries by different persons, and, even today, the Supreme Court has never held that the fact that a U.S. person owns and commercializes a foreign trademark identical to a U.S. trademark (owned by a different U.S. person) infringes the U.S. trademark. Indeed, the Court has never intimated that such a fact would even be relevant to a question of U.S. trademark infringement.

The Court also emphasized that Steele’s “operations and their effects were not confined within the territorial limits of a foreign nation,” because he purchased components from U.S. suppliers and “the spurious ‘Bulovas’ filtered through the Mexican border into this country.”37 Yet, the source of components used in foreign manufacture of goods registered under a foreign trademark is even less relevant than Steele’s nationality. Had Steele sourced his components in Switzerland, would the Court have then been satisfied that the Lanham Act could not reach his conduct? The sole rationale that could justify such an opinion is that the Court believed it was Congress’s policy to discourage the purchase of U.S. exports in foreign manufacturing operations. But, on the contrary, Congress has consistently sought, both at that time and since, to maximize U.S. manufacturing exports as a general rule.38

The third factor relied upon by the Court, the importation by purchasers into the U.S., is equally immaterial. The Lanham Act defines trademark infringement as requiring the relevant person’s “use in commerce.”39 A person does not engage in commerce vicariously except by act of his or her authorized agents, and the plaintiff had alleged no facts suggesting that the complaining purchasers were anything but independent, unaffiliated third parties. If such liability were possible, foreign sellers would infringe a U.S. trademark every time outside the U.S. they sold a good bearing a foreign trademark confusingly similar to the U.S. trademark, and their purchaser imported those goods into the U.S. without the authorization of the foreign seller. Again, the Supreme Court has never held that the Lanham Act imposes liability on the foreign seller for the decisions of an unaffiliated buyer by some newfangled extension of the law of agency. If anything, the purchasers of Steele’s watches in Mexico infringed Bulova’s U.S. trademark by bringing them into the U.S. Even so, such imports do not infringe the trademark under an exemption in the Tariff Act of 1930.40 Finding infringement against the foreign seller under such circumstances is thus doubly nonsensical.

Given the irrelevance of each of the grounds cited by the Court to sustain its decision, the question becomes whether Congress intended the Lanham Act to reach the foreign conduct based on the three facts “when viewed as a whole.”41 In other words, do the indirect connections to the U.S. cumulatively create a cause of action for trademark infringement under the Lanham Act when each individual factor cannot?

If so, the reasoning justifying the Court’s conclusion is difficult to discern. At its broadest, the holding would entitle a U.S. trademark owner to complain of infringement should any person anywhere in the world affix a mark confusingly similar to the U.S. trademark in a foreign state, with no connection to the U.S. whatsoever, because it is within Congress’s constitutional power (i.e., it is “lawful” for Congress within the U.S.) to adopt such an extrajurisdictional law.

At its narrowest, the holding would be formulated more or less as follows:

Any U.S. citizen who manufactures and sells outside the United States goods bearing a foreign trademark confusingly similar to a U.S. trademark, using at least some components acquired from U.S. suppliers, has used the U.S. trademark in commerce for purposes of the Lanham Act if at least some of the purchasers of such foreign products bring them into the United States and confuse the foreign mark with the U.S. mark.

The narrow formulation finds no support in any language of the Lanham Act except the generic reference to all commerce that Congress may lawfully regulate, but at least it is limited to the facts of the case. The problem is that the majority opinion gives no guidance on what U.S. connections other than those in the specific case could justify the extension of the Lanham Act’s prohibitions to a foreign use of a U.S.-registered trademark.

Most importantly for present purposes, the Court majority did not apply any presumption against extraterritorial application in interpreting the language of the Lanham Act. Justice Reed, in dissent, specifically drew the majority’s attention to the presumption against extraterritorial jurisdiction in the Court’s prior decisions.42 That the majority declined even to mention, much less apply, such a presumption puts the Steele decision squarely at odds with the holding of EEOC v. Aramco. It also tips the majority’s hand. By referring to Steele’s Bulova watches as “spurious,”43 the majority reveals that, regardless of the language of the Lanham Act, it had determined to find in favor of Bulova. In fact, Steele’s watches were not fakes or counterfeits; they were manufactured in Mexico under a legitimate Mexican trademark. The Court could only have believed them 
“spurious” if laboring under a misconception. Either the Court erroneously believed Mexico could not grant registration of a trademark identical to a U.S. trademark, or the Court believed that the U.S. trademark and the Mexican trademark were one and the same thing, and only the U.S. registration could be legitimate. In either case, the Court was denying Mexican sovereignty over its own intellectual property regime—precisely the kind of offense to a foreign sovereign that the presumption against extraterritoriality was intended to prevent.

On that point, it is interesting that, in seeking to apply the Civil Rights Act of 1964 to the foreign conduct of a U.S. employer relating to a U.S. employee in EEOC v. Aramco, the Equal Employment Opportunity Commission invoked Steele v. Bulova to support the argument that broad jurisdictional language should be construed to apply extraterritorially.44 The Court’s attempt to distinguish Steele from EEOC v. Aramco reveals much about the majority’s comfort with self-contradiction:

The EEOC’s attempt to analogize these cases to Steele is unpersuasive. The Lanham Act by its terms applies to “all commerce which may lawfully be regulated by Congress.” The Constitution gives Congress the power “[t]o regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes.” U.S. Const., Art. I, § 8, cl. 3. Since the Act expressly stated that it applied to the extent of Congress’ power over commerce, the Court in Steeleconcluded that Congress intended that the statute apply abroad. By contrast, Title VII’s more limited, boilerplate “commerce” language does not support such an expansive construction of congressional intent. Moreover, unlike the language in the Lanham Act, Title VII’s definition of “commerce” was derived expressly from the LMRDA, a statute that this Court had held, prior to the enactment of Title VII, did not apply abroad.45

This attempt to distinguish Steele from EEOC v. Aramco based on statutory language reflects little credit on the Court majority’s powers of reasoning or self-consistency. It is indeed difficult to imagine more “boilerplate” language than the Lanham Act’s reference to “all commerce which may lawfully be regulated by Congress.” Unless the Court imagines that, in drafting the Lanham Act, Congress foresaw every possible situation on a global scale, in which unfair competition or trademark infringement might occur and intended the Act to reach all such situations, it seems exceedingly likely that Congress’s reference to “all commerce” merely opened the door to more specific provisions of the Act extending extraterritorially if the language of those provisions indicated congressional intent to do so. To turn such broad language into an authorization to stretch every provision of the Lanham Act into a global regulatory regime, to the maximum extent of Congress’s authority, falls little short of a lampoon.46

Still more troubling, the Court’s interpretation of Steele leads to results that are plainly contrary to congressional intent. As noted, the Constitution puts no limit on Congress’s ability to regulate commerce beyond U.S. territory; it merely empowers Congress to “regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes.”47 If the Court were to take its own reasoning seriously, the Lanham Act as written would regulate trademark disputes in every foreign country in the world involving at least one U.S. citizen, or indeed exclusively involving foreign citizens and having some trivial effect on the U.S. After all, when Congress has appeared to intend an aggressive assertion of extraterritorial jurisdiction, courts have upheld such legislation despite a very tenuous relationship to the U.S. and U.S. citizens.48 The Court itself in EEOC v. Aramco did not doubt that “Congress has the authority to enforce its laws beyond the territorial boundaries of the U.S.”49

By not taking the opportunity in EEOC v. Aramco to overturn Steele v. Bulova, despite applying a canon of construction directly contradictory to the Steele decision, the Court left a naval mine in the wake of its presumption against extraterritoriality. Predictably, it has damaged the integrity of circuit court jurisprudence. In Trader Joe’s v. Hallatt, the Ninth Circuit stretched Steele to and beyond its tensile strength by concluding that a foreign national selling a genuine product in a foreign state could infringe a U.S. trademark if it potentially had “some effect” on U.S. consumers who heard negative reports of the foreign-sold products and made negative inferences about the U.S. products.50 And the Court itself navigated directly into the path of its own mine in a 2022 case.

3. Abitron v. Hetronic: the resurgent presumption against extraterritoriality

The case is Abitron Austria GmbH v. Hetronic International,51 and it involved the same general issue decided in Steele v. Bulova—the extraterritorial reach of the Lanham Act’s prohibition on trademark infringement. Hetronic International is a U.S. company that produces radio remote controls for heavy-duty construction equipment, to which its trademarks are affixed.52 During the relevant period, Hetronic had granted a license to Abitron, a foreign corporation, to distribute Hetronic’s products primarily in Europe, with some sales to the U.S.53 Abitron began manufacturing and selling products identical to Hetronic’s, and bearing Hetronic’s trademarks without authorization from the trademark owner, provoking the lawsuit on certioriari.54 Hetronic sought damages in a U.S. district court based on Abitron’s allegedly infringing acts everywhere in the world.55 Abitron argued that the Lanham Act did not reach a foreign company’s foreign sales of products subject to a U.S. trademark.56

The Supreme Court began its opinion by invoking the presumption against extraterritoriality, seeking an “unmistakable instruction” from Congress to apply the law beyond U.S. borders.57 Beginning in 2016, the Court had added a second explicit step to its Aramco analysis: if the statute was intended to apply extraterritorially, the Court would ask whether the conduct at issue in the specific case fell within the “‘focus’ of congressional concern”58 —meaning that any U.S. conduct must implicate the public policies that Congress sought to vindicate in the statute. This second step, the Court stated, “is designed to apply the presumption against extraterritoriality to claims that involve both domestic and foreign activity, separating the activity that matters from the activity that does not.”59 Mysteriously, the Court further asserted that only domestic conduct, and not foreign conduct having a domestic effect, was relevant,60 without reference to the possibility that Congress may have intended the statute to apply to foreign conduct having a substantial U.S. effect.

The Court in Abitron then concluded that the Lanham Act was not one of the “rare statute[s]” that provide a clear indication of extraterritorial intent.61 The Court noted that the relevant provisions “simply prohibit the use” of the trademark “in commerce” when the use was likely to cause confusion.62 Relying on the reasoning by which the Court had applied the Lanham Act to the defendant’s conduct in Steele,63 Hetronic had argued that the term “commerce”—defined as all commerce Congress can lawfully regulate—encompassed the foreign sales.64 The Court disagreed, even denying that references to “foreign” commerce could justify applying a statute extraterritorially.65

At this point, an observer may be forgiven for assuming that the Court, having clearly rejected the reasoning upon which the Steele decision rested, intended to overturn that decision. Yet, the Court distinguished Steele, claiming that Steele’s real ratio decidendi was that the defendant’s conduct involved “essential steps” in the U.S. and “was likely to and did cause consumer confusion in the United States.”66 The Court therefore left Steele intact.

Steele’s Mexican watches were considerably less spurious than the Supreme Court’s reasoning on this point. Which domestic steps in the Steele case, one wonders, were essential to U.S. infringement? Certainly not buying unbranded watch components from U.S. suppliers; they could have been acquired from anywhere to the same effect. As noted, nor does any language in the Lanham Act suggest that the nationality of an alleged infringer is relevant, much less essential, to the reach of the Act. Also, nationality is not a “step.”

As for whether the accused conduct was likely to cause, and actually did cause, consumer confusion in the U.S., the Court provided no basis for that conclusion in either Steele or Abitron. The Court was unable to cite a single fact to justify a prediction that U.S. consumers in general would be likely confused by the sale of watches in Mexico.67 No evidence showed that Steele was marketing primarily to U.S. consumers in Mexico, nor that he encouraged them to bring their Mexican watches into the U.S. The relevant population for consumer confusion purposes is the average, reasonably prudent U.S. consumer,68 not the average consumer in a foreign state, nor yet the average U.S. consumer buying a suspiciously inexpensive luxury watch in a foreign state. Had Steele turned on an interpretation of the relevant facts as involving a deliberate attempt by the defendant to induce U.S. consumers to travel to Mexico to buy his watches, misrepresenting them as made by or with the authorization of the U.S. trademark owner, a case for fraud directed at U.S. consumers could be made out, and a Lanham Act claim sustained on those narrow grounds. Because the Court did not do so in Steele, it created a fundamentally flawed precedent, and, because the Court tried to weakly rationalize instead of reversing that precedent in Abitron,69 it left the problematical jurisprudence in force.

4. Extraterritorial damages for IP infringement

Steele is not the only case in which the Court has implicitly renounced its presumption against extraterritoriality. It has also done so in some cases regarding the assessment of damages for infringement involving foreign sales. In WesternGeco LLC v. ION Geophysical Corp.,70 the Court confronted the question of whether infringement of a U.S. patent by a foreign person through an extraterritorial anticircumvention provision entitles the patent owner to recover lost profits for sales outside the U.S. Though it may seem like a highly technical case, the Court’s approach displayed a conception of IP jurisdiction very similar to that in Steele.

At issue in the case was section 271(f)(2) of the Patent Act, one of the few provisions of the Act by which Congress explicitly extended the concept of U.S. patent infringement beyond U.S. borders. Section 271(a) of the Act characterizes the unauthorized making, using, selling, or offering for sale in the U.S. of a patented invention as infringement, as well as the unauthorized importation of the invention into the U.S.71 It is, therefore, expressly territorial. Ordinarily, the patent owner’s damages include a reasonable royalty or possibly lost profits, with treble damages available for willful infringement.72 The Act does not explicitly limit the damages to royalties that would be paid for a license in the U.S., or to profits lost for sales within and from the U.S.., but, consistent with the fact that infringement under Section 271(a) can arise only from U.S. activity, the Federal Circuit had held that damages arising from purely foreign activity—such as sales from foreign territory to foreign territory (as opposed to the U.S.)—cannot be recovered.73

In contrast, WesternGeco involved a provision specifically targeting extraterritorial conduct. Section 271(f)(2) provides in relevant part:

Whoever without authority supplies or causes to be supplied in or from the United States any component of a patented invention that is especially made or especially adapted for use in the invention and not a staple article or commodity of commerce suitable for substantial noninfringing use, where such component is uncombined in whole or in part, knowing that such component is so made or adapted and intending that such component will be combined outside of the United States in a manner that would infringe the patent if such combination occurred within the United States, shall be liable as an infringer.74

Section 271(f)(2) does not make the foreign importer liable for infringement, but it does reach U.S. conduct intended to circumvent Section 271(a).

Section 271(a) prohibits the unauthorized sale from the U.S. of the patented invention to any purchaser within or outside the U.S. But suppose a patented invention is composed of unpatented but specialized parts. A would-be exporter could avoid infringing the U.S. patent by manufacturing these unpatented parts and selling them to a foreign purchaser for assembly outside the U.S. Section 271(f)(2) eliminates that possibility for U.S. sellers. The question before the Court in WesternGeco was whether the damages owed by the U.S. seller include lost profits from foreign sales displaced by the defendant’s exports.

Applying the presumption against extraterritoriality, the Federal Circuit had held that Section 271(f) could give rise to damages only for sales lost in the U.S.75 The Supreme Court majority disagreed. It began its analysis by strangely skipping over the RJR Nabisco76  opinion’s first step, which asks whether Congress had intended the Patent Act to afford extraterritorial damages for infringement of Section 271(f). Instead, it focused on the second step, assessing the statute’s “focus” and asking whether the conduct relevant to that focus occurred within the U.S., on the odd ground that the Court wished to avoid ruling on the claim that a general damages provision like Section 284 of the Act should ever be interpreted as extraterritorial.77 The Court was thereby able to ignore its own presumption against extraterritoriality and conclude that the “focus” of the statute was on how the defendant’s behavior in the U.S. could deprive the U.S. patent owner of profits, regardless of where in the world the profits arose.

Section 271(f)(2) prevents persons in the U.S. from facilitating foreign acts that would have infringed a U.S. patent if performed domestically. It does not explicitly confine its prohibition to foreign assembly resulting in importations of the foreign-assembled goods into the U.S. From this, the WesternGeco majority apparently concluded that the “focus” of Section 271(f)(2) is the foreign assembly, and thus the patent owner could only be made whole from such infringement if compensated for the worldwide sales lost to the foreign-assembled products.78

In a dissent in which Justice Breyer joined, Justice Gorsuch pointed out that the patent owner was seeking, and the majority granted, damages for profits lost in a market in which the U.S. patent was not in fact infringed. The dissenting opinion argued:

In measuring its damages, WesternGeco assumes it could have charged monopoly rents abroad premised on a U.S. patent that has no legal force there. Permitting damages of this sort would effectively allow U.S. patent owners to use American courts to extend their monopolies to foreign markets. That, in turn, would invite other countries to use their own patent laws and courts to assert control over our economy.79

In the dissent’s view, the statutory language of Sections 271(f)(2) and 284, interpreted in light of the presumption against extraterritoriality, would only grant the patent owner damages to the extent the foreign-assembled product was made, used, sold or offered for sale in the U.S., or imported into, the U.S. All Section 271(f)(2) was designed to accomplish was to provide U.S. patent owners with a remedy against U.S. persons who facilitate foreign-based infringement of a U.S. patent. He thus concluded that “none of this changes the bedrock rule that foreign uses of an invention (even an invention made in this country) do not infringe a U.S. patent.”80

Because Section 271(f)(2) applies only to persons within the U.S., its only real effect is to deter U.S. exporters from doing something that any foreign seller can do. Persons outside the U.S. remain free to sell components to another foreign person intent on assembling those parts—outside the U.S.—into a U.S.-patented article of manufacture. It therefore seems improbable that Congress intended to impose a worldwide burden of compensation on a U.S. violator of that provision. In practice, Section 271(f)(2) merely deprives a U.S. exporter of foreign sales in favor of foreign exporters without reducing the foreign manufacturing competition faced by the U.S. patent owner, except in the presumably rare case that a component can only be supplied from the U.S. But, more important, as Justice Gorsuch pointed out, the majority’s interpretation creates a logical anomaly: “It would allow greater recovery when a defendant exports a component of an invention in violation of § 271(f)(2) than when a defendant exports the entire invention in violation of § 271(a).”81

As in Steele, the Supreme Court majority in WesternGeco failed to apply the presumption against extraterritoriality properly to IP legislation. The dissenting interpretation of the Patent Act is based on the firm understanding of the respective jurisdictional scopes of U.S. and foreign patent legislation that the Court displayed in Microsoft Corp. v. AT&T Corp.82 WesternGeco thus provides another example of the Supreme Court applying its presumption against extraterritoriality in an inconsistent and arbitrary manner based primarily on domestic considerations.

B. Confusion About Parallel Imports

The Court has thus developed a fairly definite, indeed simplistic, general doctrine on cross-border IP infringement, although it has not applied that doctrine with consistent reasoning. On the subject of cross-border infringement through parallel imports, the Court’s doctrine has not even achieved basic coherence. Parallel imports, also known as gray market goods, are products bearing or embodying U.S.-protected IP rights that come into the U.S. from foreign states where their production or sale was authorized, but are imported without the express authorization of the owner of the U.S. IP rights. Any kind of IP—trademarks, copyrights, and patents—can be imported into the U.S. without such authorization, but each type of IP has a different statutory background relating to the territorial reach of IP rights. In consequence, courts have generally analyzed parallel imports under each form of IP law separately. The judicial practice will accordingly be followed here, and each form of IP will be discussed separately.

Regardless of the type of IP, gray market import disputes tend to arise from attempts by IP owners to offer their goods at different prices in different countries, a practice known as geographic market segmentation.83 Price distinctions of this kind reflect an attempt to maximize profits in global sales by tailoring prices to conditions of each country’s market. IP rights facilitate such market segmentation. In the absence of IP rights, a U.S. manufacturer who sells widgets in both U.S. and Ruritanian markets has no automatic legal right to prevent the reimportation into the U.S. of the widgets sold in Ruritania, or vice versa, through the practice of arbitrage. The U.S. manufacturer therefore has a limited ability to sell the widgets at different price points in the two countries, because a purchaser in the country with the lower price can resell the goods in the country with the higher price at a profit. For example, if Widgico sells its widgets in the U.S. for $300 each and in Ruritania for $200 each, nothing necessarily stops a Ruritanian purchaser from reselling the widgets in the U.S. for $275 each, pocketing the $75 profit (minus any international transportation costs and U.S. customs duties) per widget and undermining Widgico’s $300 U.S. price point. 

In theory, IP rights solve this problem for the U.S. manufacturer. If Widgico holds both a U.S. and a Ruritanian IP right to its widgets, it could segment the two markets by limiting the purchaser’s license to one country or the other. Thus, for example, Widgico could sell its (e.g., patented) widgets in the U.S. for $300 per widget, subject to a grant of a U.S. patent license only, and it could sell its widgets in Ruritania for $200 per widget, subject to a grant of a Ruritanian patent license only. If a Ruritanian purchaser tried to export its widgets to the U.S. to resell at a profit, it would find itself blocked by the absence of any right to import or resell the widgets in the U.S. It has a license to resell under the Ruritanian patent; it has acquired no right to import or sell under the U.S. patent.

IP rights thus could enable Widgico to sell its goods at the profit-maximizing point for both the U.S. and Ruritania. The result is advantageous for sellers, because they can capture more consumer surplus than would be possible using a single price for markets with potentially very different characteristics.84 Under some conditions, geographic market segmentation may benefit consumers as well, because the lower price point in some countries, such as those with a lower per capita gross domestic product, gives more consumers access to the product than would be possible at a higher price point, albeit possibly at a greater cost to consumers in wealthier countries. Whether this increases global consumer welfare depends on such factors as market conditions and the cost structure of the supplier, but the point is that price distinctions resulting from geographic market segmentation do not necessarily harm consumers in the aggregate.85

1. The First Sale (Exhaustion) Doctrine and parallel imports

Complicating this analysis is the fact that many countries have adopted a “first sale” or “exhaustion” doctrine for at least some types of IP.86 The U.S. is one such country. Under this doctrine, an IP owner is considered to reap the full advantage of the IP upon transferring ownership of the product the first time. They have no right to limit the purchaser’s ability to resell or otherwise dispose of the product once title has passed. Thus, the IP owner’s rights in a particular product are “exhausted” after the first sale, and restraints upon further alienation are forbidden.

In U.S. trademark law and patent law, the exhaustion doctrine is a creature of the common law. The Supreme Court has long held that, having transferred title to a patented product, the patent owner can no longer control the disposition of the specific product transferred:

[I]n the essential nature of things, when the patentee, or the person having his rights, sells a machine or instrument whose sole value is in its use, he receives the consideration for its use and he parts with the right to restrict that use. The article, in the language of the court, passes without the limit of the monopoly. That is to say, the patentee or his assignee having in the act of sale received all the royalty or consideration which he claims for the use of his invention in that particular machine or instrument, it is open to the use of the purchaser without further restriction on account of the monopoly of the patentees.87

In the trademark context, the doctrine has a slight wrinkle: it entitles the lawful owner of trademarked goods to transfer those goods to a third party, but only if the goods are not materially different in a manner that would create consumer confusion about the origin or qualities of the goods.88

In contrast, the exhaustion doctrine is codified in the 1976 Copyright Act. Its general provision reads in relevant part as follows:

Notwithstanding the provisions of section 106(3) [granting a copyright owner the exclusive right to distribute copies of the copyrighted work by sale, transfer, rental or lending], the owner of a particular copy or phonorecord lawfully made under this title, or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord.89

In the fields of patent and trademark, because the doctrine is judicially created, the question of whether rights are exhausted by sales or other transfers across international boundaries has been decided by courts based on their interpretations of the policies that motivated the doctrines. These decisions will be explored in parts II.B.3 and II.B.4, respectively. In the copyright context, however, the statutory basis obligates courts to focus its analysis on the language of the Copyright Act. That language has caused a great deal of judicial confusion due to the absence of explicit elucidation of what copies are “lawfully made under” Title 17 of the U.S. Code. This subject will be further explored in Part II.B.2.

The relevant question for courts in the context of parallel imports is whether a U.S. IP owner that sells patented, copyrighted, or trademarked goods to a purchaser in a foreign state has exhausted its ability to prevent importation (or reimportation) of the same goods into the U.S. In other words, although a transfer of the goods to a domestic third party may exhaust the transferor’s IP rights in the goods, does the same doctrine apply to foreign transfers? If so, the transferor has no right to use its IP to prevent the importation of the goods. If not, it has the legal right to prevent the importation.

It is helpful here to distinguish two potential parallel import scenarios. The first scenario involves a round trip of the goods, in the sense that the U.S. IP owner sells the goods to a purchaser in a foreign state (for simplicity, the purchaser is unaffiliated with the seller), after which the purchaser reimports the goods into the U.S. for sale there. In the second scenario, there is no round trip, because the goods are first sold in a foreign state, where they were manufactured. Suppose specifically that the goods are manufactured in a foreign state by an affiliate of the U.S. IP owner, but a third-party purchaser in that foreign state purchases the goods from the affiliate and imports them into the U.S. for sale there.

Whether the U.S. IP owner may use its IP rights to prevent importation of the goods in each scenario depends on one’s theory about the reach of U.S. patent, copyright, or trademark statutes. The answer may or may not differ depending on whether we are dealing with the first or second scenario, and it may or may not differ depending on what form of IP protects the goods.

Figure 1.  The two parallel import scenarios.

Credit: Aaron X. Fellmeth, Introduction to International Business Transactions 156 (2d ed. 2024)

2. Parallel imports of copyrighted works

There are sound reasons for believing that Congress never intended the copyright exhaustion doctrine to apply for foreign sales. As noted, the 1976 Copyright Act (codified in Title 17 of the U.S. Code) provides that the transferee of a copy of the copyrighted work need not seek the permission of the copyright owner to retransfer that copy, if the copy is “lawfully made under” the Act.90 However, there is no legislative history showing how section 602(a) was intended to interact with section 109(a). Moreover, section 602(a) was enacted specifically to provide copyright owners with a clear right to exclude unauthorized importation of copies in the same draft in which section 109(a) appears.91 Yet, it did not draw its exception into section 602(a). This strongly suggests that Congress never intended the exhaustion doctrine to apply to foreign sales. The right to exclude imports in section 602(a) was added in an amendment over thirty years later, specifically to strengthen the copyright owner’s ability to control imports of the copyrighted works.92

More generally, the presumption against extraterritoriality dictates that purchasers of copies of the copyrighted works outside the U.S. acquire no right under the 1976 Copyright Act, because that would make section 109(a) apply to purchasers in foreign territory. In the absence of any clear language that Congress intended foreign purchasers of copies of works copyrighted in the U.S. to acquire a right to import such copies into the U.S., such a right cannot be presumed under the EEOC v. Aramco93  doctrine. And, aside from these matters of statutory construction, there is another relevant principle: the territorial independence of IP rights, which will be taken up in Part III.B. This part will restrict itself to analyzing the Supreme Court’s jurisprudence attempting to resolve the conflict between the copyright owner’s exclusive right to import copies of the copyrighted work in section 602(a) and the exhaustion doctrine in section 109(a). In the course of that analysis, the problem of the Court conflating U.S. and foreign IP rights will appear anew.

a) Quality King v. L‘anza Research International and the round trip

The first modern case squarely to confront whether a U.S. copyright owner has the right to prevent unauthorized parallel imports of copies of the copyrighted work, Quality King Distributors v. L’anza Research International,94 centered on a product that most people do not normally associate with copyright. The imports were not books, films, or music, but bottles of shampoo and hair conditioner. The reason a hair care product dispute revolves around copyright is that the manufacturer could claim no valid patent on its product formulas, and it doubted its right to prevent the importation of the bottles based on its trademark, for reasons that will become clear in Part II.C. It therefore claimed a copyright in labels on the shampoo and conditioner bottles and tried to use that copyright to block parallel imports of the products.

The respondent, L’anza, is a California company that was trying to segment international markets by selling its hair care products abroad at prices substantially lower than those offered in the U.S. All of its products were manufactured in the U.S., and those destined for sale in Europe were sold to L’anza’s distributor in the U.K., which resold them to a Maltese distributor, which in turn resold them to the petitioner, Quality King. Quality King used undisclosed agents to import the product back into the U.S., without L’anza’s authorization, for resale at a price below the U.S. price.95 L’anza brought an action against Quality King for copyright infringement under section 602(a) of the Copyright Act, which reads, in relevant part:

Importation into the United States, without the authority of the owner of copyright under this title, of copies or phonorecords of a work that have been acquired outside the United States is an infringement of the exclusive right to distribute copies or phonorecords under section 106, actionable under section 501 . . . .96

The Court began its analysis by noting that, prior to the enactment of section 602(a), sections 106 and 107 of the Copyright Act already forbade the importation of “piratical” copies, which are intentional reproductions of the copyrighted work. L’anza argued that the logical deduction was that Congress added section 602(a) to extend the copyright owner’s exclusionary right to any copies that infringed the copyright, including non-piratical ones. The Solicitor General, in an amicus brief, pointed out that section 109(a) does not confer on the owner of a lawfully made copy any right to import those copies into the U.S. Section 109(a) creates an exception to section 106(3), not to section 602(a). Section 106(3) grants the copyright owner the exclusive right to sell or transfer ownership of copies. Neither sections 109(a), 106(3) nor their legislative history speaks to importation.97

The Court rejected these arguments on three grounds, the first two of which were not sufficiently sententious and persuasive to merit extended discussion.98 To the extent the Court’s opinion turned on questions of statutory interpretation specific to the Copyright Act, it mostly does not concern the subject at hand. However, the Court’s third argument is relevant here. The Court claimed that section 602(a) “applies to a category of copies that that are neither piratical nor ‘lawfully made under this title.’ That category encompasses copies that were ‘lawfully made’ not under the United States Copyright Act, but instead, under the law of some other country.”99 Because L’anza’s hair care products were manufactured in the U.S., they were “lawfully made” under the Copyright Act. Thus, the Court concluded, section 602(a) was intended to allow U.S. copyright owners to prohibit the importation of works produced in foreign lands under foreign copyrights.

This argument is sound as far as it goes and contains a seed of juridical wisdom important to all cases involving the extraterritorial application of IP laws. However, all it proves is that section 602(a) is not rendered nugatory if that section is found irrelevant to imports of U.S. copies first sold outside the U.S. It does nothing to support the argument that section 602(a) positively deprives the IP owner of exclusionary rights over such copies.

Whatever the flaws in the Court’s interpretation of the Copyright Act’s literal terms, its dismissal of the presumption against extraterritoriality reveals the fundamental failure in its understanding of copyright law, and indeed of the very concept of jurisdiction. The Court denied that, in applying the first sale doctrine to foreign purchasers, it was applying the Copyright Act extraterritorially, on the basis of its reading of the words “lawfully made under this title.” If the copy was “lawfully made” in the U.S., according to the Court, a foreign purchaser acquires the rights of section 109(a) to “sell or otherwise dispose” of the copy without the copyright owner’s permission.100 Because the hair care bottles at issue were indisputably “lawfully made under” the Copyright Act, by the Court’s reasoning, their purchaser benefited from section 109(a) regardless of where the sale was consummated. In other words, the Court believed that, even though the foreign purchaser acquires an importation right in a state other than the one in which it purchased the product, it does not acquire an importation right “extraterritorially,” because section 109(a) merely entitles the purchaser “to the protection of the first sale doctrine in an action in a U.S. court.”101

The logic behind this conclusion is opaque. The locus of the rights, the goods, the owner of the IP, and the owners of the goods are all separate concepts that have no necessary effect on each other. The place where enforcement takes place is entirely irrelevant. What matters for legal purposes in a sale of goods is the country or countries in which the parties intend the purchaser to have IP rights. Thus, if the Court meant to assert that no legal right can be extraterritorial if it is enforceable in a U.S. court, then the Court would be eviscerating the very concept of prescriptive extraterritoriality. U.S. legislation is almost entirely enforced in U.S. courts only, so the EEOC v. Aramco doctrine would be beside the point in nearly any possible legal dispute. This construction of the Court’s intent therefore seems unlikely. The more likely intention behind the Court’s statement was the following: The fact that the Copyright Act is granting rights to the copyright owner and the owner of “lawfully made” copies respectively under U.S. law to a foreign purchaser does not extend the protection of the law outside U.S. borders.

If so, the Court was mistaken. By definition, a law that grants rights to a person outside the state’s territory is extraterritorial, even if those rights are enforced only in U.S. courts. Again, U.S. extraterritorial laws are only ever enforced in U.S. courts. How could a Japanese court enforce a U.S. law that gave a Japanese copy owner a U.S. importation right opposable to a U.S. copyright owner, when the copyright owner brings suit in the U.S.? The Japanese defendant would necessarily raise the defense in the U.S. litigation. Again, the relevant question is where the parties intended the purchaser to exercise IP rights.

Quality King therefore raises a problem larger than the Court ignoring one of its own relevant or indeed decisive principles of statutory construction. The deeper problem is revealed in the reason the Court dismissed that principle. It failed to apply a basic concept of jurisdiction that undergirds the presumption against extraterritoriality; any legislation that establishes rights or obligations applicable outside of U.S. territory is ipso facto extraterritorial. In the case of IP law, the nature of the subject matter makes such extraterritorial legislation especially problematic. This nature will be explained in Part III. First, though, it is necessary to discuss how the Court abandoned its third argument only four years later, thereby sweeping away the only pillar of its Quality King opinion with any plausible foundation.

b) Kirtsaeng v. John Wiley & Sons: the newly extraterritorial U.S. copyright law

Quality King involved scenario 1, the round trip of goods manufactured in the U.S. and sold abroad. Kirtsaeng v. John Wiley & Sons, Inc.102 involved scenario 2, books manufactured outside the U.S., and therefore published under a foreign copyright. In Kirtsaeng, the U.S. copyright owner had published a book in the U.S., and its Asian subsidiary published a similar book in Singapore, for distribution solely in Asia. Defendant Kirtsaeng instructed his family in Thailand to purchase Asian copies of the books and export them to him in the U.S. so that he could sell them at a profit. The U.S. copyright owner brought an action for infringement under section 602(a).

Had the Court followed its own reasoning in Quality King rigorously, it would have observed that the Kirtsaeng had purchased the books under a foreign (in this case, Thai) copyright, and he therefore had no right to import the copies into the U.S. Even under the Quality King opinion’s strained reasoning that “importation” is the same as “distribution” or “disposal,” section 109(a) could grant Kirtsaeng no right to distribution or disposal, because he (or his agents) did not buy the copies in the U.S., as Justice Ginsburg pointed out in her concurring opinion in Quality King.

Instead, the Court held that section 109(a) of the 1976 Copyright Act does indeed benefit the Thai purchaser of a Singapore-published book sold in Thailand. The Court reasoned that section 109(a) protects the owner of any copy “lawfully made under this title.”103 From this, the Court leapt to the conclusion that every original expressive work produced anywhere in the world is “lawfully made under” Title 17 of the U.S. Code, and consequently section 109(a) benefits the foreign purchaser of a foreign-produced work. And, because Quality King endorsed the use of section 109(a) to derogate from section 602(a), the copyright owner is barred from exercising its right to exclude the importation of any non-piratical foreign copies, apparently other than those held by bailees or consignees.104

That the Copyright Act “applies” to such works cannot be doubted, if “applies” means that it has potentialapplication on condition that the work is reproduced in the U.S. without the author’s consent. But the Court went much further and plainly did not think through the implications of its line of reasoning. A U.S. federal criminal prohibition on narcotics possession “applies” to every foreign person in the world, but that application becomes relevant only if a foreign person travels to the U.S. and possesses a narcotic while there. The broadest construction of “applies” thus renders literally every U.S. law extraterritorial. By the same token, the Court’s reasoning renders every foreign law in the world “applicable” to U.S. persons in the U.S.

The Court’s problematical construction of the Copyright Act is further aggravated by the fact that the language of that section does not state or imply that the owner of a copy “to which this title applies” may dispose of its copies freely in the U.S. It refers only to copies “lawfully made under” the Copyright Act. In no sense are foreign copies “made under” the U.S. Copyright Act; they are “made under” the copyright law of the state in which they are first fixed in a tangible medium (in the case at hand, Singapore).

The incongruity of the Court’s interpretation of “lawfully made under this title” is still worse. As Justice Ginsburg pointed out in dissent, by the majority’s logic, a piratical copy of the book produced in Thailand, in violation of Thai copyright law, would be “lawfully made under” the U.S. Copyright Act,105 because the Copyright Act is not so extraterritorially effectual that it forbids a foreign person to copy a U.S.-copyrighted work in a foreign country. The maker and owner of that piratical copy, though subject to criminal prosecution in Thailand,106 is now entitled to import the piratical copy into the U.S. over the U.S. copyright owner’s opposition. 

Although the author of a book first published outside the U.S. may thus be entitled to a U.S. copyright under Title 17, Title 17 is inapplicable to the publication of book abroad. That publication is self-evidently governed by the laws of the state of publication, and it therefore is not “made under” Title 17, lawfully or not. It follows that a U.S. copyright owner cannot prevent a piratical foreign copy of its work from being imported into the U.S. against its wishes. In Kirtsaeng, the Court rendered section 602(a) of the Copyright Act almost entirely surplus verbiage, devoid of any significant meaning.

The application of the presumption against extraterritoriality would have resolved the scope of section 109(a) decisively against the defendant, but the Court did not even honor it with a cursory mention. It did invoke the canon that a statute should be rebuttably presumed to have been intended not to abrogate the common law,107 but that gained the Court little, because the extraterritoriality of the first sale doctrine was a matter of first impression. Nonetheless, the Court felt comfortable conferring rights under U.S. copyright law to foreign purchasers of foreign copyrighted materials and derogating from the statutory importation rights of U.S. copyright owners accordingly.108

3. Parallel imports of patented products

Kirtsaeng set the tone for subsequent cases on parallel imports of goods embodying intellectual property, including in the patent arena. As mentioned, the Patent Act contains no statutory exhaustion doctrine equivalent to section 109(a) of the Copyright Act. Its own doctrine arises from common law, which by 2013 had not conclusively pronounced on the subject of parallel imports, except to observe quite correctly that foreign patent rights and U.S. patent rights are separate and not necessarily related.109 The only relevant language in the 1952 Patent Act is found in section 154(a), which entitles the patent owner to exclude others from inter alia importing the patented invention into the U.S.110 In 2017, the Court settled the question of whether the exhaustion doctrine carved out a judicially-created exception to the patent owner’s statutory exclusionary right for imports in Impression Products v. Lexmark International.111

There, Lexmark sold its patented printer cartridges both within and outside of the U.S. Impression, a foreign purchaser, sought to recondition the cartridges and import them into the U.S. for sale without the patent owner’s authorization. Lexmark sued for infringement of its exclusive patent right to import the patented products under section 154(a). However, the Supreme Court held that the common law exhaustion doctrine trumped the Patent Act’s explicit grant of exclusive import rights to the patent owner. Citing Kirtsaeng, the Court viewed the “historic kinship between patent law and copyright law” as precluding a different result.112 It is therefore not surprising that the Court believed a common law grant of patent rights to a foreign purchaser was not an extraterritorial application of U.S. patent law.113

In dissent, Justice Ginsburg repeated her observation in Kirtsaeng that patent law, like copyright law, is territorial in nature,114 and the Court had long ago recognized that the Patent Act has no extraterritorial effect.115 For that reason, Justice Ginsburg concluded that “it makes little sense to say that [a foreign] sale exhausts an inventor’s U.S. patent rights. U.S. patent protection accompanies none of a U.S. patentee’s sales abroad—a competitor could sell the same patented product abroad with no U.S.-patent-law consequence.”116 Regardless, the status of parallel imports in copyright and patent are now aligned; in neither case may a U.S. IP owner prevent unauthorized foreign imports, because in neither case has the Court recognized the essential distinction between U.S. and foreign IP rights.

4. Confusion about trademark confusion

Trademark law differs fundamentally from copyright and patent laws, and so it might be expected that the law relating to parallel imports differs as well. It does. The purposes of copyright and utility patent law are to provide a means for promoting the proliferation of expressive works117 and the proliferation and public disclosure of industrially or commercially useful inventions,118 respectively. Trademark law, in contrast, exists primarily to help create and preserve a link in the minds of consumers between goods and services offered on the market and their commercial source.119 Trademark is therefore directed toward preventing consumer confusion about the source of the marked goods or services.120

As a result, the Lanham Trademarks Act provides the owner of a U.S. trademark with the right to prevent the importation of goods bearing that mark, or a confusingly similar mark, if the importation is not authorized by the owner. It also authorizes U.S. Customs and Border Protection (CBP) to prevent the importation of any merchandise that “shall copy or simulate” any U.S. trademark or trade name.121 The purpose of this provision is not to secure a monopoly on use of the word, logo, or other trademark for the benefit of the trademark owner, but to prevent consumers from inadvertently purchasing products from an unexpected and undesired source. It accomplishes this goal, however, by granting some control to the U.S. trademark owner over commercial uses of the mark. This aspect of the trademark law might seem to put parallel imports of trademarked goods on a footing quite different from that of copyright or patent law.

a) Lever Brothers and the focus on consumer confusion

The Supreme Court has not had occasion to rule on the applicability of the exhaustion doctrine to trademarked goods. The leading case comes instead from the U.S. Court of Appeals for the D.C. Circuit.122 That case, Lever Brothers v. United States, concerned the sale of dish soap by Lever Brothers in the U.S. and, through an affiliated company registered in the U.K., in the U.K. under identical trademarks registered in each country (SHIELD and SUNLIGHT). The U.S.-manufactured soap had a chemical composition, fragrance, and packaging different from the U.K.-made soap. The two also had different price points, leading third-party purchasers to import the U.K. soap into the U.S. for sale without the authorization of Lever Brothers as owner of the U.S. trademark.123 Lever Brothers invoked section 42 of the Lanham Act, which inter alia prohibits the importation of merchandise that “shall copy or simulate” any U.S.-registered trademark,124 to compel the U.S. Customs Service (the predecessor of CBP) to exclude the unauthorized imports. Customs refused.125

The reason for Customs’s refusal was a regulation in force at the time exempting from section 42 situations in which “[t]he foreign and domestic trademark or trade name owners are parent and subsidiary companies or are otherwise subject to common ownership or control.”126 This affiliate exception drew no distinction between goods imported by the affiliate and goods imported by a third party without the affiliate’s authorization. Instead, the exception turned on whether the goods were manufactured under a valid foreign trademark identical to a valid U.S. trademark.

The D.C. Circuit took the relevant question to be whether the imports “copy or simulate” a U.S. trademark (the language of section 42) or, in the alternative, are “genuine.”127 For these purposes, the Customs Service’s “affiliate exception” was not dispositive. Instead, the court analyzed a trilogy of Supreme Court precedents regarding face powder imports128 that demonstrate two principles. First, trademark law is “intended to protect a manufacturer’s reputation and goodwill and to prevent confusion among consumers.”129 Second, U.S. courts “reflect and enforce trademark territoriality.”130 More specifically, the court read the cases to:

clearly view trademarks as having specific territorial scope, and are at least consistent with the view that § 27 of the 1905 Act protects a domestic trademark holder from goods genuinely trademarked abroad but imported here by parties hoping to exploit consumer confusion between the domestic and foreign products.131

The court then supplemented these sound principles with the observation that the affiliation between Lever U.S. and Lever U.K. was not especially relevant for trademark purposes, because Lever U.K. was not accused of infringing the trademark of Lever U.S.; a third party had imported the unauthorized goods.132 The court concluded that “the natural, virtually inevitable reading of § 42 is that it bars foreign goods bearing a trademark identical to a valid US trademark but physically different, regardless of the trademarks’ genuine character abroad or affiliation between the producing firms.”133

The court’s conclusion depended on its conception of “genuine” marks as relating to the physical characteristics of the foreign and U.S. goods, which in turn dictated whether consumers would be confused by the source of the imports. This interpretation of the Lanham Act opened the door to U.S. Customs amending its regulations to disable the trademark owner from blocking unauthorized imports of products bearing a foreign trademark owned by one of its affiliates. The amended regulations provide that, even if the imports are materially different from the U.S.-sold products, they may be imported without the trademark owner’s authorization, as long as the importer dispelled any consumer confusion by applying a product label stating that the goods were different from those authorized for sale on the U.S. market. The U.S. Customs Service promptly put such a regulation into effect,134 and trademark owners were left with limited recourse to prevent unauthorized imports of goods bearing their trademarks, even when the foreign goods bearing the mark are substantially different from the U.S. goods bearing the identical mark.

b) Authorized circumvention of the Lever Brothers Doctrine

The D.C. Circuit’s reliance on physical material differences between the goods manufactured for sale under the U.S. trademark and those manufactured for sale under foreign trademarks proved a vulnerable link in its reasoning. In SKF USA v. International Trade Commission, the Federal Circuit held that parallel imports of physically identical goods bearing the U.S. trademark owner’s mark did not benefit from the first sale doctrine, because the U.S. trademark owner had a policy of denying post-sale services, such as technical and engineering support, to U.S. purchasers of its products first sold in foreign markets.135 The Federal Circuit held for the first time that, for purposes of the first sale doctrine, differences between the foreign and U.S. products need only be material, regardless of whether they are physical.136

Although the Federal Circuit upheld the International Trade Commission’s decision that the imports in that case were not materially different from the domestic products based on different post-sale technical support, in Beltronics USA v. Midwest Inventory Distribution, the Tenth Circuit applied the same reasoning to arrive at a different outcome based on a more consistent distinction between the contractual provisions in domestic and foreign sales.137 In Beltronics, the appellate court held that unauthorized sales do not benefit from the exhaustion doctrine when the trademark owner has a consistent policy of refusing to offer the warranty services, software upgrades, and rebates available to authorized purchasers.138

Beltronics USA manufactured radar detectors and sold them to authorized distributors who agreed to resell them at specified minimum prices. Two distributors sold the radar detectors to Midwest Inventory Distribution in violation of their distribution agreements after effacing or altering each product’s serial number to evade detection. Midwest then resold the detectors on eBay at prices below the minimum.139 However, Beltronics had a policy of providing software upgrades, rebates, product use information, service assistance, warranties, and recalls to purchasers of products with a valid serial number only.140 When purchasers of the Midwest products sought such services, they were frustrated to learn that they had not received what they expected, allegedly resulting in harm to Beltronics’s reputation and goodwill.141 Beltronics sued Midwest inter alia for trademark infringement, on the theory that Midwest’s removal of the serial numbers rendered the products it sold not “genuine.” Midwest retorted that, under the first sale doctrine, Beltronics exhausted its rights when it sold the radar detectors to its distributors.142

The absence of contractual privity between Beltronics and Midwest precluded a claim of breach of the distribution agreement. Beltronics’s claim therefore had to turn on whether trademark was “genuine” under the Lever Brothersreasoning. As in SKF USA, in which the Federal Circuit had held that imports are not genuine in relation to materially different domestically-sold goods, the Beltronics court held that the trademark on Beltronics’s products ceased to be genuine if the marked products were materially different from those sold by the trademark owner.143 The Tenth Circuit specifically held that distinctions in warranties and services afforded to domestic and foreign purchasers respectively could qualify as “material differences” that rendered the trademark on the unauthorized sales non-genuine.144

It will be helpful to observe that the Lever Brothers opinion treats foreign purchasers of goods outside the U.S., and produced under a foreign trademark, as having automatically acquired importation and resale rights under the U.S. trademark, unless parallel imports would cause consumer confusion. SKF USA and Beltronics held that consumer confusion can be caused by any material difference between the U.S.-trademarked goods and the foreign-trademarked goods, even if the trademark was applied by the trademark owner (or its affiliate) to goods that it manufactured itself.145 As a result of these holdings, the trademark owner’s ability to segment markets internationally, or for that matter even domestically (subject to any antitrust constraints), is limited only by its ability to incorporate contractual limitations on contingent services, such as post-sales support or warranties, provided to one and not the other.

In theory, because the ultimate issue is treated as consumer confusion, an importer can evade that segmentation and arbitrage the products by dispelling consumer confusion with a disclaimer communicated to the end-purchaser. However, in Beltronics, the defendant did just that, stating in its advertisement that the manufacturer “will not honor the warranty.”146 Yet, the district court found that notice insufficient because at least some purchasers overlooked the notice and were confused regardless, a holding the Tenth Circuit upheld.147

The upshot appears to be that a trademark owner’s ability to exclude parallel imports is restored in this line of cases, as long as it can write a distribution or sales agreement that provides materially different warranties to U.S.-sold and foreign-sold products. Segment the warranties internationally, and you can segment the markets internationally, at least until the unauthorized importer discovers a reliable way to call the disclaimer to the attention of consumers. Parallel imports may be blocked by a formality entirely under the control of the trademark owner, if that formality could influence customer purchasing decisions. Yet, these cat-and-mouse games of IP owners and unauthorized importers, variously endorsed or rejected in the appellate court cases described here, could have been avoided. If the courts had not cast aside the strictly territorial nature of trademark law recognized by the Supreme Court in the face powder trilogy, they would have concluded that the consideration of consumer confusion becomes relevant only if the importer has a right to import the trademarked goods under the Lanham Act. The courts instead put the consumer confusion cart before the jurisdictional horse.

III. The Principle of Territorial Independence of IP Law

A. The Bases of State Jurisdiction

To clarify the contradictions of the U.S. jurisprudence regarding extraterritoriality and parallel imports, it helps to return to first principles. The most basic of these is that the source of nearly all IP rights is municipal law. There are international treaties relating to IP rights, but (with some exceptions to be discussed in Part III.C) such treaties do not actually create IP rights. Like other ordinary legal rights, IP rights originate with the legislation of each state.

The jurisdiction of states is at once a function of international law and their own internal lawmaking processes. International law defines the extent of each state’s acceptable use of municipal law to regulate conduct, persons, and property, and each state’s legal system further narrows the state’s IP jurisdiction. Far from dictating the jurisdictional boundaries of each state’s IP legislation, international law leaves significant discretion to states, especially with regard to subject matter located within the state’s geographic territory. International law is generally interpreted as affording states discretion to exercise prescriptive, judicial, and enforcement jurisdiction freely within its territory, in the absence of a contrary rule or policy of international law.148 Such contrary rules are very rare; indeed, states have nearly plenary prescriptive jurisdiction within their territories.

Yet, states may also apply their legislation to regulate conduct outside their territories. As the Permanent Court of International Justice observed in the S.S. Lotus Case, “[r]estrictions upon the independence of States cannot [ ] be presumed.”149 It continued:

Far from laying down a general prohibition to the effect that States may not extend the application of their laws and the jurisdiction of their courts to persons, property and acts outside their territory, it leaves them in this respect a wide measure of discretion which is only limited in certain cases by prohibitive rules; as regards other cases, every State remains free to adopt the principles which it regards as best and most suitable.150

In each state’s municipal law, that state defines its prescriptive jurisdiction through legislation in accordance, first with its constitution, then in IP legislation either explicitly or under general doctrines. In general, states commonly invoke three bases for prescriptive jurisdiction over conduct, persons, and property: territorial location, nationality, and effects on the state’s territory.151 Territorial jurisdiction is usually the sole basis for state IP laws. Only rare and relatively peripheral provisions of the patent, copyright, or trademark legislation of most states define the rights and responsibilities of IP owners or accused infringers on the basis of the nationality of either. States have more commonly adopted extraterritorial legislation defining some forms of IP infringement based on the effects on the territory of the state, but effects jurisdiction in each case usually arises from a narrowly drawn, legislated exception.152

Territorial jurisdiction is incident to one of international law’s core organizing principles. We live in a world almost entirely153 divided into inhabited political states and uninhabited territories considered res universalis, such as the high seas and Antarctica. Under customary international law, a state is formally defined as an entity with a definite territory, permanent population, government, and capacity to enter into relations with other states.154 The most immediate consequences of statehood are, accordingly, the government’s primary jurisdiction over its territory and citizens.155

The decision to grant or withhold exclusionary rights to IP is thus in the first instance a product of each state’s municipal law. States are generally free, absent a binding treaty, to set the conditions for IP protection within their territories, just as they are generally free to legislate within their territories on most other matters.156 In the case of patents and trademarks, almost all states grant IP rights for their territory based on registration of the patent or trademark with a national authority. Historically, copyrights had to be registered in some states as well, but that is no longer the case due primarily due to the Berne Convention and TRIPS Agreement, as will be discussed in Part III.B.1, below. The reasons for patent and trademark registration vary from state to state, but in each case the registration traditionally results in a grant of IP rights specific to the state’s territory only.

The U.S. has been no different from other states in that regard. Although some IP statutes are more explicit than others with regard to their territorial effect, none clearly provides for a general extraterritorial IP right, and one unambiguously contemplates that the primary right applies only within U.S. territory. That statute, the Patent Act, provides in section 271 that “whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United States or imports into the United States any patented invention during the term of the patent therefor, infringes the patent.”157 Section 271 also defines a few forms of infringement that involve specific extraterritorial acts, such as the conduct discussed in WesternGeco.158 However, these exceptions are narrowly drawn against the general rule that the making, using, selling, offering, or importing of the patented invention must occur wholly within the U.S. The Patent Act thus settles the question of the geographic scope of a U.S. patent beyond peradventure.

In contrast, the 1976 Copyright Act lists the general statutory right in section 106, which was unfortunately drafted without thought to clarifying the Act’s geographic jurisdiction. It merely provides the owner of a copyright with the exclusive right to engage in and to authorize certain activities in general, such as copying or distributing copies of the work.159 Similarly, the Act defines infringement of the copyright without reference to U.S. territory, except in the case of importation.160 Congress evidently neglected to specify that the exclusive rights to copy, distribute, etc., and any infringement of that right, must occur within the U.S. to violate the Act. The result of that omission is an absence of literal language that precludes forbidding a foreign person in a foreign country from making copies of a work protected by copyright solely in the U.S. It is intuitively improbable that Congress intended to legislate copyright protection for the entire world, and, until the Kirtsaeng decision, the Supreme Court161 and all circuit courts had accordingly interpreted the Copyright Act to grant rights limited to U.S. territory.162

As noted in Part II.A, the Lanham Act similarly neglects to limit trademarks to U.S. territory. It provides that “[a]ny person” who “shall . . . use in commerce” a “reproduction, counterfeit, copy, or colorable imitation of a registered mark,” or “reproduce, counterfeit, copy, or colorably imitate a registered mark” under certain circumstances is liable to the registrant for specified remedies.163 The term “commerce” is not explicitly limited by territory; it is defined as “all commerce which may lawfully be regulated by Congress.”164 The ambiguity here is in the word “lawfully,” which, it seems likely, was originally intended to avoid intruding on state prerogatives under the Tenth Amendment to the Constitution. As discussed previously, however, read literally, as the Supreme Court did in Steele v. Bulova, the Lanham Act has no territorial limitation whatsoever, because, consistent with the Constitution, Congress may legislate extraterritorially if it wishes, even in violation of international law. Nonetheless, after decades of inconsistent circuit court approaches to assessing the extraterritoriality of the Lanham Act,165 courts now seem poised to reject Steele’sbroad reading of the Lanham Act and, consistent with Abitron v. Hetronic, hold that foreign sales of goods bearing a U.S.-registered trademark cannot infringe the Lanham Act.166 This is so precisely because, notwithstanding the Steele decision, the Act gives no “clearly expressed congressional intent to” extend U.S. trademark jurisdiction to foreign activity.167

To summarize, although the U.S. statutes relating to IP rights are superficially inconsistent in their geographic reach, that inconsistency is a result of imperfect drafting rather than a conscious congressional departure from the global norm that IP laws are limited to the state’s own territory. U.S. courts have usually interpreted the statutes accordingly, but both influential circuit courts (beginning with the D.C. Circuit in Lever Brothers) and the Supreme Court itself (in Steele v. Bulova and WesternGeco) have browsed aggressively in extraterritorial pastures. For unknown reasons, the Court in Abitron passed up an opportunity to return the wayward IP jurisprudence to the fold, nor has the Court ever developed a theory to corral the Lever Brothers line of cases.

B. The Territorial Independence of IP Laws

1. The bases for U.S. prescriptive jurisdiction respecting IP laws

The international law doctrine of prescriptive jurisdiction applies in IP law to the same extent that it applies in nearly all other fields of public policy on which a state may choose to legislate for its territory. The authority from which the U.S. Congress legislates respecting IP rights arises most directly from the Constitution’s Commerce Clause, which authorizes trademarks implicitly by empowering Congress to “regulate Commerce with foreign Nations, and among the several States . . . .”168

Trademarks help regulate domestic and international commerce by limiting the use of words, phrases, logos, and other indicators of origin to specific suppliers, thereby hindering consumer confusion about the sources of goods and services. The Intellectual Property Clause relating to patents and copyrights is more specific. It authorizes Congress to grant exclusionary rights to authors and inventors for their expressive works and inventions, respectively.169 By the same token, foreign states legislate regarding IP pursuant to the authorities of their own constitutions.

Early U.S. jurisprudence consistently and explicitly recognized the territorial independence of IP laws. In Ingenohl v. Olsen, for example, Justice Holmes, writing for the majority, observed that a trademark registered in the Philippines (then a U.S. colony) provides no rights in Hong Kong (then a British colony); rather, a trademark registered in the Philippines “would depend for its protection in Hongkong upon the law prevailing in Hongkong, and would confer no rights except by consent of that law.”170 Similarly, in Brown v. Duchesne, the Court observed that U.S. patent statutes do not “operate beyond the limits of the United States; and as the patentee’s right of property and exclusive use is derived from them, they cannot extend beyond the limits to which the law itself is confined.”171 In her dissent in Kirtsaeng, Justice Ginsburg also pointed out that “intellectual property law is territorial in nature,” meaning that different persons may own parallel IP rights in different states.172 Given the ascendance of the Aramco presumption against extraterritoriality in the last few decades,173 it is difficult to understand how the Supreme Court could repeatedly depart from its tradition of recognizing the territorial limits of IP legislation in the U.S.

One possible explanation is the progressive, albeit very incomplete, harmonization of state municipal IP laws by international treaties. This has been an ongoing process for the last 130 years, but it accelerated with the foundation of the World Intellectual Property Organization (WIPO) in 1967 and especially with the entry into force of the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement) by the contracting parties to the General Agreement on Tariffs and Trade (GATT) in 1995. Although international treaties have harmonized many areas of municipal law, including competition policy, tax law, commercial agreements, bankruptcy, and jurisdiction, very few fields of municipal law have been subjected to international harmonization at the same level of detail as IP law. 

For example, with respect to patents, the TRIPS Agreement inter alia defines the subject matter that states must protect very broadly, specifies what rights patents must grant, limits exceptions and derogations to those rights, and sets a minimum term for the patent.174 The WIPO Patent Law Treaty harmonizes state patent examination and enforcement procedures significantly,175 and the Patent Cooperation Treaty creates a system for patents in nearly every major country in the world using a single initial application.176 Global protection for copyright is even more harmonized, with the Berne Convention (and, by extension, the TRIPS Agreement)177 providing for automatic protection of expressive works without the need for registration. In addition, the TRIPS Agreement178 and WIPO Copyright Treaty179 define the subject matter, rights, procedures, and limitations of copyright in some detail.

To nonspecialists surveying the field of international IP law, and especially to nonspecialists with limited experience of international law, this harmonization might appear to establish a set of global IP legal regimes with identical, or at least comparable, protections for all types of IP. This degree of harmonization may further create an illusion that the owner of patent, trademark, and especially copyright effectively has a global IP right. However, the uniformity is in fact limited and, even if it were complete, it is not legally germane. Convergence and identity are separate concepts. Only the first describes the treaty systems for the global protection of IP rights.

2. Reaffirmation of territorial independence in international treaties

A treaty is a binding agreement between sovereign states180 committing the parties to adopt specific policies, either internally in their municipal law or practices, or externally with respect to other states or actors, or both. In some states, ratified treaties are not immediately enforceable in domestic law until implementing legislation has been adopted; in other states, treaties are directly enforceable once ratified.181 In both cases, however, the operative law is municipal, because it must be interpreted and enforced by each state party’s municipal courts using its idiosyncratic legal procedures. Municipal courts thus play a dual role (dédoublement fonctionnel in Georges Scelle’s felicitous phrase),182 defining international law for the domestic legal system while maintaining their state’s compliance with the international legal system. Different municipal legal systems grant different rights, even if the rights are identical in content. Thus, even though two states grant identical trademarks for identical classes of goods, each trademark is in fact a different entity—a bundle of legal rights effective only in the granting state.

That the IP rights harmonized by international treaties originate in each state’s municipal law is not merely a conceptual point; it is an explicit one made in the treaties themselves. Articles 4bis and 6 of the Paris Convention on the Protection of Industrial Property expressly recognize the independence of patents and trademarks granted by each state:

Article 4bis

(1) Patents applied for in the various countries of the Union by nationals of countries of the Union shall be independent of patents obtained for the same invention in other countries, whether members of the Union or not.

(2) The foregoing provision is to be understood in an unrestricted sense, in particular, in the sense that patents applied for during the period of priority are independent, both as regards the grounds for nullity and forfeiture, and as regards their normal duration.

(3) The provision shall apply to all patents existing at the time when it comes into effect.

(4) Similarly, it shall apply, in the case of the accession of new countries, to patents in existence on either side at the time of accession

Article 6

(1) The conditions for the filing and registration of trademarks shall be determined in each country of the Union by its domestic legislation.

(2) However, an application for the registration of a mark filed by a national of a country of the Union in any country of the Union may not be refused, nor may a registration be invalidated, on the ground that filing, registration, or renewal, has not been effected in the country of origin.

(3) A mark duly registered in a country of the Union shall be regarded as independent of marks registered in the other countries of the Union, including the country of origin.183

These principles are incorporated into the TRIPS Agreement by reference.184    

The same principle governs copyrights, with a minor complication. Under article 5 of the Berne Convention (also incorporated into the TRIPS Agreement),185 all state parties must grant copyright protection to qualifying expressive works without requiring any formalities.186 That means an expressive work entitled to copyright protection in any Berne Convention state or WTO member automatically receives copyright protection in every other Berne Convention state and WTO member. Nonetheless, the uniformity of some legal right to protection does not compromise the territorial independence principle. As Article 5 of the Berne Convention itself affirms, the rights conferred remain “governed exclusively by the laws of the country where protection is claimed.”187 This provision, too, is incorporated into the TRIPS Agreement.188

The far reach of copyright treaties has blinded some jurists to the nature of copyright law as conferring exclusively municipal rights. In her principally clear-sighted dissent in Impression Products v. Lexmark, Justice Ginsburg sought to distinguish copyright from patent by suggesting that copyright is indeed global in effect, if technically municipal:

But even if I subscribed to Kirtsaeng’s reasoning with respect to copyright, that decision should bear little weight in the patent context. . . . The Patent Act contains no analogue to 17 U.S.C. § 109(a), the Copyright Act first-sale provision analyzed in Kirtsaeng. More importantly, copyright protections, unlike patent protections, are harmonized across countries. Under the Berne Convention, which 174 countries have joined, members “agree to treat authors from other member countries as well as they treat their own.” Golan v. Holder, 565 U.S. 302, 308 (2012) (citing Berne Convention for the Protection of Literary and Artistic Works, Sept. 9, 1886, as revised at Stockholm on July 14, 1967, Arts. 1, 5(1), 828 U.N.T.S. 225, 231–233). The copyright protections one receives abroad are thus likely to be similar to those received at home, even if provided under each country's separate copyright regime.189

This is a misguided analysis in an otherwise insightful opinion. There is no legally meaningful difference between state jurisdictions regarding copyright law and patent law. Both the copyright laws and the patent laws of most states are partially, but not completely, harmonized. There is no more a global copyright than there is a global patent or trademark. Moreover, although Justice Ginsburg viewed the Paris Convention’s insistence that each state grants an independent patent as a confirmation of her view that patents are territorial, she failed to observe that the Berne Convention has a parallel provision relating to copyrights:

The enjoyment and the exercise of these rights shall not be subject to any formality; such enjoyment and such exercise shall be independent of the existence of protection in the country of origin of the work. Consequently, apart from the provisions of this Convention, the extent of protection, as well as the means of redress afforded to the author to protect his rights, shall be governed exclusively by the laws of the country where protection is claimed.190

The treaty provisions discussed here reaffirm that all IP rights necessarily originate in each state’s municipal laws, regardless of the extent of international harmonization or the availability of unified procedures for seeking multiple national registrations with a single application. In consequence, a grant of a patent, trademark, or copyright in the U.S. conveys no rights outside the U.S. except as provided by a federal statute, and no patent, trademark, or copyright granted by a foreign state conveys rights in the U.S., again except as provided by a federal statute (a largely hypothetical point, because no U.S. statute provides for rights by virtue of foreign patents, copyrights, or trademarks except in very specific situations).

It follows that the Supreme Court majority’s holding in Kirtsaeng, that the 1976 Copyright Act “applies” to a book published in a foreign country, and therefore that book copy is “lawfully made under” the U.S. Copyright Act, fundamentally misconstrues how IP laws (indeed, how all laws) work. Consistent with the Berne Convention and TRIPS Agreement, the 1976 Copyright Act provides that a foreign unpublished manuscript is entitled to protection in the U.S., but only under U.S. law, not under any foreign law.191 Much less do the 1952 Patent Act and Lanham Trademarks Act “apply” to foreign patents and trademarks, which confer very few rights in the U.S. (and certainly not a general exclusionary right) without a U.S. registration or other U.S. legal basis, such as common law trademark use.

The consequences of this independence are crucial to each state’s sovereignty over its own IP policy. In no treaty is harmonization of IP laws complete; each state always retains discretion to adjust IP policies to a significant degree according to its needs and preferences. A trademark granted in one state may be denied in another.192 A patent granted with certain claim language in one state may require different language in another.193 The rights and remedies vary between states as well. U.S. civil and criminal copyright infringement penalties, for example, are draconian compared to those of most other countries,194 while Japan and some European states have criminal penalties for patent infringement, unlike the U.S. and other European states.195 The variability of municipal IP laws is clear from the treaties themselves, which leave to the discretion of state parties significant latitude on many policy decisions.196 The treaties establish minimum standards of protection, after all; very rare indeed are provisions that specify maximum standards. IP treaties set floors, not ceilings, on IP protection.

By the same token, different persons may own the rights to identical IP in different countries. For example, the same or similar trademarks may be owned and used by different business firms in different countries, unaware of each other.197 Or copyrights on the same expressive work may be assigned by the author to different assignees in different countries.198 The same applies to patents, where international assignments resulting in ownership of commercialization rights by different, and sometimes unaffiliated, entities is well known.

The most telling point of discretion in municipal IP policy for purposes of the parallel import question is that the TRIPS Agreement expressly disclaims any harmonization of the exhaustion doctrine. Each state is free to apply or not apply the exhaustion doctrine to imports, as it sees fit: “For the purposes of dispute settlement under this Agreement, subject to the provisions of Articles 3 and 4 nothing in this Agreement shall be used to address the issue of the exhaustion of intellectual property rights.”199 Aside from cumulatively reaffirming the territorial independence of IP rights, this provision puts beyond dispute that nothing about an IP license (express or implied) accompanying a sale of a patented article, copyrighted copy of an expressive work, or product bearing a trademark is necessarily global.200 A state may choose to require all IP licenses accompanying product sales to be global. It might do so pursuant to an effort to create a common market with other countries, as the members of the European Union and a few others have done. Or it might do so in the belief that global market segmentation should be discouraged. But the U.S. Congress has adopted neither policy in any legislation.

In short, because IP rights are creations of state law, there are no global patents, copyrights, or trademarks. These are always the product of state municipal law or, in a few cases to be discussed, regional law. It follows that an IP license granted in one state has no necessary effect in any other. The sale of a U.S. product embodying a copyright, patent, or trademark to a foreign purchaser grants that purchaser no rights under the U.S. copyright, patent, or trademark. The rights granted are purely those of the purchaser’s state. If the U.S. seller had no right to the foreign IP, it infringes the foreign IP owner’s rights by making the sale. If the U.S. seller had rights to the foreign IP, then the license grant is governed by the licensing law of the foreign state. This is where the Supreme Court’s jurisprudence on parallel imports has gotten the law precisely backward.

3. Territorial independence, the act of state doctrine, and comity

Some states, including the U.S., have a judicial principle that reaffirms the territorial nature of governmental authority, known as the act of state doctrine. This doctrine treats as non-justiciable the sovereign acts of a foreign state within its own territory.201 The Restatement (Fourth) of Foreign Relations Law of the United States explains the principle as follows: “In the absence of a treaty or other unambiguous agreement regarding controlling legal principles, courts in the U.S. will assume the validity of an official act of a foreign sovereign performed within its own territory,” subject to modification by Congress.202 The purpose of the doctrine has been to recognize the venerable principle of sovereign equality, a corollary of which is the maxim par in parem non habet iurisdictionem.203

Grants or denials of IP protection, like all grants of legal rights effective within the state’s territory, are sovereign acts (acta iure imperii). Only a state government, or its subdivision or delegate, may grant a patent, trademark or copyright. It would be improper for any court to sit in judgment of a foreign state’s grant or denial of IP rights, and by extension, of its interpretation of its own IP laws.204 Some U.S. courts are reluctant even to entertain foreign patent infringement claims that do not necessarily raise questions of the validity of a foreign sovereign act.205 The converse equally applies; a U.S. court will not give effect to a foreign state’s attempt to expropriate a U.S. trademark, patent or copyright, because its IP law cannot extend to U.S. territory.206 The act of state doctrine thus confirms and reflects the territorial independence of each state’s IP laws.

When U.S. courts do assert jurisdiction over claims relating to the infringement of a foreign patent, copyright, or trademark, it is usually because the invalidity of the foreign grant is not raised as a defense.207 For example, in Fairchild Semiconductor Corp. v. Third Dimension (3D) Semiconductor, Inc., the Maine district court accepted jurisdiction over a licensing dispute involving a Chinese patent, because the question of patent invalidity was not raised.208 The court then assumed the aspect of adjudication relating to a right granted under foreign law, and held that although such cases are complicated by the court’s lack of expertise in any foreign law’s methods, theories, procedures, and substance, they do not necessarily require the court to evaluate the legitimacy of a foreign act of state.209 Such cases generally involve the question of infringement of a foreign IP right or the remedies available for such infringement.

In extraterritorial infringement cases, a court may nonetheless decline to adjudicate the case, or, at a minimum, may defer to the foreign courts’ interpretations of their own laws, under the doctrine of comity. Comity, as understood by the U.S. Supreme Court, is:

neither a matter of absolute obligation, on the one hand, nor of mere courtesy and good will, upon the other. But it is the recognition which one nation allows within its territory to the legislative, executive or judicial acts of another nation, having due regard both to international duty and convenience and to the rights of its own citizens or of other persons who are under the protection of its laws.210

The most difficult cases involve allegedly infringing acts, critical elements of which occurred in more than one state. The question then becomes how to identify the state in which the IP rights were infringed, if any, so that the appropriate law can be applied. The test used in any given case necessarily involves the court’s discretion and judgment. As long as the court does not merge U.S. and foreign IP rights, but treats them as separate, the principle of comity leaves the court free to adopt any legal theory that reasonably accounts for the relative interests of the U.S. and the foreign state.

For example, in one case, the Federal Circuit held that a foreign defendant who offered to sell an invention patented in the U.S. to a prospective U.S. buyer had infringed the U.S. patent, as opposed to the patent in the offeror’s state.211 To the extent that U.S. IP legislation specifically defines an act as infringing despite critical elements occurring outside U.S. territory, such legislation resolves the issue from the court’s standpoint.212 In the absence of such legislation, given that the main effect of the offer was felt on the U.S. market, the Federal Circuit was justified in using the jurisprudential principle of direct, substantial, and reasonably foreseeable effects on the U.S. to treat the claimed infringement as having occurred in the U.S.213 It would not have been justified in simply concluding, for example, that, because the plaintiff owned patents in both the foreign state and U.S., any sequence of infringing acts that occurred involving the two countries necessarily infringed the U.S. (much less the foreign) patent. The foreign acts must have caused some effect on the U.S. market for the patented technology (for a U.S. infringement claim) or the foreign market where the plaintiff holds a patent (for a foreign infringement claim).

C. The Regional Patent and Trademark Systems: Exceptions That Affirm the Rule

The territorial independence principle is subject to one major exception, but the exception merely reaffirms the rule. In a few regions, states have voluntarily entered into treaty arrangements to create genuinely multinational patents and trademarks. Most prominently, many European states have ratified the Agreement on a Unified Patent Court,214 which creates a unitary patent for all states that ratify the treaty, and all EU states have agreed to create a common EU trademark valid throughout the EU.215 Similarly, the Organisation Africaine de la Propriété Intellectuelle (OAPI) grants unitary trademarks and patents valid in multiple (mostly francophone) African countries.216

The European and OAPI systems create genuinely international patent and trademark rights within the territories of their member states. Such rights can be adjudicated as if the national boundaries within each union did not exist. However, these regional systems are the product of deliberate treaty relations, and they are ineffectual outside each regional organization. Thus, although the validity of an OAPI patent is the same in Benin and Cameroon, and infringement can occur and be adjudicated in any other OAPI state or combination of states, the same cannot be said in any state outside of the region. OAPI patents and trademarks are therefore entirely independent of patents and trademarks in Zimbabwe (not an OAPI state), and Zimbabwean patents and trademarks are independent of those granted by OAPI.                                                                                                                

IV. Two Doctrinal Dilemmas, One Cause

The root cause of the contradictions and confusion discussed here is the same: the consistent disregard by courts at the highest levels of the territorial independence of IP rights. The Supreme Court and circuit courts have repeatedly treated U.S. IP rights and foreign IP rights as assimilated, resulting in incoherent jurisprudence in all fields of IP law.

For purposes of both cross-border infringement claims and parallel imports, it is important to recall that the situs of any alleged infringement depends on two factors: the country in which the license is granted and the country in which the infringing activity took place. With regard to parallel imports, the IP license (which may be either explicit or granted implicitly by legislation or common law) accompanying every sale, lease, or other transfer of a patented, copyrighted, or trademarked good applies to the state in which the transfer took place, which is to say, the state where the lessee or transferee took delivery and where the parties understood the good would be used. There is no reason to assume that such a license is binational, much less global, in scope, unless relevant legislation or an explicit license agreement makes it so.

As for cross-border infringement, the infringing activity can very well occur in multiple states, but infringement will be defined by statute, usually to indicate that the infringing act is an unauthorized “sale,” “use,” “importation,” etc. of the IP or a good embodying it, or, in the case of trademark, bearing the mark.217 In that case, the relevant juridical activity is statutory interpretation, not a search for the nationality of the infringer, the country where the effect of the infringement was felt or in which preparatory acts were undertaken, or anything else.

A. Extraterritoriality and Territorial Independence

Recognizing and respecting the principle of territorial independence of IP laws solves the main doctrinal problems of the U.S. jurisprudence regarding the foreign effect of IP laws, made manifest in Steele v. Bulova and WesternGeco, and exacerbated by an implicit yet disclaimed overruling of the Steele decision in Abitron. The existing presumption against extraterritorial interpretations of legislation may not be the ideal formulation of a principle of statutory construction, but it is generally sound in practice. In most cases, it will be consistent with the territorial independence of IP laws, but it neglects two important juridical principles inherent in the territorial independence doctrine. Specifically, it does not recognize the relevance of the IP laws of foreign states to the question of validity of the state’s own grants of IP rights (the act of state doctrine), and it does not necessarily incorporate considerations of comity when evaluating the relevance of foreign laws relating to infringement of those IP rights. The result will inevitably be consequential confusion in interpreting IP statutes in the context of infringement cases involving multiple states.

Even the recent cases relating to extraterritorial IP infringement are difficult to reconcile into a predictable doctrine on their published reasoning. As Margaret Chon has observed:

In WesternGeco v. ION . . . the Court separately applied the [extraterritoriality] test to the liability and damages provisions of the patent act. In deciding that the damages provision did cover foreign conduct, it looked to the liability provision (section 271(f)) to which that damages provision was linked. In Abitron, however, the Court did not explore differences between sections 1114 and 1125, nor did it address the remedies sections of the Lanham Act (whether damages or injunctions, both of which are at play in the case), possibly because it was obvious that once linked to the liability provisions, these remedies provisions would not apply to purely foreign conduct.218

The problem of interpreting and applying domestic jurisdictional rules without regard to international law and the principle of territorial independence is not limited to U.S. jurisprudence. The consequences of ignoring these factors in the multinational infringement context became apparent by a misadventure of the Supreme Court of Canada several years ago.219  

Equustek was a Canadian company that had obtained an injunction from a court in British Columbia (hereinafter BC) against Canadian defendants to cease use of the Equustek’s trade secrets and to cease unfair competition, including referring to Equustek’s trademarks, on the defendants’ website. Instead of complying with the injunction, the defendants “left the jurisdiction” and incorporated shell companies in various states to continue to do business using Equustek’s trade secrets.220 The actual location of the defendants remained unknown at the time of the case. The BC court then entered an injunction with worldwide effect freezing the defendants’ assets, issued a warrant for the arrest of the individual defendant, and eventually ordered the defendants to cease doing business on any website.

With the defendants ignoring the Canadian courts’ orders and injunctions, these methods proved ineffective. Eventually, the BC court ordered Google, Inc.—a U.S. nonparty to the case that has a Canadian subsidiary—to cease indexing the defendant’s website on any Google-operated server in the world. This way, anyone, anywhere in the world, who searched for the defendants or their products on Google.com would not find them. Google, it must be emphasized, was not a defendant accused of infringing any of Equustek’s IP.

Upon being challenged for excess of jurisdiction, the Canadian Supreme Court concluded that, as long as the court has in personam jurisdiction over Google, which was uncontested, the court could order Google to comply with an injunction anywhere in the world, because “[t]he Internet has no borders—its natural habitat is global. The only way to ensure that the interlocutory injunction attained its objective was to have it apply where Google operates—globally.”221

The Canadian Supreme Court missed the main issue. The primary argument was that the injunction was equitable in nature and had to be global to be effective in the circumstances of the case. But the problem was not unfairness to Google or ensuring fairness to the plaintiff. The problem was that a Canadian court has no authority to adjudicate trade secret infringement allegedly occurring in countries other than Canada. The question entirely overlooked by the Supreme Court was whether the defendants’ marketing and sale of products violated Equustek’s trade secret rights in every other country in the world. Because the plaintiff produced no evidence that it owned the trade secret in any other country, or that it was so protected, there was no reason to assume that any other country’s laws necessarily protected Equustek’s trade secret in the identical manner of Canadian law.

The court avoided addressing this question by assuming that Canada had the authority to declare that trade secret infringement occurred under Canadian law when a seller in, say, Slovenia sold goods embodying a Canadian trade secret to a buyer in, say, Papua New Guinea. Jurisdictional problems do not disappear merely because ignoring them conduces to fairness for an aggrieved plaintiff, and because recognizing them proves extremely inconvenient to the plaintiff or the court itself. In short, the unacknowledged problem is that the Canadian Supreme Court assumed that Equustek had a right to protection of its trade secrets by Canadian courts under Canadian law in countries other than Canada. It assumed incorrectly that Canadian trade secret law applies in foreign countries. Had it recognized the territorial independence of trade secrets, it would have avoided this misapprehension, and Google would not have been subjected to the possibility of inconsistent obligations in multiple countries.

B. Parallel Imports and Territorial Independence

1. Parallel imports of trademarked goods

In the parallel import jurisprudence, ignoring the territorial independence of IP law has caused a great deal of confusion and contradiction. Had the U.S. Court of Appeals for the D.C. Circuit recognized that the U.S. and U.K. trademarks confer entirely different sets of legal rights in Lever Brothers, it would not have ventured into the substantive question of whether U.S. consumers would be confused by the sale of goods made for a foreign market under a foreign trademark.

The relevant question was never whether consumers would be confused by the marked products, because that question would be obviated by the U.S. trademark owner’s ability to block unauthorized parallel imports of goods produced and sold under a foreign trademark. Had the court not fallen into this error, the permissibility of the imports would have been a straightforward question of whether the foreign-produced goods were authorized for sale in the U.S.—implicitly or explicitly—by the owner of the U.S. trademark. Instead, the court required the question to be resolved based on an analysis of consumer confusion, incentivizing adventurism, complex legal strategies, and inevitable disputes as manufacturers try to create artificial distinctions between products sold under their respective foreign and U.S. trademarks, and unauthorized importers counter by using product labels purporting to dispel consumer confusion for legal purposes (but possibly doing precisely the opposite in reality, given the evidence that product labels are not reliably informative to consumers unless very carefully designed).222

The D.C. Circuit further complicated the analysis by putting weight on the affiliation between Lever U.S. and Lever U.K., which caused it to conclude that the U.K. trademark was “genuine” (in other words, that it did not copy or simulate the identical U.S. mark) for U.S. trademark law purposes. The flaw in the court’s conclusion that foreign trademarks are “genuine” for U.S. purposes if the foreign and U.S. products are not materially different becomes apparent when considering that, because the U.S. and U.K. trademarks are entirely separate legal rights granted under separate legal regimes, it is impossible for a foreign trademark to be “genuine” merely because the U.S. trademark owner also owns (directly or indirectly) an identical foreign trademark. Consider two hypothetical variations on the facts of the Lever Brothers case.

First, if the U.K. trademark owner had not been affiliated with the U.S. trademark owner, the court’s holding would have been indefensible even under its own reasoning. How could anyone argue that a foreign person’s U.K. trademark is “genuine” for U.S. purposes merely because it is identical to the trademark of an unrelated U.S. trademark owner? If that were the case, the concept of import infringement would entirely evaporate despite the contrary language of the Lanham Act. Yet, the importer in that case was not affiliated with either company, as the court itself recognized, so the affiliation of Lever Brothers U.S. and its U.K. affiliate was beside the point. Lever Brothers, as a multinational enterprise, was not accusing its component companies of infringing “its own” mark. The only relevant principle in that case was the territorial independence of the trademarks, and under that principle, no one had the right to import products not sold under the U.S. trademark except the U.S. trademark owner itself.

Next, consider a case that is in some ways the reverse of the facts in Lever Brothers. Suppose that Lever U.S. directly owned both U.S. and U.K. trademarks, and the two trademarks were entirely different, yet were applied to identical products. If Lever U.S. then attempted to import its U.K. merchandise into the U.S. for sale, consumers might well become confused by the fact that the same company (or affiliated companies) are selling the identical products under two different trademarks. They might wonder whether, despite appearances, the source of the two products was the same. Given that a U.S. company and its foreign affiliates are legally distinct persons, their confusion would be well justified. Yet, there is no question that Lever U.S. would be somehow infringing its own trademark.

It follows that Lever Brothers was not really a case that turned on avoiding consumer confusion, at least not as its primary juridical principle. The first relevant principle was the territorial independence of trademarks, which encompasses the U.S. trademark owner’s rights to exclude imports not licensed for distribution in the U.S. under the U.S. trademark. Preventing consumer confusion is a natural consequence of allowing the U.S. trademark owner to control which foreign goods bearing its trademark (or one confusingly similar to it) are imported into the U.S., because trademark owners have an incentive to police their own marks to avoid consumer confusion. But it is not the reason for the recognition of the territorial independence of trademarks. That recognition follows from the fact that the U.S. and foreign trademarks arise from distinct legal regimes.

2. Parallel imports of copyrighted and patented products

In Quality King and Kirtsaeng (and, by extension, Impression Products), the Supreme Court made lengthy forays into statutory interpretation, legislative history, and the public policies that Congress presumably sought to enact with the Copyright Act (and, mutatis mutandis, the Patent Act). The Court was consistently addressing the wrong questions, as Justice Ginsburg suggested in her dissent. The relevant question was never what Congress intended with regard to the statutory first sale doctrine (copyright) or what courts intended in creating the patent exhaustion doctrine in import scenarios, because there is no evidence Congress or the courts ever considered the issue in either context prior to those cases. The Court’s statutory construction in both lines of jurisprudence relies mostly on surmise and assumption.

The relevant question was more general: whether Congress intended for U.S. copyright or patent rights to apply extraterritorially, merging U.S. copyrights and patents with their foreign cognates. There was (and still is) no evidence that Congress had any such intent, and, given that absence of evidence, the Court’s own presumption against extraterritorial application of U.S. statutes from EEOC v. Aramco and its progeny would have resolved the issue decisively against a finding of international exhaustion, if the Court had begun from an understanding that U.S. patents, copyrights, and trademarks confer no rights outside U.S. territory, and vice versa for foreign patents, copyrights, and trademarks.

It would indeed be surprising if Congress had decided that U.S. law grants the licensee of a foreign copyright or patent automatic rights under the U.S. copyright or patent. Again, consider the hypothetical scenario in which the U.S. patent or copyright is owned by a person (let us call her Uma) unaffiliated with the owner of the foreign patent or copyright in, say, Finland (we will call him Felix). It is difficult to imagine by what stretch of legal reasoning the Court could conclude that, if Felix manufactures and sells a patented product in Finland, then the Finnish purchaser has a right to import the product into the U.S. for resale in defiance of Uma’s U.S. patent. Nor could the Finnish purchaser of a copyrighted work licensed by Felix in Finland import the work for resale into the U.S. regardless of Uma’s U.S. copyright. The fact in Kirtsaeng that the U.S. and foreign copyrights, and in Impression Products that the U.S. and foreign patents, were owned by the same person does not change the outcome. The distinct U.S. and foreign copyrights and patents do not magically become assimilated merely because they are owned by the same person.

C. Policy Problems Territorial Independence Does Not Solve

The territorial independence principle is not, of course, a panacea for all policy problems or difficulties of jurisprudence that may arise in international IP law. Because much commerce in goods embodying IP is international or even global in nature, the recognition of IP’s territorial nature means that discrepancies between the global community’s policy goals and the protection actually afforded by national laws (with the regional exceptions discussed in Part III.C) will arise inevitably.

In Google v. Equustek, the Supreme Court of Canada’s jurisdictional overreach resulted from the fluidity with which an IP infringer could evade the jurisdiction of national law and the limitations on courts trying to enforce that law. The Equustek decision was primarily concerned with trade secrets and unfair competition, but the challenges faced by the British Columbia court in enforcing its injunctions would have applied equally to Canadian trademark, patent, or copyright law, which could be undermined by an infringer operating over the Internet or other global medium and willing to move among jurisdictions to continue profiting from the infringement. The most direct solution, and the one creating fewest external complications, is not jurisdictional overreach, but rather cooperation between states in enforcing the evasion of judicial injunctions, such as might occur through a law enforcement cooperation treaty. Mutual assistance treaties are fairly common in criminal law enforcement.223 Extending that model to judicial orders pertaining to intellectual property rights would not face any insuperable barriers.

The U.S. Supreme Court majority opinion in Kirtsaeng furnishes another example. One reason the Court seems to have evaded its own presumption against extraterritoriality in that case was the perceived difficulty of owners of works published under a foreign copyright but now in the U.S. The Court found the submissions of the American Library Association persuasive on this point:

[L]ibrary collections contain at least 200 million books published abroad (presumably, many were first published in one of the nearly 180 copyright-treaty nations and enjoy American copyright protection under 17 U.S.C. § 104, . . . ); that many others were first published in the United States but printed abroad because of lower costs; and that a geographical interpretation will likely require the libraries to obtain permission (or at least create significant uncertainty) before circulating or otherwise distributing these books. . . .

How, the American Library Association asks, are the libraries to obtain permission to distribute these millions of books? How can they find, say, the copyright owner of a foreign book, perhaps written decades ago? They may not know the copyright holder's present address.224

The same point was made regarding U.S. persons who purchased new or used books from foreign states, and imported goods incorporating foreign-published software.225 This might be a persuasive argument to some if the case for interpreting exhaustion as international in nature were a strong one. But it was not, leaving the Court with a mere argumentum ab inconvenienti.

A more general problem might arise from the lack of clarity with which IP licenses accompany goods sold to consumers. In commercial IP licenses, territorial restrictions are commonly explicit. In consumer sales, in contrast, there is frequently no explicit license (for example, in book sales, digital music sales, sales of goods bearing trademarks, or sales of products embodying patented technologies), and so purchasers of a foreign-produced good may not reasonably be able to ascertain whether they have a global license or merely a license for the country of purchase.226 The ambiguity is a function of the seller’s failure to specify the terms of the contract, specifically of the IP license. The situation is obviously different for commercial purchasers, who can be expected to have sufficient legal sophistication to seek clarity in the license terms.

Yet, these are unpersuasive reasons for ignoring the territorial independence of IP legislation. In the first place, much of this concern is misguided for either practical or legal reasons. The customs regulations of the U.S. already exempt from IP enforcement certain non-commercial goods infringing copyrights and trademarks imported through personal luggage accompanying a traveler.227 For other situations, such doctrines as fair use alleviate much of the concern, and in still other cases, for practical reasons involving the high cost of enforcing legal rights in the U.S., enforcement of the IP rights by foreign owners will mostly be limited to cases of egregious abuse.

More importantly, to the extent the concerns are well-founded, they give rise to policy arguments best addressed by Congress, which can carve out broader fair use and other exceptions for imported goods that do not undermine domestic IP markets (as Mr. Kirtsaeng was clearly trying to do). As noted, Congress has already done so with respect to some laws affecting some classes of imports; there is no reason to doubt that Congress could be persuaded to expand such exceptions to account for the difficulties of obtaining foreign IP licenses in meritorious cases, to the extent consistent with the IP treaties to which the U.S. is a party. For example, it would not be unreasonable for Congress to adopt a statutory exception, or even for courts to carve out a judicial exception, for non-commercial purchasers who are not adequately notified by the foreign seller’s license terms that the use or importation license is restricted to the country of sale. Such solutions do not require abrogation of the territorial independence principle, with all the jurisprudential baggage a policy of ignoring the jurisdictional limits of IP rights lugs around.

V. Conclusion

There are no global patents, copyrights or trademarks. Intellectual property is the creature of municipal legislation or, in a few places, regional treaty systems. This is the basis of the territorial independence principle. When the principle is ignored, as U.S. courts have repeatedly done, it creates inevitable confusion and contradiction in both IP law and in jurisdictional doctrines. The U.S. jurisprudence relating to the extraterritorial application of IP laws and to parallel imports does not need mere tidying up or minor adjustment to resolve the occasional contradiction or implausible holding. It needs fundamental reform to conform to both international law and federal common law principles of statutory interpretation.

Both extraterritorial infringement and parallel import doctrines have been corrupted in every field of IP, and the damage can be repaired only by the overturning of existing Supreme Court and circuit court precedents and reassertion of the territorial independence principle as the starting point for legal analysis. It must be recognized that no state in the world purports to adopt IP laws that grant patents, trademarks or copyrights within foreign territory, or that purport to be generally enforceable in other states. If the U.S. Congress were to adopt legislation in defiance of this uniform practice, it can safely be assumed that the statute would be very explicit in its repudiation of traditional jurisdictional limitations. Because it has never done so, to interpret U.S. IP statutes as if Congress implicitly defied global practice and sought to apply U.S. IP law in foreign territory, much less to make foreign IP laws applicable in U.S. territory, defies basic jurisdictional doctrines.

There may be policy reasons for a legislature to adopt an international IP exhaustion doctrine. Hostility to the segmentation of global markets in IP by business firms could be one such reason. The European Union has adopted general international IP exhaustion legislation within the common market for that very reason,228 but it has not adopted a similar policy for IP marketed outside the EU. The U.S. Congress has also shown no clear propensity to limit the ability of U.S. IP owners to segment international markets, notwithstanding the Supreme Court’s insinuation to that effect by referring to antitrust doctrines disfavoring market segmentation.229 On the contrary, Congress long ago very clearly stipulated that ordinary antitrust concerns do not apply to foreign markets unless the effects of the commercial practices rebound onto the U.S. market or U.S. exporters.230  

It therefore cannot be inferred from U.S. antitrust law whether Congress did or did not intend to adopt an international IP exhaustion doctrine. Given that such a doctrine would constitute an exception to the territorial independence principle and violates the Court’s own presumption against extraterritoriality, it makes for very poor legal reasoning to adopt such a statutory interpretation and to announce that it is for Congress to adopt new IP legislation to rectify the fundamental doctrinal problems that courts themselves have created.

• 1See Breard v. Greene, 523 U.S. 371, 376 (1998) (per curiam); Chae Chan Ping v. United States, 130 U.S. 581, 600 (1889).
• 2See infra discussion in Part II.A.1.
• 3Murray v. The Schooner Charming Betsy, 6 U.S. 64, 118 (1804) [hereinafter The Charming Betsy].
• 4Equal Emp. Opportunity Comm’n v. Arabian Am. Oil Co., 499 U.S. 244, 248 (1991) [hereinafter EEOC v. Aramco].
• 5SeeAaron X. Fellmeth, Introduction to International Business Transactions 109–10 (2d ed. 2024); Aaron X. Fellmeth, Introduction to International Business Transactions 107–08 (2020).
• 6344 U.S. 280 (1952).
• 7600 U.S. 412 (2023).
• 8EEOC v. Aramco, 499 U.S. 244.
• 9Id. at 248 (quoting Foley Bros., Inc. v. Filardo, 336 U.S. 281, 285 (1949)).
• 10Id.
• 11Congress passed legislation amending Title VII and superseding the Court’s holding promptly upon publication of the Court’s Aramco decision. Civil Rights Act of 1991, Pub. L. No. 102-166, § 102, § 1977a(b)(4), 105 Stat. 1071, 1073 (codified as amended at 42 U.S.C. § 1981a(b)(4)).
• 12EEOC v. Aramco, 499 U.S. at 249 (quoting 42 U.S.C. § 2000e(g) (1990)).
• 13Id. at 250–51.
• 14Id. at 258.
• 15Id. at 248.
• 16The Charming Betsy, 6 U.S. at 118.
• 17E.g., F. Hoffmann-La Roche Ltd. v. Empagran S.A., 542 U.S. 155, 164 (2004); Weinberger v. Rossi, 456 U.S. 25, 32 (1982); McCulloch v. Sociedad Nacional de Marineros de Honduras, 372 U.S. 10, 20–21 (1963).
• 18In EEOC v. Aramco, the Court’s reasoning perhaps unwittingly, and in any case ironically, refuted its own conclusion. International human rights law forbids discrimination based on race, religion, and national origin. See Convention on the Elimination of All Forms of Discrimination Based on Race arts. 1–3, 5, opened for signature Dec. 21, 1965, 660 U.N.T.S. 195; Universal Declaration of Human Rights arts. 2, 7, G.A. Res. 217A (III), U.N. Doc. A/810, at 71 (1948). By endorsing such discrimination by U.S. employers, the Court’s interpretation of the Civil Rights Act tended to put the U.S. in conflict with international law and thereby to put the political branches in conflict with foreign states.
• 19John H. Knox, A Presumption Against Extrajurisdictionality, 104 Am. J. Int’l L. 351 (2010).
• 20See The Case of the S.S. Lotus (Fr. v. Turk.), Judgment, 1927 P.C.I.J. (ser. A) No. 10 (Sep. 7). In the S.S. Lotus Case, the Permanent Court of International Justice found no inconsistency with international law in a Turkish exercise of criminal jurisdiction over a French national aboard a French boat that had collided with a Turkish boat on the high seas. According to the court, international law leaves states “a wide measure of discretion which is only limited in certain cases by prohibitive rules; as regards other cases, every State remains free to adopt the principles which it regards as best and most suitable.” Id. at 19. Because it could identify no clear rule prohibiting the extraterritorial application of criminal law in an accident harming a Turkish boat outside any state’s jurisdiction, the exercise of that jurisdiction was approved. Id. at 19–32.
• 21See Restatement (Third) of the Foreign Rel. L. of the U.S. § 402 (A.L.I. 1987).
• 22See Knox, supra note 19, at 379–83.
• 23See id.
• 24See id. at 383–96.
• 25Morrison v. Nat’l Austl. Bank Ltd., 561 U.S. 247 (2010).
• 26Kiobel v. Royal Dutch Petroleum Co., 569 U.S. 108 (2013).
• 27RJR Nabisco v. European Community, 579 U.S. 325 (2016).
• 28See discussion in Part II.B.4, infra.
• 29344 U.S. 280.
• 30United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90, 100 (1918).
• 31Steele v. Bulova Watch Co., 344 U.S. 280, 284-85 (1952).
• 32Steele, 344 U.S. at 290 (J. Reed, dissenting) (citing 15 U.S.C. § 1114) (internal quotations omitted).
• 3315 U.S.C. § 1127.
• 34Steele, 344 U.S. at 284–85.
• 35Id. at 286.
• 36Id. at 285–86.
• 37Id. at 286.
• 38See generally Douglas A. Irwin, Clashing over Commerce: A History of U.S. Trade Policy 509–64 (2017) (describing congressional strategies for increasing U.S. exports from 1950 to 1979).
• 3915 U.S.C. § 1114(1)(a).
• 4019 U.S.C. § 1526(d) exempts from infringement a traveler who imports a single product bearing a trademark that is counterfeit or confusingly similar to a U.S. registered trademark, if the product accompanies the traveler and is for personal use.
• 41Steele, 344 U.S. at 285.
• 42Id. at 290–91, (J. Reed, dissenting) (citing Foley Bros., 336 U.S. at 285).
• 43Id. at 285.
• 44EEOC v. Aramco, 499 U.S. at 252.
• 45Id. at 252–53.
• 46In this specific context, it could also reasonably be argued that, because Section 44 of the Lanham Act specifically guarantees both foreign nationals and U.S. citizens the benefit of “any convention or treaty relating to trademarks, trade or commercial names, or the repression of unfair competition, to which the United States is also a party . . . .” 15 U.S.C. § 1126(b), the Court’s interpretation of the jurisdictional language in Steele directly contradicts the Act’s language. The Paris Convention, to which Austria, the U.S., and for that matter Mexico are parties, provides for the independence of national trademarks, as discussed hereinbelow in Part II.B.2. The principle of statutory construction known as lex specialis derogat legi generali strongly suggests that distinct statutory provisions such as this should override more general, conflicting statutory language such as that relied upon by the Supreme Court in Steele and Abitron. See, e.g., Basic Controlex Corp. v. Klockner Moeller, 202 F.3d 450, 453–54 (1st Cir. 2000). The author thanks Professor Margaret Chon for drawing his attention to this point.
• 47U.S. Const. art. I, § 8, cl. 3.
• 48E.g., Glen v. American Airlines, Inc., 7 F.4th 331, 335 (5th Cir. 2021), cert. denied, 142 S.Ct. 863 (2022) (not questioning the constitutionality of Title III of the Helms-Burton Act, which creates a right of action by U.S. persons to sue foreign nationals who trade in Cuban property formerly—but no longer—owned by the U.S. persons).
• 49EEOC v. Aramco, 499 U.S. at 248.
• 50835 F.3d 960, 971 (9th Cir. 2016). As Margaret Chon has observed with decorous understatement, that decision was “somewhat startling, even when judged against the relatively liberal extraterritorial application of the Lanham Act that it and other courts have employed.” Margaret Chon, Kondo-ing Steele v. Bulova: The Lanham Act’s Extraterritorial Reach via the Effects Test, 25 B.U. J. Sci. & Tech. L. 530, 535 (2019).
• 51600 U.S. 412.
• 52Id. at 415–16.
• 53Id. at 416.
• 54Id.
• 55Id.
• 56Id.
• 57Id. at 418.
• 58RJR Nabisco, 579 U.S. at 336.
• 59Abitron, 600 U.S. at 419.
• 60Id.
• 61Id. at 419–20.
• 62Id. at 420.
• 63Steele, 344 U.S. 280.
• 64Abitron, 600 U.S. at 420.
• 65Id.
• 66Id. at 422.
• 67As for actual confusion, although Bulova’s representatives testified that they had received “upwards of five hundred” complaints about Steele’s products, no records or other verifiable evidence corroborating the self-interested testimony presented: Bulova Watch Co. v. Steele, 194 F.2d 567, 571 (5th Cir. 1952).  Indeed, in the Steele opinion, the Court justifiably declined to rely explicitly on the alleged evidence of actual confusion.
• 68See Société des Produits Nestlé, S.A. v. Casa Helvetia, Inc., 982 F.2d 633, 641 (1st Cir. 1992) (finding that the threshold of materiality must be kept low enough to take account of potentially confusing differences that are not blatant enough to make it obvious to the average consumer that the origin of the product differs from their expectations); Montres Rolex, S.A. v. Snyder, 718 F.2d 524, 530–31 (2d Cir. 1983) (finding it to be “well-settled” that sufficient similarity should be viewed from the standpoint of the average consumer and not an expert); Checkpoint Sys., Inc. v. Check Point Software Tech., Inc., 269 F.3d 270, 281 (3d Cir. 2001) (finding that adding generic trademark terms to dominant trademark terms is usually insufficient to lead the average consumer to disassociate between products); Select Comfort Corp. v. Baxter, 996 F.3d 925, 933 (8th Cir. 2021) (finding the average consumer factor to be one within a list of nonexclusive factors); Fleischmann Distilling Corp. v. Maier Brewing Co., 314 F.2d 149, 155 (9th Cir. 1963) (quoting Stork Restaurant v. Sahati, 166 F.2d 348, 359 (9th Cir. 1948) to the effect that the law seeks to protect—amongst others—the ignorant, the unthinking and the credulous); Frehling Enters., Inc. v. International Select Group, Inc., 192 F.3d 1330, 1338 (11th Cir. 1999) (focusing the test on whether the average consumer would hold the reasonable belief that the products could be attributed to the same source).
• 69Justice Jackson, in her concurring opinion, appeared to repudiate Steele. She observed that the mere fact that U.S. persons travel to a foreign state to buy goods bearing a foreign trademark confusingly similar to a U.S. trademark, and import them back into the U.S. for personal use, does not constitute a “use in commerce” under the Lanham Act. Abitron, 600 U.S. at 429–30 (2023) (J. Jackson, concurring). In contrast, Justice Sotomayor, in her concurring opinion, threatened to aggravate the problem by implying that foreign conduct could violate the Lanham Act if any U.S. consumer confusion resulted, however indirectly, by “filter[ing]” through the U.S. border. Id.at 432, 437–38 (J. Sotomayor, concurring). This extreme view of the extraterritorial reach of the Lanham Act did not, however, command a majority.
• 70585 U.S. 407 (2018).
• 7135 U.S.C. § 271(a).
• 72Id. § 284.
• 73Carnegie Mellon Univ. v. Marvell Tech. Group, Ltd., 807 F.3d 1283, 1306–08 (Fed. Cir. 2015) (finding that the presumption against extraterritoriality demanded more than the mere accrual of damages as a result of foreign activity); Power Integrations, Inc. v. Fairchild Semiconductor Int’l, Inc., 711 F.3d 1348, 1371 (Fed. Cir. 2013) (finding it to be unreasonable to apply U.S. law to foreign conduct and harm).
• 7435 U.S.C. § 271(f)(2).
• 75WesternGeco L.L.C. v. Ion Geophysical Corp., 791 F.3d 1340 (Fed. Cir. 2015).
• 76579 U.S. at 328.
• 77WesternGeco, 585 U.S. at 413.
• 78Id. at 416–17.
• 79Id. at 418 (J. Gorsuch, dissenting).
• 80Id. at 422 (emphasis in original).
• 81Id. at 423 (emphasis added).
• 82550 U.S. 437, 456 (2007) (“[F]oreign law alone, not United States law [governs infringement] of patented inventions in foreign countries.”).
• 83This is the economic term, not to be confused with the antitrust law concept of “price discrimination” which relates more specifically to anticompetitive distinctions in price that affect a unified market. See 15 U.S.C. § 13(a) (prohibiting price discrimination between different purchasers “of like grade and quality” where the “effect is substantially to lessen competition” in the U.S. market).
• 84See Richard Schmalensee, Output and Welfare Implications of Monopolistic Third-degree Price Discrimination, 71 Am. Econ. Rev. 242 (1981).
• 85See Yongmin Chen & Marius Schwartz, Differential Pricing When Costs Differ: A Welfare Analysis, 46 RAND J. Econ. 442 (2015); Simon Cowan, Third-Degree Price Discrimination and Consumer Surplus, 60 J. Indus. Econ. 333 (2012); Inaki Aguirre, Simon Cowan & John Vickers, Monopoly Price Discrimination and Demand Curvature, 100 Am. Econ. Rev. 1601, 1611 (2010) (“The main results are that welfare is higher with discrimination when inverse demand in the low-price market is more convex than that in the other market and the price difference with discrimination is small.”); Takeshi Ikeda & Tsuyoshi Toshimitsu, Third-degree price discrimination, quality choice, and welfare, 106 Econ. Letters 54 (2010). Another factor that could increase the net economic advantage of geographic market segmentation is the diminishing marginal utility of wealth. See generally Gabriel Demange & Guy Laroque, Finance and the Economics of Uncertainty 71–72 (2006) (explaining the concept of diminishing marginal utility of wealth in the context of risk aversion). Even if market segmentation lowers prices for less wealthy consumers only to the extent that it equivalently raises prices for more wealthy consumers (which is not necessarily the case), the diminishing marginal utility of wealth suggests that this arrangement actually increases net consumer utility.
• 86In U.S. jurisprudence, these terms are effectively merged. Susy Frankel and Daniel Gervais have argued that they relate to distinct concepts, with only the first sale doctrine relating to parallel imports, see Susy Frankel & Daniel Gervais, International Intellectual Property Rules and Parallel Imports, in Research Handbook on Intellectual Property Exhaustion and Parallel Imports 85, 88–89 (Irene Calboli & Edward Lee eds., 2016). However, this proposal does not yet seem to have influenced U.S. jurisprudence on the point.
• 87Adams v. Burke, 84 U.S. 453, 456 (1873) (footnotes omitted).
• 88Prestonettes, Inc. v. Coty, 264 U.S. 359, 368 (1924) (“When the mark is used in a way that does not deceive the public we see no such sanctity in the word as to prevent its being used to tell the truth.”).
• 8917 U.S.C. § 109(a).
• 90Because piratical copies are by definition not lawfully made under the Copyright Act, section 109(a) precludes the owner of a piratical copy in the U.S. from reselling the copy to another person without the copyright owner’s consent.
• 91Copyright Act of 1976, Pub. L. No. 94-553, 90 Stat. 2541, codified as amended at 17 U.S.C. §§ 101-810.
• 92Prioritizing Resources and Organization for Intellectual Property Act of 2008, Pub. L. No. 110–403, tit. I, § 105(b), (c)(1), Oct. 13, 2008, 122 Stat. 4259, 4260.
• 93499 U.S. 244, 248.
• 94523 U.S. 135 (1998).
• 95Id. at 138–40.
• 9617 U.S.C. § 602(a).
• 9717 U.S.C. § 106(3); see Copyright Law Revision: Report together with additional views, H.R. Rep. No. 94-1476, at 61–62, 79–81 (1976) reprinted in 1976 U.S.C.C.A.N. 5659. Indeed, the legislative history suggests independent legal authority for enforcing section 602 violations that does not rely on sections 109 or 106: “The importation prohibitions of both sections 601 and 602 would be enforced under section 603, which is similar to section 109 of the statute now in effect.” Id. at 170; but see id. at 169–70, drawing a clear distinction between piratical copies that the Customs can exclude from importation and copies “lawfully made but their distribution in the United States would infringe the U.S. copyright owner’s exclusive rights[,]” which could be enjoined as the result of a private action. “[T]he mere act of importation in this situation would constitute an act of infringement and could be enjoined.”)
• 98In brief, first, the Court speculated that section 602(a) could have been intended to add “a private remedy against the importer” of a piratical copy. 523 U.S. at 146, as opposed to anyone selling or otherwise distributing a piratical copy. The Court interpreted the reference in section 602(a) to section 106 and the “right to distribute copies” as intended to mean that the exclusive importation right is a form of distribution. Id. at 149. It is not. As the Solicitor General observed, importation is neither a sale nor a “disposal.” Id. at 152. “Disposal” and “distribution” in the Copyright Act require some form of transfer of title (e.g., “sale or other transfer or ownership” under 17 U.S.C. § 106(3)). Importing copies is not a form of “disposing” of them or “distributing” them, it is the administrative act of entering them into the customs territory of the U.S. in compliance with U.S. customs regulations (which is why the statute gives U.S. Customs & Border Protection the authority to interdict the importation of copies not “lawfully made” under Title 17). 17 U.S.C. § 602(b). If the owner of copies had the unhindered right to import those copies in the U.S., then section 109(a) could ensure the importer the right to subsequently sell them, gift them, or otherwise transfer their ownership (“distribute” or “dispose of” them) to another. But section 602(a) does not assimilate importation to distribution, it merely treats the unauthorized importation of copies acquired outside the U.S. as infringement of the copyright owner’s exclusionary rights. 

  The Court replied that “In a typical commercial transaction, however, the shipper transfers ‘possession, custody, control and title to the products’ to a different person . . . .” 523 U.S. at 152. This assertion is inaccurate. Whether importation involves a transfer of title is purely a question of the contract, if any. Often there is none; the exporter and the importer are the same person, and it would be nonsensical to equate importation to transfer of title. One cannot transfer title to oneself. Other times, goods will be imported on consignment (meaning no transfer of title occurs), as the Court itself pointed out in its second argument, with unnoticed irony. Id. at 146–47. Moreover, Title 19 of the U.S. Code regulates the importation of goods into the U.S. but does not regulate their sale, which implicitly acknowledges that the concepts are unrelated. The Court’s basic error here is aggravated by the Court’s own recognition that section 501 of the Act provides a remedy for infringement against anyone who “violates any of the exclusive rights of the copyright owner as provided by sections 106 through 118 . . . or who imports copies or phonorecords into the U.S. in violation of section 602 . . . .” 17 U.S.C. § 501. As the Court admitted, that the Act treats infringement by importation as distinct from infringement by reproduction or sale strongly supports the conclusion that Congress intended section 602(a) to provide a remedy for imports lawfully made in the U.S. but not authorized for importation. 523 U.S. at 149.

  The Court’s second argument was that, because section 109(a) provided a defense only to the “owner” of a lawfully made copy, it could be that section 602(a) was intended to authorize the copyright owner to prevent the importation of lawfully made copies by non-owners, such as bailees, licensees, or consignees. Id. at 146–47. Aside from the Court apparently misapprehending what a licensee is (any right a licensee may have to import is determined by the license itself), with regard to bailees or consignees, it seems very improbable that Congress adopted section 602(a) primarily to lump copyright pirates together with authorized bailees and consignees, and to allow the copyright owner to block both from importing copies in their possession without permission.

  The Court’s only argument with any traction was that, if section 602(a) provided the basis of a separate right of action for importation of copies made in the U.S., the defenses to copyright infringement in sections 107 through 120 of the Copyright Act would not apply to the importer, because these defenses make no mention of section 602(a). For example, the importer could not benefit from the fair use defense. Id. at 150–51. This is a valid point, but it seems likely that Congress simply overlooked this fact in adopting section 602(a), intending or expecting the defenses to apply without considering how the language of section 602(a) would interact with them. Section 602(a) was not adopted as part of a comprehensive revision of the Copyright Act, and absent such revision, Congress commonly amends complex statutory regimes without fully considering the consequences of the amendment for each and every provision of the regime.

• 99523 U.S. at 147.
• 100Id. at 145.
• 101Id.
• 102568 U.S. 519 (2013).
• 103Id. at 530–32.
• 104Quality King, 523 U.S. at 547–48; see id. at 554, 555–56 (J. Kagan, concurring).
• 105Id. at 563 (J. Ginsburg, dissenting).
• 106Copyright Act B.E. 2537 (พระราชบัญญัติลิขสิทธิ์ พ.ศ. 2537), ch. 8 (1994) (Thai.), available in English at https://perma.cc/78TK-UGXG (last visited Oct. 21, 2025).
• 107Isbrandtsen Co. v. Johnson, 343 U.S. 779, 783 (1952).
• 108In Part II.D, the Court discussed the policy consequences should section 109 not be applied extraterritorially. No doubt some of these policy consequences would be regrettable, but they were not relevant to the case at hand, and in any event could be avoided by congressional amendment to the Act to clarify the scope of the section 109 defenses.
• 109Boesch v. Graff, 133 U.S. 697, 703 (1890).
• 11035 U.S.C. § 154(a).
• 111581 U.S. 360 (2017).
• 112Id. at 379 (quoting Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 439 (1984)). Although the Court has invoked this “historic kinship” in several cases, it has never clarified what it means, possibly because the Court itself does not know. Article 1, Section 8 of the Constitution does authorize Congress to grant both patents and copyrights, but aside from that coincidence and the intangible nature of the subject matter, copyrights and patents have very little in common. The two fields deal with different kinds of IP in a different manner, the former focusing on industrial or commercial utility and the latter on the opposite—human expression lacking practical functionality. Very different statutes define the respective areas of law, and each area is regulated by different international treaties (except for the TRIPS Agreement, which provides rules of international law for all major forms of intellectual property, but each in separate sections). As a result, the two fields have diverged greatly in doctrines. In consequence, copyright lawyers very rarely practice patent law, and patent lawyers very rarely practice copyright law. If the two fields are kin, they are third cousins. On the divergence of copyright and patent, see generally Edward C. Walterscheid, Divergent Evolution of the Patent Power and the Copyright Power, 9 Marq. Intell. Prop. L. Rev. 307, 356 (2005); David W. Barnes, Abuse of Supreme Court Precedent: The “Historic Kinship,” 16 Chi.-Kent J. Intell. Prop. 85 (2016); Mark Bartholomew & John Tehranian, Historical Kinship & Categorical Mischief: The Use and Misuse of Doctrinal Borrowing in Intellectual Property Law, 109 Iowa L. Rev. 51 (2023).
• 113Impression Prods., 581 U.S. at 378–80.
• 114Id. at 382 (J. Ginsburg, dissenting).
• 115Id. at 382–83 (citing Deepsouth Packing Co. v. Laitram Corp., 406 U.S. 518, 531 (1972)).
• 116Id. at 383.
• 117Aaron X. Fellmeth, Uncreative Intellectual Property Law, 27 Tex. Intell. Prop. L.J. 51, 85–87 (2019).
• 118Id. at 87–88.
• 119Id. at 88–90.
• 120Christian Louboutin S.A. v. Yves Saint Laurent Am. Holding, Inc., 696 F.3d 206, 224 n.20 (2d Cir. 2012).
• 12115 U.S.C. § 1124.
• 122Lever Brothers v. United States, 877 F.2d 101 (D.C. Cir. 1989).
• 123Id. at 102–03.
• 12415 U.S.C. § 1124.
• 125Lever Brothers, 877 F.2d at 103.
• 12619 C.F.R. § 133.21(b) (1988).
• 127Lever Brothers, 877 F.2d at 105.
• 128See generally A. Bourjois & Co. v. Katzel, 260 U.S. 689 (1923); A. Bourjois & Co. v. Aldridge, 263 U.S. 675 (1923) (per curiam); Prestonettes, Inc. v. Coty, 264 U.S. 359 (1924).
• 129Lever Brothers, 877 F.2d at 106.
• 130Id.
• 131Id. at 108.
• 132See id. at 109–10.
• 133Id. at 111.
• 13419 C.F.R. § 133.23(b).
• 135See 423 F.3d 1307–10, 1312, 1317 (Fed. Cir. 2005), cert. denied, 548 U.S. 904 (2006).
• 136See id. at 1313.
• 137Beltronics USA, Inc. v. Midwest Inventory Distrib., LLC, 562 F.3d 1067, 1072 (10th Cir. 2009).
• 138See id. at 1076.
• 139Id. at 1069.
• 140Id.
• 141Id.
• 142Id. at 1069–71, 1073.
• 143See id. at 1072.
• 144See id. at 1073.
• 145The Tenth Circuit follows the Eleventh Circuit in interpreting “material” as meaning the difference “influences the decision whether to purchase the product.” Id. at 1076.
• 146Id. at 1074.
• 147Id. at 1074–75.
• 148See Restatement (Fourth) of the Foreign Rels. L. of the U.S. § 408 & Rep. Note 1 (Am. L. Inst. 2018); Restatement (Third) of the Foreign Rels. L. of the U.S. §§ 401–04, 421, 431 (Am. L. Inst. 1987).
• 149See The Case of the S.S. “Lotus” (Fr. v. Turk.), Judgment, 1927 P.C.I.J. (ser. A) No. 10 at 18.
• 150Id. at 19.
• 151See Restatement (Fourth) of the Foreign Rels. L. of the U.S. §§ 408–10 (2018). States do sometimes invoke other bases for jurisdiction, such as passive personality jurisdiction or protective jurisdiction, see id. §§ 411–13, but these are not normally relevant to IP laws.
• 152E.g., 35 U.S.C. § 271(c),(f),(g). Arguably, the Supreme Court applied effects jurisdiction to find infringement of a U.S. trademark in Steele, 344 U.S. 280. This is Margaret Chon’s interpretation of Steele, which is reasonable in light of the Court’s literal reference to the defendant’s “operations and their effects” in the U.S., and it is also the interpretation that the Ninth Circuit seems to have adopted in Trader Joe’s, 835 F.3d 960. See Chon, supra note 50 at 108. That said, the Court’s conclusion in Steele was based on an assumption that Congress used generic language in the Lanham Act with the intention of giving the Act global effect, not on any specific language in the Act. The wording of the Act is focused on the “use” of the mark to cause mistake or deception, not on the result of confusion occurring in the U.S. regardless of where or how that confusion arises. It is not, in short, written as an Act to Prevent the Global Use of Trademarks Likely to Cause Confusion to U.S. Consumers.
• 153“Almost,” because a few inhabited territories remain artifacts of global colonialism and are neither independent states themselves nor integrated into states, such as Greenland (Denmark), Puerto Rico and Guam (U.S.), the Cayman Islands and Montserrat (U.K.), the Tahitian Islands and New Caledonia (France), and Tokelau (New Zealand).
• 154See Montevideo Convention on the Rights and Duties of States, art. 1, Dec. 26, 1933, 165 L.N.T.S. 19, O.A.S.T.S. No. 37.
• 155See 1 Lassa Oppenheim, International Law, A Treatise § 170 (Hersch Lauterpacht ed., 8th ed., 1955) (“The importance of State territory lies in the fact that it is the space within which the State exercises its supreme authority.”); see also The Schooner Exchange v. McFaddon, 11 U.S. (7 Cranch) 116, 136 (1812) (referring to “that absolute and complete jurisdiction within their respective territories which sovereignty confers”).
• 156International law and treaties can set limits to the authority of each state to legislate, for example through international human rights law or international trade law. To the extent that such limits apply to IP legislation, this will be the subject of Part III.B, below.
• 15735 U.S.C. § 271(a) (emphasis added).
• 158See35 U.S.C. § 271(f)(1); see also 35 U.S.C. § 271(c),(f)(2),(g) (providing for other forms of extraterritorial infringement).
• 159See 17 U.S.C. § 106.
• 160See 17 U.S.C. § 501(a).
• 161See United Dictionary Co. v. G & C Merriam Co., 208 U.S. 260, 264 (1908).
• 162See Willam F. Patry, 7 Patry on Copyright § 25:86 (Sept. 2025) (“Every court to have examined the issue has held that Congress did not intend the Copyright Act to be applied extraterritorially, beginning with the Supreme Court in 1908.”). Patry does not mention Kirtsaeng, however.
• 16315 U.S.C. § 1114(1).
• 16415 U.S.C. § 1127.
• 165See generally Robert Butts, Trademark Law: Interpreting the Congressional Intent of the Extraterritorial Applications of the Lanham Trademark Act, 8 Fla. J. Int’l L. 447 (1993) (describing the circuit split relating to the extraterritorial application of the Lanham Act as of 1993).
• 166See, e.g., Hetronic Int’l, Inc. v. Hetronic Ger. GmbH, 99 F.4th 1150, 1166–67 (10th Cir. 2024).
• 167RJR Nabisco v. European Cmty., 579 U.S. 325, 335 (2016).
• 168U.S. Const., art. I, § 8, cl. 3.
• 169Id. art. I, § 8, cl. 8.
• 170Ingenohl v. Walter E. Olsen & Co., 273 U.S. 541, 544 (1927).
• 171Brown v. Duchesne, 60 U.S. 183, 195 (1856).
• 172Kirtsaeng, 568 U.S. at 573 (J. Ginsburg, dissenting).
• 173See generally William S. Dodge, The New Presumption Against Extraterritoriality, 133 Harv. L. Rev. 1582 (2020) (describing inter alia the evolution of the presumption against extraterritoriality in U.S. jurisprudence and its recent revival).
• 174Agreement on Trade-Related Aspects of Intellectual Property Rights, Annex 1C, Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations, Apr. 15, 1994, 1869 U.N.T.S. 299 [hereinafter TRIPS Agreement].
• 175See Patent Law Treaty, June 1, 2000, 2340 U.N.T.S. 3.
• 176See Patent Cooperation Treaty, June 19, 1970, 1160 U.N.T.S. 231.
• 177Berne Convention on the Protection of Literary and Artistic Works, May 4, 1896, 828 U.N.T.S. 221. The relevant provisions of the Berne Convention are incorporated into the TRIPS Agreement by reference. TRIPS Agreement, supra note 174, art. 9.
• 178See TRIPS Agreement, supra note 174, Pt. II, Sec. 1.
• 179See WIPO Copyright Treaty, Dec. 20, 1996, 2186 U.N.T.S. 121.
• 180Treaties may also have intergovernmental organizations as parties, a point not relevant to this discussion.
• 181For a nuanced and empirical explanation of the range of state practice with regard to treaty implementation, see generally Pierre-Hugues Verdier & Mila Versteeg, International Law in National Legal Systems: An Empirical Investigation, 109 Am. J. Int’l L. 514 (2015).
• 182See 1 Georges Scelles, Précis de droits des gens 43, 54–56, 217 (1932); 2 Georges Scelles, Précis de droits des gens 10, 319, 450 (1934).
• 183Paris Convention for the Protection of Industrial Property arts. 4bis, 6, Mar. 20, 1883, 828 U.N.T.S. 305; see also Vanity Fair Mills Inc. v. T. Eaton Co., 234 F.2d 633, 640 (2d Cir.), cert. denied, 352 U.S. 871 (1956) (“The [Paris] Convention is not premised upon the idea that the trade-mark and related laws of each member nation shall be given extraterritorial application, but on exactly the converse principle that each nation's law shall have only territorial application.”).
• 184TRIPS Agreement, supra note 174, art. 2.
• 185Id. art. 9.
• 186Berne Convention supra note 177, art. 5(2).
• 187Id. art. 5(2).
• 188TRIPS Agreement, supra note 174, art. 9.
• 189Impression Prods., 581 U.S. at 384 (J. Ginsburg, dissenting) (citations omitted).
• 190Berne Convention, supra note 177, art. 5(2) (emphasis added).
• 191See 17 U.S.C. § 104(a) (“The works specified by sections 102 and 103, while unpublished, are subject to protection under this title without regard to the nationality or domicile of the author” clearly implies that the protection is granted under the 1976 Copyright Act).
• 192See, e.g., Alvaro Fernandez-Mora, Inconsistencies in European Trade Mark Law: The Public Policy and Morality Exclusions, 4 Intell. Prop. Q. 271, 272–73 (2020) (describing examples of trademarks successfully registered by the EU Intellectual Property Office but denied by the U.K. Intellectual Property Office).
• 193See generally Päivi Hutukka, Patent law in comparative context: Differences and similarities of patent law in the European Union, the United States and China, 30 Maastricht J. of Eur. & Comp. L. 273 (2023) (discussing differences in patent standards, procedures, and acceptable claim language in the U.S., Europe, and China).
• 194See Ioana Vasiu & Lucian Vasiu, Cross-Jurisdictional Analysis of Damage Awards in Copyright Infringement Cases, 28 J. Intell. Prop. L. 93, 97 (2021).
• 195See Magdalena Christoforou,Should Patent Infringement Be Criminal?, JIPEL Blog, June 1, 2020, https://perma.cc/Q5NK-U5FX.
• 196E.g., Berne Convention, supra note 177 arts. 2(2), 2(7), 2bis(1); TRIPS Agreement, supra note 174, art. 17.
• 197See Tamara Nanayakkara, IP and Business: Trademark Coexistence, WIPO Magazine, Nov. 8, 2006, https://perma.cc/6J8J-EZYP.
• 198See William Patry, Choice of Law and International Copyright, 48 Am. J. Comp. L. 383, 430–32 (2000).
• 199TRIPS Agreement, supra note 174, art. 6.
• 200It bears emphasis here that U.S. courts have frequently distinguished between licenses and sales as if they were two separate and mutually exclusive forms of IP transfer. This is a misapprehension. Sale is one category of license; the other is lease. In fact, all sales and leases of products embodying IP alike include a license regulating what rights are transferred with the sale. Even when there is no explicit license, there is a license implied by law stipulating what the sale or lease permits the purchaser to do with the goods and the IP they embody. See generally Lothar Determann & Aaron Fellmeth, Don’t Judge a Sale by Its License: New Approaches to Software Transfers Under the First Sale Doctrine in the United States and Europe, 36 U.S.F. L. Rev. 1 (2001) (explaining and justifying this point in light of the contrary jurisprudence).
• 201See Banco Nacional de Cuba v. Sabbatino, 376 U.S. 398, 416 (1964); Underhill v. Hernandez, 168 U.S. 250, 252–54 (1897); R. v. Bow Street Metropolitan Stipendiary Magistrate, ex parte Pinochet Ugarte, [2000] 1 A.C. 61 (H.L.), 90, 102–04 (Berwick, L.J.) (U.K.).
• 202Restatement (Fourth) of the Foreign Rels. L. of the U.S. § 441.
• 203SeeUnderhill v. Hernandez, 168 U.S. 250, 252 (1897).
• 204Vanity Fair Mills, Inc., 234 F.2d at 638–39 (holding that, where a U.S. trademark is being litigated in a U.S. court, decisions of foreign courts concerning respective trademark rights of the parties are irrelevant and inadmissible).
• 205See, e.g., Voda v. Cordis Corp., 476 F.3d 887, 898 (Fed. Cir. 2007) (holding that “[c]onsiderations of comity, judicial economy, convenience, fairness, and other exceptional circumstances constitute compelling reasons to decline” supplemental jurisdiction over a patent holder’s foreign patent infringement claims in the same case involving claims of infringement over parallel U.S. patent claims). But see, e.g., London Film Prods., Ltd. v. Intercontinental Commc’ns, Inc., 580 F. Supp. 47, 49–50 (S.D.N.Y. 1984) (hearing a foreign copyright infringement claim based on the fact that the foreign copyright did not require registration and, in the court’s opinion, therefore implicated no act of state). The obvious mistake of the district court in London Film Productions was in assuming that, because the state does not require registration of IP, it therefore is not using its sovereign authority to grant legal rights in the IP. Registration and the grant of IP rights are separate concepts, and both are sovereign acts.
• 206See Maltina Corp. v. Cawy Bottling Co., 462 F.2d 1021, 1025 (5th Cir.), cert. denied, 409 U.S. 1060 (1972).
• 207Cf. Voda v. Cordis Corp., 476 F.3d 887, 904 (Fed. Cir. 2007) (the Federal Circuit articulated the factors weighing against a trial court assuming supplemental jurisdiction over patent infringement claims. Among these was fairness, based on the fact that the act of state doctrine “would prevent our courts from inquiring into the validity of a foreign patent grant and require our courts to adjudicate patent claims regardless of validity or enforceability”).
• 208See Fairchild Semiconductor Corp. v. Third Dimension (3D) Semiconductor, Inc., 589 F.Supp.2d 84, 99–100 (D. Me. 2008).
• 209Id. at 100.
• 210Hilton v. Guyot, 159 U.S. 113, 163–64 (1895).
• 211See Transocean Offshore Deepwater Drilling, Inc. v. Maersk Contractors USA, Inc., 617 F.3d 1296, 1309–10 (Fed. Cir. 2010).
• 212E.g., 35 U.S.C. § 271(g).
• 213See generally Timothy R. Holbrook, What Counts as Extraterritorial in Patent Law?, 25 B.U. J. Sci. & Tech. L. 291, 307–13 (2019) (discussing cases in which control and beneficial use in the U.S., and effects on the U.S. market for the patented item, justified U.S. judicial jurisdiction over claims relating to cross-border infringement of U.S. patents).
• 214See Agreement on a Unified Patent Court, June 20, 2013, 2013 OJ (C 175) 1; UPC Member States, Unified Patent Ct.  https://perma.cc/J667-4DP6 (last visited Oct. 24, 2025).
• 215Regulation 2017/1001 of the European Parliament and of the Council of 14 June 2017 on the European Union trade mark, July 16, 2017, O.J. (L 154) 1. For comparative analyses of how these doctrines have affected market integration, see Irene Calboli, Market Integration and (The Limits of) the First Sale Rule in North American and European Trademark Law, 51 Santa Clara L. Rev. 1241 (2011), and Irene Calboli, The Intricate Relationship Between Intellectual Property Exhaustion and Free Movement of Goods in Regional Organizations: Comparing the EU/EEA, NAFTA, and ASEAN, 9 Queen Mary J. Intell. Prop. 22 (2019).
• 216See Bangui Agreement Relating to the Creation of an African Intellectual Property Organization, Mar. 2, 1977, (as amended in 1999 and 2015), https://perma.cc/8KWX-XQMV.
• 217This is an explicit requirement of the TRIPS Agreement, art. 28, supra note 174, with respect to patents. With respect to trademarks, the TRIPS Agreement provides in article 16 that the trademark owner has the exclusive right to prevent all unauthorized third parties from using “identical or similar signs for goods or services which are identical or similar to those in respect of which the trademark is registered” if such use would result “in a likelihood of confusion.”
• 218Margaret Chon, Inside or Outside?: Remarks on Abitron v. Hetronic (U.S. 2023), 23 Chi.-Kent J. Intell. Prop. 149, 151 (2023).
• 219See Google Inc. v. Equustek Solutions Inc., [2017] 1 S.C.R. 824 (Can.).
• 220Id. ¶¶ 5–11.
• 221Id. ¶¶ 38–41.
• 222See generally Wesley A. Magat & W. Kip Viscusi, Informational Approaches to Regulation 87–107, 161–80 (1992) (discussing studies showing that some forms of product labeling have little effect on consumer understanding or behavior). Beyond satisfying courts or customs authorities that consumers would, in theory, be “informed” of material differences between products authorized for sale in the U.S. and products intended for foreign markets, importers have limited incentive to carefully design product labels to actually dispel consumer confusion.
• 223For example, the U.S. has mutual legal assistance treaties for criminal law enforcement with seventy-six other countries. See U.S. Dep’t of Just., Crim. Div., Office of Int’l Aff., Mutual Legal Assistance Treaties of the United States (MLATs) (2022), https://perma.cc/7H25-SVKG. Australia has such agreements with thirty-one states. See Australian Government Attorney-General’s Department, Bilateral Treaties on Mutual Assistance in Criminal Matters (2024), https://perma.cc/S3T9-Q5LQ.
• 224Kirtsaeng v. John Wiley & Sons, Inc., 568 U.S. 519, 541 (2013).
• 225See id. at 541–42.
• 226See, e.g., id. at 526. In Kirtsaeng, the license limitation was printed on the books themselves, so the defendant could not even avail himself of the claim of ignorance about the scope of the copyright license.
• 22717 U.S.C. § 602(a)(3)(B),(C); 19 U.S.C. § 1526(d). There are no such exceptions for patented goods, because U.S. Customs and Border Protection has no statutory authority to regulate the importation of patented goods in personal luggage.
• 228E.g., Directive (EU) 2015/2436 of the European Parliament and of the Council of 16 December 2015 to approximate the laws of the Member States relating to trade marks, art. 15, EU Doc. 32015L2436, O.J. L 336, Dec. 23, 2015, at 1.
• 229See Kirtsaeng, 568 U.S. at 552–53 (“Congress enacted a copyright law that [through the ‘first sale’ doctrine] limits copyright holders’ ability to divide domestic markets. And that limitation is consistent with antitrust laws that ordinarily forbid market divisions.”). The Court correctly identified the policy as relating to domestic markets, then erroneously applied the same policy without differentiation to international markets.
• 230See 15 U.S.C. § 6a.