Trademark lawyers often tell the story in one form or another of the Coca-Cola lawyer who spoke in 1986 of the value of the company’s goodwill as symbolized by its brand: “The production plants and inventories of The Coca-Cola Company could go up in flames overnight. Yet, on the following morning there is not a bank in Atlanta, New York, or anywhere else, that would not lend this Company the funds necessary for rebuilding, accepting as security only the inherent good will in its trademarks ‘Coca-Cola’ and ‘Coke.”’1 The story was and remains no exaggeration. In 2016, Interbrand estimated the value of the Coca-Cola brand to be $73.1 billion2—as against fixed assets in 2016 of approximately $10.6 billion.3
apple, google, coke, microsoft, samsung, toyota, mcdonalds, louis vuitton, nike, pepsi, facebook, visa, citi, starbucks, mastercard. Instantly recognizable by a very large proportion of humanity, these are among the most valuable and influential signs in the world, rivalling in significance many religious and national symbols. They are only the most notorious of the millions of brand names that populate the modern marketplace. Trademark law regulates these brand names, from the multi-billion dollar global brands to the name of the local shop down the street. Without trademark protection, many would cease to exist.
In this introductory chapter, we first review the early history of trademarks and trademark law, including the first great Supreme Court trademark case, the so-called Trade-Mark Cases. We then critically consider the varied policy justifications for trademark protection. We conclude by briefly situating trademark law within the larger scheme of intellectual property law.
In the excerpt that follows, Mark McKenna surveys the origins of American trademark law from seventeenth-century English case law up through nineteenth-century American case law. Some of the cases he mentions may seem far removed in their facts and reasoning from the present-day world of the global internet and multi-billion dollar brands. But as you will see, the early history of trademark law implicates questions that continue to concern courts and trademark law policymakers. What is the proper rationale for trademark protection? Are trademark rights simply a form of pernicious monopoly rights? Is trademark law intellectual property law or is it unfair competition law? What should qualify for trademark protection? What role should intent or “bad faith” play in the determination of liability for trademark infringement? Should consumers have standing to sue when they are confused by one company’s use of a mark similar to another company’s mark?
Mark P. McKenna, The Normative Foundations of Trademark Law, 82 Notre Dame L. Rev. 1839, 1849-62 (2007) (some footnotes altered or omitted)
II. A SECOND LOOK AT EARLY TRADEMARK PROTECTION
 Use of markings to identify and distinguish one's property dates to antiquity, and regulations regarding use of those marks almost as long…. Because nineteenth-century American courts explicitly drew on English law…, a full account necessarily begins in England.
A. Medieval Marks as Liabilities
 Scholars have identified a number of ways in which individuals and producers historically used distinguishing marks. Most basically, merchants used marks to demonstrate ownership of physical goods, much in the way that ranchers use cattle brands to identify their cattle. Use of marks to indicate ownership of goods was particularly important for owners whose goods moved in transit, as those marks often allowed owners to claim goods that were lost. Producers relied on identifying marks, for example, to demonstrate ownership of goods recovered at sea.4
 Marks also were quite important to the operation of the guild system in medieval England. Local guilds often developed reputations for the quality of their products. When they did, the names of the towns or regions in which those guilds operated became repositories of goodwill. To maintain that goodwill, guilds needed to be able to restrict membership and identify and punish members who produced defective products. Guilds therefore required their members to affix distinguishing marks to their products so they could police their ranks effectively.5
 Importantly, guilds required members to display their marks for the purpose of developing and maintaining the collective goodwill of the guild; marks were not used for the purpose of establishing individual producer goodwill. Indeed, intraguild competition was strictly forbidden. Moreover, guild regulations were not motivated primarily by a concern for consumers. Even in the cutlers' trade, where marks seem to have been viewed most analogously to modern trademarks,6 regulation was intended not for the protection of purchasers, but for “guidance of those exercising control or working in rivalry.”7 In fact, though it is not clear how often mark owners sought enforcement of their marks during this period, whatever enforcement mark owners did pursue seems to have been motivated by their concern about being held responsible for products they did not make.
B. English Trademark Cases
 Commentators often cite the 1618 case of Southern v. How8 as the first English case dealing with replication of another’s identifying mark. According to Popham’s report of that case,
an action upon the case was brought in the Common Pleas by a clothier, that whereas he had gained great reputation for his making of his cloth, by reason whereof he had great utterance to his great benefit and profit, and that he used to set his mark to his cloth, whereby it should be known to be his cloth: and another clothier perceiving it, used the same mark to his ill-made cloth on purpose to deceive him, and it was resolved that the action did well lie.9
 There are, however, good reasons to doubt the reliability of Popham’s report, as Frank Schechter ably demonstrated in his seminal work The Historical Foundations of the Law Relating to Trade-Marks. Popham’s is only one of five known reports of the case,10 and the other reports do not corroborate Popham’s account of the clothier’s case. Some of the reports contain no reference at all to the clothier’s case,11 and at least one of the reports suggests that it was the deceived customer who brought the action rather than the merchant.12 Nevertheless, much to the detriment of trademark law’s conceptual development, Popham’s characterization of Southern played a prominent role in early English law. In fact, several English judges deciding trademark cases in the eighteenth and nineteenth centuries relied on Popham’s report of the case for the proposition that cases based on use of another’s mark could be brought as actions on the case, sounding in deceit.
1. Trademarks in Courts of Law and Equity
 The first reported English decision clearly involving a claim based on use of a party’s trademark was the court of equity’s 1742 decision in Blanchard v. Hill.13 The plaintiff in that case, a maker of playing cards, sought an injunction
to restrain the defendant from making use of the Great Mogul as a stamp upon his cards, to the prejudice of the plaintiff, upon a suggestion, that the plaintiff had the sole right to this stamp, having appropriated it to himself, conformable to the charter granted to the cardmakers’ company by King Charles the First.14
 The factual context of Blanchard is particularly noteworthy; the plaintiff was seeking protection of a mark for playing cards pursuant to a royal charter, and charters granting exclusive rights to cardmakers had been at the center of a long political struggle between Parliament and the Crown. Marks played an important role in the contested charter scheme because cardmakers were required to use their seals so that exclusivity could be enforced,15 a fact that clearly colored the court’s view of the case….
 The Blanchard decision, however, should not be read as a categorical condemnation of claims based on use of a competitor’s mark. Rather, Lord Hardwicke was focused on cases in which the plaintiff’s claim of exclusive rights emanated from a monopoly granted by royal charter. In fact, his decision in Blanchard specifically distinguished the plaintiff’s claim in that case from the clothier’s claim referenced in Popham’s report of Southern. Unlike the plaintiff in Blanchard, who claimed the exclusive right to use his Mogul mark without qualification, the clothier in Southern based his case on the defendant’s “fraudulent design, to put off bad cloths by this means, or to draw away customers from the other clothier.”16 When the defendant intended to pass off its goods as those of the plaintiff, Lord Hardwicke implied, an injunction might well be appropriate.
 Despite the initial reluctance of courts of equity to recognize exclusive rights in trademarks and Lord Hardwicke’s clear suggestion that claimants pursue such claims at law, the first reported trademark decision by an English common law court was the 1824 decision in Sykes v. Sykes.17 In that case, the court upheld a verdict for the plaintiff against defendants who marked their shot-belts and powder-flasks with the words “Sykes Patent” in imitation of the plaintiff’s use of the same mark for its shot-belts and powder-flasks.18 After specifically noting that the plaintiff’s sales had decreased after the defendants began selling their identically labeled products, the court concluded that the defendants had violated the plaintiff’s rights by marking their goods so as “to denote that they were of the genuine manufacture of the plaintiff” and “[selling] them to retail dealers, for the express purpose of being resold, as goods of the plaintiff’s manufacture.”19
 A number of common law cases following the Sykes decision recognized claims in similar circumstances, imposing liability when a producer sought to pass off its goods as those of a competitor.20 Those cases generally were brought as actions on the case, in the nature of deceit.21 Yet one must be careful not to read those cases through modern lenses—despite the form of action, courts in these early cases invariably described the defendant as having practiced fraud against the plaintiff.22
 Like their counterparts in courts of law, courts of equity became more solicitous of trademark claims in the first part of the nineteenth century. Of particular significance, courts very early on concluded that, where a claimant could demonstrate an exclusive right to use a particular mark, equity would intervene to protect a property interest and evidence of fraudulent intent was not necessary….
 As Lord Westbury said in Leather Cloth Co. v. American Leather Cloth Co.23 rejecting any contention that courts of equity based their jurisdiction on fraud,24
The true principle, therefore, would seem to be, that the jurisdiction of the Court in the protection given to trade marks rests upon property, and that the Court interferes by injunction, because that is the only mode by which property of this description can be effectually protected.25
Significantly, Lord Westbury reached this conclusion after noting that, even when a party held out his goods as those of another, the other had no right to complain unless the act caused him some pecuniary loss or damage.26 “Imposition on the public, occasioned by one man selling his goods as the goods of another, cannot be the ground of private right of action or suit.”27 The court in Levy v. Walker28 was even more explicit that the protection of trademarks was intended to protect producers and not primarily for the benefit of consumers: “The Court interferes solely for the purpose of protecting the owner of a trade or business from a fraudulent invasion of that business by somebody else. It does not interfere to prevent the world outside from being misled into anything.”29
C. Early American Trademark Jurisprudence
1. Trademark Law Targets Dishonest Trade Diversion
 As noted above, I read the decisions of the English common law courts and courts of equity as reflecting the same fundamental concern. In both types of cases, courts were singularly focused on the harm to a producer from improper diversion of its trade, and they worked with existing forms of action to remedy that harm. American courts had the same focus when they began deciding trademark cases, and they repeatedly made clear that the purpose of trademark law was to protect a party from illegitimate attempts to divert its trade.30
 In Coats v. Holbrook,31 for example, the court said that a person is not allowed to imitate the product of another and “thereby attract to himself the patronage that without such deceptive use of such names ... would have inured to the benefit of that other person.”32
 […] Moreover, ... American courts concluded very early on that this protection in many cases was based on a property right,33 following essentially the approach of English courts of equity.
2. Trademarks and Unfair Competition
 Because the purpose of trademark protection traditionally was to prevent trade diversion by competitors, it has long been regarded as a species of the broader law of unfair competition, and even more broadly, as part of the law governing other fraudulent (and unfair) business practices. This view of trademark protection as a species of unfair competition was not, as some have suggested,34 a post hoc conflation of two branches of the law. From the very beginning, trademark cases and those only “analogous” to trademark cases were grounded in the same fundamental principle—that no person has the right to pass off his goods as those of another….
 At some point in the late nineteenth century, American courts began to use the term “unfair competition” slightly differently. Those courts divided the universe of distinguishing marks into “technical trademarks,” which were protected in actions for trademark infringement, and “trade names,” which could only be protected in actions for unfair competition. Arbitrary or fanciful terms applied to particular products were considered technical trademarks,35 while surnames, geographic terms, descriptive terms were considered trade names.36 ….
 In practice, cases of trademark infringement and those of unfair competition differed primarily in terms of what the plaintiff had to prove. Use of another’s technical trademark was unlikely to have a legitimate explanation and could be condemned categorically. Trademark infringement plaintiffs therefore did not have to prove intent. Use of another’s trade name, on the other hand, may have had an innocent purpose, such as description of the product’s characteristics or its geographic origin. As a result, in contrast to trademark infringement plaintiffs, unfair competition claimants had to prove that the defendant intended to pass off its products as those of the plaintiff.
Comments and Questions
1. “Technical trademarks”, “trade names”, and intent. In addressing the role of intent in late nineteenth century American unfair competition law, McKenna cites The Restatement (Third) of Unfair Competition. The Restatement explains:
In both England and the United States [in the late nineteenth century], the property conception of trademark rights extended only to certain designations. When the defendant imitated a designation that was clearly distinctive of the plaintiff’s goods, the natural inference that the defendant intended to deceive prospective purchasers eventually led to a conclusive presumption of fraud. Thus, in the case of words or other symbols invented by the plaintiff or arbitrary designations that had no apparent relation to the plaintiff’s goods except as an indication of source, the courts began to protect the plaintiff’s “property” interest in the mark without regard to the presence of any fraudulent intent. Such marks were characterized as “trademarks,” and cases involving the unauthorized use of these marks were designated as actions for “trademark infringement.” The focus of the inquiry thus shifted from an analysis of the defendant’s conduct to a consideration of the nature of the plaintiff’s right. Less distinctive marks that had nevertheless come to be recognized by prospective consumers as indications of source were called “trade names.” Although not recognized as “property” in the same sense as technical “trademarks,” protection for “trade names” remained available through the action for “unfair competition,” with its historical emphasis on the fraudulent character of the defendant’s conduct.
 The initial emphasis on fraud and property rights has generally given way to a more explicit analysis of the propriety of the defendant’s conduct as a means of competition, and the technical distinctions between the actions for trademark infringement and unfair competition have now been abandoned.
Restatement (Third) Of Unfair Competition § 9, cmt. d (1995).
2. Production marks. As the McKenna excerpt explains, local guilds required production marks not just to aid in asserting their monopoly but also to fix liability for poorly-made goods that might tarnish the reputation of the guild. An early example of quality enforcement—and of trademark adjudication—comes to us in the remarkable story of the fourteenth-century bladesmith John Odinsay. Odinsay was accused of making a sword that broke during combat when one Sir Peter Harpdon used it to defend himself from highway brigands (led by Geoffrey Tete Noir) while travelling through Bordeaux in 1345. Sir Peter recovered from his wounds in that skirmish and went on to fight next to the Black Prince in the Battle of Crecy in 1346. But upon his return to London, he pursued the matter of the broken sword. The hallmark suggested that Odinsay had made it (and the penalties for such faulty craftsmanship would have ruined Odinsay and his family), but the mark turned out to be a forgery. The London bladesmiths’ guild discovered that several of its members’ marks were being forged, perhaps by smiths in nearby cities. See Thomas D. Drescher, The Transformation and Evolution of Trademarks—From Signals to Symbols to Myth, 82 Trademark Reporter 301, 313-18 (1992).
Rep. Fritz Lanham, 1880-1965
Excerpt from Restatement (Third) of Unfair Competition § 9 (1995)
 e. Trademark legislation. The federal government and each of the states have enacted legislation protecting trademarks. The statutes generally provide a mechanism for the registration of trademarks, describe the types of marks that may be registered, and specify the procedural and substantive advantages afforded to the owner of a trademark registration. The statutes, however, do not ordinarily preempt the protection of trademarks at common law.
 Although several states had earlier enacted legislation to prevent the fraudulent use of trademarks, the first federal trademark statute was not enacted until 1870. This initial attempt at federal protection proved short-lived, however, when in 1879 the Supreme Court in the Trade-Mark Cases, 100 U.S. (10 Otto) 82 (1879), held that the statute had been unconstitutionally grounded on the patent and copyright clause of the Constitution. A second federal statute was enacted in 1881, but in reaction to the Trade-Mark Cases, registration under the act was limited to marks used in commerce with foreign nations and the Indian tribes. The first modern federal trademark registration statute was the Trademark Act of 1905, grounded on the commerce clause. In a continuation of the distinction that had developed at common law between technical “trademarks” and “trade names,” the Act of 1905 limited registration to fanciful and arbitrary marks, except for marks that had been in actual use for 10 years preceding passage of the statute.
 To clarify and strengthen the rights of trademark owners, the Act of 1905 was replaced by the Trademark Act of 1946 (effective July 5, 1947), 15 U.S.C.A. §§ 1051-1127, commonly known as the Lanham Act. The Lanham Act is generally declarative of existing law, incorporating the principal features of common law trademark protection. However, among the major innovations of the Lanham Act were the adoption of a constructive notice rule that effectively expanded the geographic scope of trademark rights, and an attempt to provide a measure of security to trademark owners in the form of “incontestable” rights in certain trademarks. The Lanham Act in § 43(a) also added a general proscription against false designations and representations that has come to serve as a federal law of deceptive marketing.
 Statutes in every state also provide for the registration of trademarks. In 1949 the United States (now International) Trademark Association prepared a Model State Trademark Bill patterned after the federal registration system. The Model Bill, revised in 1964 and 1992, provides the basis for much of the current state legislation.
From Edward S. Rogers, The Lanham Act and the Social Function of Trademarks, 14 Law & Contemp. Probs. 173, 180-83 (1949)
 The prospect of getting anything through Congress in 1937 was not encouraging. Our committee[, the Trade Mark Committee of the Patent Section of the American Bar Association,] kept notes and I had a scrapbook in which I stuck ideas that came in from all sorts of places. More as a matter of convenience than anything else, I cast those notes and ideas in the form of a draft statute.
 In the winter of 1937 the Commissioner of Patents asked me to come to Washington to see him. He said he had had a conference with Fritz Lanham, who was chairman of the subcommittee of the House Patent Committee dealing with trademarks, and asked me to see Mr. Lanham, which I did. Mr. Lanham said that a large number of piecemeal amendments to the 1905 Act had been proposed and that he had been studying the Act and couldn’t make head or tail of it; that if it were amended piecemeal it would make incomprehensible what had hitherto been merely obscure. So he asked if anywhere around there was a skeleton draft of a new act that could be used as a sort of clotheshorse to hang things on. I told him I had such a draft and he asked me to leave it with him, which I, of course, was glad to do.
 I supposed that Mr. Lanham was just going to study this memorandum and skeleton—it was hardly more than that—and begin to hold hearings. I was surprised when, on January 19, 1938, he introduced it as H.R. 9041.
 Immediately bar associations appointed committees which did thoughtful and conscientious work, with the result that we now have a new Trade-Mark Act. Since the last Act was passed in 1905 and the new Act in 1946—forty-one years later—I suspect we are going to have to live with the Lanham Act for a long time.
 Whenever there was a hearing before any committee on the trade-mark bill, sooner or later there appeared zealous men from the Department of Justice who raised all manner of objections. They asserted that trade-marks are monopolistic and any statutory protection of them plays into the hands of big business and should be discouraged. In vain it was pointed out that what is now big business started as little business—that trade-marks are not, like patents and copyrights, a government grant of an exclusive right, that trade-marks are visible reputation and symbols of good will, that trade-marks are the antithesis of monopoly, and that to protect them is to insure the one whose goods or services they distinguish against fraud and misrepresentation.
 No progress seemed to be made with the Department's representatives, who were against not only the protection of trade-marks but trade-marks as an institution….
Comments and Questions
1. The Long Road to the Lanham Act. In his influential treatise, J. Thomas McCarthy records the fate of legislative efforts through the war years leading to the Lanham Act of 1946:
Hearings on the bill and the various forms in which it was reintroduced were held in March 1938, March 1939, June 1939, and passed the House and Senate in 1939 and 1940. However, the Senate moved to reconsider the bill on June 23, 1940 and it was returned to the calendar and died. In the 77th Congress a reintroduced bill passed the Senate in 1941 and the House in 1942, but the bill died upon being referred back to Committee in 1942. Hearings were held in the 78th Congress in 1943 and 1944, but the bill was not passed. Finally, the 1945 version of the bill (H.R. 1654) was passed by the 79th Congress.
McCarthy on Trademarks and Unfair Competition § 5.4 (2015). If we date the Lanham Act from its first draft in 1937 (or indeed back to the so-called Vestal Bill of 1931), then the Act is nearly 80 years old. This may help to explain the existence of certain especially abstruse statutory sections that the student will confront through the course of studying U.S. trademark law.
The Lanham Act has been amended numerous times since its July 5, 1947 effective date. Listed here are some of the more important amendments, many of which we will refer to through the course of this casebook.
The limiting phrase “purchasers as to the source of origin of such goods or services” was deleted from Lanham Act § 32. 1962 Pub. L. No. 87-772, 76 Stat. 769. This arguably significantly broadened the scope of anti-infringement protection under the Act.
The following sentence was added to Lanham Act § 35: “The court in exceptional cases may award reasonable attorney fees to the prevailing party.” 1975 Pub. L. No. 93-600, 88 Stat. 1955.
Congress finally changed the name of the “Patent Office” to the “Patent and Trademark Office.” 1975 Pub. L. No. 93-596, 88 Stat. 1949.
The Court of Customs and Patent Appeals became the Court of Appeals for the Federal Circuit. Pub. L. 97-164, 96 Stat. 25.
The Trademark Counterfeiting Act of 1984 was enacted, 1984 Pub. L. No. 98-473, 98 Stat. 1837, amending Lanham §§ 34, 35, and 36, and establishing criminal trademark anti-counterfeiting penalties in 18 U.S.C. § 2320.
The Trademark Law Revision Act of 1988 (TLRA) was enacted, effective November 16, 1989. Pub. L. No. 100-667, 102 Stat. 3935. The TLRA established the “intent-to-use” basis for registration and federal statutory “constructive use” for purposes of priority. It also significantly rewrote Lanham Act § 43(a).
The Federal Trademark Dilution Act (FTDA), enacted and effective January 16, 1996, established a federal cause of action for anti-dilution protection in Lanham Act § 43(c). Pub. L. No. 104-98, 109 Stat. 985. The FTDA has been replaced by the Trademark Dilution Revision Act of 2006.
The Anticounterfeiting Consumer Protection Act of 1996 further enhanced procedures to combat and penalties for trademark counterfeiting. Pub. L. No. 104-153, 110 Stat 1386. The Act also introduced statutory damages for counterfeiting
The Anticybersquatting Consumer Protection Act (ACPA) established Lanham Act § 43(d) to combat the cybersquatting of domain names confusingly similar to or dilutive of trademarks. Pub. L. No. 106-113, 113 Stat. 1501.
The Madrid Protocol Implementation Act (MPIA), enacted Nov. 2, 2002 and effective Nov. 2, 2003, established Lanham Act §§ 60-74. 116 Stat. 1758, 1913 Pub. L. No. 107-273. With the MPIA, the U.S. became a member of the Madrid System of international trademark registration.
The Trademark Dilution Revision Act of 2006 (TDRA) significantly rewrote Lanham Act § 43(c). Pub. L. No. 109-312, 120 Stat. 1730. It replaced the FTDA of 1996.
The Prioritizing Resources and Organization for Intellectual Property Act of 2008 (PRO-IP Act) enhanced civil damages and criminal penalties for trademark counterfeiting. Pub. L. No. 110-313, 122 Stat. 3014
Comments and Questions
1. “The Last Best Place.” One of the stranger moments in the history of U.S. trademark legislation involves the phrase “The Last Best Place.” Between 2001 and 2004, a Nevada business named Last Best Beef, LLC filed eight applications at the PTO to register the phrase “The Last Best Place” in connection with various goods and services. In 2005, Congress passed and the President signed into law an appropriations bill with a rider that consisted of the following language: “Notwithstanding any other provision of this Act, no funds appropriated under this Act shall be used to register, issue, transfer, or enforce any trademark of the phrase ‘The Last Best Place.’” Upon learning of this statutory command in an appropriations bill that covered the PTO, the PTO suspended all consideration of Last Best Beef’s trademark applications and no further applications for the phrase have since been filed. What? In 1988, a Montana writer had entitled an anthology of Montana-oriented poetry and prose “The Last Best Place.” The phrase was soon taken up by Montana businesses and state government. In 2005, Montana Senator Conrad Burns attached the rider to the appropriations bill on the ground that the phrase “belongs to the State of Montana.” See John L. Welch, Montana Senator Again Blocks “LAST BEST PLACE” Registrations, The TTABlog, Feb. 27, 2009, http://thettablog.blogspot.com/2009/02/montana-senator-max-baucus-announced.html. See also The Last Best Beef, LLC v. Dudas, 506 F.3d 333 (4th Cir. 2007) (not seeing a problem with any of this).
Probably the most oft-quoted passage from the Trade-Mark Cases is the paragraph in which the Supreme Court compared trademarks to the two other most significant forms of intellectual property, copyrights and patents (paragraph 11 in the excerpt above). Consider whether what Justice Miller wrote in 1879 about the development of brand names is still accurate today:
The ordinary trade-mark has no necessary relation to invention or discovery. The trade-mark recognized by the common law is generally the growth of a considerable period of use, rather than a sudden invention. It is often the result of accident rather than design, and when under the act of Congress it is sought to establish it by registration, neither originality, invention, discovery, science, nor art is in any way essential to the right conferred by that act. If we should endeavor to classify it under the head of writings of authors, the objections are equally strong. In this, as in regard to inventions, originality is required. And while the word writings may be liberally construed, as it has been, to include original designs for engravings, prints, &c., it is only such as are original, and are founded in the creative powers of the mind. The writings which are to be protected are the fruits of intellectual labor, embodied in the form of books, prints, engravings, and the like. The trade-mark may be, and generally is, the adoption of something already in existence as the distinctive symbol of the party using it. At common law the exclusive right to it grows out of its use, and not its mere adoption. By the act of Congress this exclusive right attaches upon registration. But in neither case does it depend upon novelty, invention, discovery, or any work of the brain. It requires no fancy of imagination, no genius, no laborious thought. It is simply founded on priority of appropriation.
Id. at 94.
By 1942, the Court was describing trademarks and the role of trademark law in different terms. In Mishawaka Rubber & Woolen Mfg. Co. v. S.S. Kresge Co., 316 U.S. 203 (1942), Justice Frankfurter explained:
The protection of trade-marks is the law's recognition of the psychological function of symbols. If it is true that we live by symbols, it is no less true that we purchase goods by them. A trade-mark is a merchandising short-cut which induces a purchaser to select what he wants, or what he has been led to believe he wants. The owner of a mark exploits this human propensity by making every effort to impregnate the atmosphere of the market with the drawing power of a congenial symbol. Whatever the means employed, the aim is the same—to convey through the mark, in the minds of potential customers, the desirability of the commodity upon which it appears. Once this is attained, the trade-mark owner has something of value. If another poaches upon the commercial magnetism of the symbol he has created, the owner can obtain legal redress.
Id. at 205.
By the 1980s, American courts were describing trademarks and trademark law in yet different terms, terms which still resonate today. Reflecting the rise of the Chicago School economic analysis of law, Judge Easterbrook described the economic benefits of trademarks and trademark protection in Scandia Down Corp. v. Euroquilt, Inc., 772 F.2d. 1423 (7th Cir. 1985):
Trademarks help consumers to select goods. By identifying the source of the goods, they convey valuable information to consumers at lower costs. Easily identified trademarks reduce the costs consumers incur in searching for what they desire, and the lower the costs of search the more competitive the market. A trademark also may induce the supplier of goods to make higher quality products and to adhere to a consistent level of quality. The trademark is a valuable asset, part of the “goodwill” of a business. If the seller provides an inconsistent level of quality, or reduces quality below what consumers expect from earlier experience, that reduces the value of the trademark. The value of a trademark is in a sense a “hostage” of consumers; if the seller disappoints the consumers, they respond by devaluing the trademark. The existence of this hostage gives the seller another incentive to afford consumers the quality of goods they prefer and expect.
Id. at 1429-30.
Which description of trademarks most accurately reflects their characteristics in the present day? Are they often adopted, in the terms of the Trade-Mark Cases, as “the result of accident rather than design”? Can we say of the development of trademarks, as of the legal conditions leading to their protection, that “no fancy of imagination, no genius, no laborious thought” is required? Or is it rather that, through the development of a brand name, “[t]he owner of a mark...mak[es] every effort to impregnate the atmosphere of the market with the drawing power of a congenial symbol”? Is the consumer in some sense a victim of these machinations of the trademark owner, who through the “commercial magnetism” of the trademark “induces the purchaser to select what he wants, or what he has been led to believe he wants”? Or is it finally not consumers who are victims of the trademark, but the trademark who is a “hostage” of consumers, whom it serves by enabling them to find what they desire and to insist on “the quality of goods they prefer and expect”?
The Economic Justification for Trademark Protection
These differing accounts of the trademark and trademark law are probably all more or less true, depending on the trademark, product, and consumer at issue. But it is well-accepted that the last account, based on the economic analysis of law, is currently by far the dominant account of trademark law. In Qualitex Co. v. Jacobson Products Co., Inc., 514 U.S. 159 (U.S. 1995), Justice Breyer cited, among other sources, William Landes & Richard Posner, The Economics of Trademark Law, 78 Trademark Rep. 267, 271-272 (1988), in support of the following statement of the purposes of trademark law:
[T]rademark law, by preventing others from copying a source-identifying mark, reduces the customer's costs of shopping and making purchasing decisions, for it quickly and easily assures a potential customer that this item—the item with this mark—is made by the same producer as other similarly marked items that he or she liked (or disliked) in the past. At the same time, the law helps assure a producer that it (and not an imitating competitor) will reap the financial, reputation-related rewards associated with a desirable product. The law thereby encourages the production of quality products, and simultaneously discourages those who hope to sell inferior products by capitalizing on a consumer's inability quickly to evaluate the quality of an item offered for sale.
Id. at 163-64 (citations omitted).
The current orthodox view of trademarks, then, is that they (1) minimize consumer search costs, and (2) provide incentives to producers to produce consistent levels of product quality. This latter benefit of trademarks is especially important for certain types of products. In general, products may be understood to possess three types of characteristics: “search” characteristics, such as color or price, which can be inspected prior to purchase; “experience” characteristics, such as taste, which can only be verified through use of the product; and “credence” characteristics, such as durability, which can only be confirmed over time. See Phillip Nelson, Advertising as Information, 82 J. Pol. Econ. 729 (1974). For products such as medicine, automobiles or high-technology goods, the readily-apparent “search” characteristics of which say little about the quality of the product, consumers may rely heavily on the trademark attached to the product in making their purchasing decision. It follows that in a market without reliable source-identification for such products, producers would have little incentive to invest in the production of products of high quality. This is because they would likely be undercut by competitors who would offer cheaper products of lower quality. See George A. Ackerlof, The Market for "Lemons": Quality Uncertainty and the Market Mechanism, 84 Q.J. of Econ. 488 (1970).
As indications of quality, trademarks signify and allow firms to develop commercial goodwill, which for many firms may be by far their most valuable asset. The concept of goodwill encompasses the reputation of the firm and its products and the probability, based on this reputation, that consumers will continue to patronize the firm in the future. A nineteenth-century court described goodwill in these terms:
When an individual or a firm or a corporation has gone on for an unbroken series of years conducting a particular business, and has been so scrupulous in fulfilling every obligation, so careful in maintaining the standard of goods dealt in, so absolutely honest and fair in all business dealings that customers of the concern have become convinced that their experience in the future will be as satisfactory as it has been in the past, while such customers' good report of their own experience tends continually to bring new customers to the same concern, there has been produced an element of value quite as important—in some cases, perhaps far more important—than the plant or machinery with which the business is carried on.
Washburn v. National Wall-Paper Co., 81 F. 17, 20 (2d Cir. 1897).
Criticisms of the Economic Justification for Trademark Protection
Yet the example of Coca-Cola and brands like it may lead many readers to doubt the sufficiency of the economic account of trademark law, focused as it is on search costs and incentives to produce quality goods. After all, many trademarks, such as coke, do more than merely indicate the source of the goods to which they are affixed, and strictly speaking, some trademarks don’t even do that. A t-shirt bearing the trademark arsenal is not intended to indicate and is not read by consumers to indicate that Arsenal soccer players knitted the shirt themselves. The trademark primarily functions instead as a “badge of support for or loyalty or affiliation to the trademark proprietor.” Arsenal Football Club Plc v. Matthew Reed, Case C-206/01,  ETMR 19, ¶ 15. This same function may be attributed to many trademarks, and not simply to high-fashion marks such as polo or prada, but also to more mundane marks such as pepsi or ford, whose owners have quite consciously sought to build “consumption communities”37 around these brands. See Int'l Order of Job's Daughters v. Lindeburg & Co., 633 F.2d 912, 918 (9th Cir. 1980) (recognizing that “[w]e commonly identify ourselves by displaying emblems expressing allegiances. Our jewelry, clothing, and cars are emblazoned with inscriptions showing the organizations we belong to, the schools we attend, the landmarks we have visited, the sports teams we support, the beverages we imbibe”). In such situations, the mark itself is often the primary product characteristic that the consumer wishes to acquire, and the underlying material good, if any, is merely a means of conveying that characteristic,38 and an alibi for the consumption of that characteristic. We typically think of a trademark as supplementary in relation to the goods to which it is affixed, as something added to preexisting goods. But certain doctrines in trademark law may make sense only if one appreciates that for certain brands, this relation is reversed. The brand is prior and the physical goods are supplementary to it, supporting and enhancing the brand’s value, so that a firm (for example, a fashion house) may first design a brand and then produce or license goods consistent with that brand.
Even when the consumer is interested in the quality of the material good, the trademark may contribute to deleterious “artificial product differentiation,” as when consumers pay a premium for branded versions of pharmaceuticals when lower-cost generic versions are required by government regulation to meet exactly the same quality standards as the more expensive branded versions. This argument, which associates trademarks with the purported evils of some forms of advertising, first gained significant influence with the publication in 1933 of the economist Edward Chamberlin’s book The Theory of Monopolistic Competition, which systematically formulated the artificial product differentiation view.39 Chamberlin’s work proved to especially influential in mid-twentieth century trademark commentary40 and is reflected to some degree in Justice Frankfurter’s discussion of trademarks in Mishawaka Rubber. Other courts sometimes picked up on Chamberlin’s ideas. See, e.g., Smith v. Chanel, Inc., 402 F.2d 562, 567 (2d Cir. 1968) (proposing that, through the trademark, “economically irrational elements are introduced into consumer choices; and the trademark owner is insulated from the normal pressures of price and quality competition. In consequence the competitive system fails to perform its function of allocating available resources efficiently.”).
In recent decades, however, mainstream economic thought has grown increasingly hostile towards, even dismissive of, the argument that, as Landes and Posner characterize it, trademarks “promote social waste and consumer deception” through “the power of brand advertising to bamboozle the public and thereby promote monopoly.”41 Instead, economists have come to view trademarks and advertising in a much more positive light. See George Stigler, The Economics of Information, 69 J. Pol. Econ. 213 (1961). The consensus view now is that advertising cheaply conveys information to consumers, particularly with respect to “experience goods.” See Phillip Nelson, Advertising as Information, 82 J. Pol. Econ. 729 (1974). Advertising also signals that the advertiser believes its goods to be of sufficiently high quality to benefit from advertising. “The higher quality brand will, other things being equal, have a comparative advantage in acquiring more customers by advertising—since it will retain a larger fraction of them on repeat sales.” See Jack Hirshleifer, Where Are We in the Theory of Information?, 63 Am. Econ. Rev. Proc. 31, 38 (1973).42
Despite the current consensus in economic and legal thought that advertising serves important informational functions in markets, criticisms of branding and advertising remain influential in popular thought. See, e.g., Naomi Klein, No Logo: Taking Aim at Brand Bullies (2000); Juliet B. Schor, Born to Buy: The Commercialized Child and the New Consumer Culture (2005). For readers sympathetic to these criticisms, two questions arise with respect to trademarks and trademark law. First, is it fair to apply general criticisms of advertising to trademarks specifically? Though trademarks are usually central to most forms of advertising, aren’t trademarks themselves mere informational devices? Second, and related, how, if at all, can trademark law be modified to limit such alleged harms as artificial product differentiation or the “bamboozl[ing]” of the public? Stated differently, how can trademark law continue to promote the ability of marks to inform consumers without also promoting the ability of marks to persuade? How practically speaking can trademark law minimize persuasion but still preserve information?43 Any serious criticism of the role that trademark law plays in perpetuating status consumption or introducing “economically irrational elements” into purchasing decisions should be able to answer these questions. Perhaps limiting the scope of trademark rights or the kinds of commercial signifiers that can be protected as trademarks would lessen the persuasive impact of strong brands. But it may be that minor modifications to trademark law will not help to ameliorate the effects of deeply-engrained consumption practices, and efforts to reform these practices will be more effective if undertaken elsewhere.
Opponents of overly expansive trademark rights (and defendants in trademark cases) may find more traction by appealing to what is arguably the true overarching goal of trademark law, one which subsumes the goals of lowering consumer search costs and incentivizing consistent levels of product quality. Trademark law’s overarching goal is to foster competition, primarily by enabling the efficient communication of information in the marketplace. When trademark law overprotects, it impedes the optimal flow of information to consumers, tends to give undue market power to incumbents, and can significantly disrupt the efficient operation of the patent and copyright systems (a possibility which we will address in a moment). The argument from competition speaks the language of mainstream economics, but often does so in favor of limiting rather than expanding trademark property rights.
Comments and Questions
1. Trademark law and “property.” Critics of the expansion in the subject matter and scope of trademark protection in recent decades often accuse the law of having lost its purportedly traditional focus on consumer protection and having instead embraced a property-rights rationale for trademark protection. Elsewhere in the article excerpted above in Part A, McKenna directly challenges this view:
[T]rademark law was not traditionally intended to protect consumers. Instead, trademark law, like all unfair competition law, sought to protect producers from illegitimate diversions of their trade by competitors. Courts did focus on consumer deception in these cases, but only because deception distinguished actionable unfair competition from mere competition, which was encouraged. In fact, courts denied relief in many early trademark cases despite clear evidence that consumers were likely to be confused by the defendant's use. Invariably they did so because the plaintiff could not show that the defendant's actions were likely to divert customers who otherwise would have gone to the plaintiff.
Moreover, American courts protected producers from illegitimately diverted trade by recognizing property rights. This property-based system of trademark protection was largely derived from the natural rights theory of property that predominately influenced courts during the time American trademark law developed in the nineteenth century….
Critics cannot continue simply to claim that modern law is illegitimate because it does not seek to protect consumers. Because it never really did.
Mark P. McKenna, The Normative Foundations of Trademark Law, 82 Notre Dame L. Rev. 1839, 1841, 1916 (2007). For an alternative reading of the history of American trademark law, see Robert G. Bone, Hunting Goodwill: A History of the Concept of Goodwill in Trademark Law, 86 B.U. L. Rev. 547 (2006).
As the excerpt above from the Trade-Mark Cases suggests, when seen from the perspective of trademark law, copyright law and patent law can appear to be closely similar to each other and quite different from trademark law—so much so that it is not unreasonable to ask why trademark is grouped with patent and copyright under the rubric of “intellectual property law” rather than separated out as some hybrid of competition law and intellectual property law. As the table at the conclusion of this section summarizes, both copyright and patent are based on the Intellectual Property Clause of the Constitution, which empowers Congress “[t]o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.” U.S. Const. art. I, § 8, cl. 8. The Constitution thus requires copyright and patent to promote innovation, human creativity, or more generally, human “Progress,” with patent focusing primarily on incentivizing the invention of new technologies, such as new pharmaceuticals, better machines, or more efficient methods of manufacture, and copyright focusing on incentivizing the production of “works of authorship,” such as novels, music, and motion pictures.
Inventions and works of authorship share important characteristics that make intellectual property protections useful. Both tend to be expensive to develop, but once developed, they are relatively inexpensive to reproduce in copies. It can cost $1 billion to develop a successful pharmaceutical and bring it to market and potentially only a few dollars per copy to manufacture it. The consumption of inventions and works of authorship also tends to be “non-rivalrous.” A potentially unlimited number of people can benefit equally from the same idea or listen each to his or her own copy of the same recording of the same musical work. Finally, without recourse to prohibitions established by law, it is often exceedingly difficult to exclude people from and thus charge a price for the benefit of an invention or work of authorship. This condition has only intensified with improvements in reproduction and distribution technologies, whether they take the form of ever more flexible assembly lines, automated manufacture, or the reproduction of digital files on a home computer or the internet.
To address these problems, patent law and copyright law provide limited terms of protection to qualifying works, with patent’s term significantly shorter in duration than copyright’s. In essence, the public makes a bargain with inventors and authors. To incentivize them, we give them exclusive rights in their innovations so that they can recoup the costs of and perhaps profit from their innovating activity, but in exchange, we eventually claim their innovations for the public domain, where these innovations become free for all, including subsequent inventors and authors, to use.
In contrast to copyright and patent law, trademark law is based not on the Intellectual Property Clause, but the Commerce Clause. Its goal is not to promote the progress of “Science and useful Arts” but rather to promote fair and efficient competition. Its term of protection is unlimited in time provided that the trademark owner continues to use the trademark in commerce. And the utilization of trademarks is arguably rivalrous. If two firms share the same trademark for the same type of product in the same marketplace, the utility of both trademarks will be severely diminished.
For all of the differences among copyright, patent, and trademark law, note that these separate regimes of intellectual property law can simultaneously protect the same thing. For example, a logo might qualify for both copyright and trademark protection. A particular product feature, such as the shape of a mobile phone, might qualify for trademark protection and design patent protection. A particular furniture design might qualify for trademark protection, design patent protection, and copyright protection as well.
These overlapping regimes of exclusive rights can create significant problems in intellectual property law, some of which we will engage later in this casebook. For example, what should happen when the term of copyright protection in a particular work of authorship expires, but that expression also functions as a trademark? Should trademark law allow the Walt Disney Company to continue to assert exclusive rights in images of Mickey Mouse after its copyright in those images has expired? More significantly, should companies be able to assert trademark rights in product features that also qualify for utility patent protection, or at least that perform some mechanical function in addition to serving as designations of source?
Comments and Questions
1. Do We Want to Incentivize More Trademarks? We generally seek through patent and copyright law to incentivize the production of more patentable inventions and more copyrightable works of authorship. Should we similarly design trademark law to incentivize the production of more trademarks? Is there anything intrinsically valuable about trademarks? Do more trademarks indicate or themselves constitute “Progress”?
Utility Patent Law
Design Patent Law
Protectable Subject Matter
Designations of commercial source, including brand names, logos, product packaging, and product configurations
Designations of commercial source, including brand names, logos, product packaging, and product configurations
Inventions, including processes, machines, manufactured articles, and compositions of matter
Ornamental designs for articles of manufacture
Intellectual Property Clause
Intellectual Property Clause
Intellectual Property Clause
Lanham Act, 15 U.S.C. § 1051 et seq.
Copyright Act, 17 U.S.C. §101 et seq.
Patent Act, 35 U.S.C. § 101 et seq.
Patent Act, , 35 U.S.C. § 101 et seq.
Basic Requirements for Protection
Distinctive of source; used in commerce; not functional
Fixed in a tangible medium of expression; originality
Novel, non-obvious, and useful
Ornamental, novel, and non-obvious
Term of Protection
Registration lasts 10 years; perpetually renewable as long as the mark is distinctive and used in commerce
Life of the author plus 70 years; for works for hire, 95 years from date of publication
20 years from filing date of patent application
14 years from grant of patent
How Rights Are Acquired
“Common law” rights through use in commerce; registered rights through registration at PTO
Through fixation; registration not required
Patent application at PTO
Patent application at PTO
Quoted in Thomas D. Drescher, The Transformation and Evolution of Trademarks—From Signals to Symbols to Myth, 82 Trademark Rep. 301, 301-02 (1992).
See Interbrand, Best Global Brands 2016 (2016), http://www.bestglobalbrandsinterbrand.com/2014best-brands/best-global-brands/2016/ranking/cocacola/.
The Coca-Cola Company, Balance Sheet, http://finance.yahoowww.nasdaq.com /q/bs?s=KO+Balance+Sheet&annual/symbol/ko/financials?query=balance-sheet.
Owners also carved identifying marks into the beaks of swans they were allowed to own by royal privilege. See Frank I. Schechter, The Historical Foundations Of The Law Relating To Trade-Marks 35-37 (1925)
Not coincidentally, these mandatory marks also made it possible for the Crown to regulate conduct, particularly in the printing industry, where the Crown policed heresy and piracy. See id. at 63-77.
There are some examples in the cutlers' trade of the government treating marks as property that could be passed by will and of owners advertising to suppress piracy. See id. at 119-20.
Id. at 120 (quoting Robert Eadon Leader, History of the Cutlers of Hallamshire 110 (1906)).
Southern, (1618) Pop. at 144, 79 Eng. Rep. at 1244
See (1659) Bridg.J. at 125-28, 123 Eng. Rep. at 1248-50; 2 Rolle at 5-6, 81 Eng. Rep. at 621-23.
Croke’s Report stated: “Dodderidge cited a case to be adjudged 33 Eliz. in the Common Pleas: a clothier of Gloucestershire sold very good cloth, so that in London if they saw any cloth of his mark, they would buy it without searching thereof; and another who made ill cloths put his mark upon it without his privity; and an action upon the case was brought by him who bought the cloth, for this deceit; and adjudged maintainable.” (1659) Cro. Jac. at 471, 79 Eng. Rep. at 402 (emphasis added). The second of Rolle’s Reports, 2 Rolle 28, 81 Eng. Rep. at 637, is somewhat ambiguous, but that report also suggests it may have been the purchaser who brought the case.
Id. at 484, 26 Eng. Rep. at 692-93.
See The Case of Monopolies, (1603) 11 Co. Rep. 84b, 88b, 77 Eng. Rep. 1260, 1266 (K.B.) (calling the playing card monopoly granted by Queen Elizabeth under her royal prerogative an “odious monopoly”).
Blanchard, 2 Atk. at 485, 26 Eng. Rep. at 693.
Sykes, 3 B. & C. at 543, 107 Eng. Rep. at 835.
See, e.g., Blofeld v. Payne, (1833) 4 B. & Ad. 410, 411-12, 110 Eng. Rep. 509, 510 (K.B.).
See, e.g., Edelsten v. Edelsten, (1863) 1 De. G.J. & S. 185, 199, 46 Eng. Rep. 72, 78 (Ch.) (stating that in actions for trademark infringement “[a]t law the proper remedy is by an action on the case for deceit: and proof of fraud on the part of the Defendant is of the essence of the action”).
See Blofeld, 4 B. & Ad. at 412, 110 Eng. Rep. at 510 (upholding the verdict for the plaintiff and holding that the defendant’s use of envelopes resembling those of plaintiff’s, and containing the same words, was a “fraud against the plaintiff”).
Id. at 141, 46 Eng. Rep. at 870.
Id. at 142, 46 Eng. Rep. at 870.
Id. at 140, 46 Eng. Rep. at 870.
Id. at 141, 46 Eng. Rep. at 870.
Id. at 448.
Like its English predecessor, American trademark law was predominantly a product of judicial decision. Prior to the Act of July 8, 1870, ch. 230, 16 Stat. 198, 210, statutory protection, to the extent it existed, was at the state level and highly trade-specific. Massachusetts, for example, specifically regulated the use of marks on sailcloth. See Schechter, supra note 23, at 130-32. The Supreme Court declared the first two attempts at federal trademark legislation unconstitutional. See The Trade-Mark Cases, 100 U.S. 82, 99 (1879) (invalidating the trademark legislation of 1870 and the Act of Aug. 14, 1876, ch. 274, 19 Stat. 141 (which imposed criminal sanctions against one who fraudulently used, sold or counterfeited trademarks)). Even after Congress began legislating again in this area, however, trademark law remained fundamentally a creature of common law. Indeed, the Lanham Act, ch. 540, 60 Stat. 427 (1946), is widely noted to have generally codified common law.
Id. at 717.
See, e.g., The Trade-Mark Cases, 100 U.S. 82, 92 (1879); Blackwell v. Armistead, 3 F. Cas. 546, 548 (C.C.W.D. Va. 1872) (No. 1474); Derringer v. Plate, 29 Cal. 292, 294-95 (1865); Avery & Sons v. Meikle & Co., 4 Ky. L. Rptr. 759, 764-65 (1883);
See, e.g., Robert G. Bone, Hunting Goodwill: A History of the Concept of Goodwill in Trademark Law, 86 B.U. L. REV. 547, 572 (2006). (arguing that courts’ shift to recognizing goodwill as the relevant property interest helped “to unify, at the level of general principle, the distinct but closely related torts of trademark infringement and unfair competition”).
See Restatement (Third) of Unfair Competition § 9 (1995); see also 1 McCarthy, supra note 13, § 4:4, at 4-4 (defining technical trademarks as marks that were “fanciful, arbitrary, distinctive, non-descriptive in any sense and not a personal name”).
For further discussion of the trademark “merchandising right,” see Stacey L. Dogan & Mark A. Lemley, The Merchandising Right: Fragile Theory or Fait Accompli?, 54 EMORY L.J. 461 (2005).
See also Joan Robinson, The Economics of Imperfect Competition 89 (1933).
See generally Glynn S. Lunney, Trademark Monopolies, 48 EMORY L.J. 367, 367-69 (1999) (discussing the influence of Chamberlin’s work on trademark commentary). See also Sherwin Rosen, Advertising, Information, and Product Differentiation, in ISSUES IN ADVERTISING: THE ECONOMICS OF PERSUASION 161-91 (David G. Tuerck ed., 1978) (summarizing the artificial product differentiation view). See generally Beverly W. Pattishall, Trademarks and the Monopoly Phobia, 50 MICH. L. REV. 967 (1952) (criticizing the artificial product differentiation view).
Landes & Posner, supra, at 276-77.
See also Mark Lemley, The Modern Lanham Act and the Death of Common Sense, 108 YALE L.J. 1687, 1690 (1999) (paraphrasing, though not necessarily endorsing, this theory as “In effect, ‘we advertise, and therefore we must sell a good of sufficiently high quality that we can afford this high-cost expenditure.’”).
See Barton Beebe, Search and Persuasion in Trademark Law, 103 MICH. L. REV. 2020 (2005).
In order to qualify for trademark protection under U.S. federal law, a trademark must meet three basic requirements: (1) the trademark must be “distinctive” of the source of the goods or services to which it is affixed, (2) the trademark must not be disqualified from protection by various statutory bars to protection, the most significant of which is that the trademark not be “functional,” and (3) the trademark must be used in commerce.
Note what is missing from this list of basic requirements for trademark protection. First, in order to qualify for protection under the Lanham Act, a trademark does not need to be registered at the PTO (though, as we will discuss in Part I.D, there are significant benefits to registration). Lanham Act § 32, 15 U.S.C. § 1114, protects registered marks from unauthorized uses that are likely to cause consumer confusion as to the true source of the unauthorized user’s goods. Lanham Act § 43(a), 15 U.S.C. § 1125(a), does the same for unregistered marks. (And Section 43(c), 15 U.S.C. § 1125(c), protects both registered and unregistered marks from trademark dilution). As a matter of tradition, trademark lawyers sometimes refer to unregistered mark protection under § 43(a) as “common law” protection of trademarks even though this protection is based on statutory federal law.
Second, a protectable trademark need not manifest itself in any particular form.1 Consider the extraordinary variety of forms that trademarks (here, all registered) may take:
The reader may be surprised to see that trademark rights can cover such a wide array of subject matter. This Part covers how these various marks have managed to qualify for trademark protection and why various other marks have failed to qualify. Section I.A devotes a great deal of attention to what is by far the most important requirement for trademark protection: that the trademark be “distinctive.” Section I.B then turns to the various statutory bars to protection, including the functionality bar, which disqualify marks from protection under the Lanham Act. Section I.C seeks to make sense of the “use in commerce” requirement for trademark protection. Section I.D reviews why it is worthwhile to register a mark at the PTO and how the registration process works. Section I.E addresses the geographic scope of the protection of registered and unregistered marks.
Lanham Act § 45; 15 U.S.C. § 1127
The term “trademark” includes any word, name, symbol, or device, or any combination thereof… used by a person… to identify and distinguish his or her goods, including a unique product, from those manufactured or sold by others and to indicate the source of the goods, even if that source is unknown.
The § 45 definition of the term “trademark” emphasizes that a protectable trademark must be distinctive of source — it must “identify and distinguish… goods… and… indicate the source of the goods.” Note that in order to qualify for protection, a trademark need not indicate the precise manufacturing source of the goods. For example, the trademark tide for laundry detergent need not indicate in exactly which factory the particular bottle of laundry detergent was made or that Proctor & Gamble ultimately owns the tide brand. Instead, consumers need only know that all products bearing the same trademark originate in or are sponsored by a single, albeit “unknown” or “anonymous,” source.3 This is sometimes known as the “anonymous source” theory of trademark protection.
A trademark will qualify as distinctive if either (1) it is “inherently distinctive” of source or (2) it has developed “acquired distinctiveness” of source. A mark is inherently distinctive if “its intrinsic nature serves to identify a particular source.” Wal-Mart Stores, Inc. v. Samara Bros., Inc., 529 U.S. 205 210 (2000) (alterations omitted). The underlying assumption is that as a matter of consumer literacy, consumers will almost instantly recognize, even when they encounter the mark for the first time, that an inherently distinctive mark is a designation of source. After all, how else would a modern consumer make sense of the word “apple” as used in the sale of electronics that have nothing to do with apples? Inherently distinctive marks “almost automatically tell a customer that they refer to a brand,” Qualitex Co. v. Jacobson Products Co., Inc., 514 U.S. 159 162-63 (1995) (emphasis in original), and “immediately … signal a brand or a product ‘source.’” Id. at 163.
Marks that lack inherent distinctiveness may nevertheless qualify as distinctive if they have developed “acquired distinctiveness,” otherwise known as “secondary meaning.” Over time, consumers may come to identify what might have seemed merely a description of the good or service (e.g., “American Airlines”) or merely a decoration on a product (e.g., three stripes on the side of an athletic shoe) as a designation of the source of that product. Indeed, consumers may come to identify the configuration of the product itself as a signifier of its source.
Here in Section I.A, we will spend considerable time reviewing how courts determine if a commercial sign qualifies as inherently distinctive or as possessing acquired distinctiveness. Before proceeding, two things should be kept in mind. First, some of the opinions below address the registrability of the marks at issue at the PTO while other opinions address the protectability under § 43(a) of marks that have never been registered. Recall that registration is not a prerequisite for trademark protection under the Lanham Act. Many significant trademark cases over past decades have involved unregistered marks. The important point for our purposes in this subsection is that the basic doctrine relating to the registrability of a mark is essentially the same as the doctrine relating to whether it may be protected regardless of its registration status. We may use opinions from either context to understand the distinctiveness requirement in trademark law.
Second, this subsection will first consider distinctiveness doctrine as it relates to verbal marks, and will then proceed to the more difficult area of distinctiveness doctrine that covers non-verbal marks, such as logos, colors, product packaging, and product configuration (i.e., the shape of the product itself).
Even if a trademark is distinctive of source, it will still be denied protection if it falls within one of the statutory bars established under Lanham Act § 2, 15 U.S.C. §1052. We review the most important of these statutory bars here.
It is important to note that, strictly speaking, the § 2 statutory bars are bars only to the registration of a mark at the PTO. Recall however that the Lanham Act will protect both registered marks under § 32, 15 U.S.C. § 1114, and unregistered marks under § 43(a), 15 U.S.C. § 1125(a). This framework raises a question that the law has not yet definitively answered: if a mark is refused registration or its registration is cancelled under one of the statutory bars established in § 2, can the owner of the mark nevertheless seek protection of the mark under § 43(a)? For example, if a mark consists of the flag of a foreign nation and thus is barred from registration under Lanham Act § 2(b), could the owners of the mark nevertheless claim exclusive rights in the mark under § 43(a)? Though scholarly opinion remains divided, the better view would appear to be that a mark unregistrable under § 2 should be unprotectable under § 43(a). See Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 768 (1992) (“[I]t is common ground that § 43(a) protects qualifying unregistered trademarks and that the general principles qualifying a mark for registration under § 2 of the Lanham Act are for the most part applicable in determining whether an unregistered mark is entitled to protection under § 43(a).”); Renna v. Cnty. of Union, N.J., No. 11 Civ. 3328, 2014 WL 2435775, at *8 (D.N.J. May 29, 2014) (“It follows that such unregistrable marks, not actionable as registered marks under Section 32, are not actionable under Section 43, either.”). Cf. Matal v. Tam, No. 15-1293, slip op. at 4 n. 1 (U.S. June 19, 2017) (“We need not decide today whether respondent could bring suit under § 43(a) if his application for federal registration had been lawfully denied under the disparagement clause.”).
We will not review the specifics of the registration process until Section II.D. However, in order to complete our picture of what marks qualify for protection, be they registered or unregistered, we will necessarily review opinions in this section that involve questions of registration. Thus, the reader will need to tolerate references to certain aspects of the registration process that will not become clear until Section II.D.
Lanham Act § 2; 15 U.S.C. § 1052
No trademark by which the goods of the applicant may be distinguished from the goods of others shall be refused registration on the principal register on account of its nature unless it--
(a) Consists of or comprises immoral, deceptive, or scandalous matter; or matter which may disparage70 or falsely suggest a connection with persons, living or dead, institutions, beliefs, or national symbols, or bring them into contempt, or disrepute; or a geographical indication which, when used on or in connection with wines or spirits, identifies a place other than the origin of the goods and is first used on or in connection with wines or spirits by the applicant on or after one year after the date on which the WTO Agreement (as defined in section 3501(9) of Title 19) enters into force with respect to the United States.
(b) Consists of or comprises the flag or coat of arms or other insignia of the United States, or of any State or municipality, or of any foreign nation, or any simulation thereof.
(c) Consists of or comprises a name, portrait, or signature identifying a particular living individual except by his written consent, or the name, signature, or portrait of a deceased President of the United States during the life of his widow, if any, except by the written consent of the widow.
(d) Consists of or comprises a mark which so resembles a mark registered in the Patent and Trademark Office, or a mark or trade name previously used in the United States by another and not abandoned, as to be likely, when used on or in connection with the goods of the applicant, to cause confusion, or to cause mistake, or to deceive: Provided, That if the Director determines that confusion, mistake, or deception is not likely to result from the continued use by more than one person of the same or similar marks under conditions and limitations as to the mode or place of use of the marks or the goods on or in connection with which such marks are used, concurrent registrations may be issued to such persons when they have become entitled to use such marks as a result of their concurrent lawful use in commerce prior to (1) the earliest of the filing dates of the applications pending or of any registration issued under this chapter; (2) July 5, 1947, in the case of registrations previously issued under the Act of March 3, 1881, or February 20, 1905, and continuing in full force and effect on that date; or (3) July 5, 1947, in the case of applications filed under the Act of February 20, 1905, and registered after July 5, 1947. Use prior to the filing date of any pending application or a registration shall not be required when the owner of such application or registration consents to the grant of a concurrent registration to the applicant. Concurrent registrations may also be issued by the Director when a court of competent jurisdiction has finally determined that more than one person is entitled to use the same or similar marks in commerce. In issuing concurrent registrations, the Director shall prescribe conditions and limitations as to the mode or place of use of the mark or the goods on or in connection with which such mark is registered to the respective persons.
(e) Consists of a mark which (1) when used on or in connection with the goods of the applicant is merely descriptive or deceptively misdescriptive of them, (2) when used on or in connection with the goods of the applicant is primarily geographically descriptive of them, except as indications of regional origin may be registrable under section 1054 of this title, (3) when used on or in connection with the goods of the applicant is primarily geographically deceptively misdescriptive of them, (4) is primarily merely a surname, or (5) comprises any matter that, as a whole, is functional.
(f) Except as expressly excluded in subsections (a), (b), (c), (d), (e)(3), and (e)(5) of this section, nothing in this chapter shall prevent the registration of a mark used by the applicant which has become distinctive of the applicant's goods in commerce. The Director may accept as prima facie evidence that the mark has become distinctive, as used on or in connection with the applicant's goods in commerce, proof of substantially exclusive and continuous use thereof as a mark by the applicant in commerce for the five years before the date on which the claim of distinctiveness is made. Nothing in this section shall prevent the registration of a mark which, when used on or in connection with the goods of the applicant, is primarily geographically deceptively misdescriptive of them, and which became distinctive of the applicant's goods in commerce before December 8, 1993.
A mark which would be likely to cause dilution by blurring or dilution by tarnishment under section 1125(c) of this title, may be refused registration only pursuant to a proceeding brought under section 1063 of this title. A registration for a mark which would be likely to cause dilution by blurring or dilution by tarnishment under section 1125(c) of this title, may be canceled pursuant to a proceeding brought under either section 1064 of this title or section 1092 of this title.
In Matal v. Tam, No. 15–1293, 582 U.S. __ (U.S. June 19, 2017), which is excerpted below, the Supreme Court held that the Lanham Act §2(a) prohibition on the registration of marks that “may disparage…persons” was invalid under the Free Speech Clause of the First Amendment. Tam is significant for a number of reasons specific to trademark law. First, it abrogated a half-century of PTO practice and federal court case law applying the §2(a)’s “disparagement clause.” Second, Tam raises significant questions about whether the Lanham Act §2(a) prohibition on the registration of any mark that “consists of or comprises…scandalous matter” is also unconstitutional. The Federal Circuit is expected to address the constitutionality of the prohibition on the registration of scandalous marks in the currently-pending appeal of In re Brunetti, Serial No. 85310960, 2014 WL 3976439 (TTAB Aug. 1, 2014) (affirming examiner’s refusal to register fuct for athletic apparel as scandalous). Third, Tam also arguably raises significant questions about whether antidilution law, which we cover in Part II.C below, is constitutional. May the government restrict non-deceptive speech that “impairs the distinctiveness of the famous mark,” 15 U.S.C. § 1125(c)(2)(B), or that “harms the reputation of the famous mark,” 15 U.S.C. § 1125(c)(2)(C)? Fourth and finally, Tam brings to an end the appeal to the Fourth Circuit of Blackhorse v. Pro-Football, Inc., 111 U.S.P.Q.2d 1080, 2014 WL 2757516 (TTAB June 18, 2014), in which five Native Americans successfully petitioned to cancel various trademark registrations consisting in whole or in part of the term redskins for professional football-related services on the ground that at the time of their registration they were disparaging of Native Americans and thus obtained contrary to Lanham Act §§ 14(c) and 2(a), 15 U.S.C. §§ 1064(c) & 1052(a). (If you strongly support Tam’s registration of the slants, what is your position on the government’s registration of the term “redskins” by a professional football team in the nation’s capital?)
Matal v. Tam
No. 15–1293, 582 U.S. __ (U.S. June 19, 2017)
 Justice ALITO announced the judgment of the Court and delivered the opinion of the Court with respect to Parts I, II, and III–A, and an opinion with respect to Parts III–B, III–C, and IV, in which THE CHIEF JUSTICE, Justice THOMAS, and Justice BREYER join.
 This case concerns a dance-rock band’s application for federal trademark registration of the band’s name, “The Slants.” “Slants” is a derogatory term for persons of Asian descent, and members of the band are Asian–Americans. But the band members believe that by taking that slur as the name of their group, they will help to “reclaim” the term and drain its denigrating force.
 The Patent and Trademark Office (PTO) denied the application based on a provision of federal law prohibiting the registration of trademarks that may “disparage ... or bring ... into contemp[t] or disrepute” any “persons, living or dead.” 15 U.S.C. § 1052(a). We now hold that this provision violates the Free Speech Clause of the First Amendment. It offends a bedrock First Amendment principle: Speech may not be banned on the ground that it expresses ideas that offend.
 “The principle underlying trademark protection is that distinctive marks—words, names, symbols, and the like—can help distinguish a particular artisan’s goods from those of others.” B & B Hardware, Inc. v. Hargis Industries, Inc., 135 S.Ct. 1293, 1299 (2015); see also Wal–Mart Stores, Inc. v. Samara Brothers, Inc., 529 U.S. 205, 212 (2000). A trademark “designate [s] the goods as the product of a particular trader” and “protect[s] his good will against the sale of another’s product as his.” United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90, 97 (1918); see also Hanover Star Milling Co. v. Metcalf, 240 U.S. 403, 412–413 (1916). It helps consumers identify goods and services that they wish to purchase, as well as those they want to avoid. See Wal–Mart Stores, supra, at 212–213; Park ‘N Fly, Inc. v. Dollar Park & Fly, Inc., 469 U.S. 189, 198 (1985).
 “[F]ederal law does not create trademarks.” B & B Hardware, supra, at ––––, 135 S.Ct., at 1299. Trademarks and their precursors have ancient origins, and trademarks were protected at common law and in equity at the time of the founding of our country. 3 J. McCarthy, Trademarks and Unfair Competition § 19:8 (4th ed. 2017) (hereinafter McCarthy); see Trade–Mark Cases, 100 U.S. 82, 92 (1879). For most of the 19th century, trademark protection was the province of the States. See Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 780–782 (1992) (Stevens, J., concurring in judgment); id., at 785 (THOMAS, J., concurring in judgment). Eventually, Congress stepped in to provide a degree of national uniformity, passing the first federal legislation protecting trademarks in 1870. See Act of July 8, 1870, §§ 77–84, 16 Stat. 210–212. The foundation of current federal trademark law is the Lanham Act, enacted in 1946. See Act of July 5, 1946, ch. 540, 60 Stat. 427. By that time, trademark had expanded far beyond phrases that do no more than identify a good or service. Then, as now, trademarks often consisted of catchy phrases that convey a message.
 Under the Lanham Act, trademarks that are “used in commerce” may be placed on the “principal register,” that is, they may be federally registered. 15 U.S.C. § 1051(a)(1). And some marks “capable of distinguishing [an] applicant’s goods or services and not registrable on the principal register ... which are in lawful use in commerce by the owner thereof” may instead be placed on a different federal register: the supplemental register. § 1091(a). There are now more than two million marks that have active federal certificates of registration. PTO Performance and Accountability Report, Fiscal Year 2016, p. 192 (Table 15), https://www.uspto.gov/sites/default/files/ documents/USPTOFY16PAR.pdf (all Internet materials as last visited June 16, 2017). This system of federal registration helps to ensure that trademarks are fully protected and supports the free flow of commerce. “[N]ational protection of trademarks is desirable,” we have explained, “because trademarks foster competition and the maintenance of quality by securing to the producer the benefits of good reputation.” San Francisco Arts & Athletics, Inc. v. United States Olympic Comm., 483 U.S. 522, 531 (1987) (internal quotation marks omitted); see also Park ‘N Fly, Inc., supra, at 198 (“The Lanham Act provides national protection of trademarks in order to secure to the owner of the mark the goodwill of his business and to protect the ability of consumers to distinguish among competing producers”).
 Without federal registration, a valid trademark may still be used in commerce. See 3 McCarthy § 19:8. And an unregistered trademark can be enforced against would-be infringers in several ways. Most important, even if a trademark is not federally registered, it may still be enforceable under § 43(a) of the Lanham Act, which creates a federal cause of action for trademark infringement. See Two Pesos, supra, at 768 (“Section 43(a) prohibits a broader range of practices than does § 32, which applies to registered marks, but it is common ground that § 43(a) protects qualifying unregistered trademarks” (internal quotation marks and citation omitted)).94 Unregistered trademarks may also be entitled to protection under other federal statutes, such as the Anticybersquatting Consumer Protection Act, 15 U.S.C. § 1125(d). See 5 McCarthy § 25A:49, at 25A–198 (“[T]here is no requirement [in the Anticybersquatting Act] that the protected ‘mark’ be registered: unregistered common law marks are protected by the Act”). And an unregistered trademark can be enforced under state common law, or if it has been registered in a State, under that State’s registration system. See 3 id., § 19:3, at 19–23 (explaining that “[t]he federal system of registration and protection does not preempt parallel state law protection, either by state common law or state registration” and “[i]n the vast majority of situations, federal and state trademark law peacefully coexist”); id., § 22:1 (discussing state trademark registration systems).
 Federal registration, however, “confers important legal rights and benefits on trademark owners who register their marks.” B & B Hardware, 135 S.Ct., at 1317 (internal quotation marks omitted). Registration on the principal register (1) “serves as ‘constructive notice of the registrant’s claim of ownership’ of the mark,” ibid. (quoting 15 U.S.C. § 1072); (2) “is ‘prima facie evidence of the validity of the registered mark and of the registration of the mark, of the owner’s ownership of the mark, and of the owner’s exclusive right to use the registered mark in commerce on or in connection with the goods or services specified in the certificate,’” B & B Hardware, 135 S.Ct., at 1300 (quoting § 1057(b)); and (3) can make a mark “‘incontestable’” once a mark has been registered for five years,” ibid. (quoting §§ 1065, 1115(b)); see Park ‘N Fly, 469 U.S., at 193. Registration also enables the trademark holder “to stop the importation into the United States of articles bearing an infringing mark.” 3 McCarthy § 19:9, at 19–38; see 15 U.S.C. § 1124.
 The Lanham Act contains provisions that bar certain trademarks from the principal register. For example, a trademark cannot be registered if it is “merely descriptive or deceptively misdescriptive” of goods, § 1052(e)(1), or if it is so similar to an already registered trademark or trade name that it is “likely ... to cause confusion, or to cause mistake, or to deceive,” § 1052(d).
 At issue in this case is one such provision, which we will call “the disparagement clause.” This provision prohibits the registration of a trademark “which may disparage ... persons, living or dead, institutions, beliefs, or national symbols, or bring them into contempt, or disrepute.” § 1052(a).95 This clause appeared in the original Lanham Act and has remained the same to this day. See § 2(a), 60 Stat. 428.
 When deciding whether a trademark is disparaging, an examiner at the PTO generally applies a “two-part test.” The examiner first considers “the likely meaning of the matter in question, taking into account not only dictionary definitions, but also the relationship of the matter to the other elements in the mark, the nature of the goods or services, and the manner in which the mark is used in the marketplace in connection with the goods or services.” Trademark Manual of Examining Procedure § 1203.03(b)(i) (Apr. 2017), p. 1200–150, http://tmep.uspto.gov. “If that meaning is found to refer to identifiable persons, institutions, beliefs or national symbols,” the examiner moves to the second step, asking “whether that meaning may be disparaging to a substantial composite96 of the referenced group.” Ibid. If the examiner finds that a “substantial composite, although not necessarily a majority, of the referenced group would find the proposed mark ... to be disparaging in the context of contemporary attitudes,” a prima facie case of disparagement is made out, and the burden shifts to the applicant to prove that the trademark is not disparaging. Ibid. What is more, the PTO has specified that “[t]he fact that an applicant may be a member of that group or has good intentions underlying its use of a term does not obviate the fact that a substantial composite of the referenced group would find the term objectionable.” Ibid.
 Simon Tam is the lead singer of “The Slants.” In re Tam, 808 F.3d 1321, 1331 (C.A.Fed.2015) (en banc), as corrected (Feb. 11, 2016). He chose this moniker in order to “reclaim” and “take ownership” of stereotypes about people of Asian ethnicity. Ibid. (internal quotation marks omitted). The group “draws inspiration for its lyrics from childhood slurs and mocking nursery rhymes” and has given its albums names such as “The Yellow Album” and “Slanted Eyes, Slanted Hearts.” Ibid.
 Tam sought federal registration of “THE SLANTS,” on the principal register, but an examining attorney at the PTO rejected the request, applying the PTO’s two-part framework and finding that “there is ... a substantial composite of persons who find the term in the applied-for mark offensive.” The examining attorney relied in part on the fact that “numerous dictionaries define ‘slants’ or ‘slant-eyes’ as a derogatory or offensive term.” The examining attorney also relied on a finding that “the band’s name has been found offensive numerous times”—citing a performance that was canceled because of the band’s moniker and the fact that “several bloggers and commenters to articles on the band have indicated that they find the term and the applied-for mark offensive.”
 Tam contested the denial of registration before the examining attorney and before the PTO’s Trademark Trial and Appeal Board (TTAB) but to no avail. Eventually, he took the case to federal court, where the en banc Federal Circuit ultimately found the disparagement clause facially unconstitutional under the First Amendment’s Free Speech Clause. The majority found that the clause engages in viewpoint-based discrimination, that the clause regulates the expressive component of trademarks and consequently cannot be treated as commercial speech, and that the clause is subject to and cannot satisfy strict scrutiny. See 808 F.3d, at 1334–1339. The majority also rejected the Government’s argument that registered trademarks constitute government speech, as well as the Government’s contention that federal registration is a form of government subsidy. See id., at 1339–1355. And the majority opined that even if the disparagement clause were analyzed under this Court’s commercial speech cases, the clause would fail the “intermediate scrutiny” that those cases prescribe. See id., at 1355–1357.
 Several judges wrote separately, advancing an assortment of theories. Concurring, Judge O’Malley agreed with the majority’s reasoning but added that the disparagement clause is unconstitutionally vague. See id., at 1358–1363. Judge Dyk concurred in part and dissented in part. He argued that trademark registration is a government subsidy and that the disparagement clause is facially constitutional, but he found the clause unconstitutional as applied to THE SLANTS because that mark constitutes “core expression” and was not adopted for the purpose of disparaging Asian–Americans. See id., at 1363–1374. In dissent, Judge Lourie agreed with Judge Dyk that the clause is facially constitutional but concluded for a variety of reasons that it is also constitutional as applied in this case. See id., at 1374–1376. Judge Reyna also dissented, maintaining that trademarks are commercial speech and that the disparagement clause survives intermediate scrutiny because it “directly advances the government’s substantial interest in the orderly flow of commerce.” See id., at 1376–1382.
 The Government filed a petition for certiorari, which we granted in order to decide whether the disparagement clause “is facially invalid under the Free Speech Clause of the First Amendment.” Pet. for Cert. i; see sub. nom. Lee v. Tam, 579 U.S. ––––, 137 S.Ct. 30 (2016).
[The Court rejected Tam’s argument that Lanham Act § 2(a) prohibits the registration of marks that disparage only “persons,” which, Tam argued, “includes only natural and juristic persons,” not “non-juristic entities such as racial and ethnic groups.”]
 Because the disparagement clause applies to marks that disparage the members of a racial or ethnic group, we must decide whether the clause violates the Free Speech Clause of the First Amendment. And at the outset, we must consider three arguments that would either eliminate any First Amendment protection or result in highly permissive rational-basis review. Specifically, the Government contends (1) that trademarks are government speech, not private speech, (2) that trademarks are a form of government subsidy, and (3) that the constitutionality of the disparagement clause should be tested under a new “government-program” doctrine. We address each of these arguments below.
 The First Amendment prohibits Congress and other government entities and actors from “abridging the freedom of speech”; the First Amendment does not say that Congress and other government entities must abridge their own ability to speak freely. And our cases recognize that “[t]he Free Speech Clause ... does not regulate government speech.” Pleasant Grove City v. Summum, 555 U.S. 460, 467 (2009); see Johanns v. Livestock Marketing Assn., 544 U.S. 550, 553 (2005) (“[T]he Government’s own speech ... is exempt from First Amendment scrutiny”); Board of Regents of Univ. of Wis. System v. Southworth, 529 U.S. 217, 235 (2000).
 As we have said, “it is not easy to imagine how government could function” if it were subject to the restrictions that the First Amendment imposes on private speech. Summum, supra, at 468; see Walker v. Texas Div., Sons of Confederate Veterans, Inc., 135 S.Ct. 2239, 2245–2247 (2015). “‘[T]he First Amendment forbids the government to regulate speech in ways that favor some viewpoints or ideas at the expense of others,’” Lamb’s Chapel v. Center Moriches Union Free School Dist., 508 U.S. 384, 394 (1993), but imposing a requirement of viewpoint-neutrality on government speech would be paralyzing. When a government entity embarks on a course of action, it necessarily takes a particular viewpoint and rejects others. The Free Speech Clause does not require government to maintain viewpoint neutrality when its officers and employees speak about that venture.
 Here is a simple example. During the Second World War, the Federal Government produced and distributed millions of posters to promote the war effort.97 There were posters urging enlistment, the purchase of war bonds, and the conservation of scarce resources.98 These posters expressed a viewpoint, but the First Amendment did not demand that the Government balance the message of these posters by producing and distributing posters encouraging Americans to refrain from engaging in these activities.
 But while the government-speech doctrine is important—indeed, essential—it is a doctrine that is susceptible to dangerous misuse. If private speech could be passed off as government speech by simply affixing a government seal of approval, government could silence or muffle the expression of disfavored viewpoints. For this reason, we must exercise great caution before extending our government-speech precedents.
 At issue here is the content of trademarks that are registered by the PTO, an arm of the Federal Government. The Federal Government does not dream up these marks, and it does not edit marks submitted for registration. Except as required by the statute involved here, 15 U.S.C. § 1052(a), an examiner may not reject a mark based on the viewpoint that it appears to express. Thus, unless that section is thought to apply, an examiner does not inquire whether any viewpoint conveyed by a mark is consistent with Government policy or whether any such viewpoint is consistent with that expressed by other marks already on the principal register. Instead, if the mark meets the Lanham Act’s viewpoint-neutral requirements, registration is mandatory. Ibid. (requiring that “[n]o trademark ... shall be refused registration on the principal register on account of its nature unless” it falls within an enumerated statutory exception). And if an examiner finds that a mark is eligible for placement on the principal register, that decision is not reviewed by any higher official unless the registration is challenged. See §§ 1062(a), 1071; 37 C.F.R § 41.31(a) (2016). Moreover, once a mark is registered, the PTO is not authorized to remove it from the register unless a party moves for cancellation, the registration expires, or the Federal Trade Commission initiates proceedings based on certain grounds. See 15 U.S.C. §§ 1058(a), 1059, 1064; 37 C.F.R. §§ 2.111(b), 2.160.
 In light of all this, it is far-fetched to suggest that the content of a registered mark is government speech. If the federal registration of a trademark makes the mark government speech, the Federal Government is babbling prodigiously and incoherently. It is saying many unseemly things. See App. to Brief for Pro–Football, Inc., as Amicus Curiae. It is expressing contradictory views.99 It is unashamedly endorsing a vast array of commercial products and services. And it is providing Delphic advice to the consuming public.
 For example, if trademarks represent government speech, what does the Government have in mind when it advises Americans to “make.believe” (Sony),100 “Think different” (Apple),101 “Just do it” (Nike),102 or “Have it your way” (Burger King)103 ? Was the Government warning about a coming disaster when it registered the mark “EndTime Ministries”104?
 The PTO has made it clear that registration does not constitute approval of a mark. See In re Old Glory Condom Corp., 26 USPQ 2d 1216, 1220, n. 3 (T.T.A.B.1993) (“[I]ssuance of a trademark registration ... is not a government imprimatur”). And it is unlikely that more than a tiny fraction of the public has any idea what federal registration of a trademark means. See Application of National Distillers & Chemical Corp., 49 C.C.P.A. (Pat.) 854, 863, 297 F.2d 941, 949 (1962) (Rich, J., concurring) (“The purchasing public knows no more about trademark registrations than a man walking down the street in a strange city knows about legal title to the land and buildings he passes” (emphasis deleted)).
 None of our government speech cases even remotely supports the idea that registered trademarks are government speech. In Johanns, we considered advertisements promoting the sale of beef products. A federal statute called for the creation of a program of paid advertising “‘to advance the image and desirability of beef and beef products.’” 544 U.S., at 561 (quoting 7 U.S.C. § 2902(13)). Congress and the Secretary of Agriculture provided guidelines for the content of the ads, Department of Agriculture officials attended the meetings at which the content of specific ads was discussed, and the Secretary could edit or reject any proposed ad. 544 U.S., at 561. Noting that “[t]he message set out in the beef promotions [was] from beginning to end the message established by the Federal Government,” we held that the ads were government speech. Id., at 560. The Government’s involvement in the creation of these beef ads bears no resemblance to anything that occurs when a trademark is registered.
 Our decision in Summum is similarly far afield. A small city park contained 15 monuments. 555 U.S., at 464. Eleven had been donated by private groups, and one of these displayed the Ten Commandments. Id., at 464–465. A religious group claimed that the city, by accepting donated monuments, had created a limited public forum for private speech and was therefore obligated to place in the park a monument expressing the group’s religious beliefs.
 Holding that the monuments in the park represented government speech, we cited many factors. Governments have used monuments to speak to the public since ancient times; parks have traditionally been selective in accepting and displaying donated monuments; parks would be overrun if they were obligated to accept all monuments offered by private groups; “[p]ublic parks are often closely identified in the public mind with the government unit that owns the land”; and “[t]he monuments that are accepted ... are meant to convey and have the effect of conveying a government message.” Id., at 472.
 Trademarks share none of these characteristics. Trademarks have not traditionally been used to convey a Government message. With the exception of the enforcement of 15 U.S.C. § 1052(a), the viewpoint expressed by a mark has not played a role in the decision whether to place it on the principal register. And there is no evidence that the public associates the contents of trademarks with the Federal Government.
 This brings us to the case on which the Government relies most heavily, Walker, which likely marks the outer bounds of the government-speech doctrine. Holding that the messages on Texas specialty license plates are government speech, the Walker Court cited three factors distilled from Summum. 135 S.Ct., at 2246–2247. First, license plates have long been used by the States to convey state messages. 135 S.Ct., at 2248–2249. Second, license plates “are often closely identified in the public mind” with the State, since they are manufactured and owned by the State, generally designed by the State, and serve as a form of “government ID.” 135 S.Ct., at 2249 (internal quotation marks omitted). Third, Texas “maintain[ed] direct control over the messages conveyed on its specialty plates.” 135 S.Ct., at 2249. As explained above, none of these factors are present in this case.
 In sum, the federal registration of trademarks is vastly different from the beef ads in Johanns, the monuments in Summum, and even the specialty license plates in Walker. Holding that the registration of a trademark converts the mark into government speech would constitute a huge and dangerous extension of the government-speech doctrine. For if the registration of trademarks constituted government speech, other systems of government registration could easily be characterized in the same way.
 Perhaps the most worrisome implication of the Government’s argument concerns the system of copyright registration. If federal registration makes a trademark government speech and thus eliminates all First Amendment protection, would the registration of the copyright for a book produce a similar transformation? See 808 F.3d, at 1346 (explaining that if trademark registration amounts to government speech, “then copyright registration” which “has identical accoutrements” would “likewise amount to government speech”).
 The Government attempts to distinguish copyright on the ground that it is “‘the engine of free expression,’” Brief for Petitioner 47 (quoting Eldred v. Ashcroft, 537 U.S. 186, 219, 123 S.Ct. 769, 154 L.Ed.2d 683 (2003)), but as this case illustrates, trademarks often have an expressive content. Companies spend huge amounts to create and publicize trademarks that convey a message. It is true that the necessary brevity of trademarks limits what they can say. But powerful messages can sometimes be conveyed in just a few words.
 Trademarks are private, not government, speech.
 We next address the Government’s argument that this case is governed by cases in which this Court has upheld the constitutionality of government programs that subsidized speech expressing a particular viewpoint. These cases implicate a notoriously tricky question of constitutional law. “[W]e have held that the Government ‘may not deny a benefit to a person on a basis that infringes his constitutionally protected ... freedom of speech even if he has no entitlement to that benefit.’” Agency for Int’l Development v. Alliance for Open Society Int’l, Inc., 133 S.Ct. 2321, 2328 (2013) (some internal quotation marks omitted). But at the same time, government is not required to subsidize activities that it does not wish to promote. Ibid. Determining which of these principles applies in a particular case “is not always self-evident,” 133 S.Ct., at 2330, but no difficult question is presented here.
 Unlike the present case, the decisions on which the Government relies all involved cash subsidies or their equivalent. In Rust v. Sullivan, 500 U.S. 173 (1991), a federal law provided funds to private parties for family planning services. In National Endowment for Arts v. Finley, 524 U.S. 569 (1998), cash grants were awarded to artists. And federal funding for public libraries was at issue in United States v. American Library Assn., Inc., 539 U.S. 194 (2003). In other cases, we have regarded tax benefits as comparable to cash subsidies. See Regan v. Taxation With Representation of Wash., 461 U.S. 540 (1983); Cammarano v. United States, 358 U.S. 498 (1959).
 The federal registration of a trademark is nothing like the programs at issue in these cases. The PTO does not pay money to parties seeking registration of a mark. Quite the contrary is true: An applicant for registration must pay the PTO a filing fee of $225–$600. 37 C.F.R. § 2.6(a)(1). (Tam submitted a fee of $275 as part of his application to register THE SLANTS. App. 18.) And to maintain federal registration, the holder of a mark must pay a fee of $300–$500 every 10 years. § 2.6(a)(5); see also 15 U.S.C. § 1059(a). The Federal Circuit concluded that these fees have fully supported the registration system for the past 27 years. 808 F.3d, at 1353.
 The Government responds that registration provides valuable non-monetary benefits that “are directly traceable to the resources devoted by the federal government to examining, publishing, and issuing certificates of registration for those marks.” But just about every government service requires the expenditure of government funds. This is true of services that benefit everyone, like police and fire protection, as well as services that are utilized by only some, e.g., the adjudication of private lawsuits and the use of public parks and highways.
 Trademark registration is not the only government registration scheme. For example, the Federal Government registers copyrights and patents. State governments and their subdivisions register the title to real property and security interests; they issue driver’s licenses, motor vehicle registrations, and hunting, fishing, and boating licenses or permits.
 Finally, the Government urges us to sustain the disparagement clause under a new doctrine that would apply to “government-program” cases. For the most part, this argument simply merges our government-speech cases and the previously discussed subsidy cases in an attempt to construct a broader doctrine that can be applied to the registration of trademarks. The only new element in this construct consists of two cases involving a public employer’s collection of union dues from its employees. But those cases occupy a special area of First Amendment case law, and they are far removed from the registration of trademarks.
 Having concluded that the disparagement clause cannot be sustained under our government-speech or subsidy cases or under the Government’s proposed “government-program” doctrine, we must confront a dispute between the parties on the question whether trademarks are commercial speech and are thus subject to the relaxed scrutiny outlined in Central Hudson Gas & Elec. Corp. v. Public Serv. Comm’n of N. Y., 447 U.S. 557 (1980). The Government and amici supporting its position argue that all trademarks are commercial speech. They note that the central purposes of trademarks are commercial and that federal law regulates trademarks to promote fair and orderly interstate commerce. Tam and his amici, on the other hand, contend that many, if not all, trademarks have an expressive component. In other words, these trademarks do not simply identify the source of a product or service but go on to say something more, either about the product or service or some broader issue. The trademark in this case illustrates this point. The name “The Slants” not only identifies the band but expresses a view about social issues.
 We need not resolve this debate between the parties because the disparagement clause cannot withstand even Central Hudson review.105 Under Central Hudson, a restriction of speech must serve “a substantial interest,” and it must be “narrowly drawn.” Id., at 564–565 (internal quotation marks omitted). This means, among other things, that “[t]he regulatory technique may extend only as far as the interest it serves.” Id., at 565. The disparagement clause fails this requirement.
 It is claimed that the disparagement clause serves two interests. The first is phrased in a variety of ways in the briefs. Echoing language in one of the opinions below, the Government asserts an interest in preventing “‘underrepresented groups’” from being “‘bombarded with demeaning messages in commercial advertising.’” Brief for Petitioner 48 (quoting 808 F.3d, at 1364 (Dyk, J., concurring in part and dissenting in part)). An amicus supporting the Government refers to “encouraging racial tolerance and protecting the privacy and welfare of individuals.” Brief for Native American Organizations as Amici Curiae 21. But no matter how the point is phrased, its unmistakable thrust is this: The Government has an interest in preventing speech expressing ideas that offend. And, as we have explained, that idea strikes at the heart of the First Amendment. Speech that demeans on the basis of race, ethnicity, gender, religion, age, disability, or any other similar ground is hateful; but the proudest boast of our free speech jurisprudence is that we protect the freedom to express “the thought that we hate.” United States v. Schwimmer, 279 U.S. 644, 655 (1929) (Holmes, J., dissenting).
 The second interest asserted is protecting the orderly flow of commerce. See 808 F.3d, at 1379–1381 (Reyna, J., dissenting); Brief for Petitioner 49; Brief for Native American Organizations as Amicus Curiae 18–21. Commerce, we are told, is disrupted by trademarks that “involv[e] disparagement of race, gender, ethnicity, national origin, religion, sexual orientation, and similar demographic classification.” 808 F.3d, at 1380–1381 (opinion of Reyna, J.). Such trademarks are analogized to discriminatory conduct, which has been recognized to have an adverse effect on commerce. See ibid.; Brief for Petitioner 49; Brief for Native American Organizations as Amici Curiae 18–20.
 A simple answer to this argument is that the disparagement clause is not “narrowly drawn” to drive out trademarks that support invidious discrimination. The clause reaches any trademark that disparages any person, group, or institution. It applies to trademarks like the following: “Down with racists,” “Down with sexists,” “Down with homophobes.” It is not an anti-discrimination clause; it is a happy-talk clause. In this way, it goes much further than is necessary to serve the interest asserted.
 The clause is far too broad in other ways as well. The clause protects every person living or dead as well as every institution. Is it conceivable that commerce would be disrupted by a trademark saying: “James Buchanan was a disastrous president” or “Slavery is an evil institution”?
 There is also a deeper problem with the argument that commercial speech may be cleansed of any expression likely to cause offense. The commercial market is well stocked with merchandise that disparages prominent figures and groups, and the line between commercial and non-commercial speech is not always clear, as this case illustrates. If affixing the commercial label permits the suppression of any speech that may lead to political or social “volatility,” free speech would be endangered.
* * *
 For these reasons, we hold that the disparagement clause violates the Free Speech Clause of the First Amendment. The judgment of the Federal Circuit is affirmed.
It is so ordered.
Justice GORSUCH took no part in the consideration or decision of this case.
Justice KENNEDY, with whom Justice GINSBURG, Justice SOTOMAYOR, and Justice KAGAN join, concurring in part and concurring in the judgment.
 The Patent and Trademark Office (PTO) has denied the substantial benefits of federal trademark registration to the mark THE SLANTS. The PTO did so under the mandate of the disparagement clause in 15 U.S.C. § 1052(a), which prohibits the registration of marks that may “disparage ... or bring ... into contemp[t] or disrepute” any “persons, living or dead, institutions, beliefs, or national symbols.”
 As the Court is correct to hold, § 1052(a) constitutes viewpoint discrimination—a form of speech suppression so potent that it must be subject to rigorous constitutional scrutiny. The Government’s action and the statute on which it is based cannot survive this scrutiny.
 The Court is correct in its judgment, and I join Parts I, II, and III–A of its opinion. This separate writing explains in greater detail why the First Amendment’s protections against viewpoint discrimination apply to the trademark here. It submits further that the viewpoint discrimination rationale renders unnecessary any extended treatment of other questions raised by the parties.
 Those few categories of speech that the government can regulate or punish—for instance, fraud, defamation, or incitement—are well established within our constitutional tradition. See United States v. Stevens, 559 U.S. 460, 468 (2010). Aside from these and a few other narrow exceptions, it is a fundamental principle of the First Amendment that the government may not punish or suppress speech based on disapproval of the ideas or perspectives the speech conveys. See Rosenberger v. Rector and Visitors of Univ. of Va., 515 U.S. 819, 828–829 (1995).
 The First Amendment guards against laws “targeted at specific subject matter,” a form of speech suppression known as content based discrimination. Reed v. Town of Gilbert, 576 U.S. ––––, ––––, 135 S.Ct. 2218, 2230 (2015). This category includes a subtype of laws that go further, aimed at the suppression of “particular views ... on a subject.” Rosenberger, 515 U.S., at 829. A law found to discriminate based on viewpoint is an “egregious form of content discrimination,” which is “presumptively unconstitutional.” Id., at 829–830.
 At its most basic, the test for viewpoint discrimination is whether—within the relevant subject category—the government has singled out a subset of messages for disfavor based on the views expressed. See Cornelius v. NAACP Legal Defense & Ed. Fund, Inc., 473 U.S. 788, 806 (1985) (“[T]he government violates the First Amendment when it denies access to a speaker solely to suppress the point of view he espouses on an otherwise includible subject”). In the instant case, the disparagement clause the Government now seeks to implement and enforce identifies the relevant subject as “persons, living or dead, institutions, beliefs, or national symbols.” 15 U.S.C. § 1052(a). Within that category, an applicant may register a positive or benign mark but not a derogatory one. The law thus reflects the Government’s disapproval of a subset of messages it finds offensive. This is the essence of viewpoint discrimination.
 The Government disputes this conclusion. It argues, to begin with, that the law is viewpoint neutral because it applies in equal measure to any trademark that demeans or offends. This misses the point. A subject that is first defined by content and then regulated or censored by mandating only one sort of comment is not viewpoint neutral. To prohibit all sides from criticizing their opponents makes a law more viewpoint based, not less so. Cf. Rosenberger, supra, at 831–832 (“The ... declaration that debate is not skewed so long as multiple voices are silenced is simply wrong; the debate is skewed in multiple ways”). The logic of the Government’s rule is that a law would be viewpoint neutral even if it provided that public officials could be praised but not condemned. The First Amendment’s viewpoint neutrality principle protects more than the right to identify with a particular side. It protects the right to create and present arguments for particular positions in particular ways, as the speaker chooses. By mandating positivity, the law here might silence dissent and distort the marketplace of ideas.
 The Government next suggests that the statute is viewpoint neutral because the disparagement clause applies to trademarks regardless of the applicant’s personal views or reasons for using the mark. Instead, registration is denied based on the expected reaction of the applicant’s audience. In this way, the argument goes, it cannot be said that Government is acting with hostility toward a particular point of view. For example, the Government does not dispute that respondent seeks to use his mark in a positive way. Indeed, respondent endeavors to use The Slants to supplant a racial epithet, using new insights, musical talents, and wry humor to make it a badge of pride. Respondent’s application was denied not because the Government thought his object was to demean or offend but because the Government thought his trademark would have that effect on at least some Asian–Americans.
 The Government may not insulate a law from charges of viewpoint discrimination by tying censorship to the reaction of the speaker’s audience. The Court has suggested that viewpoint discrimination occurs when the government intends to suppress a speaker’s beliefs, Reed, 135 S.Ct., at 2229–2230, but viewpoint discrimination need not take that form in every instance. The danger of viewpoint discrimination is that the government is attempting to remove certain ideas or perspectives from a broader debate. That danger is all the greater if the ideas or perspectives are ones a particular audience might think offensive, at least at first hearing. An initial reaction may prompt further reflection, leading to a more reasoned, more tolerant position.
 Indeed, a speech burden based on audience reactions is simply government hostility and intervention in a different guise. The speech is targeted, after all, based on the government’s disapproval of the speaker’s choice of message. And it is the government itself that is attempting in this case to decide whether the relevant audience would find the speech offensive. For reasons like these, the Court’s cases have long prohibited the government from justifying a First Amendment burden by pointing to the offensiveness of the speech to be suppressed.
 The Government’s argument in defense of the statute assumes that respondent’s mark is a negative comment. In addressing that argument on its own terms, this opinion is not intended to imply that the Government’s interpretation is accurate. From respondent’s submissions, it is evident he would disagree that his mark means what the Government says it does. The trademark will have the effect, respondent urges, of reclaiming an offensive term for the positive purpose of celebrating all that Asian–Americans can and do contribute to our diverse Nation. While thoughtful persons can agree or disagree with this approach, the dissonance between the trademark’s potential to teach and the Government’s insistence on its own, opposite, and negative interpretation confirms the constitutional vice of the statute.
 The parties dispute whether trademarks are commercial speech and whether trademark registration should be considered a federal subsidy. The former issue may turn on whether certain commercial concerns for the protection of trademarks might, as a general matter, be the basis for regulation. However that issue is resolved, the viewpoint based discrimination at issue here necessarily invokes heightened scrutiny.
 “Commercial speech is no exception,” the Court has explained, to the principle that the First Amendment “requires heightened scrutiny whenever the government creates a regulation of speech because of disagreement with the message it conveys.” Sorrell v. IMS Health Inc., 564 U.S. 552, 566 (2011) (internal quotation marks omitted). Unlike content based discrimination, discrimination based on viewpoint, including a regulation that targets speech for its offensiveness, remains of serious concern in the commercial context. See Bolger v. Youngs Drug Products Corp., 463 U.S. 60, 65, 71–72 (1983).
 To the extent trademarks qualify as commercial speech, they are an example of why that term or category does not serve as a blanket exemption from the First Amendment’s requirement of viewpoint neutrality. Justice Holmes’ reference to the “free trade in ideas” and the “power of ... thought to get itself accepted in the competition of the market,” Abrams v. United States, 250 U.S. 616, 630 (1919) (dissenting opinion), was a metaphor. In the realm of trademarks, the metaphorical marketplace of ideas becomes a tangible, powerful reality. Here that real marketplace exists as a matter of state law and our common-law tradition, quite without regard to the Federal Government. These marks make up part of the expression of everyday life, as with the names of entertainment groups, broadcast networks, designer clothing, newspapers, automobiles, candy bars, toys, and so on. See Brief for Pro–Football, Inc., as Amicus Curiae 8 (collecting examples). Nonprofit organizations—ranging from medical-research charities and other humanitarian causes to political advocacy groups—also have trademarks, which they use to compete in a real economic sense for funding and other resources as they seek to persuade others to join their cause. See id., at 8–9 (collecting examples). To permit viewpoint discrimination in this context is to permit Government censorship.
 This case does not present the question of how other provisions of the Lanham Act should be analyzed under the First Amendment. It is well settled, for instance, that to the extent a trademark is confusing or misleading the law can protect consumers and trademark owners. See, e.g., FTC v. Winsted Hosiery Co., 258 U.S. 483, 493 (1922) (“The labels in question are literally false, and ... palpably so. All are, as the Commission found, calculated to deceive and do in fact deceive a substantial portion of the purchasing public”). This case also does not involve laws related to product labeling or otherwise designed to protect consumers. See Sorrell, supra, at 579, (“[T]he government’s legitimate interest in protecting consumers from commercial harms explains why commercial speech can be subject to greater governmental regulation than noncommercial speech” (internal quotation marks omitted)). These considerations, however, do not alter the speech principles that bar the viewpoint discrimination embodied in the statutory provision at issue here.
 It is telling that the Court’s precedents have recognized just one narrow situation in which viewpoint discrimination is permissible: where the government itself is speaking or recruiting others to communicate a message on its behalf. See Legal Services Corporation v. Velazquez, 531 U.S. 533, 540–542 (2001); Board of Regents of Univ. of Wis. System v. Southworth, 529 U.S. 217, 229, 235 (2000); Rosenberger, 515 U.S., at 833. The exception is necessary to allow the government to stake out positions and pursue policies. See Southworth, supra, at 235. But it is also narrow, to prevent the government from claiming that every government program is exempt from the First Amendment. These cases have identified a number of factors that, if present, suggest the government is speaking on its own behalf; but none are present here.
 There may be situations where private speakers are selected for a government program to assist the government in advancing a particular message. That is not this case either. The central purpose of trademark registration is to facilitate source identification. To serve that broad purpose, the Government has provided the benefits of federal registration to millions of marks identifying every type of product and cause. Registered trademarks do so by means of a wide diversity of words, symbols, and messages. Whether a mark is disparaging bears no plausible relation to that goal. While defining the purpose and scope of a federal program for these purposes can be complex, see, e.g., Agency for Int’l Development v. Alliance for Open Society Int’l, Inc., 133 S.Ct. 2321, 2328 (2013), our cases are clear that viewpoint discrimination is not permitted where, as here, the Government “expends funds to encourage a diversity of views from private speakers,” Velazquez, supra, at 542 (internal quotation marks omitted).
* * *
 A law that can be directed against speech found offensive to some portion of the public can be turned against minority and dissenting views to the detriment of all. The First Amendment does not entrust that power to the government’s benevolence. Instead, our reliance must be on the substantial safeguards of free and open discussion in a democratic society.
 For these reasons, I join the Court’s opinion in part and concur in the judgment.
Justice THOMAS, concurring in part and concurring in the judgment.
 I also write separately because “I continue to believe that when the government seeks to restrict truthful speech in order to suppress the ideas it conveys, strict scrutiny is appropriate, whether or not the speech in question may be characterized as ‘commercial.’” Lorillard Tobacco Co. v. Reilly, 533 U.S. 525, 572 (2001) (THOMAS, J., concurring in part and concurring in judgment); see also, e.g., 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484, 518 (1996) (same). I nonetheless join Part IV of Justice ALITO’s opinion because it correctly concludes that the disparagement clause, 15 U.S.C. § 1052(a), is unconstitutional even under the less stringent test announced in Central Hudson Gas & Elec. Corp. v. Public Serv. Comm’n of N. Y., 447 U.S. 557 (1980).
Comments and Questions
1. Marijuana marks. To qualify for federal registration, a mark must be used in commerce for goods or services that are legal under federal law. For this reason, the T.T.A.B. has affirmed the refusal of registration of the mark herbal access for “retail store services featuring herbs” when such services consisted of the sale of marijuana in Washington state, under whose law such sales are legal. In re Brown, 119 USPQ2d 1350 (TTAB 2016). See also In re JJ206, LLC, dba JuJu Joints, 120 USPQ2d 1568 (TTAB 2016).
Lanham Act § 45, 15 U.S.C. § 1127
The term “use in commerce” means the bona fide use of a mark in the ordinary course of trade, and not made merely to reserve a right in a mark. For purposes of this chapter, a mark shall be deemed to be in use in commerce--
(1) on goods when--
(A) it is placed in any manner on the goods or their containers or the displays associated therewith or on the tags or labels affixed thereto, or if the nature of the goods makes such placement impracticable, then on documents associated with the goods or their sale, and
(B) the goods are sold or transported in commerce, and
(2) on services when it is used or displayed in the sale or advertising of services and the services are rendered in commerce, or the services are rendered in more than one State or in the United States and a foreign country and the person rendering the services is engaged in commerce in connection with the services.
The word “commerce” means all commerce which may lawfully be regulated by Congress.
The trademark owner must make a “use in commerce” of its mark, as that phrase is defined in Lanham Act § 45, in order for the mark to qualify for registration under § 1 of the Lanham Act106 or for protection as an unregistered mark under § 43(a). See § 1(a)(1) (providing that “[t]he owner of a trademark used in commerce” may apply for registration of that mark); § 1(b) (providing that the owner of mark filed on an intent-to-use basis must file a “[v]erified statement that trademark is used in commerce” in order for the mark to proceed to registration); International Bancorp, LLC v. Societe des Bains de Mer et du Cercle des Estrangers a Monaco, 329 F.3d 359 (4th Cir. 2003) (assuming, without citing a statutory basis, that for an unregistered mark to qualify for protection under Section 43(a), it must be “use[d] in commerce”).
We consider here what kinds of uses of a mark will satisfy § 45’s definition of “use in commerce.” Both of the opinions below emerge out of priority disputes, i.e., disputes over who by virtue of their being the first to “use in commerce” a particular mark can claim exclusive rights in that mark. (We will address priority in more detail in Part I.E on the geographic scope of trademark rights). But underlying the priority issue in both cases is the more basic question of how much and what kind of use will make the mark registrable or otherwise protectable under the Lanham Act. The first opinion, Aycock Engineering, Inc. v. Airflite, Inc., 560 F.3d 1350 (Fed. Cir. 2009), involves the question of whether Aycock made sufficient use in commerce of its mark to justify registration of the mark at the PTO. The second opinion, Planetary Motion, Inc. v. Techsplosion, Inc., 261 F.3d 1188 (11th Cir. 2001), involves the question of whether the assignee of the unregistered mark coolmail for email services could benefit from the priority date established by the assignor’s pre-assignment use of the mark.
To avoid ambiguity, it may be useful to recognize from the start the several different aspects of the concept of “use in commerce” in U.S. trademark law, only one of which we will focus on in this subsection.
The student is strongly advised to distinguish between these various aspects of “use in commerce” as we proceed, particularly the difference between “use in commerce” by the plaintiff for purposes of establishing the plaintiff’s trademark rights versus “use in commerce” by the defendant for purposes of establishing the defendant’s trademark infringement.
In reading Aycock Engineering and Planetary Motion, consider the following questions:
How helpful is the “totality of the circumstances” test referenced in Planetary Motion? Can you think of a better alternative test?
Aycock Engineering, Inc. v. Airflite, Inc.
560 F.3d 1350 (Fed. Cir. 2009)
O'GRADY, District Judge:
 In 1970, Respondent–Appellant Aycock Engineering, Inc. (“Aycock Engineering”) applied for a service mark, which was registered at the United States Patent and Trademark Office (“USPTO”) in 1974 after examination. In 2007, however, the USPTO Trademark Trial and Appeal Board (“TTAB”) declared the registration void because it failed to meet the “use in commerce” element of the Lanham Act. Aycock Engineering now appeals the TTAB's ruling. The question presented herein is whether the use in commerce requirement is met when an applicant uses a service mark in the preparatory stages of a service's development, but never offers the service to the public. We hold that it is not.
 In the late 1940s, William Aycock conceived of and began work on a service involving chartering flights in the air taxi industry. At that time, the common practice for air taxi companies was to lease entire airplanes, not individual seats. Consequently, individual passengers not belonging to a larger party faced more difficulty and expense in chartering a flight. Mr. Aycock intended, through his service, to allow solo passengers to arrange flights on chartered aircraft for less cost.
 Mr. Aycock did not plan on operating the chartered air taxi services himself. Instead, his goal was to develop a system where he would serve his customers by acting as the middleman, or “communication link,” between the customer and one of the air taxi service operators he contracted with to provide flights on an individual seat basis. Mr. Aycock planned to advertise his service, which he called the AIRFLITE service, to the public and to have those interested in using the service call a toll-free phone number to schedule reservations. After learning of customers' travel plans, Mr. Aycock would then arrange for the air taxi service to fly his customers with similar travel plans to their destinations. Mr. Aycock believed that in order for his service to become operational, he needed at least 300 air taxi operators in the United States to agree to participate in his air-taxi-operator network.107
 In the years after conceiving of the idea for his service, Mr. Aycock worked toward offering the service to the public. In the mid–1960s, he formed Aycock Engineering—the corporate entity under which his service would operate. He also sought and obtained two toll-free telephone numbers that the public could use to make reservations. In March of 1970, Mr. Aycock invited virtually all air taxi operators certified by the Federal Aviation Administration (“FAA”) to join his operation by, inter alia, distributing flyers with in-depth information about his AIRFLITE service. He eventually entered into contracts with some of those air taxi service operators.108 Under these contracts, air taxi operators agreed to participate in the AIRFLITE service and even paid modest initiation fees to Mr. Aycock. Furthermore, Mr. Aycock filed a service mark application on August 10, 1970 for the term AIRFLITE.
 Despite his efforts, Mr. Aycock's operation never got off the ground. While he estimated that he needed at least 300 air service operators under contract to make his service operational, Mr. Aycock never had more than twelve (4% of his minimum goal) under contract at any time throughout his company's history. And while Mr. Aycock advertised to air taxi operators, he never marketed the AIRFLITE service to the general public. More specifically, the record does not suggest that Mr. Aycock ever gave the public an opportunity to use the toll-free phone numbers to book reservations, or that he ever spoke with a member of the general public about making a reservation. Finally, and most notably, Mr. Aycock never arranged for a single passenger to fly on a chartered flight.109
 Mr. Aycock's AIRFLITE mark, which he applied for on August 10, 1970, was registered by the USPTO on April 30, 1974 on the Supplemental Register110 after a prosecution that involved considerable negotiation between Mr. Aycock and the trademark examining attorney…. The recitation of services for the AIRFLITE service mark eventually agreed upon by the USPTO and Mr. Aycock was “[a]rranging for individual reservations for flights on airplanes.” Id. at 729. Mr. Aycock's application to renew his AIRFLITE service mark was granted by the USPTO on April 27, 1994.
 In 2001, Airflite, Inc., the Petitioner–Appellee, filed a petition for cancellation alleging, inter alia, that Aycock Engineering did not use its AIRFLITE mark prior to registration in connection with the services identified in its registration. In that proceeding, the TTAB agreed with Airflite, Inc. and cancelled the AIRFLITE registration, finding that Mr. Aycock failed to render the service described in its registration in commerce. Airflite, Inc. v. Aycock Eng'g, Inc., Cancellation 92032520, 2007 WL 2972237, at *7 (TTAB Oct. 4, 2007) (“TTAB Decision ”).
D. Use Requirement
 Under § 45 of the Lanham Act, a service mark is any “word, name, symbol or device, or any combination thereof used by a person, or which a person has a bona fide intention to use in commerce ... to identify and distinguish the services of one person ... from the services of others.” 15 U.S.C. § 1127 (2006). The definition of “service mark” is virtually identical to the definition of “trademark.” But while service marks apply to intangible services, trademarks are used to distinguish tangible goods. See Chance v. Pac–Tel Teletrac Inc., 242 F.3d 1151, 1156 (9th Cir.2001).
 “It is clear from the wording of the Lanham Act that applications for service mark registrations are subject to the same statutory criteria as are trademarks.” 3 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 19:82 (4th ed.2008) [hereinafter McCarthy]; see 15 U.S.C. § 1053 (2006). One such statutory criterion that applies to both trademarks and service marks is the “use in commerce” requirement.… The registration of a mark that does not meet the use requirement is void ab initio. See Gay Toys, Inc. v. McDonald's Corp., 585 F.2d 1067, 1068 (CCPA 1978); 3 McCarthy § 19:112.
 Despite the seeming harmony and simplicity in the application of the use requirement to trademarks and service marks, opportunity exists for confusion in this area of the law. Different statutory requirements apply to applications filed before November 16, 1989, as compared to those filed after. This is because in 1988, Congress passed the Trademark Law Revision Act (“TLRA”). The TLRA altered the burden that applicants must meet before satisfying the use element by requiring an applicant to make a “bona fide use of [the] mark in the ordinary course of trade.” Trademark Law Revision Act of 1988, Pub.L. No. 100–667, 102 Stat. 3935 (effective November 16, 1989) (codified at 15 U.S.C. § 1127 (2006)).
 This “bona fide use” language was intended to eliminate “token uses,” which occurred when applicants used marks in conjunction with selling goods or offering services for the sole purpose of obtaining registration, and with no intention of legitimately using the mark in commerce until a later date. See Blue Bell, Inc. v. Jaymar–Ruby, Inc., 497 F.2d 433, 437 (2d Cir.1974). Before 1989, a “token use” was sufficient to satisfy the use requirement and qualify a mark for registration. See Id.
 In addition to eliminating token uses, the 1988 TLRA made other changes to the use requirement. Before 1989, an applicant only qualified for registration if he was using his mark in commerce at the time he filed his application at the USPTO. WarnerVision Entm't Inc. v. Empire of Carolina, Inc., 101 F.3d 259, 260 (2d Cir.1996). But after 1989, an applicant could begin the registration process even when his mark was not in use in commerce at the time of the filing, so long as he had a “bona fide intention to use the mark in commerce” at a later date. 15 U.S.C. § 1051(b) (2006). Applicants filing these “intent to use” applications are only granted registration, however, if they file a verified statement of commercial use proving eventual use of the mark in commerce. Id. § 1051(d).
 Because the mark at issue here is a service mark, the use requirement relating to service mark applications, as opposed to trademark applications, guides our analysis. Furthermore, the application at issue in this case was filed in 1970. Therefore, this case must be decided according to the service mark use requirement that appeared in the Lanham Act in 1970 (i.e., the pre–1989 version). See 3 McCarthy § 19:112. However, for the reasons stated below, our holding in this case also applies to the current (and post–1989) service mark use requirement.
E. Use Requirement for Service Marks
 With the exception of the 1988 TLRA statutory language eliminating token uses and permitting intent-to-use applications, the service mark use requirement as it appeared in 1970 is materially identical to the post–1989 version. The use provision of the Lanham Act in force in 1970 stated that a service mark was in use in commerce “when it is used or displayed in the sale or advertising of services, and the services are rendered in commerce, or the services are rendered in more than one State or in this and a foreign country and the person rendering the services is engaged in commerce in connection therewith.” Pub.L. No. 87–772, 76 Stat. 769 (1962). Therefore, like the current use requirement, a service mark applicant seeking to meet the pre–1989 version had to (1) use the mark in the sale or advertising of a service and (2) show that the service was either rendered in interstate commerce or rendered in more than one state or in this and a foreign country by a person engaged in commerce.
 Courts, as well as the TTAB, have interpreted the pre–1989 statutory language in analogous cases. Without question, advertising or publicizing a service that the applicant intends to perform in the future will not support registration. In re Cedar Point, Inc., 220 USPQ 533, 536 (TTAB 1983) (quoting Intermed Commc'ns, Inc. v. Chaney, 197 USPQ 501, 507–08 (TTAB 1977)); Greyhound Corp. v. Armour Life Ins. Co., 214 USPQ 473, 474 (TTAB 1982). Instead, the advertising or publicizing must relate to “an existing service which has already been offered to the public.” Greyhound, 214 USPQ at 474. Furthermore, “[m]ere adoption (selection) of a mark accompanied by preparations to begin its use are insufficient ... for claiming ownership of and applying to register the mark.” Intermed, 197 USPQ at 507; see Blue Bell, 497 F.2d at 437. “At the very least,” in order for an applicant to meet the use requirement, “there must be an open and notorious public offering of the services to those for whom the services are intended.” Intermed, 197 USPQ at 507.
 In Intermed, the TTAB rejected a service mark application for failing to meet the use in commerce requirement even where the applicant had performed many pre-application service-oriented activities involving the public. Id. at 508–09. The applicant in that case sought to register a mark intended to identify an international medical services operation. Id. at 502. The applicant's plan was to build the international service from an already operating United States-based medical service. Id. at 503. The applicant intended to, and did use the United States-based operation as a fundraising affiliate of the new international operation. Id. at 504. Additionally, the applicant communicated with and solicited the support of the Iranian government regarding the service before the application was filed. Id. The applicant also issued a detailed announcement using the service mark term before the filing date designed to inform and update individuals about the service's status. Id. Finally, and also before the date of application, the applicant hired a fundraising firm to raise money for the service. Id. at 508.
 Despite these activities, the TTAB held that the applicant failed to meet the use requirement because the services described in the application were not “offered, promoted, advertised or rendered ... in commerce.” Intermed, 197 USPQ at 504. The TTAB stated that “[t]he statute requires not only the display of the mark in the sale or advertising of services but also the rendition of those services in order to constitute use of the service mark in commerce.” Id. At 507–08. The TTAB further explained that adopting a mark accompanied by mere “preparations to begin its use” is insufficient for service mark registration, and that in order for the use requirement to be met, there must be “an open and notorious public offering of the services to those for whom the services are intended.” Id. at 507.
 In 1983, the TTAB again rejected a service mark application because it failed to meet the use requirement. Cedar Point, 220 USPQ at 533. In Cedar Point, the Cedar Point amusement park, which had been in business for decades, was preparing to open a new water park addition in mid-May of 1980. Id. at 535. One preparatory step taken by Cedar Point before opening day was the filing of a service mark application to register the mark “OCEANA” for its new water park service. Id. Cedar Point also distributed nearly 700,000 water park advertisement brochures containing the OCEANA mark during the months preceding the grand opening. Id.
 The TTAB emphasized the fact that Cedar Point filed its service mark application with the USPTO before it opened the water park's doors and offered those services to the public. Id. at 535–36. The TTAB then explained that the use of a mark in connection with the advertising of services intended to be “available at some time in the future, but not yet available at the time of filing” does not qualify the mark for registration. Id. at 535. Therefore, Cedar Point's water park advertising campaign, which was ongoing at the time the application was filed, was insufficient on its own to support registration. Id. As a result, the TTAB held that the “applicant's mark ‘OCEANA’ was not in ‘use in commerce’ ... at the time of the filing of [the] application” and that the application was thus void ab initio. Id. at 537.
 Interestingly, Cedar Point filed for its service mark roughly one month before the scheduled opening of the new water park. Id. at 535. With the application date being so close to the opening date, it is indisputable that Cedar Point had taken numerous steps toward constructing the water park by the time the application was filed. Nevertheless, the TTAB found none of these preparatory steps sufficient to satisfy the use in commerce requirement.
 The TTAB also addressed the use in commerce issue in the 1982 Greyhound case. Greyhound, 214 USPQ at 473. In that case, the applicant, a life insurance company, filed a service mark application in November of 1979. Id. at 474. Before the filing date, the applicant advertised its services by disseminating informational letters and posters using the service mark. Id. Despite this activity, the TTAB held that the service described in the application was not rendered in commerce and thus declared the application void ab initio. Id. at 475. The TTAB explained that “it is well settled that advertising of a service, without performance of a service, will not support registration.... The use in advertising which creates a right in a service mark must be advertising which relates to an existing service which has already been offered to the public.” Id. at 474.
 We find the reasoning of these cases persuasive. The language of the statute, by requiring that the mark be “used or displayed in the sale or advertising of services, and the services are rendered in commerce,” makes plain that advertisement and actual use of the mark in commerce are required; mere preparations to use that mark sometime in the future will not do. Thus, we hold that an applicant's preparations to use a mark in commerce are insufficient to constitute use in commerce. Rather, the mark must be actually used in conjunction with the services described in the application for the mark.
 But [Aycock's] activities, even taken together, do not constitute a service that falls within the scope of our definition of the recitation of services. As mentioned earlier, it is our view that the service described in Mr. Aycock's service mark application covers only the arranging of flights between an air taxi operator and a passenger, and not preparatory efforts to arrange a network of air taxi operators. The activities described above, however, were merely preparatory steps that Mr. Aycock took toward his goal of one day, as he described, operating a “communication service between persons desiring to charter aircraft” that “put[ ] individuals desiring air transportation in contact with people rendering that service.” J.A. 736, 749.
 In order for Mr. Aycock to satisfy the use requirement, more was required. Mr. Aycock had to develop his company to the point where he made an open and notorious public offering of his AIRFLITE service to intended customers. See Intermed, 197 USPQ at 507. However, at no point in time did Mr. Aycock give a potential customer the chance to use his AIRFLITE service. He never arranged for a single flight between a customer and an air taxi operator. This is because Mr. Aycock, as stated in his deposition, believed he needed at least 300 air taxi operators under contract before his service could become operational. Reasonably, because he never had more than twelve air taxi operators under contract at any one time, Mr. Aycock chose not to open his doors to the public.
[The court affirmed the TTAB’s cancellation of Aycock’s mark. Note that Judge Newman dissented on the ground that, notwithstanding the description of services listed in the registration that was finally agreed to by Aycock, “it is inappropriate now to construe the registration so as to exclude the actual use of the mark as was explained in the examination, shown in the specimens, and fully explored in the public record of the prosecution.” Aycock Eng’g, 560 F.3d at 1365 (Newman, J., dissenting).]
Planetary Motion, Inc. v. Techsplosion, Inc.
261 F.3d 1188 (11th Cir. 2001)
 Planetary Motion, Inc. (“Planetary Motion” or “Appellee”) sued Techsplosion, Inc. and Michael Gay a/k/a Michael Carson (respectively “Techsplosion” and “Carson”; collectively “Appellants”) for infringement and dilution of an unregistered trademark under Section 43(a) and (c) of the Federal Trademark Act, 15 U.S.C. § 1051 et seq. (1994) (“Lanham Act”), and for violation of Florida's unfair competition law. Fla. Stat. Ann. § 495.151 (West 2000). Finding that Planetary Motion had established priority of use and a likelihood of confusion, the United States District Court for the Southern District of Florida entered summary judgment in favor of Planetary Motion. We affirm the judgment….
I. Development and Distribution of the “Coolmail” Software
 In late 1994, Byron Darrah (“Darrah”) developed a UNIX-based program (the “Software”) that provides e-mail users with notice of new e-mail and serves as a gateway to the users' e-mail application. On December 31, 1994, Darrah distributed the Software over the Internet by posting it on a UNIX user site called “Sunsite,” from which it could be downloaded for free. Darrah had named the Software “Coolmail” and this designation appeared on the announcement sent to the end-users on Sunsite as well as on the Software user-manual, both of which accompanied the release.
 The Software was distributed without charge to users pursuant to a GNU General Public License that also accompanied the release. A GNU General Public License allows users to copy, distribute and/or modify the Software under certain restrictions, e.g., users modifying licensed files must carry “prominent notices” stating that the user changed the files and the date of any change. After the release of the Software, Darrah received correspondence from users referencing the “Coolmail” mark and in some cases suggesting improvements. In 1995, Darrah released two subsequent versions of the Software under the same mark and also pursuant to the GNU General Public License.
 In early 1995, a German company named S.u.S.E. GmbH sought permission from Darrah to include the Software in a CD-ROM package sold as a compilation of Unix-based programs. Darrah consented and, pursuant to the GNU licensing agreement, S.u.S.E. distributed the Software in its compilation product and in subsequent versions thereof. S.u.S.E. sold and continues to sell the software compilation in stores in the United States and abroad, as well as over the Internet.
II. Launch of Techsplosion's “CoolMail” E-mail Service
 In 1998, Appellant Carson formed Techsplosion, for the purpose of operating a business based on an e-mail service that he had developed. On April 16, 1998, Techsplosion began offering the e-mail service on the Internet under the mark “CoolMail.” Two days later, Techsplosion activated the domain name “coolmail.to”. Techsplosion delivered an e-mail solicitation under the “CoolMail” mark to approximately 11,000 members of the Paramount Banner Network, an Internet advertising network, also created and operated by Carson. Techsplosion charged no fee to subscribe to the service and generated revenues through the sale of banner advertisements on its web site.
III. Planetary Motion's E-mail Service & Application for Trademark Registration
 Appellee Planetary Motion is a computer software and telecommunications company that developed and owns an electronic mail service called “Coolmail.” As part of its service, Planetary Motion enables a person to check e-mail via telephone without logging onto a computer. On April 24, 1998, Planetary Motion filed three intent-to-use applications to register the mark “Coolmail” with the United States Patent and Trademark Office. Though Planetary Motion was aware that Darrah's Software also bore the mark “Coolmail,” it represented in its applications that it was not aware of any mark upon which its proposed registered mark would infringe. Planetary Motion launched its Coolmail e-mail service to subscribers on June 8, 1998.
IV. Planetary Motion's Complaint and Subsequent Acquisition of Darrah's Rights
 On April 22, 1999, Planetary Motion filed a complaint against Techsplosion. In the complaint, Planetary Motion alleged infringement of the alleged mark “Coolmail” for use in connection with e-mail services. Planetary alleged federal trademark infringement and unfair competition under Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a), as well as injury to business reputation and dilution under Florida Statute § 495.151.
 On June 10, 1999, Techsplosion filed an Answer, Affirmative Defenses, and Counterclaims. The counterclaims alleged infringement of the mark “Coolmail” for use in connection with e-mail services. Techsplosion alleged unfair competition, false designation, description, and representation under the Lanham Act, common trademark infringement, common law unfair competition, and injury to business reputation and dilution.
 In July of 1999, Planetary Motion purchased from Darrah all rights, title, and interest to the Software including all copyrights, trademarks, patents and other intellectual property rights.111 On August 31, 1999, Planetary filed an Amended Verified Complaint, adding a claim for dilution under Section 43(c) of the Lanham Act, 15 U.S.C. § 1125(c), and alleging violation of trademark rights assigned from Darrah.
V. Disposition of Planetary Motion's Complaint
 On January 31, 2000, the district court entered an Order granting Planetary Motion's motion for summary judgment and denying Carson's and Techsplosion's motion for summary judgment. The district court based the Order on two findings: (1) that the alleged mark was affixed to Darrah's software, and that Darrah's distribution of the software over the Internet constituted a “transport in commerce,” resulting in the creation of trademark rights and priority, and (2) there was a likelihood of confusion because the marks “are essentially the same.” The district court did not reach the issue of whether Techsplosion's use of “CoolMail” in connection with its e-mail service diluted Planetary Motion's mark.
 On the same date, the district court entered final judgment granting Planetary Motion permanent injunctive relief. See 15 U.S.C. § 1116. The order also awarded Planetary Motion profits and damages, as well as attorney fees and costs, pursuant to section 35 of the Lanham Act, 15 U.S.C. § 1117. [Techsplosion appealed.]
 Section 43(a) of the Lanham Act forbids unfair trade practices involving infringement of trade dress, service marks, or trademarks, even in the absence of federal trademark registration. Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 768 (1992)…. To prevail under this section, a claimant must show (1) that it had prior rights to the mark at issue and (2) that the defendant had adopted a mark or name that was the same, or confusingly similar to its mark, such that consumers were likely to confuse the two. Lone Star Steakhouse & Saloon, Inc. v. Longhorn Steaks, Inc., 106 F.3d 355, 360 (11th Cir.1997) (citing Conagra Inc. v. Singleton, 743 F.2d 1508, 1512 (11th Cir.1984)), modified, 122 F.3d 1379 (1997). Appellants argue that the district court erred in finding that Planetary Motion had established both elements. Appellants also dispute the scope of injunctive relief, as well as the award of attorney fees and costs.
I. Prior Use in Commerce
 Under common law, trademark ownership rights are “appropriated only through actual prior use in commerce.” Tally-Ho, Inc. v. Coast Community College Dist., 889 F.2d 1018, 1022 (11th Cir.1989) (citation omitted). Under the Lanham Act,112 the term “use in commerce” is defined in … 15 U.S.C. § 1127.113 The district court found that because the statute is written in the disjunctive (i.e., “sale or transport”), Darrah's wide distribution of the Coolmail software over the Internet, even absent any sales thereof, was sufficient to establish ownership rights in the “CoolMail” mark. Appellants contend that “transport in commerce” alone—here, Darrah's free distribution of software over the Internet “with no existing business, no intent to form a business, and no sale under the mark”—is insufficient to create trademark rights. Appellants' Brief at 13. Appellants' argument lacks merit.
 The parties do not make clear the two different contexts in which the phrase “use in commerce” is used. The term “use in commerce” as used in the Lanham Act “denotes Congress's authority under the Commerce Clause rather than an intent to limit the [Lanham] Act's application to profit making activity.” United We Stand Am., Inc. v. United We Stand, Am. N.Y., Inc., 128 F.3d 86, 92-93 (2d Cir.1997) (citation omitted), cert. denied, 523 U.S. 1076 (1998); U.S. Const., Art. I, § 8, cl. 3. Because Congress's authority under the Commerce Clause extends to activity that “substantially affects” interstate commerce, United States v. Lopez, 514 U.S. 549, 559 (1995), the Lanham Act's definition of “commerce” is concomitantly broad in scope: “all commerce which may lawfully be regulated by Congress.” 15 U.S.C. § 1127. See also Steele v. Bulova Watch Co., 344 U.S. 280, 283-84 (1952); Larry Harmon Pictures Corp. v. Williams Rest. Corp., 929 F.2d 662, 666 (Fed.Cir.) (allowing registration for an intrastate provider of restaurant services with an undefined interstate clientele), cert. denied, 502 U.S. 823 (1991). The distribution of the Software for end-users over the Internet satisfies the “use in commerce” jurisdictional predicate. See, e.g., Planned Parenthood Fed'n of Am., Inc. v. Bucci, 42 U.S.P.Q.2d 1430, 1434 (S.D.N.Y.1997) ( “The nature of the Internet indicates that establishing a typical home page on the Internet, for access to all users, would satisfy the Lanham Act's ‘in commerce’ requirement.”), aff'd, 152 F.3d 920 (2d Cir.), cert. denied, 525 U.S. 834 (1998).
 Nevertheless, the use of a mark in commerce also must be sufficient to establish ownership rights for a plaintiff to recover against subsequent users under section 43(a). See New England Duplicating Co. v. Mendes, 190 F.2d 415, 417-18 (1st Cir.1951) (after finding “use in commerce” jurisdiction predicate satisfied, court noted that “[t]he question remains whether the plaintiff has established that he was the ‘owner’ of the mark, for under [15 U.S.C. § 1051] only the ‘owner’ of a mark is entitled to have it registered.”). The court in Mendes set forth a two part test to
determine whether a party has established “prior use” of a mark sufficient to establish ownership:
[E]vidence showing, first, adoption,114 and, second, use in a way sufficiently public to identify or distinguish the marked goods in an appropriate segment of the public mind as those of the adopter of the mark, is competent to establish ownership, even without evidence of actual sales.115
Id. at 418. See also New West, 595 F.2d at 1200.116
 Courts generally must inquire into the activities surrounding the prior use of the mark to determine whether such an association or notice is present. See, e.g., Johnny Blastoff, Inc. v. L.A. Rams Football Co., 188 F.3d 427, 433 (7th Cir.1999) (“The determination of whether a party has established protectable rights in a trademark is made on a case by case basis, considering the totality of the circumstances.”), cert. denied, 528 U.S. 1188, (2000). Under the “totality of circumstances” analysis, a party may establish “use in commerce” even in the absence of sales. “[A]lthough evidence of sales is highly persuasive, the question of use adequate to establish appropriation remains one to be decided on the facts of each case ....” New West, 595 F.2d at 1200 (quoting Mendes, 190 F.2d at 418). The court in New West recognized that “mere advertising by itself may not establish priority of use,” but found that promotional mailings coupled with advertiser and distributor solicitations met the Mendes “public identification” ownership requirement. Id. at 1200. Thus, contrary to Appellants' assertions, the existence of sales or lack thereof does not by itself determine whether a user of a mark has established ownership rights therein.117 Compare Marvel Comics Ltd. v. Defiant, 837 F.Supp. 546, 549 (S.D.N.Y.1993) (finding announcement of “Plasmer” title to 13 million comic book readers and promotion at annual trade convention sufficient to establish trademark ownership rights, notwithstanding lack of any sales) with WarnerVision Entm't Inc. v. Empire of Carolina Inc., 915 F.Supp. 639, 645-46 (S.D.N.Y.) (finding toy manufacturer's promotional efforts insufficient to establish priority of use where only a few presentations were made to industry buyers, even though one resulted in a sale to a major toy retailer), aff'd in part, vacated in part, 101 F.3d 259 (2d Cir.1996).118
 Similarly, not every transport of a good is sufficient to establish ownership rights in a mark. To warrant protection, use of a mark “need not have gained wide public recognition,” but “[s]ecret, undisclosed internal shipments are generally inadequate.” Blue Bell, Inc. v. Farah Mfg. Co., 508 F.2d 1260, 1265 (5th Cir.1975).119 In general, uses that are de minimis may not establish trademark ownership rights. See, e.g., Paramount Pictures Corp. v. White, 31 U.S.P.Q.2d 1768, 1772-73, 1994 WL 484936 (Trademark Tr. & App. Bd.1994) (finding no bona fide use in ordinary course of trade where mark was affixed to a game consisting of three pieces of paper and distributed for the purpose of promoting musical group).
 We find that, under these principles, Darrah's activities under the “Coolmail” mark constitute a “use in commerce” sufficiently public to create ownership rights in the mark. First, the distribution was widespread, and there is evidence that members of the targeted public actually associated the mark Coolmail with the Software to which it was affixed. Darrah made the software available not merely to a discrete or select group (such as friends and acquaintances, or at a trade show with limited attendance), but to numerous end-users via the Internet. The Software was posted under a filename bearing the “Coolmail” mark on a site accessible to anyone who had access to the Internet. End-users communicated with Darrah regarding the Software by referencing the “Coolmail” mark in their e-mails. Appellants argue that only technically-skilled UNIX-users made use of the Software, but there is no evidence that they were so few in number to warrant a finding of de minimis use.
 Third,120 the mark served to identify the source of the Software. The “Coolmail” mark appeared in the subject field and in the text of the announcement accompanying each release of the Software, thereby distinguishing the Software from other programs that might perform similar functions available on the Internet or sold in software compilations.121 The announcements also apparently indicated that Darrah was the “Author/Maintainer of Coolmail” and included his e-mail address. The user manual also indicated that the Software was named “Coolmail.”122 The German company S.u.S.E. was able to locate Darrah in order to request permission to use his Software in its product under the mark “Coolmail.” Appellants do not assert that S.u.S.E. was unaware that the Software was called “Coolmail” when it contacted Darrah.
 Fourth, other potential users of the mark had notice that the mark was in use in connection with Darrah's Software. In investigating whether the mark Coolmail existed before submitting its trademark registration application for its e-mail service, Planetary Motion was able to discover that Darrah was using the mark to designate his Software product.
 Fifth, the Software was incorporated into several versions of a product that was in fact sold worldwide and specifically attributed ownership of the Software to Darrah under the “Coolmail” mark. Any individual using the S.u.S.E. product, or competitor of S.u.S.E., that wanted to know the source of the program that performed the e-mail notification function, could do so by referring to the user manual accompanying the product. There is no support for the argument that for a trademark in software to be valid, the mark must appear on the box containing the product incorporating it, that the mark must be displayed on the screen when the program is running, or that the software bearing the mark be a selling point for the product into which it is incorporated. There is no requirement that the public come to associate a mark with a product in any particular way or that the public be passive viewers of a mark for a sufficient public association to arise.
 Sixth, software is commonly distributed without charge under a GNU General Public License. The sufficiency of use should be determined according to the customary practices of a particular industry. See S. Rep. 100-515 at 44 (1988) (“The committee intends that the revised definition of ‘use in commerce’ [see note 13, supra ] be interpreted to mean commercial use which is typical in a particular industry.”) (emphasis added). That the Software had been distributed pursuant to a GNU General Public License does not defeat trademark ownership, nor does this in any way compel a finding that Darrah abandoned his rights in trademark. Appellants misconstrue the function of a GNU General Public License. Software distributed pursuant to such a license is not necessarily ceded to the public domain and the licensor purports to retain ownership rights, which may or may not include rights to a mark.123
 Appellants also rely on DeCosta v. Columbia Broad. Sys., Inc., 520 F.2d 499, 513 (1st Cir.1975), cert. denied, 423 U.S. 1073 (1976), to argue that Darrah is an eleemosynary individual and therefore unworthy of protection under unfair competition laws. The DeCosta court did not hold that the that the absence of a profit-oriented enterprise renders one an eleemosynary individual, nor did it hold that such individuals categorically are denied protection. Rather, the DeCosta court expressed “misgivings” of extending common law unfair competition protection, clearly available to eleemosynary organizations, to eleemosynary individuals.124 Id. The court's reluctance to extend protection to eleemosynary individuals was based on an apparent difficulty in establishing a line of demarcation between those eleemosynary individuals engaged in commerce and those that are not. But as the sufficiency of use to establish trademark ownership is inherently fact-driven, the court need not have based its decision on such a consideration. Mendes, 190 F.2d at 418. Common law unfair competition protection extends to non-profit organizations because they nonetheless engage in competition with other organizations. See Girls Clubs of Am., Inc. v. Boys Clubs of Am., Inc., 683 F.Supp. 50 (S.D.N.Y.1988), aff'd, 859 F.2d 148 (2d Cir.). Thus, an eleemosynary individual that uses a mark in connection with a good or service may nonetheless acquire ownership rights in the mark if there is sufficient evidence of competitive activity.
 One individual can invest time, effort and money in developing software or other technologically-based goods or services that would be of interest to a multitude of users, other developers, and retail establishments. In fact, the program was of sufficient interest for S.u.S.E. to put effort into including it in its own software which was sold for profit, including the effort of obtaining Darrah's permission under the GNU General Public License.
 Here, Darrah's activities bear elements of competition, notwithstanding his lack of an immediate profit-motive. By developing and distributing software under a particular mark, and taking steps to avoid ceding the Software to the public domain, Darrah made efforts to retain ownership rights in his Software and to ensure that his Software would be distinguishable from other developers who may have distributed similar or related Software. Competitive activity need not be fueled solely by a desire for direct monetary gain. Darrah derived value from the distribution because he was able to improve his Software based on suggestions sent by end-users. Just as any other consumers, these end-users discriminate among and share information on available software. It is logical that as the Software improved, more end-users used his Software, thereby increasing Darrah's recognition in his profession and the likelihood that the Software would be improved even further.
 In light of the foregoing, the use of the mark in connection with the Software constitutes significant and substantial public exposure of a mark sufficient to have created an association in the mind of public.
[The court went on to find a likelihood of confusion between Planetary Motion’s and Techsplosion’s marks. The court affirmed the terms of the permanent injunction but found the award of attorney fees to be an abuse of discretion.]
Comments and Questions
1. The “totality of the circumstances” test. Aycock Engineering stands for the proposition that, as a general matter, the use of a mark merely in pre-sales advertising or merely in preparation to sell goods or services does not constitute “use in commerce” under § 45. Yet Planetary Motion points out that the actual sale of goods or services bearing the mark may also be insufficient to constitute use in commerce. So how can courts determine what kind and degree of pre-sales and/or sales activity can satisfy the use in commerce requirement? Most courts have adopted some form of a “totality of the circumstances” test, sometimes heavily influenced by the equities of the case. See La Societe Anonyme des Parfums Le Galion v. Jean Patou, Inc., 495 F.2d 1265, 1274 n. 11 (2d Cir.1974) (“[T]he balance of the equities plays an important role in deciding whether defendant's use is sufficient to warrant trademark protection.”). In Chance v. Pac-Tel Teletrac Inc., 242 F.3d 1151 (9th Cir. 2001), for example, the Ninth Circuit summarized the factors that might be relevant to a totality of the circumstances analysis of use in commerce sufficient to justify rights:
Accordingly, we hold that the totality of the circumstances must be employed to determine whether a service mark has been adequately used in commerce so as to gain the protection of the Lanham Act. In applying this approach, the district courts should be guided in their consideration of non-sales activities by factors we have discussed, such as the genuineness and commercial character of the activity, the determination of whether the mark was sufficiently public to identify or distinguish the marked service in an appropriate segment of the public mind as those of the holder of the mark, the scope of the non-sales activity relative to what would be a commercially reasonable attempt to market the service, the degree of ongoing activity of the holder to conduct the business using the mark, the amount of business transacted, and other similar factors which might distinguish whether a service has actually been “rendered in commerce”.
Id. at 1159. Applying these factors, the Ninth Circuit found that the October 1989 mailing by Allen Chance (“Chance”) of 35,000 postcards promoting his teletrac tracking service that led to 128 telephone responses but no sales was not sufficient to establish use in commerce. Meanwhile,
Pac–Tel, in contrast, had significant activities even prior to [Chance’s] post card mailing. The record demonstrates that as early as June 1989, Pac–Tel began using the mark on a continuous basis. As early as 1984, a Pac–Tel predecessor company was using the mark as part of its business name. Pac–Tel began a public relations campaign using the mark to introduce its new service in July 1989. In September 1989, it sent out brochures to potential customers. In early fall 1989, it conducted interviews with major newspapers including the Wall Street Journal, Washington Post and Chicago Tribune which resulted in a number of stories that mentioned the service mark. During this time the service was marketed to potential customers who managed large vehicle fleets through a slide presentation using the mark. While the district court found that Pac–Tel's first use was in April 1990, when it began making its service available on a commercial basis for the first time on the Los Angeles school buses, the totality of the record demonstrates that its first use of the mark was significantly earlier and clearly predated [Chance]'s first use
Id. at 1160.
Another example of the application of the totality of the circumstances test, along with a strong grounding in the balance of the equities, is Johnny Blastoff, Inc. v. Los Angeles Rams Football Co., 188 F.3d 427 (7th Cir. 1999). When the Los Angeles Rams announced that they were moving to St. Louis, Rodney Rigsby, proprietor of Johnny Blastoff, Inc., had the bright idea somehow to claim ownership of the st. louis rams mark before the football team could. He filed a State of Wisconsin trademark application on February 22, 1995, and two federal intent-to-use registration applications in March 10, 1995. The court found that the football team’s use in commerce preceded these dates. Here is the core of the court’s analysis:
On January 17, 1995, Georgia Frontiere, the owner of the Rams, and St. Louis Mayor Freeman Bosley held a press conference at which they announced the Rams' intention to relocate from Los Angeles to St. Louis. The press conference story received extensive national and local press, including the St. Louis Dispatch's publication, on January 18, 1995, of a sixteen-page pullout section of the newspaper entitled “St. Louis Rams.” Vendors sold unlicensed “St. Louis Rams” merchandise in the St. Louis area in January of 1995, and by February of 1995, more than 72,000 personal seat licenses for the St. Louis Rams' home games had been received. By the time Blastoff registered the “St. Louis Rams” mark in Wisconsin in February of 1995, a significant portion of the public associated the mark with the Rams football club. However, Blastoff asserts that the defendants had not sufficiently used the mark “St. Louis Rams” to be given priority. Blastoff argues that at the January 17, 1995, press conference, none of the defendants used the words “St. Louis Rams,” and thus, this term was rendered an “unarticulated idea for a team name,” which is not protectable. Blastoff also states that newspaper and media coverage is insufficient to establish priority. Finally, Blastoff contends that the football club “operated publicly and exclusively as [the] ‘L.A. Rams' ” as late as February 8, 1995.
For the purpose of establishing public identification of a mark with a product or service, the fact-finder may rely on the use of the mark in “advertising brochures, catalogs, newspaper ads, and articles in newspapers and trade publications,” T.A.B. Systems v. Pactel Teletrac, 77 F.3d 1372, 1375 (Fed.Cir.1996), as well as in media outlets such as television and radio. See In re Owens–Corning Fiberglas Corp., 774 F.2d 1116, 1125 (Fed.Cir.1985). In addition, courts have recognized that “abbreviations and nicknames of trademarks or names used only by the public give rise to protectable rights in the owners of the trade name or mark which the public modified.” Nat'l Cable Television Assoc. v. Am. Cinema Editors, Inc., 937 F.2d 1572, 1577 (Fed.Cir.1991). Such public use of a mark is deemed to be on behalf of the mark's owners. See id. Blastoff has failed to demonstrate any equivalent use of the mark “St. Louis Rams” by February of 1995, when the defendants established, by use and public association, their priority in the mark. Blastoff's insignificant and very limited use of the mark prior to February of 1995, consisting of the development of the “Tower City Rams” design, along with the production of a swatch of material with “St. Louis Rams” embroidery, is insufficient to establish a link between the mark and its products. Furthermore, the owner's use of a trademark is relevant in establishing public identification of a mark with a product or service. Georgia Frontiere, owner of the Rams, in announcing her intention to move the franchise to St. Louis from Los Angeles, implicitly adopted the exact phrase “St. Louis Rams” on the date of her press conference. This Court's decision in Indianapolis Colts, Inc. v. Metropolitan Baltimore Football Club Ltd., 34 F.3d 410, 413 (7th Cir.1994), is strong support for the proposition that the Rams organization and the NFL had a long-established priority over the use of the “Rams” name in connection with the same professional football team, regardless of urban affiliation.
Id. at 435.
2. “Stealing” someone else’s idea for a trademark. Because use, rather than invention, is the basis for trademark rights under the Lanham Act, there is no remedy under the Act for the “theft” of an idea for a trademark. In American Express Co. v. Goetz, 515 F.3d 156 (2d Cir. 2008), cert. denied, 129 S. Ct. 176 (U.S. 2008), the declaratory defendant Stephen Goetz developed the slogan “My Life. My Card.” for a credit card and sought to interest various credit card providers in using it and his consulting services. On July 30, 2004, Goetz mailed a proposal to American Express urging American Express to adopt the mark. American Express never responded. In November, 2004, however, American Express launched a global campaign based on the phrase “My Life. My Card.”
When Goetz threatened suit, American Express filed for a declaration of non-infringement. Documents produced in the litigation showed that the advertising firm Ogilvy Group first proposed the mark to American Express on July 22, 2004, and Goetz eventually conceded that Ogilvy had developed and American Express had adopted the mark without any knowledge of his proposal.
The district court granted summary judgment to American Express and the Second Circuit affirmed. What drove the outcome of the litigation was not the priority of invention issue, however. Instead, it was the simple fact that Goetz never made a qualifying use in commerce of the mark: “[C]onstruing all the facts in Goetz's favor, the only reasonable conclusion that can be drawn is that My Life, My Card was a component of Goetz's business proposal to the credit card companies rather than a mark designating the origin of any goods or services he offered to them.” Id. at 160.
3. Trademark trolls and the use in commerce requirement. The use in commerce prerequisite for trademark rights has the salutary effect of limiting the ability of bad faith agents to exploit the trademark registration system in the way that “non-practicing entities” arguably exploit the patent system. In Central Mfg., Inc. v. Brett, 492 F.3d 876 (9th Cir. 2007), the defendant George Brett (and brothers) manufactured a hybrid wood-metal bat under the trademark stealth. Plaintiff Central Mfg., of which the then-notorious trademark troll Leo Stoller was president and sole shareholder, sued for infringement of its own mark stealth, which it had registered in 1985 for “[s]porting goods, specifically, tennis rackets, golf clubs, tennis balls, basketballs, baseballs, soccer balls, golf balls, cross bows, tennis racket strings and shuttle cocks.” When Brett challenged Stoller to produce any evidence of use in commerce of the mark, Stoller’s documents failed to persuade the district court. For example: “Plaintiffs produced a table of ‘Stealth Brand Baseball Sales’ between 1996 and 2003, but could provide absolutely no information to justify the lump sum ‘sales’ figures listed. There is no way for this Court to know that this alleged sales sheet bears any relation to reality and is not simply something Plaintiffs generated on a home computer for the purposes of this litigation.” Id. at 883 (quoting Central Mfg. Co. v. Brett, 2006 WL 681058 (N.D.Ill. Mar 15, 2006)). The Seventh Circuit affirmed:
Stoller has repeatedly sought ways to get around trademark law's prohibition on the stockpiling of unused marks, and this case is no different. It is unfathomable that a company claiming to have engaged in thousands of dollars of sales of a product for more than a decade would be unable to produce even a single purchase order or invoice as proof. Self-serving deposition testimony is not enough to defeat a motion for summary judgment. By exposing Central's failure to make bona fide use of the “Stealth” mark for baseballs, Brett Brothers met its burden to overcome the presumption afforded by the 1985 registration, and summary judgment in its favor was the appropriate course.
Id. at 883. Brett was also awarded attorney fees.
In December 2010, Stoller was indicted for fraud charges related to statements made in his bankruptcy filings. In November 2014, he was sentenced to 20 months in a federal prison. See http://en.wikipedia.org/wiki/Leo_Stoller.
Use, rather than registration, is the basis of federal trademark rights in the United States (subject to one exception noted below). As explained previously in this Part, the Lanham Act will protect a trademark owner’s exclusive rights in any trademark it is using in commerce regardless of whether the mark is registered provided that the unregistered mark meets the various substantive requirements for registration established by the Act. In other words, if the mark as used in commerce could be registered, it will be protected even if it is not registered. Conversely, the Lanham Act will not protect a trademark registrant’s exclusive rights in its registered mark if it no longer uses its mark in commerce and cannot prove an intent to resume use in the near future. On this basis, it is often said that the U.S. trademark system is a “use-based” system in contrast to the “registration-based” systems more common around the world.125 In the United States, registration merely records the preexistence of externally established rights.126
The U.S. registration system is different in another significant respect. Unlike most foreign registration systems, which review applications only for compliance with formal requirements, the PTO reviews applications to ensure that they meet both formal requirements (which are largely set forth in Lanham Act § 1) and substantive requirements (largely found in Lanham Act § 2). These substantive requirements include both “absolute grounds” for refusal of registration, such as that the mark is deceptive or scandalous, and “relative grounds” for refusal, such as that the mark is confusingly similar with a previously registered mark.
A trademark applicant at the PTO must claim at least one “filing basis” for its application among the five that are provided by the Lanham Act. These filing bases are:
Note that the first four filing bases are not mutually exclusive; the § 66(a) filing basis, by contrast, may not be combined with other filing bases. Note also that while the §§ 1(a), 44(d), and 44(e) filing bases have been available since the effective date of the original Lanham Act on July 5, 1947, the § 1(b), or “ITU,” filing basis became available with the effective date of the Trademark Law Revision Act (TLRA) on November 16, 1989,127 and the § 66(a) filing basis became available with the effective date of the Madrid Protocol Implementation Act on November 2, 2003.128 Most trademark applications at the PTO are now filed under the Lanham Act § 1(b) intent to use basis.
Lanham Act §§ 44 & 66(a), 15 U.S.C. §§ 1126 & 1141f, constitute important though relatively obscure exceptions to the general rule that a trademark must be used in commerce for it to be federally registered. As noted in In re Cyber-Blitz Trading Services, 47 U.S.P.Q.2d 1638 (Comm'r Pats. 1998),
[o]ne significant difference between Section 1(b) and 44 of the Trademark Act is that Applicants who rely on Section 1(b) as a filing basis must establish use of the mark prior to registration, or the application will become abandoned. In contrast, Applicants who rely solely on Section 44 are not required to demonstrate use in order to obtain registration. Crocker National Bank v. Canadian Imperial Bank of Commerce, 223 USPQ 909 (TTAB 1984). In fact, the first time evidence of use usually is required for Section 44 Applicants is upon the filing of an Affidavit of Continued Use under Section 8 of the Trademark Act, 15 U.S.C. §1058. This does not occur until five to six years after registration.
Id. at 1639-40. See also TMEP § 1009. The reasoning of Cyber-Blitz also applies to § 66(a) applications.129
For a sense of scale, the figure below shows the number of trademark applications at the PTO per year for each filing basis from 1981 through 2010. What might explain the significant rise in applications in the period 1999-2000?
For marks already being used in commerce, a successful application proceeds though at least five basic stages: (1) application, (2) examination, (3) publication in the PTO’s Official Gazette, (4) opposition, and (5) registration. Intent-to-use applications proceed through two additional stages following opposition and preceding registration: (4.a) the issuance by the PTO of a “Notice of Allowance” and (4.b) the filing by the applicant of a Statement of Use showing that the applicant has begun to make actual use of the mark in commerce. We review each of these stages below. But first we consider why a trademark owner should federally register its mark.
Comments and Questions
1. The Phenomenon of “Submarine Trademarks.” Below is the registration certificate for the iPhone mark. Note the priority date claimed: “Priority claimed under Sec. 44(d) on Trinidad/Tobago Application No. 37090, filed 3-27-2006.” Rather than file an application for the mark at the USPTO on March 27, 2006, Apple instead filed on that date in Trinidad & Tobago through a shell company. As the registration certificate indicates, on September 26, 2006 (six months after March 27, 2006 minus a day), Apple then took advantage of Lanham Act § 44(d) to assert the priority date of their Trinidad & Tobago application in the U.S. Why would Apple, like many other consumer-oriented high-technology and fashion companies, engage in such a circuitous route to registration? See Carsten Fink, Andrea Fosfuri, Christian Helmers, & Amanda Myers, “Submarine Trademarks” (working paper).
We consider in this section the geographical extent of rights in registered and unregistered marks. Because the case law excerpted below on the geographical extent of rights in unregistered marks assumes knowledge of the regime relating to registered marks, we begin first with registered marks.
See Jerome Gilson & Anne Gilson LaLonde, Cinnamon Buns, Marching Ducks, and Cherry-Scented Racecar Exhaust: Protecting Nontraditional Trademarks, 95 TRADEMARK REP. 773 (2005).
See Christina S. Monteiro, A Nontraditional Per-Spectrum: The Touch of Trademarks, INTA BULL., June 15, 2010, at 4.
See McCarthy § 3.9 (“[T]he “source” identified by a trademark need not be known by name to the buyer. It may be anonymous in the sense that the buyer does not know, or care about, the name of the corporation that made the product or the name of the corporation which distributes it. But the buyer is entitled to assume that all products carrying the same trademark are somehow linked with or sponsored by that single, anonymous source.”).
See, e. g., W. E. Bassett Co. v. Revlon, Inc., 435 F.2d 656 (2 Cir. 1970). A Commentator has illuminated the distinction with an example of the “Deep Bowl Spoon”:
“Deep Bowl” identifies a significant characteristic of the article. It is “merely descriptive” of the goods, because it informs one that they are deep in the bowl portion . . . . It is not, however, “the common descriptive name” of the article (since) the implement is not a deep bowl, it is a spoon . . . . “Spoon” is not merely descriptive of the article it identifies the article (and therefore) the term is generic.
Fletcher, Actual Confusion as to Incontestability of Descriptive Marks, 64 Trademark Rep. 252, 260 (1974). On the other hand, “Deep Bowl” would be generic as to a deep bowl.
We note that, in contrast with its position in this case, in other litigation NVE has asserted that its own mark, “6 Hour POWER,” is an “inherently distinctive” mark. See Complaint at ¶ 12, N.V.E., Inc. v. N2G Distrib., Inc. & Alpha Performance Labs, No. 2:08–cv–01824 (D.N.J. Apr. 14, 2008) (“The 6 HOUR POWER mark distinguishes NVE as the source of these products, is inherently distinctive, and has also become distinctive through the acquisition of secondary meaning.” (emphasis added)).
Qualitex, 514 U.S. at 162.
155 F.3d 526, 536 (5th Cir.1998) (internal quotation marks omitted), abrogation on other grounds recognized by Eppendorf–Netheler–Hinz GMBH v. Ritter GMBH, 289 F.3d 351, 356 (5th Cir.2002).
Pebble Beach, 155 F.3d at 541.
See Pebble Beach, 155 F.3d at 541–52 (prominent display of golf hole's trade dress in advertising supported finding of secondary meaning as a designator of source).
Bd. of Supervisors, 438 F.Supp.2d at 658.
See also Thomas & Betts Corp. v. Panduit Corp., 65 F.3d 654, 663 (7th Cir.1995). We also note that the record does contain survey evidence compiled by the Universities indicating that approximately thirty percent of consumers interviewed believed two of Smack's t-shirts were produced or sponsored by the Universities. We have indicated that survey evidence often may be the most direct and persuasive evidence of secondary meaning. Sugar Busters LLC v. Brennan, 177 F.3d 258, 269 (5th Cir.1999). Nevertheless, Smack moved in limine to exclude the Universities' survey evidence, and the district court found it unnecessary to rule on the motion because of the other evidence in the record. Because no party has raised the issue, we express no opinion on the correctness of the district court's belief and merely note the presence of the survey evidence in the record.
Cf. Taco Cabana Int'l, Inc. v. Two Pesos, Inc., 932 F.2d 1113, 1121 (5th Cir.1991) (“An owner may license its trademark or trade dress and retain proprietary rights if the owner maintains adequate control over the quality of goods and services that the licensee sells with the mark or dress.”).
Is “App Store” a generic term for an online platform selling apps? See Apple, Inc. v. Amazon.com Inc., No. 11 Civ. 1327, 2011 WL 2638191, at *7 (N.D. Cal. July 6, 2011) (“The court assumes without deciding that the ‘App Store’ mark is protectable as a descriptive mark that has arguably acquired secondary meaning.” But the court found, on Apple’s preliminary injunction motion, that Amazon’s use of “App Store” to describe its app store did not create a likelihood of confusion.).
A “Teflon” survey refers to the format of the survey used in E. I. du Pont de Nemours & Co. v. Yoshida International, Inc., 393 F. Supp. 502, 185 USPQ 597 (E.D.N.Y. 1975) to demonstrate that “Teflon” was not generic.
The District Court instructed the jury: “ ‘[T]rade dress' is the total image of the business. Taco Cabana's trade dress may include the shape and general appearance of the exterior of the restaurant, the identifying sign, the interior kitchen floor plan, the decor, the menu, the equipment used to serve food, the servers' uniforms and other features reflecting on the total image of the restaurant.” 1 App. 83–84. The Court of Appeals accepted this definition and quoted from Blue Bell Bio–Medical v. Cin–Bad, Inc., 864 F.2d 1253, 1256 (CA5 1989): “The ‘trade dress' of a product is essentially its total image and overall appearance.” See 932 F.2d 1113, 1118 (CA5 1991). It “involves the total image of a product and may include features such as size, shape, color or color combinations, texture, graphics, or even particular sales techniques.” John H. Harland Co. v. Clarke Checks, Inc., 711 F.2d 966, 980 (CA11 1983). Restatement (Third) of Unfair Competition § 16, Comment a (Tent.Draft No. 2, Mar. 23, 1990).
Secondary meaning is used generally to indicate that a mark or dress “has come through use to be uniquely associated with a specific source.” Restatement (Third) of Unfair Competition § 13, Comment e (Tent.Draft No. 2, Mar. 23, 1990). “To establish secondary meaning, a manufacturer must show that, in the minds of the public, the primary significance of a product feature or term is to identify the source of the product rather than the product itself.” Inwood Laboratories, Inc. v. Ives Laboratories, Inc., 456 U.S. 844, 851, n. 11, 102 S.Ct. 2182, 2187, n. 11, 72 L.Ed.2d 606 (1982).
Section 43(a) replaced and extended the coverage of § 3 of the Trademark Act of 1920, 41 Stat. 534, as amended. Section 3 was destined for oblivion largely because it referred only to false designation of origin, was limited to articles of merchandise, thus excluding services, and required a showing that the use of the false designation of origin occurred “willfully and with intent to deceive.” Ibid. As a result, “[a]lmost no reported decision can be found in which relief was granted to either a United States or foreign party based on this newly created remedy.” Derenberg, Federal Unfair Competition Law at the End of the First Decade of the Lanham Act: Prologue or Epilogue?, 32 N.Y.U.L.Rev. 1029, 1034 (1957).
This is clear from the fact that the cause of action created by this section is available only to a person doing business in the locality falsely indicated as that of origin. See n. 1, supra.
The common-law tort of passing off has been described as follows:
“Beginning in about 1803, English and American common law slowly developed an offshoot of the tort of fraud and deceit and called it ‘passing off’ or ‘palming off.’ Simply stated, passing off as a tort consists of one passing off his goods as the goods of another. In 1842 Lord Langdale wrote:
“ ‘I think that the principle on which both the courts of law and equity proceed is very well understood. A man is not to sell his own goods under the pretence that they are the goods of another man....’
“In 19th century cases, trademark infringement embodied much of the elements of fraud and deceit from which trademark protection developed. That is, the element of fraudulent intent was emphasized over the objective facts of consumer confusion.” 1 J. McCarthy, Trademarks and Unfair Competition § 5.2, p. 133 (2d ed. 1984) (McCarthy) (footnotes omitted).
2 id., § 27:3, p. 345.
See, e.g., Germain, Unfair Trade Practices Under § 43(a) of the Lanham Act: You've Come a Long Way Baby—Too Far, Maybe?, 64 Trademark Rep. 193, 194 (1974) (“It is submitted that the cases have applied Section 43(a) to situations it was not intended to cover and have used it in ways that it was not designed to function”).
The United States Trademark Association Trademark Review Commission Report and Recommendations to USTA President and Board of Directors, 77 Trademark Rep. 375, 426 (1987). [In the body of his opinion, Justice Stevens appears to have misnamed the United States Trademark Association, which was the predecessor organization of the International Trademark Association].
Consistent with the common-law background of § 43(a), the Second Circuit has said that proof of secondary meaning is required to establish a claim that the defendant has traded on the plaintiff's good will by falsely representing that his goods are those of the plaintiff. See, e.g., Crescent Tool Co. v. Kilborn & Bishop Co., 247 F. 299 (1917). To my knowledge, however, the Second Circuit has not explained why “inherent distinctiveness” is not an appropriate substitute for proof of secondary meaning in a trade dress case. Most of the cases in which the Second Circuit has said that secondary meaning is required did not involve findings of inherent distinctiveness. For example, in Vibrant Sales, Inc. v. New Body Boutique, Inc., 652 F.2d 299 (1981), cert. denied, 455 U.S. 909, 102 S.Ct. 1257, 71 L.Ed.2d 448 (1982), the product at issue—a velcro belt—was functional and lacked “any distinctive, unique or non-functional mark or feature.” 652 F.2d, at 305. Similarly, in Stormy Clime Ltd. v. ProGroup, Inc., 809 F.2d 971, 977 (1987), the court described functionality as a continuum, and placed the contested rainjacket closer to the functional end than to the distinctive end. Although the court described the lightweight bag in LeSportsac, Inc. v. K mart Corp., 754 F.2d 71 (1985), as having a distinctive appearance and concluded that the District Court's finding of nonfunctionality was not clearly erroneous, id., at 74, it did not explain why secondary meaning was also required in such a case.
The Senate Report elaborated on these two goals:
“The purpose underlying any trade-mark statute is twofold. One is to protect the public so it may be confident that, in purchasing a product bearing a particular trade-mark which it favorably knows, it will get the product which it asks for and wants to get. Secondly, where the owner of a trade-mark has spent energy, time, and money in presenting to the public the product, he is protected in his investment from its misappropriation by pirates and cheats. This is the well-established rule of law protecting both the public and the trade-mark owner.” S.Rep. No. 1333, 79th Cong., 2d Sess., 3 (1946).
By protecting trademarks, Congress hoped “to protect the public from deceit, to foster fair competition, and to secure to the business community the advantages of reputation and good will by preventing their diversion from those who have created them to those who have not. This is the end to which this bill is directed.” Id., at 4.
Forty years later, the USTA Trademark Review Commission assessed the state of trademark law. The conclusion that it reached serves as a testimonial to the success of the Act in achieving its goal of uniformity: “The federal courts now decide, under federal law, all but a few trademark disputes. State trademark law and state courts are less influential than ever. Today the Lanham Act is the paramount source of trademark law in the United States, as interpreted almost exclusively by the federal courts.” Trademark Review Commission, 77 Trademark Rep., at 377.
“When several acts of Congress are passed touching the same subject-matter, subsequent legislation may be considered to assist in the interpretation of prior legislation upon the same subject.” Tiger v. Western Investment Co., 221 U.S. 286, 309 (1911); see NLRB v. Bell Aerospace Co. Division of Textron, Inc., 416 U.S. 267, 275 (1974); Red Lion Broadcasting Co. v. FCC, 395 U.S. 367, 380–381 (1969); United States v. Stafoff, 260 U.S. 477, 480 (1923) (opinion of Holmes, J.).
As the Senate Report explained, revision of § 43(a) is designed “to codify the interpretation it has been given by the courts. Because Section 43(a) of the Act fills an important gap in federal unfair competition law, the committee expects the courts to continue to interpret the section. “As written, Section 43(a) appears to deal only with false descriptions or representations and false designations of geographic origin. Since its enactment in 1946, however, it has been widely interpreted as creating, in essence, a federal law of unfair competition. For example, it has been applied to cases involving the infringement of unregistered marks, violations of trade dress and certain nonfunctional configurations of goods and actionable false advertising claims.” S.Rep. No. 100–515, p. 40 (1988) U.S.Code Cong. & Admin.News 1988, pp. 5577, 5605.
“Trademark protection is important to both consumers and producers. Trademark law protects the public by making consumers confident that they can identify brands they prefer and can purchase those brands without being confused or misled. Trademark laws also protec[t] trademark owners. When the owner of a trademark has spent conside[r]able time and money bringing a product to the marketplace, trademark law protects the producer from pirates and counterfeiters.” Id., at 4.
One commentator has noted that marks consisting of symbols and designs are typically arbitrary with respect to their associated goods and services where the marks are “nonrepresentational”:
Nonverbal and nonrepresentational designs and figures are perfectly acceptable as trademarks. Indeed, they have the advantage of being totally arbitrary, and so cannot be descriptive of the goods or services. The only problem which may be encountered is the question of whether such designs or figures are regarded by the public as identifying indicia or merely as decorations. Especially is this true of such simple figures as rectangles, diamonds, circles, triangles, or lines.
LOUIS ALTMAN & MALLA POLLACK, 3 CALLMAN ON UNFAIR COMPETITION, TRADEMARKS AND MONOPOLIES § 18:24 (4th ed.2010) (footnotes omitted). Under this reasoning, nonverbal marks—even though “arbitrary”—must still be shown to serve as identifying indicia. Professor McCarthy appears to share the view that such marks are arbitrary when they are nonrepresentational. See 1 MCCARTHY ON TRADEMARKS § 7:36, at 7–91 (“A picture that is merely a representation of the goods themselves is regarded as merely descriptive of the goods.”).
As noted above, the district court omitted discussion of the fourth factor, which by its terms applies only when a party seeks trademark protection for a background design typically accompanied by words. See Amazing Spaces, 665 F.Supp.2d at 736. Similarly, we will not consider the fourth Seabrook Foods factor.
The Second Circuit has also used a related test that asks “whether [the mark] is ‘likely to serve primarily as a designator of origin of the product.’ ” Knitwaves, Inc. v. Lollytogs Ltd., 71 F.3d 996, 1008 (2d Cir.1995) (quoting Duraco Prods., Inc. v. Joy Plastic Enters., 40 F.3d 1431, 1449 (3d Cir.1994)). Although the Knitwaves court quoted the Duraco Products court for part of its test, it rejected the requirements that the product configuration “be ‘unusual and memorable’ and ‘conceptually separable from the product.’ ” Id. at 1009 n. 6 (quoting Duraco Prods., 40 F.3d at 1449); see also Stuart Hall Co. v. Ampad Corp., 51 F.3d 780, 787–88 (8th Cir.1995) (rejecting the Duraco Products test as inconsistent with Two Pesos by imposing additional, non-statutory requirements). The First Circuit has, in turn, criticized the Knitwaves test as inconsistent with Two Pesos because it fails to keep the inherent distinctiveness inquiry separate from the secondary meaning inquiry. See I.P. Lund Trading, 163 F.3d at 41. This case does not require us to delve into the propriety of these variations on the Seabrook Foods test, and we do not do so.
We note, of course, that the Wal–Mart Court was urged by the respondent in that case and by the United States as amicus curiae to adopt the Seabrook Foods test writ large for product design but declined to do so. Id. at 213–14, 120 S.Ct. 1339. The Court's concern was that “[s]uch a test would rarely provide the basis for summary disposition of an anticompetitive strike suit.” Id. at 214, 120 S.Ct. 1339. However, as discussed below, we are of the opinion that the relevant portions of the Seabrook Foods test do provide a basis for summary disposition in this case. Because we conclude that the Star Symbol is not inherently distinctive under the Seabrook Foods test, we do not address whether it constitutes a “reasonably clear test,” id. at 213, such that it would be preferable to the Abercrombie test in the ordinary trademark or service mark dispute.
The interrelationship between these inquiries is also reflected in Professor McCarthy's discussion of common geometric shapes:
Most common geometric shapes are regarded as not being inherently distinctive, in view of the common use of such shapes in all areas of advertising. Thus, such ordinary shapes as circles, ovals, squares, etc., either when used alone or as a background for a word mark, cannot function as a separate mark unless (1) the shape is likely to create a commercial impression on the buyer separate from the word mark or any other indicia and (2) the shape is proven to have secondary meaning.... The rationale is that such designs have been so widely and commonly used as mere decorative graphic elements that the origin-indicating ability of such designs has been diminished.
1 MCCARTHY ON TRADEMARKS § 7:29, at 7–73–74 (footnotes omitted).
The parties dispute the scope of the “relevant market”—specifically, whether the district court correctly considered use of a similar or identical star design beyond the self-storage service industry. Amazing Spaces contends that we should limit our analysis to the self-storage services industry, while Metro argues that we may take into account uses of star designs in a larger context. The second Seabrook Foods factor refers to uniqueness or unusualness “in a particular field,” 568 F.2d at 1344, and the Second Circuit has stated that a stylized design may be protectable when it “is not commonplace but rather unique or unusual in the relevant market,” Star Indus., 412 F.3d at 382. Similarly, the third factor refers to whether a mark is commonly used as ornamentation for a “particular class of goods.” Seabrook Foods, 568 F.2d at 1344. In contrast, the First Circuit, in considering whether a red heart on the chest of a teddy bear was inherently distinctive, appeared to consider the broader use of red hearts in determining whether the use at issue was unique or unusual. See Wiley, 762 F.2d at 142 (“[T]he record contains so many examples of use of a red heart motif on teddy bears and other stuffed animals, not to mention all manner of other toys and paraphernalia, that no reasonable argument on this point can be made.” (emphasis added)). The rule in the RESTATEMENT asks whether, “because of the nature of the designation and the context in which it is used, prospective purchasers are likely to perceive it as a designation that ... identifies goods or services produced or sponsored by a particular person.” RESTATEMENT § 13(a), at 104 (emphasis added). It further explains that
[c]ommonplace symbols and designs are not inherently distinctive since their appearance on numerous products makes it unlikely that consumers will view them as distinctive of the goods or services of a particular seller. Thus, unless the symbol or design is striking, unusual, or otherwise likely to differentiate the products of a particular producer, the designation is not inherently distinctive.
Id. § 13 cmt. d, at 107. Finally, and most importantly, the Lanham Act defines “service mark” as a mark used “to identify and distinguish the services of one person ... from the services of others and to indicate the source of the services.” Lanham Act § 45, 15 U.S.C. § 1127. Because a mark must distinguish one person's services from another, we agree that our inquiry is whether the Star Symbol identifies and distinguishes Amazing Spaces's self-storage services from others' self-storage services. This does not mean, however, that we must blind ourselves to uses beyond the self-storage services industry: the fact that the same or a similar star is used in countless other ways certainly bears on whether it is “likely that prospective purchasers will perceive [a given star design] as an indication of source” within a particular industry because a “[c]ommonplace symbol[']s ... appearance on numerous products makes it unlikely that consumers will view [it] as distinctive of the goods or services of a particular seller.” RESTATEMENT § 13 cmt. d, at 107.
Under this analysis, use by third parties of a design bears on whether the design is inherently distinctive, not merely whether the design “is a ‘strong’ or a ‘weak’ [ ]mark.” Exxon Corp. v. Tex. Motor Exchange of Houston, Inc., 628 F.2d 500, 504 (5th Cir.1980); cf. Union Nat'l Bank of Tex., Laredo, Tex., 909 F.2d at 848 n. 25 (noting that widespread industry use can render a mark not inherently distinctive, but third-party use otherwise typically affects the issue of whether there is a likelihood of confusion between marks).
This is what differentiates the Star Symbol from the examples of registered marks containing stars that Amazing Spaces cites to support the protectability of five-pointed stars. The Dallas Cowboys star is stylized through the inclusion of a white border. The star in the Wal–Mart registration is a plain, five-pointed star, but the registered mark consists of more than just the star—the mark is the words “Wal” and “Mart” on either side of the star. The LanChile Airlines star is set against a circle that is 50% filled in, and it is adjacent to the words “LanChile Airlines.” Finally, the USA Truck mark is a complex design consisting of a white star within a blue circle, set against a white rectangle with blue borders and a red stripe running across the middle. Each of these marks contains elements distinguishing it from the commonplace stars in the record. See Union Nat'l Bank of Tex., Laredo, Tex., 909 F.2d at 848 n. 25 (noting that the appropriate inquiry is whether the mark as a whole is protectable, not whether its component parts are individually protectable (citing Estate of P.D. Beckwith v. Comm'r of Patents, 252 U.S. 538, 40 S.Ct. 414, 64 L.Ed. 705 (1919))).
It is well known that the law of “functionality” has been applied in both a “utilitarian” sense and in terms of “aesthetics.” See e.g., Vuitton et Fils S.A. v. J. Young Enterprises, Inc., 644 F.2d 769, 210 USPQ 351 (9th Cir. 1981); International Order of Job's Daughters v. Lindeburg and Co., 633 F.2d 912, 208 USPQ 718 (9th Cir. 1980); Famolare, Inc. v. Melville Corp., 472 F.Supp. 738, 203 USPQ 68 (D.Hawaii 1979). Recognition of this provides an explanation for the statement that, “the term ‘functional’ is not to be treated as synonymous with the literal significance of the term ‘utilitarian’.” J.C. Penney Co. v. H.D. Lee Mercantile Co., 120 F.2d 949, 954, 50 USPQ 165, 170 (8th Cir. 1941). It will be so treated, however, where the issue is one of “utilitarian functionality” and not “aesthetic functionality.” The PTO does not argue in this case that appellant's container configuration is aesthetically functional, notwithstanding appellant's argument that its design was adopted, in part, for aesthetic reasons.
TrafFix suggests that there may be a requirement under Qualitex to inquire into a “significant non-reputation-related disadvantage” in aesthetic functionality cases, because aesthetic functionality was “the question involved in Qualitex.” 121 S.Ct. at 1262. This statement has been criticized because “aesthetic functionality was not the central question in the Qualitex case.” J. Thomas McCarthy, 1 McCarthy on Trademarks and Unfair Competition § 7:80, 7–198 (4th ed.2001). We need not decide what role, if any, the determination of a “significant non-reputation-related disadvantage” plays in aesthetic functionality cases, because aesthetic functionality is not at issue here.
For example, a feature may be found functional where the feature “affects the cost or quality of the device.” TrafFix, 121 S.Ct. at 1263.
[The Jay Franco opinion is excerpted below in our discussion of aesthetic functionality.]
The Draft Restatement's Illustrations expressly reject Pagliero. Illustration 6 reads as follows:
A manufactures china. Among the products marketed by A is a set of china bearing a particular “overall” pattern covering the entire upper surface of each dish. Evidence indicates that aesthetic factors play an important role in the purchase of china, that A's design is attractive to a significant number of consumers, and that the number of alternative patterns is virtually unlimited. In the absence of evidence indicating that similarly attractive “overall” patterns are unavailable to competing manufacturers, A's pattern design is not functional under the rule stated in this Section.
Draft Restatement Illustrations 7 and 8 reflect this aspect of the rule. They read as follows:
7. The facts being otherwise as stated in Illustration 6, A's design consists solely of a thin gold band placed around the rim of each dish. Evidence indicates that a significant number of consumers prefer china decorated with only a gold rim band. Because the number of alternative designs available to satisfy the aesthetic desires of these prospective purchasers is extremely limited, the rim design is functional under the rule stated in this Section.
8. A is the first seller to market candy intended for Valentine's Day in heart-shaped boxes. Evidence indicates that the shape of the box is an important factor in the appeal of the product to a significant number of consumers. Because there are no alternative designs capable of satisfying the aesthetic desires of these prospective purchasers, the design of the box is functional under the rule stated in this Section.
See Wallace Int'l Silversmiths, Inc. v. Godinger Silver Art Co., 916 F.2d 76, 80 (2d Cir.1990) (noting that the term “functionality” as commonly understood seems to imply “only utilitarian considerations”).
In LeSportsac, K Mart challenged the trade dress of a backpack composed of “parachute nylon and trimmed in cotton carpet tape with matching cotton-webbing straps. The zippers used to open and close the bags [we]re color coordinated with the bags themselves, and usually [we]re pulled with hollow rectangular metal sliders.” LeSportsac, 754 F.2d at 74.
In Warner Brothers, we cited as examples Kellogg Co. v. National Biscuit Co., 305 U.S. 111, 122, 59 S.Ct. 109, 83 L.Ed. 73 (1938), in which the pillow shape of a shredded wheat biscuit was deemed functional because the cost of the cereal would be increased and its quality lessened by any other form, and Fisher Stoves Inc. v. All Nighter Stove Works, Inc., 626 F.2d 193, 195 (1st Cir.1980), in which a two-tier woodstove design was deemed functional because it improved the operation of the stove. See Warner Bros., Inc., 724 F.2d at 331.
See, e.g., Industria Arredamenti Fratelli Saporiti v. Charles Craig, Ltd., 725 F.2d 18, 19 (2d Cir.1984) (interlocking design of couch cushions was a visual “label” but served a utilitarian purpose by keeping cushions in place and was therefore functional).
In 1938, the Restatement of Torts stated that “[a] feature of goods is functional ... if it affects their purpose, action or performance, or the facility or economy of processing, handling or using them; it is non-functional if it does not have any of such effects.” Restatement of Torts § 742 (1938). In the official comment to that Section, the Restatement explained several ways in which goods or their features might be functional. With regard to “goods [that] are bought largely for their aesthetic value,” the Restatement suggested that “their features may be functional because they definitely contribute to that value and thus aid the performance of an object for which the goods are intended.” Id. § 742, cmt. a. This was the first time that a commentator had proposed that an aesthetic product feature might be functional. See 1 McCarthy on Trademarks § 7:79 (4th ed.).
The doctrine of aesthetic functionality remains controversial in our sister circuits, which have applied the doctrine in varying ways (and some not at all). For example, the Seventh Circuit has applied the doctrine of aesthetic functionality liberally, holding that “[f]ashion is a form of function.” See Jay Franco & Sons, Inc. v. Franek, 615 F.3d 855, 860 (7th Cir. 2010). The Sixth Circuit recently discussed the doctrine, but made clear that it has not yet decided whether or not to adopt it. See Maker's Mark Distillery, Inc. v. Diageo N. Am., Inc., 679 F.3d 410, 417–19 (6th Cir. 2012). The Ninth Circuit has applied the doctrine inconsistently. See 1 McCarthy on Trademarks § 7:80 (4th ed.) (collecting cases). The Fifth Circuit rejects the doctrine of aesthetic functionality entirely. Bd. of Supervisors for La. State Univ. Agric. & Mech. Coll. v. Smack Apparel Co., 550 F.3d 465, 487–88 (5th Cir. 2008) (arguing that the Supreme Court has recognized the aesthetic functionality doctrine only in dicta, and that therefore the Fifth Circuit's long-standing rejection of the doctrine was not abrogated by Qualitex and TrafFix).
See Wallace Int'l Silversmiths, 916 F.2d at 80 (“We rejected Pagliero [ 's ‘important ingredient’ formulation] in [Le]Sportsac and reiterate that rejection here.” (internal citation omitted)); Mark P. McKenna, (Dys)functionality, 48 Hous. L.Rev. 823, 851 (2011) ( “Courts that apply the aesthetic functionality doctrine today overwhelmingly rely on the test the Supreme Court endorsed in TrafFix [rather than the Pagliero test], ... asking whether exclusive use of the claimed feature put competitors at a significant non-reputation-related disadvantage.”).
The intellectual property protection of fashion design has been for years a subject of controversy among commentators. Some have proposed working within the confines of the current intellectual property system, while others have advocated that fashion design may be an appropriate area for sui generis statutory protection. See generally C. Scott Hemphill & Jeannie Suk, The Law, Culture, and Economics of Fashion, 61 Stan. L.Rev. 1147 (2009); see also id. at 1184–90. (Indeed, suggested legislation creating such protection has been considered several times by Congress, although not adopted. See, e.g., Design Piracy Prohibition Act, H.R.2033, 110th Cong. § 2(c) (2007); Design Piracy Prohibition Act, S.1957, 110th Cong. § 2(c) (2007).) Still other commentators have suggested that intellectual property protection of fashion design would be damaging to the industry and should be avoided. See Kal Raustiala & Christopher Sprigman, The Piracy Paradox: Innovation and Intellectual Property in Fashion Design, 92 Va. L.Rev. 1687, 1775–77 (2006). It is arguable that, in the particular circumstances of this case, the more appropriate vehicle for the protection of the Red Sole Mark would have been copyright rather than trademark. See generally Kieselstein–Cord v. Accessories by Pearl, Inc., 632 F.2d 989, 993–94 (2d Cir.1980) (addressing the broad issue of aesthetically functional copyrights and holding that decorative belt buckles that were used principally for ornamentation could be copyrighted because the primary ornamental aspect of the buckles was conceptually separate from their subsidiary utilitarian function); Laura A. Heymann, The Trademark/Copyright Divide, 60 SMU L.Rev. 55 (2007). However, because Louboutin has chosen to rely on the law of trademarks to protect his intellectual property, we necessarily limit our review to that body of law and do not further address the broad and complex issue of fashion design protection.
The trademark system, in this way, stands in sharp contrast to the copyright system. Copyright, unlike trademark, rewards creativity and originality even if they interfere with the rights of an existing copyright holder. In the copyright system there is a defense to infringement known as “independent creation”: if a writer or musician, through the creative process, independently arrives at an arrangement of words or notes that is the subject of a copyright, he may market the result of his creativity despite the existing copyright. See Feist Publ'ns, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 346, 111 S.Ct. 1282, 113 L.Ed.2d 358 (1991) (requesting that the reader “assume that two poets, each ignorant of the other, compose identical poems. Neither work is novel, yet both are original and, hence, copyrightable”); Procter & Gamble Co. v. Colgate–Palmolive Co., 199 F.3d 74, 77–78 (2d Cir.1999). The trademark system, unlike the copyright system, aims to prevent consumer confusion even at the expense of a manufacturer's creativity: in trademark, if a branding specialist produces a mark that is identical to one already trademarked by another individual or corporation, he must “go back to the drawing board.” See Blendco, Inc. v. Conagra Foods, Inc., 132 Fed.Appx. 520, 523 (5th Cir.2005) (although defendant's allegedly independent creation of infringing mark tended to show that infringement was not willful, defendant remained liable for damages); Tuccillo v. Geisha NYC, LLC, 635 F.Supp.2d 227 (E.D.N.Y.2009) (same).
15 U.S.C. § 1119 provides that “[i]n any action involving a registered mark the court may determine the right to registration, order the cancellation of registrations, in whole or in part, restore canceled registrations, and otherwise rectify the register with respect to the registrations of any party to the action. Decrees and orders shall be certified by the court to the Director, who shall make appropriate entry upon the records of the Patent and Trademark Office, and shall be controlled thereby.” (emphasis added).
No one likes this statutory phrase “primarily geographically deceptively misdescriptive” from Section 2(e)(3), but we appear to be stuck with it. In her opinion in In re Miracle Tuesday, LLC, 695 F.3d 1339 (Fed. Cir. 2012), Judge O’Malley took pains to distance her own elegant prose from the statutory language: “The phrase ‘primarily geographically deceptively misdescriptive’ is a statutory term of art in the trademark context; we neither take responsibility for nor endorse the split infinitives or absence of necessary commas its use in this opinion requires.” Id. at 1342 n. 2. Where possible, this casebook will drop “primarily” and simply speak of “geographically deceptively misdescriptive” marks.
The Gilson treatise explains why the difference between the two categories might matter:
The test for determining whether a mark is deceptive under Section 2(a) is now the same as that for determining whether a mark is primarily geographically deceptively misdescriptive under Section 2(e)(3). The difference comes with respect to registrability: Geographically deceptive marks cannot be registered on either the Principal or Supplemental Register, while primarily geographically deceptively misdescriptive marks may be registered on the Principal Register if the marks became distinctive of the goods or services before December 8, 1993, and they may be registered on the Supplemental Register if they have been in use in commerce since before December 8, 1993
“Royal” is defined, inter alia, as “of or pertaining to a king, queen, other sovereign” and informally as “a royal person; member of the royalty,” or “a member of England’s royal family.” The Random House Dictionary of the English Language (Unabridged), p. 1677 (2nd ed. 1977). The Board may take judicial notice of dictionary definitions., Univ. of Notre Dame du Lac v. J.C. Gourmet Food Imp. Co., 213 USPQ 594 (TTAB 1982), aff’d, 703 F.2d 1372, 217 USPQ 505 (Fed. Cir. 1983). See also Dictionary.com attached to the October 27, 2011 Office Action.
There is an important exception to the general rule that a trademark owner must make use in commerce of its mark in order for the mark to qualify for federal registration. As discussed more fully in Part II.D below, Lanham Act § 44, 15 U.S.C. § 1126, provides that foreign applicants applying under a § 44 filing basis need not show actual use in commerce prior to obtaining registration. See MCCARTHY § 29:14.
[The Supplemental Register is reserved for marks that are capable of, but have not yet developed, source distinctiveness. See Lanham Act § 23, 15 U.S.C. § 1091.]
The assignee of a trade name or service mark “steps into the shoes of the assignor.” Premier Dental Prods. Co. v. Darby Dental Supply Co., 794 F.2d 850, 853 (3d Cir.), cert. denied, 479 U.S. 950 (1986). Appellants do not contest the validity of the assignment from Darrah, nor do they dispute that in purchasing rights to Darrah's software, Planetary Motion succeeded to all rights possessed by Darrah.
“In the absence of registration, rights to a mark traditionally have depended on the very same elements that are now included in the statutory definition: the bona fide use of a mark in commerce that was not made merely to reserve a mark for later exploitation.” Allard Enters., Inc. v. Advanced Programming Res., Inc., 146 F.3d 350, 357 (6th Cir.1998). Common law and statutory trademark infringements are merely specific aspects of unfair competition. New West Corp. v. NYM Co. of Cal., Inc., 595 F.2d 1194, 1201 (9th Cir.1979) (citing, inter alia, Dresser Indus., Inc. v. Heraeus Engelhard Vacuum, Inc., 395 F.2d 457, 461 (3d Cir.), cert. denied, 393 U.S. 934, 89 S.Ct. 293, 21 L.Ed.2d 270 (1968)).
Appellants appear to have conceded that if Darrah sent out original programs and related manuals, this would satisfy the affixation requirement:
MR. GIGLIOTTI [counsel for Techsplosion]: [The mark] has to be on the product or on the associated documentation. It is on neither.
THE COURT: It is not on the associated documentation[?] How about the original programs Darrah sent out and manuals that went with it, and all that material, wasn't that enough for affixation?
MR. GIGLIOTTI: Yes, Your Honor, that is affixation; however, he did not meet the sale requirement.
R3-85-19 to 20.
In any case, the affixation requirement is met because the Software was distributed under a filename that is also the claimed mark, was promoted under the same mark, was accompanied by a user manual bearing the mark, and was sold in a compilation under the mark.
This ownership test is not for the purpose of establishing the “use in commerce” jurisdictional predicate of the Lanham Act. See, e.g., Univ. of Fla. v. KPB, Inc., 89 F.3d 773, 776 n. 4 (11th Cir.1996). See supra discussion in text.
This ownership requirement parallels the statutory definition of “trademark”: “any word, name, symbol, or device, or any combination thereof ... used by a person ... to identify and distinguish his or her goods ... from those manufactured or sold by others ....” 15 U.S.C. § 1127. The Seventh Circuit has held that a higher quantum of use may be necessary to establish ownership rights under common law than under the statute because the notice function of registration is lacking. See Zazu Designs v. L'Oreal, S.A., 979 F.2d 499, 503-04 (7th Cir.1992). In addition, the continuity of a user's commercial activities in connection with the mark is also relevant to determining whether use is sufficient to establish common law ownership. Circuit City Stores, Inc. v. CarMax, Inc., 165 F.3d 1047, 1054-55 (6th Cir.1999) (“A party establishes a common law right to a trademark only by demonstrating that its use of the mark was ‘deliberate and continuous, not sporadic, casual or transitory.’ ”).
Appellants cite Future Domain Corp. v. Trantor Sys. Ltd., 27 U.S.P.Q.2d 1289, 1293, 1993 WL 270522 (N.D.Cal.1993) for the proposition that there must be a sale in order to satisfy the “use in commerce” requirement. Future Domain, however, turned not on the existence of sales but whether the extent of the purported mark owner's activities created a public association between the mark and the product. There, the court determined that a computer software manufacturer's promotion of a mark at a trade show—where at most 7,000 persons actually received or requested information about the mark and where no orders were taken—was not sufficient to create such an association. Id. at 1293-95
Courts applying the “totality of circumstances” approach routinely have found evidence of a few sales of goods to which the mark had been affixed insufficient to establish trademark ownership. For example, in Zazu Designs, 979 F.2d at 503-04, the plaintiff hair salon had sold a few bottles of shampoo bearing the mark “Zazu” both over the counter and mailed over state lines. The court found that such limited sales “neither link the Zazu mark with [the plaintiff's] product in the minds of consumers nor put other producers on notice.” Id. at 503.
In Bonner v. City of Prichard, 661 F.2d 1206 (11th Cir.1981) (en banc), the Eleventh Circuit adopted as binding precedent all decisions handed down by the former Fifth Circuit prior to October 1, 1981.
Darrah: The Coolmail name always comes with the documentation that comes with the software.
* * *
Q: What documentation are you talking about?
A: There's a user manual that comes with it.
* * *
Q: Does it say “Coolmail” on page 1?
Q: Where does it say “Coolmail” on page 1?
A: At the top.
... and on the header of every page.
Q: What does it say, exactly?
A: I'm not sure if it says this verbatim, it's “Coolmail,” space, then the version number.
R2-47-Exh. 3 at 68, 72 to 73.
It is unlikely that the plaintiff's activities in De Costa—costumed performances and distribution of his picture at local rodeos, parades, hospitals, etc.—would generate a “public association” sufficient to confer him common law trademark ownership rights. The court assumed arguendo, however, that the plaintiff's activities did warrant protection, and went on to find that the evidence did not support a finding of likelihood of confusion.
See, e.g., Graeme B. Dinwoodie, (National) Trademark Laws and the (Non-National) Domain Name System, 21 U. PA. J. INT’L ECON. L. 495, 496 (2000) (“[F]or over a century the United States has steadfastly resisted adoption of a registration-based system of trademark priority and has adhered instead to a use-based philosophy.”); see also William M. Landes & Richard A. Posner, Trademark Law: An Economic Perspective, 30 J.L. & Econ. 265, 282 (1987) (comparing the American use-based system to other nations’ registration-based systems, and concluding that the former is more economically efficient).
See, e.g., Keebler Co. v. Rovira Biscuit Corp., 624 F.2d 366, 372 (1st Cir. 1980) (“[F]ederal registration . . . does not create the underlying right in a trademark. That right, which accrues from the use of a particular name or symbol, is essentially a common law property right . . . .”).
Trademark Law Revision Act of 1988, Pub. L. No. 100-667, 102 Stat. 3935 (codified as amended in scattered sections of the U.S.C.).
However, as is suggested by Dragon Bleu (SARL) v. VENM, LLC, 112 U.S.P.Q.2d 1925 (TTAB 2014), which dealt with § 66(a) registrations, if the § 44 or § 66(a) registrant is accused of having abandoned its mark in the U.S., the registrant may be required to present evidence of use sooner than five years after the date of registration. On the issue of trademark abandonment, see Part III.D.
In his lengthy dissent, Justice Stevens expressed his dismay that the decision of a single trademark examiner in an ex parte proceeding a decade earlier, followed by the registrant’s perfunctory filing of a declaration of incontestability, could somehow prevent the Court from striking from the Principal Register an “inherently unregistrable” mark. See Park ‘N Fly, 469 U.S. at 206–07 (Stevens, J., dissenting). He also added his own opinion of trademark quality at the PTO for good measure. See id. at 212 (“No matter how dedicated and how competent administrators may be, the possibility of error is always present, especially in nonadversary proceedings.”). In a footnote to this statement, Justice Stevens quoted a PTO official who testified to Congress that “at any one time, about 7 percent of our 25 million documents are either missing or misfiled.” Id. at 212 n.12 (quoting Hearing Before the Subcomm. on Patents, Copyrights & Trademarks of the S. Comm. on the Judiciary, 98th Cong. 5 (1983) (statement of Gerald J. Mossinghoff, Assistant Secretary and Comm’r of Patents and Trademarks)).
At the examination stage, an examiner may not refuse registration on the ground that the applied-for mark will dilute another mark. See Lanham Act § 2(f), 15 U.S.C. § 1052(f) (“A mark which when used would cause dilution under section 43(c) may be refused registration only pursuant to a[n opposition] proceeding brought under section 13.”).
Sections 8 and 9 add a sixth-month grace period to these deadlines. See Lanham Act §§ 8(c) & 9(a), 15 U.S.C. §§ 1058(c) & 1059(a). Thus, for example, the first Section 8 affidavit must be filed, strictly speaking, during the sixth year following registration or the six-month grace period following that sixth year.
May registrants take advantage of their rights under Lanham Act § 43(a), dealing with unregistered marks, to claim profits and damages even where the registrant did not provide statutory notice? McCarthy suggests that the answer is no:
The more problematic question is whether a registrant who proves infringement under both § 32(1) (registered mark) and § 43(a) (unregistered mark) can avoid the notice limitation imposed by § 29 by claiming all of its damages fall under the § 43(a) count. A strict reading of the statutory language of § 29 would, in the author's opinion, lead to the conclusion that such a registrant cannot avoid the § 29 damage limitation by using § 43(a). Section 29 does not distinguish between the kind of statutory infringement that a registrant proves. Rather, § 29 simply states that no profits and damages shall be recovered “under the provisions of this Act” unless statutory or actual notice was given.
MCCARTHY § 19:144.
As of June 2016. See http://www.wipo.int/madrid/en/members/.
Note that Lanham Act § 14 can be read only to apply to cancellation petitions brought before the PTO. If the registrant has not obtained incontestable status for the mark, challengers in federal court are arguably not limited by Lanham Act § 14 in the grounds on which they can challenge the validity—or at least the enforceability—of the mark. But see MCCARTHY § 30:112 (arguing that § 14’s five-year limit on grounds for cancellation applies to federal courts as well).
Section 33(b) enumerates seven categories of defenses to an action to enforce an incontestable mark. See 15 U.S.C. § 1115(b), quoted ante, at 662, n. 3. In addition, a defendant is free to argue that a mark should never have become incontestable for any of the four reasons enumerated in § 15. 15 U.S.C. § 1065. Moreover, § 15 expressly provides that an incontestable mark may be challenged on any of the grounds set forth in subsections (c) and (e) of § 14, 15 U.S.C. § 1064, and those sections, in turn, incorporate the objections to registrability that are defined in §§ 2(a), 2(b), and 2(c) of the Act. 15 U.S.C. §§ 1052(a), (b), and (c).
See 2 J. McCarthy, Trademarks and Unfair Competition § 26:4 (2d ed. 1984); Restatement of Torts § 732 comment a (1938).
248 U.S. at 97. It goes without saying that the underlying policy upon which this function is grounded is the protection of the public in its purchase of a service or product. See, e.g. In re Canadian Pacific Ltd., 754 F.2d 992, 994 (Fed.Cir. 1985).
Specifically, Grupo Gigante asserted the following causes of action: (1) improper use of a well-known mark, under Article 6 bis of the Paris Convention; (2) unfair competition, under Article 10 bis of the Paris Convention; (3) trademark infringement, under § 43(a) of the Lanham Act, 15 U.S.C. § 1125(a); (4) false designation of origin, misrepresentation, and unfair competition, under § 43(a) of the Lanham Act, 15 U.S.C. § 1125(a); (5) violation of the Federal Trademark Dilution Act of 1996, 15 U.S.C. § 1125(c); (6) common law unfair competition; (7) unfair competition under California law; (8) dilution under California law; and (9) common law misappropriation.
J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition, § 29:2, at 29–6 (4th ed. 2002) (internal footnote omitted).
See Person's Co., Ltd. v. Christman, 900 F.2d 1565, 1569–70 (Fed.Cir. 1990); Buti v. Perosa, S.R.L., 139 F.3d 98, 103–05 (2d Cir.1998); Fuji Photo Film Co., Inc. v. Shinohara Shoji Kabushiki Kaisha, 754 F.2d 591, 599 (5th Cir. 1985).
Fuji Photo, 754 F.2d at 599; see also Person's, 900 F.2d at 1569.
See, e.g., Int'l Bancorp, LLC v. Societe des Bains de Mer, 329 F.3d 359, 370 (4th Cir. 2003).
Int'l Bancorp, 329 F.3d at 389 n. 9 (Motz, J., dissenting) (“Nor does the ‘famous marks' doctrine provide SBM any refuge. That doctrine has been applied so seldom (never by a federal appellate court and only by a handful of district courts) that its viability is uncertain.”).
Vaudable v. Montmartre, Inc., 20 Misc.2d 757, 193 N.Y.S.2d 332 (N.Y.Sup.Ct.1959).
Id. at 334.
Id. at 335.
Id. at 334 (emphasis added).
Wal–Mart Stores, Inc. v. Samara Bros., Inc., 529 U.S. 205, 211 (2000) (internal quotation and editing omitted).
Good faith may also be an issue in such cases. See Hanover Star, 240 U.S. at 415, 36 S.Ct. 357 (excepting from the general Tea Rose–Rectanus principle cases in which “the second adopter has selected the mark with some design inimical to the interests of the first user, such as to take the benefit of the reputation of his goods, to forestall the extension of his trade, or the like.”). Good faith is not raised in this appeal (perhaps because the appeal comes up on summary judgment) and is irrelevant to our analysis.
As McCarthy has noted, traces of the territoriality principle appear in Justice Holmes's opinion for the U.S. Supreme Court in A. Bourjois & Co. v. Katzel, 260 U.S. 689, 692, 43 S.Ct. 244, 67 L.Ed. 464 (1923). McCarthy, supra, at § 29:1, p. 29–4; see also Philip Morris Inc. v. Allen Distribs., Inc., 48 F.Supp.2d 844, 850 (S.D.Ind. 1999) (identifying Bourjois as marking the shift from “the ‘universality’ principle [to] a ‘territoriality principle’ that recognizes a separate legal existence for a trademark in each country whose laws afford protection to the mark”).
Fuji Photo, 754 F.2d at 599; Person's, 900 F.2d at 1569.
See Ingenohl v. Walter E. Olsen & Co., Inc., 273 U.S. 541, 544, (1927) (“A trademark started elsewhere would depend for its protection in Hongkong upon the law prevailing in Hongkong and would confer no rights except by the consent of that law.”); Fuji Photo, 754 F.2d at 599 (“[T]rademark rights exist in each country solely according to that country's statutory scheme.”).
Paris Convention for the Protection of Industrial Property, Mar. 20, 1883, as revised at Stockholm, July 14, 1967, art. 6(3), 21 U.S.T. 1583, § 6(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.”).
Expert surveys can provide the most persuasive evidence of secondary meaning. Comm. for Idaho's High Desert, Inc. v. Yost, 92 F.3d 814, 822 (9th Cir.1996). “However, survey data is not a requirement and secondary meaning can be, and often is, proven by circumstantial evidence.” 5 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition, § 32:190, at 32–319 to 32–320 (4th ed.2002).
There are no other cases that directly guide us here. Although international trademark law has recognized both the territoriality principle and the exception for famous and well-known marks since 1925, remarkably, no case addressed meaningfully the exception before the district court's decision below. Since that decision, only one case has confronted the issue. Empresa Cubana del Tabaca v. Culbro Corp., 70 U.S.P.Q.2d 1650, 2004 WL 602295 (S.D.N.Y.2004). Empresa Cubana adhered closely to the reasoning and conclusion of the district court in this case. Id. at 1676–77.
Although the term “famous marks” is often used to describe marks that qualify for protection under the federal anti-dilution statute, see 15 U.S.C. § 1125(c), the “famous marks” doctrine is, in fact, a different and distinct “legal concept under which a trademark or service mark is protected within a nation if it is well known in that nation even though the mark is not actually used or registered in that nation,” 4 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition, § 29.2, at 29–164 (4th ed.2002). Thus, the famous marks doctrine might more aptly be described as the famous foreign marks doctrine. It is in this latter sense that we reference the famous marks doctrine on this appeal.
ITC's amended complaint also charged defendants with false designation of origin in violation of the Lanham Act, 15 U.S.C. § 1125(a), and deceptive acts and practices in violation of New York General Business Law § 349, but it appears to have abandoned those claims in otherwise opposing defendants' motion for summary judgment. See ITC Ltd. v. Punchgini, Inc., 373 F.Supp.2d at 278.
The “territoriality principle” stands in contrast to the so-called “universality principle,” which posits that “if a trademark [is] lawfully affixed to merchandise in one country, the merchandise would carry that mark lawfully wherever it went and could not be deemed an infringer although transported to another country where the exclusive right to the mark was held by someone other than the owner of the merchandise.” Osawa & Co. v. B & H Photo, 589 F.Supp. at 1171. The universality principle has been rejected in American trademark law. See American Circuit Breaker Corp. v. Or. Breakers, Inc., 406 F.3d at 581 (citing A. Bourjois & Co. v. Katzel, 260 U.S. 689, 43 S.Ct. 244, 67 L.Ed. 464 (1923)).
The reach of Article 6bis was extended to service marks by Article 16(2) of the Agreement on Trade–Related Aspects of Intellectual Property Rights (“TRIPs”), see generally Uruguay Round Agreements Act, Pub.L. No. 103–465, 108 Stat. 4809 (1994) (codified as amended at scattered sections of the United States Code), which states that “Article 6bis of the Paris Convention shall apply, mutatis mutandis, to services.”
Article 10bis of the Paris Convention requires member states to “assure to nationals [of other member states] effective protection against unfair competition.” Paris Convention, art. 10bis.
The Trademark Board's primary function is to determine whether trademarks are registerable and to conduct opposition and cancellation proceedings by which interested parties can dispute the claims of applicants and registrants. See 15 U.S.C. §§ 1051, 1063–64.
In Empresa Cubana, however, we did observe, in dictum, that “[t]o the extent that a foreign entity attempts to utilize the famous marks doctrine as [a] basis for its right to a U.S. trademark and seeks to prevent another entity from using the mark in the United States, the claim should be brought under Section 43(a).” Id. at 480 n. 10.
See, e.g., Pub.L. No. 103–465, 514, 108 Stat. 4809, 4976 (amending 17 U.S.C. § 104A, governing copyrights in restored works, to comport with TRIPs); Pub.L. No. 103–465, 532, 108 Stat. 4809, 4983 (amending 35 U.S.C. § 154, governing United States patents, to comport with TRIPs). Significantly, Congress has enacted legislation to implement TRIPs Article 16(3), which contemplates the extension of anti-dilution protection to certain famous marks. See Federal Trademark Dilution Act of 1995, Pub.L. No. 104–98, 109 Stat. 985 (1995) (codified at 15 U.S.C. § 1125(c)); see H. Rep. 104–374, reprinted in 1995 U.S.C.C.A.N. 1029 (indicating that anti-dilution act was intended to make United States law consistent with terms of TRIPs and Paris Convention). No comparable legislation exists with respect to Article 16(2).
Section 44(b) states:
Any person whose country of origin is a party to 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, or extends reciprocal rights to nationals of the United States by law, shall be entitled to the benefits of this section under the conditions expressed herein to the extent necessary to give effect to any provision of such convention, treaty or reciprocal law, in addition to the rights to which any owner of a mark is otherwise entitled by this chapter.
15 U.S.C. § 1126(b).
Section 44(h) states:
Any person designated in subsection (b) of this section as entitled to the benefits and subject to the provisions of this chapter shall be entitled to effective protection against unfair competition, and the remedies provided in this chapter for infringement of marks shall be available so far as they may be appropriate in repressing acts of unfair competition.
Id. § 1126(h).
See, e.g., 4 McCarthy, supra, § 29:4, at 29–12; Graeme B. Dinwoodie et al., International Intellectual Property Law and Policy 108 (2001).
"One of the most obvious forms" of palming off "occurs when the copier of an article overtly and explicitly misrepresents its source[, for example], where [a] defendant . . . substituted its product for plaintiff's when customers specifically asked for plaintiff's product" (Remco Indus., Inc. v Toyomenka, Inc., 286 F Supp 948, 954 [SD NY 1968], affd 397 F2d 977 [2d Cir 1968] [citation omitted]; see also Shaw v Time-Life Records, 38 NY2d 201, 206  [defendant "could not 'palm off' its records as being the personal work of" plaintiff under New York unfair competition law]).
A plaintiff who relies only on foreign commercial activity may face difficulty proving a cognizable false association injury under § 43(a). A few isolated consumers who confuse a mark with one seen abroad, based only on the presence of the mark on a product in this country and not other misleading conduct by the mark holder, would rarely seem to have a viable § 43(a) claim.
The story is different when a defendant, as alleged here, has—as a cornerstone of its business—intentionally passed off its goods in the United States as the same product commercially available in foreign markets in order to influence purchases by American consumers. See M. Kramer Mfg. Co. v. Andrews, 783 F.2d 421, 448 (4th Cir.1986) (“[E]vidence of intentional, direct copying establishes a prima facie case of secondary meaning sufficient to shift the burden of persuasion to the defendant on that issue.”). Such an intentional deception can go a long way toward establishing likelihood of confusion. See Blinded Veterans, 872 F.2d at 1045 (“Intent to deceive ... retains potency; when present, it is probative evidence of a likelihood of confusion.”).