808 F.3d 1321
United States Court of Appeals,
In re Simon Shiao TAM.
Dec. 22, 2015.
As Corrected Feb. 11, 2016.
Attorneys and Law Firms
*1326 Ronald D. Coleman, Archer & Greiner, P.C., Hackensack, NJ, argued for appellant. Also represented by Joel Geoffrey MacMull; John C. Connell, Haddonfield, NJ; Darth M. Newman, Martin Law Firm LLC, Pittsburgh, PA.
Daniel Tenny, Appellate Staff, Civil Division, United States Department of Justice, Washington, DC, argued for appellee Michelle K. Lee. Also represented by Benjamin C. Mizer, Mark R. Freeman, Joshua Marc Salzman; Nathan K. Kelley, Thomas W. Krause, Molly R. Silfen, Christina Hieber, Thomas L. Casagrande, Office of the Solicitor, United States Patent and Trademark Office, Alexandria, VA.
Lee Rowland, Speech, Privacy & Technology, American Civil Liberties Union Foundation, New York, NY, argued for amici curiae American Civil Liberties Union, American Civil Liberties Union of Oregon, American Civil Liberties Union of the National Capital Area. Also represented by Esha Bhandari, Brett Max Kaufman; Arthur B. Spitzer, American Civil Liberties Union of the National Capital Area, Washington, DC; Mathew W. Dos Santos, ACLU of Oregon, Portland, OR.
Jeffrey Joseph Lopez, Drinker Biddle & Reath LLP, Washington, DC, for amici curiae Amanda Blackhorse, Marcus Briggs–Cloud, Phillip Gover, Jillian Pappan, Courtney Tsotigh. Also represented by Jesse A. Witten.
*1327 Megan Leef Brown, Wiley Rein, LLP, Washington, DC, for amici curiae Cato Institute, The Rutherford Institute. Also represented by Christopher J. Kelly, Joshua S. Turner, Jennifer L. Elgin, Dwayne D. Sam; Cato Institute also represented by Ilya Shapiro, Cato Institute, Washington DC; The Rutherford Institute also represented by Douglas R. McKusick, John W. Whitehead, Charlottesville, VA.
Marc J. Randazza, Randazza Legal Group, Las Vegas, NV, for amicus curiae First Amendment Lawyers Association. Also represented by Ronald D. Green, Jr.
Charanjit Brahma, Troutman Sanders LLP, San Francisco, CA, for amici curiae Fred T. Korematsu Center for Law and Equality, National Asian Pacific American Bar Association, South Asian Bar Association of Washington, DC. National Asian Pacific American Bar Association also represented by George C. Chen, Bryan Cave LLP, Phoenix, AZ.
Hugh C. Hansen, Fordham University School of Law, New York, NY, as amicus curiae pro se.
Lawrence Kurt Nodine, Ballard Spahr LLP, Atlanta GA, for amicus curiae International Trademark Association. Also represented by Robert D. Carroll, Goodwin Procter LLP, Boston, MA.
Robert Lloyd Raskopf, Quinn Emanuel Urquhart & Sullivan, LLP, New York, NY, for amicus curiae Pro–Football, Inc. Also represented by Sanford Ian Weisburst, Todd Anten.
Phillip R. Malone, Juelsgaard Intellectual Property and Innovation Clinic, Mills Legal Clinic, Stanford Law School, Stanford, CA, for amicus curiae Public Knowledge. Also represented by Jeffrey Theodore Pearlman.
Richard L. Stanley, Law Office of Richard L. Stanley, Houston TX, for amicus curiae Richard L. Stanley.
Before PROST, Chief Judge, NEWMAN, LOURIE, DYK, MOORE, O’MALLEY, REYNA, WALLACH, TARANTO, CHEN, HUGHES, and STOLL, Circuit Judges.
Opinion for the court filed by Circuit Judge MOORE, in which Chief Judge PROST and Circuit Judges NEWMAN, O’MALLEY, WALLACH, TARANTO, CHEN, HUGHES, and STOLL join.
Concurring opinion filed by Circuit Judge O’MALLEY, in which Circuit Judge WALLACH joins.
Opinion concurring in part and dissenting in part filed by Circuit Judge DYK, in which Circuit Judges LOURIE and REYNA join with respect to parts I, II, III, and IV.
Dissenting opinion filed by Circuit Judge LOURIE.
Dissenting opinion filed by Circuit Judge REYNA.
MOORE, Circuit Judge.
Section 2(a) of the Lanham Act bars the Patent and Trademark Office (“PTO”) from registering scandalous, immoral, or disparaging marks. 15 U.S.C. § 1052(a). The government enacted this law—and defends it today—because it disapproves of the messages conveyed by disparaging marks. It is a bedrock principle underlying the First Amendment that the government may not penalize private speech merely because it disapproves of the message it conveys. That principle governs even when the government’s message-discriminatory penalty is less than a prohibition.
Courts have been slow to appreciate the expressive power of trademarks. Words—even a single word—can be powerful. Mr. Simon Shiao Tam named his band THE SLANTS to make a statement *1328 about racial and cultural issues in this country. With his band name, Mr. Tam conveys more about our society than many volumes of undisputedly protected speech. Another rejected mark, STOP THE ISLAMISATION OF AMERICA, proclaims that Islamisation is undesirable and should be stopped. Many of the marks rejected as disparaging convey hurtful speech that harms members of oft-stigmatized communities. But the First Amendment protects even hurtful speech.
The government cannot refuse to register disparaging marks because it disapproves of the expressive messages conveyed by the marks. It cannot refuse to register marks because it concludes that such marks will be disparaging to others. The government regulation at issue amounts to viewpoint discrimination, and under the strict scrutiny review appropriate for government regulation of message or viewpoint, we conclude that the disparagement proscription of § 2(a) is unconstitutional. Because the government has offered no legitimate interests justifying § 2(a), we conclude that it would also be unconstitutional under the intermediate scrutiny traditionally applied to regulation of the commercial aspects of speech. We therefore vacate the Trademark Trial and Appeal Board’s (“Board”) holding that Mr. Tam’s mark is unregistrable, and remand this case to the Board for further proceedings.
I. The Lanham Act
Congress enacted the Lanham Act in 1946 to provide a national system for registering and protecting trademarks used in interstate and foreign commerce. Congress’s purpose in enacting the Lanham Act was to advance the two related goals of trademark law. First, the purpose of the Lanham Act 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.” Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 782 n. 15, 112 S.Ct. 2753, 120 L.Ed.2d 615 (1992) (Stevens, J., concurring) (quoting S.Rep. No. 791333, at 3 (1946)). Second, the Lanham Act ensures that a markholder can protect “his investment from ... misappropriation by pirates and cheats.” Id.; see also Inwood Labs., Inc. v. Ives Labs., Inc., 456 U.S. 844, 854 n. 14, 102 S.Ct. 2182, 72 L.Ed.2d 606 (1982) (“By applying a trademark to goods produced by one other than the trademark’s owner, the infringer deprives the owner of the goodwill which he spent energy, time, and money to obtain. At the same time, the infringer deprives consumers of their ability to distinguish among the goods of competing manufacturers.” (citations omitted)).
“Registration is significant. The Lanham Act confers important legal rights and benefits on trademark owners who register their marks.” B & B Hardware, Inc. v. Hargis Ind., Inc., ––– U.S. ––––, 135 S.Ct. 1293, 1300, 191 L.Ed.2d 222 (2015) (quotation marks omitted). These benefits—unavailable in the absence of federal registration—are numerous, and include both substantive and procedural rights. The holder of a federal trademark has a right to exclusive nationwide use of that mark where there was no prior use by others. See 15 U.S.C. §§ 1072, 1115. Because the common law grants a markholder the right to exclusive use only in the geographic areas where he has actually used his mark, see 5 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 26:32 (4th ed.) (hereinafter “McCarthy”), holders of a federally registered trademark have an important substantive right they could not otherwise obtain. Also, a registered mark is presumed *1329 to be valid, 15 U.S.C. § 1057(b), and the mark becomes incontestable (with certain exceptions) after five years of consecutive post-registration use, id. § 1065; see also B & B Hardware, 135 S.Ct. at 1310 (“Incontestability is a powerful protection.”). A markholder may sue in federal court to enforce his trademark, 15 U.S.C. § 1121, and he may recover treble damages if he can show infringement was willful, id. § 1117. He may also obtain the assistance of U.S. Customs and Border Protection in restricting importation of infringing or counterfeit goods, id. § 1124, 19 U.S.C. § 1526, and qualify for a simplified process for obtaining recognition and protection of his mark in countries that have signed the Paris Convention, see id. § 1141b (Madrid Protocol); Paris Convention for the Protection of Industrial Property art. 6 quinquies, July 14, 1967, 21 U.S.T. 1583, 828 U.N.T.S. 305. Lastly, registration operates as a complete defense to state or common law claims of trademark dilution. 15 U.S.C. § 1125(c)(6).
Under the Lanham Act, the PTO must register source-identifying trademarks unless the mark falls into one of several categories of marks precluded from registration. Id. § 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....” (emphasis added)). Many of these categories bar the registration of deceptive or misleading speech, because such speech actually undermines the interests served by trademark protection and, thus, the Lanham Act’s purposes in providing for registration. For example, a mark may not be registered if it resembles a registered mark such that its use is likely to “cause confusion, or to cause mistake, or to deceive,” § 2(d), or if it is “deceptively misdescriptive,” § 2(e). These restrictions on registration of deceptive speech do not run afoul of the First Amendment. See Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n, 447 U.S. 557, 563, 100 S.Ct. 2343, 65 L.Ed.2d 341 (1980) (“The government may ban forms of communication more likely to deceive the public than to inform it.”); see also Friedman v. Rogers, 440 U.S. 1, 13, 15–16, 99 S.Ct. 887, 59 L.Ed.2d 100 (1979); Ohralik v. Ohio State Bar Ass’n, 436 U.S. 447, 462–63, 98 S.Ct. 1912, 56 L.Ed.2d 444 (1978).
Section 2(a), however, is a hodgepodge of restrictions. Among them is the bar on registration of a mark that “[c]onsists of or comprises immoral, deceptive, or scandalous matter; or matter which may disparage or falsely suggest a connection with persons, living or dead, institutions, beliefs, or national symbols, or bring them into contempt or disrepute.” Section 2(a) contains proscriptions against deceptive speech, for example, the prohibition on deceptive matter or the prohibition on falsely suggesting a connection with a person or institution. But other restrictions in § 2(a) differ in that they are based on the expressive nature of the content, such as the ban on marks that may disparage persons or are scandalous or immoral. These latter restrictions cannot be justified on the basis that they further the Lanham Act’s purpose in preventing consumers from being deceived. These exclusions from registration do not rest on any judgment that the mark is deceptive or likely to cause consumer confusion, nor do they protect the markholder’s investment in his mark. They deny the protections of registration for reasons quite separate from any ability of the mark to serve the consumer and investment interests underlying trademark protection. In fact, § 2(a)’s exclusions can undermine those interests because they can even be employed in cancellation proceedings challenging a mark many years after its issuance *1330 and after the markholder has invested millions of dollars protecting its brand identity and consumers have come to rely on the mark as a brand identifier.
This case involves the disparagement provision of § 2(a).1 Section 2(a)’s ban on the federal registration of “immoral” or “scandalous” marks originated in the trademark legislation of 1905. See Act of Feb. 20, 1905, ch. 592, § 5(a), 33 Stat. 724, 725. The provision barring registration based on disparagement first appeared in the Lanham Act in 1946. Pub.L. 79–489, § 2(a), 60 Stat. 427, 428 (codified at 15 U.S.C. § 1052(a)). It had no roots in the earlier trademark statute or the common law. There were few marks rejected under the disparagement provision following enactment of the Lanham Act. Only in the last several decades has the disparagement provision become a more frequent ground of rejection or cancellation of trademarks. Marks that the PTO has found to be disparaging include: REDSKINS, Pro–Football, Inc. v. Blackhorse, No. 1–14–CV–01043–GBL, 112 F.Supp.3d 439, 2015 WL 4096277 (E.D.Va. July 8, 2015) (2014 PTO cancellation determination currently on appeal in Fourth Circuit); STOP THE ISLAMISATION OF AMERICA, In re Geller, 751 F.3d 1355 (Fed.Cir.2014); THE CHRISTIAN PROSTITUTE (2013); AMISHHOMO (2013); MORMON WHISKEY (2012); KHORAN for wine, In re Lebanese Arak Corp., 94 U.S.P.Q.2d 1215 (T.T.A.B. Mar. 4, 2010); HAVE YOU HEARD THAT SATAN IS A REPUBLICAN? (2010); RIDE HARD RETARD (2009); ABORT THE REPUBLICANS (2009); HEEB, In re Heeb Media, LLC, 89 U.S.P.Q.2d 1071 (T.T.A.B. Nov. 26, 2008); SEX ROD, Bos. Red Sox Baseball Club L.P. v. Sherman, 88 U.S.P.Q.2d 1581 (T.T.A.B. Sept. 9, 2008) (sustaining an opposition on multiple grounds, including disparagement); MARRIAGE IS FOR FAGS (2008); DEMOCRATS SHOULDN’T BREED (2007); REPUBLICANS SHOULDN’T BREED (2007); 2 DYKE MINIMUM (2007); WET BAC/WET B.A.C. (2007); URBAN INJUN (2007); SQUAW VALLEY, In re Squaw Valley Dev. Co., 80 U.S.P.Q.2d 1264 (T.T.A.B. June 2, 2006); DON’T BE A WET BACK (2006); FAGDOG (2003); N.I.G.G.A. NATURALLY INTELLIGENT GOD GIFTED AFRICANS (1996); a mark depicting a defecating dog, Greyhound Corp. v. Both Worlds, Inc., 6 U.S.P.Q.2d 1635 (T.T.A.B. Mar. 30, 1988) (found to disparage Greyhound’s trademarked running dog logo); an image consisting of the national symbol of the Soviet Union with an “X” over it, In re Anti–Communist World Freedom Cong., Inc., 161 U.S.P.Q. 304 (T.T.A.B. Feb. 24, 1969); DOUGH–BOY for “a prophylactic preparation for the prevention of venereal diseases,” Doughboy Indus., Inc. v. Reese Chem. Co., 88 U.S.P.Q. 227 (T.T.A.B. Jan. 25, 1951).
A disparaging mark is a mark which “dishonors by comparison with what is inferior, slights, deprecates, degrades, or affects or injures by unjust comparison.” Geller, 751 F.3d at 1358 (alterations omitted). To determine if a mark is disparaging under § 2(a), a trademark examiner of the PTO considers:
*1331 (1) What is 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; and
(2) If that meaning is found to refer to identifiable persons, institutions, beliefs or national symbols, whether that meaning may be disparaging to a substantial composite of the referenced group.
Trademark Manual of Exam. Proc. (“TMEP”) § 1203.03(b)(i) (Jan.2015 ed.) (citing Geller, 751 F.3d at 1358). If the examiner “make[s] a prima facie showing that a substantial composite, although not necessarily a majority, of the referenced group would find the proposed mark, as used on or in connection with the relevant goods or services, to be disparaging in the context of contemporary attitudes,” the burden shifts to the applicant for rebuttal. Id. If the applicant fails to rebut the prima facie case of disparagement, the examiner refuses to register the mark. The Trademark Manual of Examining Procedure does not require an examiner who finds a mark disparaging to consult her supervisor or take any further steps to ensure the provision is applied fairly and consistently across the agency. Compare TMEP § 1203.03 (no discussion of action to take if examiner finds mark disparaging), with TMEP § 1203.01 (requiring examiner who finds a mark scandalous or immoral to consult his supervisor). A single examiner, with no input from her supervisor, can reject a mark as disparaging by determining that it would be disparaging to a substantial composite of the referenced group.
II. Facts of This Case
Mr. Tam is the “front man” for the Asian–American dance-rock band The Slants. Mr. Tam named his band The Slants to “reclaim” and “take ownership” of Asian stereotypes. J.A. 129–30. The band draws inspiration for its lyrics from childhood slurs and mocking nursery rhymes, J.A. 130, and its albums include “The Yellow Album” and “Slanted Eyes, Slanted Hearts.” The band “feel[s] strongly that Asians should be proud of their cultural heri[ta]ge, and not be offended by stereotypical descriptions.” J.A. 52. With their lyrics, performances, and band name, Mr. Tam and his band weigh in on cultural and political discussions about race and society that are within the heartland of speech protected by the First Amendment.
On November 14, 2011, Mr. Tam filed the instant application (App. No. 85/472,044) seeking to register the mark THE SLANTS for “Entertainment in the nature of live performances by a musical band,” based on his use of the mark since 2006.2 The examiner refused to register Mr. Tam’s mark, finding it likely disparaging to “persons of Asian descent” under § 2(a). The examiner found that the mark likely referred to people of Asian descent in a disparaging way, explaining that the term “slants” had “a long history of being used to deride and mock a physical feature” of people of Asian descent. J.A. 42. And even though Mr. Tam may have chosen the mark to “reappropriate the disparaging term,” the examiner found that a substantial *1332 composite of persons of Asian descent would find the term offensive. J.A. 43.
The Board affirmed the examiner’s refusal to register the mark. The Board wrote that “it is abundantly clear from the record not only that THE SLANTS ... would have the ‘likely meaning’ of people of Asian descent but also that such meaning has been so perceived and has prompted significant responses by prospective attendees or hosts of the band’s performances.” In re Tam, No. 85472044, 2013 WL 5498164, at *5 (T.T.A.B. Sept. 26, 2013) (“Board Opinion ”). To support its finding that the mark likely referred to people of Asian descent, the Board pointed to dictionary definitions, the band’s website, which displayed the mark next to “a depiction of an Asian woman, utilizing rising sun imagery and using a stylized dragon image,” and a statement by Mr. Tam that he selected the mark in order to “own” the stereotype it represents. Id. The Board also found that the mark is disparaging to a substantial component of people of Asian descent because “[t]he dictionary definitions, reference works and all other evidence unanimously categorize the word ‘slant,’ when meaning a person of Asian descent, as disparaging,” and because there was record evidence of individuals and groups in the Asian community objecting to Mr. Tam’s use of the word. Id. at *7. The Board therefore disqualified the mark for registration under § 2(a).
Mr. Tam appealed, arguing that the Board erred in finding the mark disparaging and that § 2(a) is unconstitutional. On appeal, a panel of this Court affirmed the Board determination that the mark is disparaging.3 In re Tam, 785 F.3d 567, 570–71 (Fed.Cir.2015) (“Panel Opinion ”), reh’g en banc granted, opinion vacated, 600 Fed.Appx. 775 (Fed.Cir.2015) (“En Banc Order ”). Although the term “slants” has several meanings, the panel found that substantial evidence supported the Board’s finding that the mark likely refers to people of Asian descent. Panel Op. at 570–71. This included an article in which Mr. Tam described the genesis of the band’s name by explaining: “I was trying to think of things that people associate with Asians. Obviously, one of the first things people say is that we have slanted eyes....” Id. at 570 (quoting J.A. 130). Moreover, the band’s Wikipedia page stated that the band’s name is “derived from an ethnic slur for Asians.” Id. (quoting J.A. 57). The Wikipedia entry quoted Mr. Tam: “We want to take on these stereotypes that people have about us, like the slanted eyes, and own them. We’re very proud of being Asian—we’re not going to hide that fact. The reaction from the Asian community has been positive.” J.A. 57. The record included an image from the band’s website in which the mark THE SLANTS is set against Asian imagery. Id. (citing J.A. 59). Finally, the record included unrebutted evidence that both individuals and Asian groups have perceived the term as referring to people of Asian descent. Id. at 570–71 (citing, e.g., J.A. 95 (“[Mr. Tam] was initially slated to give the keynote address at the 2009 Asian American Youth Leadership Conference in Portland. But some conference supporters and attendees felt the name of the band was offensive and racist, and out of respect for these opinions the conference organizers decided to choose someone less controversial.”)).
The panel also found that substantial evidence supported the Board’s finding that the mark is disparaging to a substantial composite of people of Asian descent. Panel Op. at 571. It noted that the definitions in evidence universally characterize the word “slant” as disparaging, offensive, or an ethnic slur when used to refer to a person of Asian descent, including the dictionary *1333 definitions provided by Mr. Tam. Id. The record also included a brochure published by the Japanese American Citizens League describing the term “slant,” when used to refer to people of Asian descent, as a “derogatory term” that is “demeaning” and “cripple[s] the spirit.” Id. (quoting J.A. 48–49). Finally, the record included news articles and blog posts discussing the offensive nature of the band’s name. Id. (citing Board Op. at *2–3; J.A. 45, 51, 94–98, 100).
Having found the mark disparaging under § 2(a), the panel held that binding precedent foreclosed Mr. Tam’s arguments that § 2(a) is unconstitutional, including Mr. Tam’s argument that § 2(a) violates the First Amendment on its face. Panel Op. at 572–73. As the panel explained, in McGinley, our predecessor court held that the refusal to register a mark under § 2(a) does not bar the applicant from using the mark, and therefore does not implicate the First Amendment. Id. at 572 (citing In re McGinley, 660 F.2d 481, 484 (C.C.P.A.1981)). The entirety of the McGinley analysis was:
With respect to appellant’s First Amendment rights, it is clear that the PTO’s refusal to register appellant’s mark does not affect his right to use it. No conduct is proscribed, and no tangible form of expression is suppressed. Consequently, appellant’s First Amendment rights would not be abridged by the refusal to register his mark.
660 F.2d at 484 (citations omitted). In subsequent cases, panels of this Court relied on the holding in McGinley. See In re Fox, 702 F.3d 633, 635 (Fed.Cir.2012); In re Boulevard Entm’t, Inc., 334 F.3d 1336, 1343 (Fed.Cir.2003); In re Mavety Media Grp., 33 F.3d 1367, 1374 (Fed.Cir.1994). Additional views by the panel’s authoring judge questioned whether the en banc court should reconsider the constitutionality of § 2(a) en banc. Panel Op. at 573–85 (Moore, J., additional views).
More than thirty years have passed since the decision in McGinley, and in that time both the McGinley decision and our reliance on it have been widely criticized.4 Id. at 573–74. Furthermore, the McGinley analysis was cursory, without citation *1334 to legal authority, and decided at a time when the First Amendment had only recently been applied to commercial speech. Id. at 574, 581 (citing Cent. Hudson, 447 U.S. at 566, 100 S.Ct. 2343). First Amendment jurisprudence on the unconstitutional conditions doctrine and the protection accorded to commercial speech has evolved significantly since the McGinley decision. Id. at 574; see also id. at 574–580 (describing evolution of commercial speech doctrine and unconstitutional conditions doctrine).
Other courts’ reliance on the reasoning in McGinley further reinforces the importance of taking this case en banc. Without analysis, the Fifth Circuit wrote that “[w]e join our sister circuit in rejecting [the applicant’s] argument that prohibiting him from registering a mark with the PTO violates his [F]irst [A]mendment rights.” Test Masters Educ. Servs., Inc. v. Singh, 428 F.3d 559, 578 n. 9 (5th Cir.2005). And a district court in the Eastern District of Virginia relied upon McGinley when it concluded that the cancellation of trademark registrations under § 2(a) did not implicate the First Amendment. Pro–Football, Inc., 112 F.Supp.3d at 454–57, 2015 WL 4096277, at *8–10 (“[T]he Court agrees with the Federal Circuit and Fifth Circuit and holds that Section 2(a) of the Lanham Act does not implicate the First Amendment.”).
For these reasons, we sua sponte ordered rehearing en banc. We asked the parties to file briefs on the following issue:
Does the bar on registration of disparaging marks in 15 U.S.C. § 1052(a) violate the First Amendment?
En Banc Order at 775. In addition to the parties’ briefs, we received ten amicus briefs. We heard oral argument on October 2, 2015.
I. Section 2(a)’s Denial of Important Legal Rights to Private Speech Based on Disapproval of the Message Conveyed Is Subject to, and Cannot Survive, Strict Scrutiny
Strict scrutiny is used to review any governmental regulation that burdens private speech based on disapproval of the message conveyed. Section 2(a), which denies important legal rights to private speech on that basis, is such a regulation. It is therefore subject to strict scrutiny. It is undisputed that it cannot survive strict scrutiny.
A. The Disparagement Provision, Which Discriminates Based on Disapproval of the Message, Is Not Content or Viewpoint Neutral
“Content-based regulations are presumptively invalid.” R.A.V. v. City of St. Paul, 505 U.S. 377, 382, 112 S.Ct. 2538, 120 L.Ed.2d 305 (1992); see also Ashcroft v. ACLU, 542 U.S. 656, 660, 124 S.Ct. 2783, 159 L.Ed.2d 690 (2004). “Content-based laws—those that target speech based on its communicative content—are presumptively unconstitutional and may be justified only if the government proves that they are narrowly tailored to serve compelling state interests.” Reed v. Town of Gilbert, ––– U.S. ––––, 135 S.Ct. 2218, 2226, 192 L.Ed.2d 236 (2015); see also Police Dep’t of Chi. v. Mosley, 408 U.S. 92, 95, 92 S.Ct. 2286, 33 L.Ed.2d 212 (1972) (“[A]bove all else, the First Amendment means that the government has no power to restrict expression because of its message, its ideas, its subject matter, or its content.”). A regulation is content based even when its reach is defined simply by the topic (subject matter) of the covered speech. See Reed, 135 S.Ct. at 2230.
Viewpoint-based regulations, targeting the substance of the viewpoint *1335 expressed, are even more suspect. They are recognized as a particularly “egregious form of content discrimination,” id., though they have sometimes been discussed without being cleanly separated from topic discrimination, see, e.g., Mosley, 408 U.S. at 95, 92 S.Ct. 2286. Such measures “raise[ ] the specter that the government may effectively drive certain ideas or viewpoints from the marketplace.” Simon & Schuster, Inc. v. Members of N.Y. State Crime Victims Bd., 502 U.S. 105, 116, 112 S.Ct. 501, 116 L.Ed.2d 476 (1991); see also Sorrell v. IMS Health Inc., ––– U.S. ––––, 131 S.Ct. 2653, 2667, 180 L.Ed.2d 544 (2011); Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819, 828, 115 S.Ct. 2510, 132 L.Ed.2d 700 (1995). “The First Amendment requires heightened scrutiny whenever the government creates ‘a regulation of speech because of disagreement with the message it conveys.’ ” Sorrell, 131 S.Ct. at 2664 (quoting Ward v. Rock Against Racism, 491 U.S. 781, 791, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989)). This is true whether the regulation bans or merely burdens speech. “[H]eightened judicial scrutiny is warranted” when an act “is designed to impose a specific, content-based burden on protected expression.” Id.; see also Rosenberger, 515 U.S. at 828, 115 S.Ct. 2510 (“[T]he government offends the First Amendment when it imposes financial burdens on certain speakers based on the content of their expression.”). “The distinction between laws burdening and laws banning speech is but a matter of degree. The Government’s content-based burdens must satisfy the same rigorous scrutiny as its content-based bans.” United States v. Playboy Entm’t Grp., Inc., 529 U.S. 803, 812, 120 S.Ct. 1878, 146 L.Ed.2d 865 (2000). “Lawmakers may no more silence unwanted speech by burdening its utterance than by censoring its content.” Sorrell, 131 S.Ct. at 2664; see also infra at 1340.
It is beyond dispute that § 2(a) discriminates on the basis of content in the sense that it “applies to particular speech because of the topic discussed.” Reed, 135 S.Ct. at 2227. Section 2(a) prevents the registration of disparaging marks—it cannot reasonably be argued that this is not a content-based restriction or that it is a content-neutral regulation of speech. And the test for disparagement—whether a substantial composite of the referenced group would find the mark disparaging—makes clear that it is the nature of the message conveyed by the speech which is being regulated. If the mark is found disparaging by the referenced group, it is denied registration. “Listeners’ reaction to speech is not a content-neutral basis for regulation.” Forsyth Cty. v. Nationalist Movement, 505 U.S. 123, 134, 112 S.Ct. 2395, 120 L.Ed.2d 101 (1992).
And § 2(a) does more than discriminate on the basis of topic. It also discriminates on the basis of message conveyed, “the idea or message expressed,” Reed, 135 S.Ct. at 2227; it targets “viewpoints [in] the marketplace,” Simon & Schuster, 502 U.S. at 116, 112 S.Ct. 501. It does so as a matter of avowed and undeniable purpose, and it does so on its face.5
*1336 First, the government enacted and continues to defend § 2(a) “because of disagreement with the message [disparaging marks] convey [ ].” Sorrell, 131 S.Ct. at 2664. When the government refuses to register a mark under § 2(a), it does so because it disapproves of “the message a speaker conveys” by the mark. Reed, 135 S.Ct. at 2227. Underscoring its hostility to these messages, the government repeatedly asserts in its briefing before this court that it ought to be able to prevent the registration of “the most vile racial epithets and images,” Appellee’s En Banc Br. 1, and “to dissociate itself from speech it finds odious,” id. 41. The legislative history of § 2(a) reinforces this conclusion. See Hearings on H.R. 4744 Before the Sub-comm. on Trademarks of the House Comm. on Patents, 76th Cong., 1st Sess. 18–21 (1939) (statement of Rep. Thomas E. Robertson) (Rep. Maroney) (“[W]e would not want to have Abraham Lincoln gin.”); id. (Rep. Rogers) (stating that a mark like “Abraham Lincoln gin ought not to be used,” and that § 2(a) “would take care of [such] abuses”). From its enactment in 1946 through its defense of the statute today, the government has argued that the prohibited marks ought not to be registered because of the messages the marks convey. When the government discriminates against speech because it disapproves of the message conveyed by the speech, it discriminates on the basis of viewpoint. Sorrell, 131 S.Ct. at 2664.
The legal significance of viewpoint discrimination is the same whether the government disapproves of the message or claims that some part of the populace will disapprove of the message. This point is recognized in the Supreme Court’s long-standing condemnation of government impositions on speech based on adverse reactions among the public. See, e.g., Snyder v. Phelps, 562 U.S. 443, 460–61, 131 S.Ct. 1207, 179 L.Ed.2d 172 (2011); R.A.V., 505 U.S. 377, 112 S.Ct. 2538, 120 L.Ed.2d 305; Texas v. Johnson, 491 U.S. 397, 414, 109 S.Ct. 2533, 105 L.Ed.2d 342 (1989).
Second, the disparagement provision at issue is viewpoint discriminatory on its face. The PTO rejects marks under § 2(a) when it finds the marks refer to a group in a negative way, but it permits the registration of marks that refer to a group in a positive, non-disparaging manner. In this case the PTO refused to register Mr. Tam’s mark because it found the mark “disparaging” and “objectionable” to people of Asian descent. Tam, 2013 WL 5498164, at *6. But the PTO has registered marks that refer positively to people of Asian descent. See, e.g., CELEBRASIANS, ASIAN EFFICIENCY. Similarly, the PTO has prohibited the registration of marks that it found disparaged other groups. See, e.g., Pro–Football, 112 F.Supp.3d 439, 2015 WL 4096277 (affirming cancellation of REDSKINS); Geller, 751 F.3d 1355 (affirming rejection of STOP THE ISLAMISATION OF AMERICA); Lebanese Arak Corp., 94 U.S.P.Q.2d 1215 (refusing to register KHORAN for wine); Heeb Media, 89 U.S.P.Q.2d 1071 (refusing to register HEEB); Squaw Valley Dev. *1337 Co., 80 U.S.P.Q.2d 1264 (refusing to register SQUAW VALLEY for one class of goods, but registering it for another). Yet the government registers marks that refer to particular ethnic groups or religions in positive or neutral ways—for example, NAACP, THINK ISLAM, NEW MUSLIM COOL, MORMON SAVINGS, JEWISHSTAR, and PROUD 2 B CATHOLIC.
The government argues that § 2(a) is viewpoint neutral because it does not eliminate any particular view-point—only particular words. Appellee’s En Banc Br. 39–40. It argues that under § 2(a), two marks with diametrically opposed viewpoints will both be refused, so long as those marks use the same disparaging term. Id. 39–40. It points to Mr. Tam—who does not seek to express an anti-Asian viewpoint—as proof. It cites a statement in R.A.V. that a hypothetical statute that prohibited “odious racial epithets ... to proponents of all views” would not be viewpoint discriminatory. Id. 40 (quoting 505 U.S. at 391, 112 S.Ct. 2538); see also Ridley v. Mass. Bay Transp. Auth., 390 F.3d 65, 90–91 (1st Cir.2004) (holding that “guidelines prohibiting demeaning or disparaging ads are themselves viewpoint neutral”).
The R.A.V. statement does not apply here. The government’s starting point—that it rejects marks conveying diametrically opposed viewpoints, if they contain the same offensive word—is incorrect. The PTO looks at what message the referenced group takes from the applicant’s mark in the context of the applicant’s use, and it denies registration only if the message received is a negative one. Thus, an applicant can register a mark if he shows it is perceived by the referenced group in a positive way, even if the mark contains language that would be offensive in another context. For example, the PTO registered the mark DYKES ON BIKES, U.S. Reg. No. 3,323,803, after the applicant showed the term was often enough used with pride among the relevant population. In Squaw Valley, the Board allowed the registration of the mark SQUAW VALLEY in connection with one of the applied-for classes of goods (namely, skiing-related products), but not in connection with a different class of goods. 80 U.S.P.Q.2d at *22. Section 2(a) does not treat identical marks the same. A mark that is viewed by a substantial composite of the referenced group as disparaging is rejected. It is thus the viewpoint of the message conveyed which causes the government to burden the speech. This form of regulation cannot reasonably be argued to be content neutral or viewpoint neutral.
The government’s argument also fails because denial of registration under § 2(a) turns on the referenced group’s perception of a mark. Speech that is offensive or hostile to a particular group conveys a distinct viewpoint from speech that carries a positive message about the group. STOP THE ISLAMISATION OF AMERICA and THINK ISLAM express two different viewpoints. Under § 2(a), one of these viewpoints garners the benefits of registration, and one does not. The government enacted § 2(a), and defends it today, because it is hostile to the messages conveyed by the refused marks. Section 2(a) is a viewpoint-discriminatory regulation of speech, created and applied in order to stifle the use of certain disfavored messages. Strict scrutiny therefore governs its First Amendment assessment—and no argument has been made that the measure survives such scrutiny.
B. The Disparagement Provision Regulates the Expressive Aspects of the Mark, Not Its Function As Commercial Speech
The government cannot escape strict scrutiny by arguing that § 2(a) regulates *1338 commercial speech. True, trademarks identify the source of a product or service, and therefore play a role in the “dissemination of information as to who is producing and selling what product, for what reason, and at what price.” Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, Inc., 425 U.S. 748, 765, 96 S.Ct. 1817, 48 L.Ed.2d 346 (1976). But they very commonly do much more than that. And, critically, it is always a mark’s expressive character, not its ability to serve as a source identifier, that is the basis for the disparagement exclusion from registration. The disparagement provision must be assessed under First Amendment standards applicable to what it targets, which is not the commercial-speech function of the mark.
This case exemplifies how marks often have an expressive aspect over and above their commercial-speech aspect. Mr. Tam explicitly selected his mark to create a dialogue on controversial political and social issues. With his band name, Mr. Tam makes a statement about racial and ethnic identity. He seeks to shift the meaning of, and thereby reclaim, an emotionally charged word. He advocates for social change and challenges perceptions of people of Asian descent. His band name pushes people. It offends. Despite this—indeed, because of it—Mr. Tam’s band name is expressive speech.
Importantly, every time the PTO refuses to register a mark under § 2(a), it does so because it believes the mark conveys an expressive message—a message that is disparaging to certain groups. STOP THE ISLAMISATION OF AMERICA is expressive. In refusing to register the mark, the Board explained that the “mark’s admonition to ‘STOP’ Islamisation in America ‘sets a negative tone and signals that Islamization is undesirable and is something that must be brought to an end in America.’ ” Geller, 751 F.3d at 1361. And by finding HEEB and SQUAW VALLEY disparaging, the PTO necessarily did so based on its finding that the marks convey an expressive message over and above their function as source identifiers—namely, an expressive message disparaging Jewish and Native American people. It was these expressive messages that the government found objectionable, and that led the government to refuse to register or to cancel the marks. In doing so, the government made moral judgments based solely and indisputably on the marks’ expressive content. Every single time registration is refused or cancelled pursuant to the disparagement provision, it is based upon a determination by the government that the expressive content of the message is unsuitable because it would be viewed by the referenced group as disparaging them.
“Commercial speech is no exception” to the need for heightened scrutiny of content-based impositions seeking to curtail the communication of particular information or messages. Sorrell, 131 S.Ct. at 2664. Indeed, “[a] consumer’s concern for the free flow of commercial speech often may be far keener than his concern for urgent political dialogue.” Id. (internal quotation marks omitted). Strict scrutiny must apply to a government regulation that is directed at the expressive component of speech. That the speech is used in commerce or has a commercial component should not change the inquiry when the government regulation is entirely directed to the expressive component of the speech. This is not a government regulation aimed at the commercial component of speech. See Va. State Bd. of Pharmacy, 425 U.S. at 765, 96 S.Ct. 1817 (commercial speech involves the “dissemination of information as to who is producing and selling what product, for what reason, and at what price”); see id. at 762, 96 S.Ct. 1817 (defining *1339 “commercial speech” as speech that does “no more than propose a commercial transaction”); Bd. of Trs. of State Univ. of N.Y. v. Fox, 492 U.S. 469, 473–74, 109 S.Ct. 3028, 106 L.Ed.2d 388 (1989); City of Cincinnati v. Discovery Network, Inc., 507 U.S. 410, 423, 113 S.Ct. 1505, 123 L.Ed.2d 99 (1993).
In R.A.V., the Supreme Court explained the key point: under First Amendment law, government measures often affect speech that has a dual character, and when they do, which First Amendment standard is applicable depends on which aspect of the speech is targeted by the measure being reviewed. See 505 U.S. at 385, 112 S.Ct. 2538 (“The proposition that a particular instance of speech can be proscribable on the basis of one feature (e.g., obscenity) but not on the basis of another (e.g., opposition to the city government) is commonplace and has found application in many contexts.”). In particular, commercial speech that is “inextricably intertwined” with expressive speech is treated as expressive speech under the First Amendment when the expressive aspect is being regulated. Riley v. Nat’l Fed’n of the Blind, 487 U.S. 781, 796, 108 S.Ct. 2667, 101 L.Ed.2d 669 (1988). Here, § 2(a) targets speech that is of “public concern,” because it “can be fairly considered as relating to any matter of political, social, or other concern to the community.” Snyder v. Phelps, 562 U.S. 443, 453, 131 S.Ct. 1207, 179 L.Ed.2d 172 (2011) (internal quotation marks omitted). It therefore “occupies the highest rung of the hierarchy of First Amendment values, and is entitled to special protection.” Id. at 452, 131 S.Ct. 1207 (internal quotation marks omitted).
Because § 2(a) discriminates on the basis of the content of the message conveyed by the speech, it follows that it is presumptively invalid, and must satisfy strict scrutiny to be found constitutional. “In the ordinary case it is all but dispositive to conclude that a law is content-based and, in practice, viewpoint-discriminatory.” Sorrell, 131 S.Ct. at 2667. The government here does not even argue that § 2(a) satisfies strict scrutiny.
II. Section 2(a) Is Not Saved From Strict Scrutiny Because It Bans No Speech or By Government–Speech or Government–Subsidy Doctrines
Faced with the daunting prospect of defending a content- and viewpoint-discriminatory regulation of speech, the government argues that § 2(a) does not implicate the First Amendment at all. First, the government suggests that § 2(a) is immune from First Amendment scrutiny because it prohibits no speech, but leaves Mr. Tam free to name his band as he wishes and use this name in commerce. Second, the government suggests that trademark registration is government speech, and thus the government can grant and reject trademark registrations without implicating the First Amendment. Finally, the government argues that § 2(a) merely withholds a government subsidy for Mr. Tam’s speech and is valid as a permissible definition of a government subsidy program. We reject each of the government’s arguments.
A. Strict Scrutiny Applies to § 2(a), Which Significantly Chills Private Speech on Discriminatory Grounds, Though It Does Not Ban Speech
The government argues that § 2(a) does not implicate the First Amendment because it does not prohibit any speech. Appellee’s En Banc Br. 17. The government’s argument is essentially the same as that of our predecessor court in McGinley: “it is clear that the PTO’s refusal to register appellant’s mark does not affect his right to use it. No conduct is *1340 proscribed, and no tangible form of expression is suppressed.” 660 F.2d at 484 (citations omitted). But the First Amendment’s standards, including those broadly invalidating message discrimination, are not limited to such prohibitions. See Pitt News v. Pappert, 379 F.3d 96, 111–12 (3d Cir.2004) (Alito, J.) (“The threat to the First Amendment arises from the imposition of financial burdens that may have the effect of influencing or suppressing speech, and whether those burdens take the form of taxes or some other form is unimportant.”).
The point has been recognized in various doctrinal settings. “For if the government could deny a benefit to a person because of his constitutionally protected speech or associations, his exercise of those freedoms would in effect be penalized and inhibited. This would allow the government to produce a result which it could not command directly.” Perry v. Sindermann, 408 U.S. 593, 597, 92 S.Ct. 2694, 33 L.Ed.2d 570 (1972) (internal quotation marks and alterations omitted). This premise—that denial of a benefit would chill exercise of the constitutional right—undergirds every unconstitutional conditions doctrine case, discussed infra. See, e.g., Speiser v. Randall, 357 U.S. 513, 518, 78 S.Ct. 1332, 2 L.Ed.2d 1460 (1958) (“It is settled that speech can be effectively limited by the exercise of the taxing power. To deny an exemption to claimants who engage in certain forms of speech is in effect to penalize them for such speech.” (citation omitted)); Bd. of Cty. Comm’rs v. Umbehr, 518 U.S. 668, 674, 116 S.Ct. 2342, 135 L.Ed.2d 843 (1996) (loss of a valuable benefit “in retaliation for speech may chill speech on matters of public concern”); Legal Servs. Corp. v. Velazquez, 531 U.S. 533, 545, 121 S.Ct. 1043, 149 L.Ed.2d 63 (2001); Rosenberger, 515 U.S. at 835, 115 S.Ct. 2510 (explaining that “[v]ital First Amendment speech principles are at stake here,” including danger arising “from the chilling of individual thought and expression”).
The general principle is clear: “Lawmakers may no more silence unwanted speech by burdening its utterance than by censoring its content.” Sorrell, 131 S.Ct. at 2664. “[T]he government’s ability to impose content-based burdens on speech raises the specter that the government may effectively drive certain ideas or viewpoints from the marketplace.” Simon & Schuster, 502 U.S. at 116, 112 S.Ct. 501. A law may burden speech even when it does so indirectly. In Sorrell, the challenged statute did not directly ban speech, but rather forbade certain pharmaceutical marketing executives from obtaining and using information that could help them market their products more effectively. 131 S.Ct. at 2659–60. The Court found that the state “ha[d] burdened a form of protected expression,” while leaving “unburdened those speakers whose messages are in accord with its own views.” Id. at 2672.
Here, too, § 2(a) burdens some speakers and benefits others. And while it is true that a trademark owner may use its mark in commerce even without federal registration, it has been widely recognized that federal trademark registration bestows truly significant and financially valuable benefits upon markholders. B & B Hardware, 135 S.Ct. at 1300; Park ‘N Fly, Inc. v. Dollar Park & Fly, Inc., 469 U.S. 189, 199–200, 105 S.Ct. 658, 83 L.Ed.2d 582 (1985) (valuable new rights were created by the Lanham Act); McCarthy at § 19:9, :11 (“Registration of a mark on the federal Principal Register confers a number of procedural and substantive legal advantages over reliance on common law rights.... Registration on the Principal Register should be attempted if it is at all possible.”); McCarthy at § 2:14 (“Businesspeople *1341 regard trademarks as valuable assets and are willing to pay large sums to buy or license a well-known mark.”); Lee Ann W. Lockridge, Abolishing State Trademark Registrations, 29 Cardozo Arts & Ent. L.J. 597, 605 (2011) (“[T]he incentives to pursue federal registration ... are now so significant as to make federal registration indispensable for any owner making an informed decision about its trademark rights. A federal registration is the only rational choice.”); Susan M. Richey, The Second Kind of Sin: Making the Case for a Duty to Disclose Facts Related to Genericism and Functionality in the Trademark Office, 67 Wash. & Lee L.Rev. 137, 174 (2010) (“Federal registration has evolved into a powerful tool for trademark holders....”); Patricia Kimball Fletcher, Joint Registration of Trademarks and the Economic Value of a Trademark System, 36 U. Miami L.Rev. 297, 298–99 (1982) (“Federal registration under the Lanham Act is advantageous, however, because it increases the owner’s legal rights in the mark, making the mark itself more valuable. Thus, trademark owners have significant legal and economic interests in obtaining federal registration of trademarks.”).
Denial of these benefits creates a serious disincentive to adopt a mark which the government may deem offensive or disparaging. Br. of Amici Curiae ACLU 12 (“If a group fears that its chosen name will be denied federal trademark protection by the government’s invocation of Section 2(a), it will be less likely to adopt the name, at least in part because the associative value of the trademark itself is lessened when it is unlikely that a group will be the exclusive holder of that mark.”); Br. of Amicus Curiae Pro–Football, Inc. 15 (“Section 2(a) certainly works to chill speech.... Through it, the Government uses threatened denial of registration to encourage potential registrants not to use ‘disparaging’ names. Faced with the possibility of being denied a registration—or worse, cancellation after years of investment-backed brand development—new brand owners are more likely to avoid brand names that may be arguably controversial for fear of later being deemed ‘disparaging.’ ”); Br. of Amicus Curiae First Amendment Lawyers Ass’n 7 (“Individuals and businesses refrain from using certain terms as trademarks for fear the PTO might see the terms as immoral, scandalous, or derogatory, in violation of section 2(a). Such self-censorship narrows the spectrum of speech in the public marketplace.”); Br. of Amici Curiae Rutherford Inst. 12 (“Denial of registration indisputably has the effect of placing applicants at a legal and financial disadvantage.”); Jeffrey Lefstin, Does the First Amendment Bar Cancellation of Redskins?, 52 Stan. L.Rev. 665, 678 (2000) (“[I]t is clear that section 2(a) of the Lanham Act, by denying the valuable registration right to scandalous or disparaging trademarks, imposes a financial disincentive to the use of such marks in commercial communication.”); Michelle B. Lee, Section 2(a) of the Lanham Act as a Restriction on Sports Team Names: Has Political Correctness Gone Too Far?, 4 Sports L.J. 65, 69 (1997) (“Use [of disparaging marks] is discouraged by cancellation of registration by a loss of the benefits that go along with it. These benefits go well beyond those granted by the common law, and a loss of them will remove advantages which make the property more valuable.”).
For those reasons, the § 2(a) bar on registration creates a strong disincentive to choose a “disparaging” mark. And that disincentive is not cabined to a clearly understandable range of expressions. The statute extends the uncertainty to marks that “may disparage.” 15 U.S.C. § 1052(a). The uncertainty as to what might be deemed disparaging is not only evident on its face, given the subjective-reaction element and shifting usages in *1342 different parts of society.6 It is confirmed by the record of PTO grants and denials over the years, from which the public would have a hard time drawing much reliable guidance.7
Such uncertainty of speech-affecting standards has long been recognized as a First Amendment problem, e.g., in the overbreadth doctrine. See Broadrick v. Oklahoma, 413 U.S. 601, 613, 615, 93 S.Ct. 2908, 37 L.Ed.2d 830 (1973). It has also been recognized as a problem under Fifth Amendment vagueness standards as they have been specially applied in the First Amendment setting.8 All we need say about the uncertainty here, however, is that it contributes significantly to the chilling effect on speech.
The disincentive to choose a particular mark extends to any mark that could require *1343 the expenditure of substantial resources in litigating to obtain registration in the first place. And the disincentive does not stop there, because the disparagement determination is not a onetime matter. Even if an applicant obtains a registration initially, the mark may be challenged in a cancellation proceeding years later. Thus, after years of investment in promoting a registered mark and coming to be known by it, a mark’s owner may have to (re)litigate its character under § 2(a) and might lose the registration. This effectively forces the mark’s owner to find a new mark and make substantial new investments in educating the public that the products known by the old mark are now known by the new mark and, more generally, in establishing recognition of the new mark. The “disparagement” standard steers applicants away from choosing a mark that might result in these problems any time in the future.
Not surprisingly, “those who are denied registration under Section 2(a) often abandon the denied application and file a new one, indicating that they have changed their name rather than bear the costs of using a ‘disparaging’ mark or challenge the PTO’s determination.” Br. of Amicus Curiae Pro–Football, Inc. 15. In many cases, as soon as a trademark examiner issues a rejection based upon disparagement, the applicant immediately abandons the trademark application. See, e.g., AMISHHOMO (abandoned 2013); MORMON WHISKEY (abandoned 2012); HAVE YOU HEARD THAT SATAN IS A REPUBLICAN? (abandoned 2010); DEMOCRATS SHOULDN’T BREED (abandoned 2008); REPUBLICANS SHOULDN’T BREED (abandoned 2008); 2 DYKE MINIMUM (abandoned 2007); WET BAC/WET B.A.C. (abandoned 2007); DON’T BE A WET BACK (abandoned 2006); FAGDOG (abandoned 2003).
The importance of the benefits of federal trademark registration explains the strength of the incentive to avoid marks that are vulnerable under § 2(a). For example, the holder of a federally registered trademark has a right to exclusive nationwide use of that mark anywhere there is not already a prior use that proceeds registration. See 15 U.S.C. §§ 1072, 1115. In the absence of federal registration, if a trademark owner has any common law rights, they are “limited to the territory in which the mark is known and recognized by those in the defined group of potential customers.” McCarthy at § 26:2. Without the recognition of nationwide constructive use conferred by federal registration, a competitor can swoop in and adopt the same mark for the same goods in a different location. Without federal registration, the applicant does not have prima facie evidence of the mark’s validity or its ownership or exclusive use of the mark. 15 U.S.C. § 1057(b). And a common law trademark can never become incontestable. Id. § 1065. Without federal registration, a trademark user cannot stop importation of goods bearing the mark, or recover treble damages for willful infringement. Id. §§ 1117, 1124. The common law provides no rights like these.
Contrary to the suggestion by the government, Mr. Tam is likely also barred from registering his mark in nearly every state. Three years after the enactment of the Lanham Act, the United States Trademark Association prepared the Model State Trademark Act—a bill patterned on the Lanham Act in many respects. McCarthy at § 22:5. The Model Act contains language barring a mark from registration if it “consists of or comprises matter which may disparage ... persons, living or dead, institutions, beliefs, or national symbols, or bring them *1344 into contempt, or disrepute.” 1964 Model State Trademark Act, § 2. Following the lead of the federal government, virtually all states have adopted the Model Act and its disparagement provision. McCarthy at § 22:5. Thus, not only are the benefits of federal registration unavailable to Mr. Tam, so too are the benefits of trademark registration in nearly all states.9
The government argues that the denial of Mr. Tam’s registration “does not eliminate any common-law rights that might exist in [his] mark.” Appellee’s En Banc Br. 20. But as the government’s use of “might” indicates, it is unclear whether Mr. Tam could actually enforce any common law rights to a disparaging mark.10 The 1964 Model State Trademark Act, which most states have adopted, provides that “[n]othing herein shall adversely affect the rights or the enforcement of rights in marks acquired in good faith at any time at common law.” § 14. However, the term “mark” is defined as “any trademark or service mark entitled to registration under this Act whether registered or not.” § 1.C (emphasis added). Common law rights to a mark may thus be limited to marks “entitled to registration.” Whether a user of an unregistrable, disparaging mark has any enforceable common law rights is at best unclear. See Justin G. Blankenship, The Cancellation of Redskins as a Disparaging Trademark: Is Federal Trademark Law an Appropriate Solution for Words That Offend?, 72 U. Colo. L.Rev. 415, 451 (2001) (“[A]ny mark that is canceled under section 2(a) of the Lanham Act for being scandalous or disparaging is unlikely to find much protection under common law principles either, although this will ultimately be determined by state courts applying their own common law principles.”); Llewellyn Joseph Gibbons, Semiotics of the Scandalous and the Immoral and the Disparaging: Section 2(A) Trademark Law After Lawrence v. Texas, 9 Marq. Intell. Prop. L.Rev. 187, 232 (2005) (“[A]s immoral, scandalous, and/or disparaging marks may not be registered under either state or federal law, nor do they enjoy common law protection, there appears to be no way of establishing a legally recognized property right in these marks.”); Stephen Baird, Moral Intervention in the Trademark Arena: Banning the Registration of Scandalous and Immoral Trademarks, 83 TRADEMARK REPORTER 661, 795 (1993) (disparaging marks are presumably “unprotect[a]ble pursuant to state common law”). The Restatement (Third) of Unfair Competition notes that the Lanham Act and the Model State Trademark Bill both prohibit registration of disparaging marks and that adoption and use of such marks may preclude enforcement under the common law doctrine of unclean hands. Restatement (Third) of Unfair Competition § 32 cmt. c(1995). The government has not pointed to a single case where the common-law holder of a disparaging mark was able to enforce that mark, nor could we find one. The government’s suggestion that Mr. Tam has common-law rights to his mark appears illusionary.11
*1345 Whether Mr. Tam has enforceable common-law rights to his mark or could bring suit under § 43(a) does not change our conclusion. Federal trademark registration brings with it valuable substantive and procedural rights unavailable in the absence of registration. These benefits are denied to anyone whose trademark expresses a message that the government finds disparages any group, Mr. Tam included. The loss of these rights, standing alone, is enough for us to conclude that § 2(a) has a chilling effect on speech. Denial of federal trademark registration on the basis of the government’s disapproval of the message conveyed by certain trademarks violates the guarantees of the First Amendment.
B. Trademark Registration Is Not Government Speech
The government suggests, and several amici argue, that trademark registration is government speech, and as such outside the coverage of the First Amendment. See Appellee’s En Banc Br. 41–42; Br. of Amici Curiae Nat’l Asian Pacific Am. Bar Ass’n 19–22; Br. of Amici Curiae Blackhorse 13–23. “The Free Speech Clause restricts government regulation of private speech; it does not regulate government speech.” Pleasant Grove City v. Summum, 555 U.S. 460, 467, 129 S.Ct. 1125, 172 L.Ed.2d 853 (2009). Although we find it difficult to understand the government’s precise position as to how trademark registration constitutes government speech, we conclude that there is no government speech at issue in the rejection of disparaging trademark registrations that would insulate § 2(a) from First Amendment review.
Wisely, the government does not argue that a mark-holder’s use or enforcement of its federally registered trademark is government speech. Use of a mark by its owner is clearly private speech. Trademarks identify the source of a product, and are often closely associated with the actual product. A mark’s purpose—to identify the source of goods—is antithetical to the notion that a trademark is tied to the government. The fact that COCA COLA and PEPSI may be registered trademarks does not mean the government has endorsed these brands of cola, or prefers them over other brands. We see no reason that a markholder’s use of its mark constitutes government speech.
Instead, the government appears to argue that trademark registration and the accoutrements of registration—such as the registrant’s right to attach the ® symbol to the registered mark, the mark’s placement on the Principal Register, and the issuance of a certificate of registration—amount to government speech. See Oral Argument at 52:40–53:07; 54:20–54:32. This argument is meritless. *1346 Trademark registration is a regulatory activity. These manifestations of government registration do not convert the underlying speech to government speech. And if they do, then copyright registration would likewise amount to government speech. Copyright registration has identical accoutrements—the registrant can attach the © symbol to its work, registered copyrights are listed in a government database, and the copyright owner receives a certificate of registration. The logical extension of the government’s argument is that these indicia of registration convert the underlying speech into government speech unprotected by the First Amendment. Thus, the government would be free, under this logic, to prohibit the copyright registration of any work deemed immoral, scandalous, or disparaging to others. This sort of censorship is not consistent with the First Amendment or government speech jurisprudence.
In Walker v. Texas Division, Sons of Confederate Veterans, Inc., the Supreme Court detailed the indicia of government speech. ––– U.S. ––––, 135 S.Ct. 2239, 192 L.Ed.2d 274 (2015). The Court concluded that specialty license plates were government speech, even though a state law allowed individuals, organizations, and nonprofit groups to request certain designs. The Court found several considerations weighing in favor of this holding. It emphasized that “the history of license plates shows that, insofar as license plates have conveyed more than state names and vehicle identification numbers, they long have communicated messages from the States.” Id. at 2248. It stressed that “[t]he State places the name ‘TEXAS’ in large letters at the top of every plate,” that “the State requires Texas vehicle owners to display license plates, and every Texas license plate is issued by the State,” that “Texas also owns the designs on its license plates,” and that “Texas dictates the manner in which drivers may dispose of unused plates.” Id. As a consequence, the Court reasoned, “Texas license plate designs ‘are often closely identified in the public mind with the State.’ ” Id. (quoting Summum, 555 U.S. at 472, 129 S.Ct. 1125 (alteration omitted)). Amidst all of its other aspects of control, moreover, “Texas maintains direct control over the messages conveyed on its specialty plates.” Id. at 2249. “Indeed, a person who displays a message on a Texas license plate likely intends to convey to the public that the State has endorsed that message.” Id.
The government’s argument in this case that trademark registration amounts to government speech is at odds with the Supreme Court’s analysis in Walker and unmoored from the very concept of government speech. When the government registers a trademark, the only message it conveys is that a mark is registered. The vast array of private trademarks are not created by the government, owned or monopolized by the government, sized and formatted by the government, immediately understood as performing any government function (like unique, visible vehicle identification), aligned with the government, or (putting aside any specific government-secured trademarks) used as a platform for government speech. There is simply no meaningful basis for finding that consumers associate registered private trademarks with the government.
Indeed, the PTO routinely registers marks that no one can say the government endorses. See, e.g., RADICALLY FOLLOWING CHRIST IN MISSION TOGETHER, U.S. Reg. No. 4,759,522; THINK ISLAM, U.S. Reg. No. 4,719,002 (religious marks); GANJA UNIVERSITY, U.S. Reg. No. 4,070,160 (drug-related); CAPITALISM SUCKS DONKEY BALLS, U.S. Reg. No. 4,744,351; TAKE YO PANTIES OFF, U.S. Reg. No. 4,824,028; and MURDER 4 HIRE, U.S. *1347 Reg. No. 3,605,862. As the government itself explains, “the USPTO does not endorse any particular product, service, mark, or registrant” when it registers a mark. Appellee’s En Banc Br. 44. For decades, the government has maintained that:
[J]ust as the issuance of a trademark registration by this Office does not amount to government endorsement of the quality of the goods to which the mark is applied, the act of registration is not a government imprimatur or pronouncement that the mark is a “good” one in an aesthetic, or any analogous, sense.
In re Old Glory Condom Corp., 26 U.S.P.Q.2d 1216, 1219–20 n. 3 (T.T.A.B. Mar. 3, 1993); see also McCarthy at § 19:3.50 (“[G]overnment registration of a mark is neither a government endorsement of the quality of the goods to which the mark is applied nor a government pronouncement that the mark is a good or reliable one in any moral or commercial sense.”); Jeffrey Lefstin, Does the First Amendment Bar Cancellation of Redskins?, 52 Stan. L.Rev. 665, 684 (2000) (“The overwhelming majority of the public encounters trademarks in their roles as product identifiers, not as the beneficiaries of a federal registration scheme. The public is unlikely to believe that a registered trademark designation accompanying a word or logo on a product reflects government endorsement.”). Trademarks are understood in society to identify the source of the goods sold, and to the extent that they convey an expressive message, that message is associated with the private party that supplies the goods or services. Trademarks are not understood to convey a government message or carry a government endorsement.
The government argues that use of the ® symbol, being listed in a database of registered marks, and having been issued a registration certificate makes trademark registration government speech. These incidents of registration do not convert private speech into government speech. The government does not own the trademark designs or the underlying goods to which the trademark is affixed as the state owned the license plates in Walker. Markholders are not even required to use the ® symbol on their goods. 15 U.S.C. § 1111. And if simply affixing the ® symbol converted private speech into government speech then the government would be free to regulate the content, viewpoint, and messages of registered copyrights. A copyright registration likewise allows the copyright owner to affix a © symbol, 17 U.S.C. § 401, but this symbol does not convert the copyrighted work into government speech or permit the government to grant some copyrights and deny others on account of the work’s message. Just as the public does not associate the copyrighted works Nigger: The Strange Career of a Troublesome Word or Fifty Shades of Grey with the government, neither does the public associate individual trademarks such as THE SLANTS with the government.
Similarly, a registered mark’s placement on the Principal Register or publication in the PTO’s Official Gazette does not morph the private expression being registered into government expression. As a preliminary matter, it is not entirely clear what the Principal Register is. There is apparently no government-published book of all trademark registrations; instead, the Principal Register is at most an internet database hosted on the PTO’s website. See U.S. Patent and Trademark Office, Search Trademark Database, available at http://www.uspto.gov/trademarks-application-process/search-trademark-database. If being listed in a government database or published in a list of registrations were enough to convert private speech to government speech, nearly every action the government takes—every parade permit *1348 granted, every property title recorded, every hunting or fishing license issued—would amount to government speech. The government could record recipients of parade permits in an official database or publish them weekly, thus insulating content-based grants of these permits from judicial review. Governmental agencies could assign TV and radio licenses and states could refuse to license medical doctors with no First Amendment oversight by “registering” these licenses in an online database, or by allowing licensees to display a mark by their name. The fact that the government records a trademark in a database of all registered trademarks cannot possibly be the basis for concluding that government speech is involved.
Finally, the issuance of a registration certificate signed by the Director with the seal of the United States Patent and Trademark Office does not convert private expression or registration into government speech. This is a certificate, a piece of paper, which the trademark owner is free to do with as it wishes. The government maintains no control over the certificates. The government does not require companies to display their trademark registration certificate, or dictate the manner in which markholders may dispose of unused registration certificates. It is not public like license plates or monuments. When copyrights are granted, the copyright owner receives a similar registration certificate with the seal and signed by the Registrar of Copyrights. 17 U.S.C. § 410(a). And patents issue “in the name of the United States of America, under the seal of the Patent and Trademark Office,” with a gold seal and red ribbon attached. 35 U.S.C. § 153; see also U.S. Patent and Trademark Office, Patent Process Overview, available at http://www.uspto.gov/patents-getting-started/patentprocess-overview# step7 (explaining that patent grants are issued with “a gold seal and red ribbon on the cover”). These certificates do not convert the registered subject matter into government speech such that the government is free to regulate its content. The public simply does not view these registration certificates as the government’s expression of its ideas or as the government’s endorsement of the ideas, inventions, or trademarks of the private speakers to whom they are issued.
In short, the act of registration, which includes the right (but not the obligation) to put an ® symbol on one’s goods, receiving a registration certificate, and being listed in a government database, simply cannot amount to government speech. The PTO’s processing of trademark registrations no more transforms private speech into government speech than when the government issues permits for street parades, copyright registration certificates, or, for that matter, grants medical, hunting, fishing, or drivers licenses, or records property titles, birth certificates, or articles of incorporation. To conclude otherwise would transform every act of government registration into one of government speech and thus allow rampant viewpoint discrimination. When the government registers a trademark, it regulates private speech. It does not speak for itself.
C. Section 2(a) Is Not a Government Subsidy Exempt from Strict Scrutiny
We reject the government’s argument that § 2(a)’s message-based discrimination is merely the government’s shaping of a subsidy program. The government’s defense is contrary to the long-established unconstitutional conditions doctrine. The Supreme Court has repeatedly invalidated denials of “benefits” based on message-based disapproval of private speech that is not part of a government-speech program. In such circumstances, denial of an otherwise-available benefit is *1349 unconstitutional at least where, as here, it has a significant chilling effect on private speech. Bd. of Cty. Comm’rs, 518 U.S. at 674, 116 S.Ct. 2342 (1996) (explaining that “the threat of the loss of [a valuable financial benefit] in retaliation for speech may chill speech on matters of public concern”); id. (“[r]ecognizing that constitutional violations may arise from the deterrent, or ‘chilling,’ effect of governmental efforts that fall short of a direct prohibition against the exercise of First Amendment rights”) (citations and alterations omitted).
Under the unconstitutional conditions doctrine:
[E]ven though a person has no ‘right’ to a valuable governmental benefit and even though the government may deny him the benefit for any number of reasons, there are some reasons upon which the government may not rely. It may not deny a benefit to a person on a basis that infringes his constitutionally protected interests—especially, his interest in freedom of speech.
Perry, 408 U.S. at 597, 92 S.Ct. 2694. The Supreme Court, applying this doctrine, held that a state college could not refuse to retain a professor because of his public criticism of that college’s policy, even though the professor had no right to reemployment, and even though the government had not directly prohibited the professor from speaking. Id. at 597–98, 92 S.Ct. 2694. This is because “[t]o deny [a benefit] to claimants who engage in certain forms of speech is in effect to penalize them for such speech.” Speiser v. Randall, 357 U.S. 513, 518, 78 S.Ct. 1332, 2 L.Ed.2d 1460 (1958); Perry, 408 U.S. at 597, 92 S.Ct. 2694 (“For if the government could deny a benefit to a person because of his constitutionally protected speech or associations, his exercise of those freedoms would in effect be penalized and inhibited.”).
Since Perry, the Supreme Court has wrestled with how to apply the unconstitutional conditions doctrine while protecting Congress’s ability to direct government spending. The Spending Clause of the U.S. Constitution, which grants Congress the power “[t]o lay and collect Taxes, Duties, Imposts and Excises, to pay the Debts and provide for the common Defence and general Welfare of the United States,” U.S. Const. Art. I, § 8, cl. 1, “provides Congress broad discretion to tax and spend for the ‘general Welfare,’ including by funding particular state or private programs or activities.” Agency for Int’l Dev. v. Alliance for Open Soc’y Int’l, Inc., ––– U.S. ––––, 133 S.Ct. 2321, 2327–28, 186 L.Ed.2d 398 (2013). This includes “the authority to impose limits on the use of such funds to ensure they are used in the manner Congress intends,” even when these limits exclude protected speech or other constitutionally protected conduct. Id. at 2328 (citing Rust v. Sullivan, 500 U.S. 173, 195 n. 4, 111 S.Ct. 1759, 114 L.Ed.2d 233 (1991)). The Court reasoned that “if a party objects to a condition on the receipt of federal funding,” it can always decline the funds. Id.
“[W]hen the Government appropriates public funds to establish a program it is entitled to define the limits of that program.” United States v. Am. Library Ass’n, 539 U.S. 194, 211, 123 S.Ct. 2297, 156 L.Ed.2d 221 (2003) (quoting Rust, 500 U.S. at 194, 111 S.Ct. 1759). For purposes of a message-discriminatory condition on the grant of government funds, the Supreme Court has said that the government can “disburse[ ] public funds to private entities to convey a governmental message.” Legal Servs. Corp. v. Velazquez, 531 U.S. 533, 541, 121 S.Ct. 1043, 149 L.Ed.2d 63 (2001) (citation omitted). When it does so, “it may take legitimate and appropriate steps to ensure *1350 that its message is neither garbled nor distorted by the grantee.” Id. Therefore, “viewpoint-based funding decisions can be sustained in instances ... in which the government used private speakers to transmit specific information pertaining to its own program.” Id. (citations omitted).
Thus, in Rust, the government could prohibit the expenditure of public federal family planning funds on abortion-related counseling because the government distributed those funds to promote the conveying of a particular message. Rosenberger, 515 U.S. at 833, 115 S.Ct. 2510 (citing Rust, 500 U.S. at 194, 111 S.Ct. 1759); Velazquez, 531 U.S. at 541, 121 S.Ct. 1043 (noting that Rust must be understood as resting on the conclusion that it involved “government speech”). Relatedly, although there was no majority opinion in American Library Ass’n, the Court upheld a specific congressional determination not to give money for technology to be used for supporting particular speech (pornography) in particular circumstances (in public libraries where non-user patrons likely would inadvertently see it), even then only upon confirming the minor nature of the burden on the user patrons involved. 539 U.S. at 211–12, 123 S.Ct. 2297 (upholding conditioning public libraries’ receipt of federal subsidies on their use of Internet filtering software, because Congress was entitled to insist that “public funds be spent for the purposes for which they were authorized” (quotation marks omitted)). Earlier, the Court had recognized that tax exemptions or deductions were a form of subsidy for First Amendment analysis. Regan v. Taxation with Representation of Wash., 461 U.S. 540, 544, 103 S.Ct. 1997, 76 L.Ed.2d 129 (1983) (“Both tax exemptions and tax-deductibility are a form of subsidy that is administered through the tax system.”); id. (explaining that tax-exempt status “has much the same effect as a cash grant to an organization”).
The government’s discretion to direct its spending, while broad, is not unbounded, and the limits take account of the real-world effect on the speech of those subject to the conditions. If a program arises under the Spending Clause, Congress is free to attach “conditions that define the limits of the government spending program—those that specify the activities Congress wants to subsidize.” Agency for Int’l Dev., 133 S.Ct. at 2328. However, Congress does not have the authority to attach “conditions that seek to leverage funding to regulate speech outside the contours of the program itself.” Id. “Congress cannot recast a condition on funding as a mere definition of its program in every case, lest the First Amendment be reduced to a simple semantic exercise.” Velazquez, 531 U.S. at 547, 121 S.Ct. 1043. The Court held that Congress could not restrict appropriations aimed at combating the spread of HIV/AIDS to only organizations having policies affirmatively opposing prostitution and sex trafficking, which would make such organizations unable to convey a contrary message. Agency for Int’l Dev., 133 S.Ct. at 2330–31. The Court struck down Congress’s conditioning of funding to public broadcasters on their refraining from editorializing, even with their non-federal money. FCC v. League of Women Voters, 468 U.S. 364, 104 S.Ct. 3106, 82 L.Ed.2d 278 (1984). And in Regan, the Court, in upholding the subsidy of certain organizations for lobbying, took pains to note the relatively easy work-around for other unsubsidized organizations to achieve a comparable position for lobbying and the absence of any attempt to suppress ideas. 461 U.S. at 548, 103 S.Ct. 1997; see Leathers v. Medlock, 499 U.S. 439, 452, 111 S.Ct. 1438, 113 L.Ed.2d 494 (1991) (discussing Regan).
The government argues that trademark registration is a form of government *1351 subsidy that the government may refuse where it disapproves of the message a mark conveys. It contends: “Congress has at least as much discretion to determine which terms and symbols should be registered and published by a federal agency as it would to determine which private entities should receive federal funds.” Appellee’s En Banc Br. 29. But as already described, trademark registration is not a program through which the government is seeking to get its message out through recipients of funding (direct or indirect). And for the reasons described above, the denial of registration has a major chilling effect on private speech, because the benefits of registration are so substantial. Nor is there a ready work-around to maintain private speech without significant disadvantage. Markholders cannot, for example, realistically have two brand names, one inoffensive, non-disparaging one (which would be able to secure registration) and a second, expressive, disparaging one (which would be unregistrable and unprotectable).
In any event, the scope of the subsidy cases has never been extended to a “benefit” like recognition of legal rights in speakers against private interference. The cases cannot be extended to any “program” conferring legal rights on the theory that the government is free to distribute the legal rights it creates without respecting First Amendment limits on content and viewpoint discrimination. Not surprisingly, the subsidy cases have all involved government funding or government property.
The government cites Ysursa v. Pocatello Education Ass’n, 555 U.S. 353, 129 S.Ct. 1093, 172 L.Ed.2d 770 (2009), and Davenport v. Washington Education Ass’n, 551 U.S. 177, 127 S.Ct. 2372, 168 L.Ed.2d 71 (2007), in support of its subsidy defense of § 2(a). Appellee’s En Banc Br. 28–29. But they are inapposite. Both Davenport and Ysursa center on challenges to the constitutionality of state laws limiting the ability of public-sector unions to spend on political speech non-members’ money the unions obtain through the government’s affirmative use of its own payroll systems. Davenport, 551 U.S. at 180, 127 S.Ct. 2372 (considering constitutionality of law prohibiting payroll deductions for political spending unless the union had the affirmative consent of the non-member); Ysursa, 555 U.S. at 355, 129 S.Ct. 1093 (considering constitutionality of law completely prohibiting payroll deductions for political spending). Even in the context of use of government property, the Court focused on the absence of viewpoint discrimination, holding that the programs placed a “reasonable, viewpoint-neutral limitation” on the unions’ abilities to enlist the government’s aid in acquiring the money of government employees for spending on political speech to which particular employees might object. Davenport, 551 U.S. at 189, 127 S.Ct. 2372; see also Ysursa, 555 U.S. at 361 n. 3, 129 S.Ct. 1093. The prohibitions were not “aimed at the suppression of dangerous ideas.” Ysursa, 555 U.S. at 359, 129 S.Ct. 1093 (alterations omitted); see also Davenport, 551 U.S. at 190, 127 S.Ct. 2372 (“Quite obviously, no suppression of ideas is afoot.”).
These cases do not speak to Congress’s power to enact viewpoint-discriminatory regulations like § 2(a). The government does not shy away from the fact that the purpose of § 2(a) is to discourage, and thereby eliminate, disparaging marks, particularly marks that include “the most vile racial epithets,” “religious insults,” “ethnic caricatures,” and “misogynistic images.” Appellee’s En Banc Br. 1–3. On its face, § 2(a) is aimed at the suppression of dangerous ideas, unlike the provisions in Ysursa and Davenport. Moreover Ysursa and Davenport both took place in “the unique context of public-sector agency- *1352 shop arrangements,” where the government was “act[ing] in a capacity other than as regulator.” Davenport, 551 U.S. at 188, 190, 127 S.Ct. 2372. Thus, the risk that the government “may effectively drive certain ideas or viewpoints from the marketplace [was] attenuated.” Id. at 188, 127 S.Ct. 2372. Section 2(a) is regulation of speech that targets expressive content and thereby threatens to drive ideas or viewpoints from the marketplace.
In determining if a condition on a favorable government action is unconstitutional, courts—both before and after Davenport and Ysursa—have distinguished between government actions that implicate the government’s power to spend and government actions that do not. For example, the Ninth Circuit considered the constitutionality of a treaty under which certain “educational, scientific and cultural audio-visual materials” were granted various benefits, including exemption from import duties. Bullfrog Films, Inc. v. Wick, 847 F.2d 502, 503 (9th Cir.1988). The government argued, as it does here, that the regulations stemming from the treaty did not “punish or directly obstruct [filmmakers’] ability to produce or disseminate their films,” but amount to “the government simply declining to pay a subsidy.” Id. at 509. The Ninth Circuit rejected the government’s “benign characterization” of the regulations and held that the trade benefits were not a subsidy because “no Treasury Department funds [were] involved.” Id. at 509. Because the trade benefits were not a subsidy, the Ninth Circuit held that the unconstitutional conditions doctrine applied, and found the treaty and implementing regulations unconstitutional. Id. at 511.
The Fifth Circuit, sitting en banc, recently considered the constitutionality of a Texas law allowing charitable organizations to hold bingo games so long as the resulting funds were not used for lobbying. Dep’t of Tex., Veterans of Foreign Wars v. Tex. Lottery Comm’n, 760 F.3d 427, 430 (5th Cir.2014) (en banc). The Texas Lottery Commission argued that the restrictions were constitutional because they fell within the state government’s spending power, which is analogous to the federal government’s spending power. Id. at 434. The Fifth Circuit agreed that “the government may attach certain speech restrictions to funds linked to the public treasury—when either granting cash subsidies directly from the public coffers ... or approving the withholding of funds that otherwise would go to the public treasury.” Id. at 435. But it found the Texas bingo program “wholly distinguishable ... because no public monies or ‘spending’ by the state are involved.” Id. at 436. Reasoning that the bingo program’s primary function is regulatory, further “underscor [ing] the incongruity of [applying] the ‘subsidy’ paradigm to the bingo program,” the Fifth Circuit applied the unconstitutional conditions doctrine and found the lobbying provision unconstitutional. Id. at 437–41.
Similarly, the D.C. Circuit recently held that a presidential directive barring lobbyists from serving on international trade advisory committees implicated the First Amendment. Autor v. Pritzker, 740 F.3d 176, 177 (D.C.Cir.2014). The government argued that “when [it] appropriates public funds to establish a program, its decision not to use program funds to subsidize the exercise of a fundamental right does not infringe” the First Amendment. Id. at 182 (quotations and alterations omitted). The D.C. Circuit rejected this argument because membership in the advisory committees was a non-financial—albeit valuable—benefit. Id. at 182–83. Explaining that “[t]he Supreme Court has never extended the [spending exception] to situations not involving financial benefits,” the D.C. Circuit found the directive could be an unconstitutional *1353 condition, and remanded for further consideration. Id. at 183–84.
Trademark registration does not implicate Congress’s power to spend or to control use of government property.12 Trademark registration is not a subsidy. The benefits of trademark registration, while valuable, are not monetary. Unlike a subsidy consisting of, for example, HIV/AIDS funding, or tax exemptions, a trademark registration does not directly affect the public fisc. Instead, a registered trademark redefines the nature of the markholder’s rights as against the rights of other citizens, depriving others of their ability to use the mark. Like the programs in Bullfrog and Texas Lottery Commission, the system of trademark registration is a regulatory regime, not a government subsidy program.
The government also argues that because the PTO is funded by appropriations, any government spending requirement is met here. Appellee’s En Banc Br. 29–30 (citing 35 U.S.C. § 42(c)(1)-(2)). Trademark registration fees are collected and, “[t]o the extent and in the amounts provided in advance in appropriations Acts,” made available “to carry out the activities of the [PTO].” 35 U.S.C. § 42(c)(1). However, since 1991 these appropriations have been funded entirely by registration fees, not the taxpayer. Figueroa v. United States, 466 F.3d 1023, 1028 (Fed.Cir.2006); see also 56 Fed.Reg. 65147 (1991); Omnibus Budget Reconciliation Act of 1990, Pub.L. No. 101–508, S. 10101, 1990 U.S.C.C.A.N. (104 Stat.) 1388. The fact that registration fees cover all of the operating expenses associated with registering marks is further evidence that, despite conveying valuable benefits, trademark registration is not a government subsidy.
While PTO operations are fully underwritten by registration fees, some federal funds are nonetheless spent on the registration and enforcement of trademarks. For example, PTO employee benefits, including pensions, health insurance, and life insurance, are administered by the Office of Personnel Management and funded from the general treasury. Figueroa, 466 F.3d at 1028. And registering a trademark may lead to additional government spending, such as when the trademark owner seeks to enforce the trademark through the federal courts and U.S. Customs and Border Patrol. This spending, however, is attenuated from the benefits bestowed by registration. Trademark registration does not implicate the Spending Clause merely because of this attenuated spending, else every benefit or regulatory program provided by the government would implicate the Spending Clause. The Copyright Office is only partially funded by user fees, but copyright registration is nonetheless not a subsidy. Copyright Office Fees: Registration, Recordation and Related Services; Special Services; Licensing Division Services; FOIA Services, 79 Fed.Reg. 15910–01 (Mar. 24, 2014) (setting fees to recover “a significant part of the costs to the Office of registering copyright claims”). It would be unreasonable to argue that the government subsidizes an author when it grants him a copyright. Similarly, the programs in Bullfrog and Texas Lottery Commission were likely funded in some part by the government—perhaps also by government benefits paid to employees administering the programs—but the Ninth Circuit and the Fifth Circuit considered only whether the conditioned benefits were paid for by government spending, and not whether the programs were subsidized in more indirect *1354 ways. And while the government argued in Autor that the government had appropriated public funds to establish the international trade advisory committees, 740 F.3d at 182, the D.C. Circuit nonetheless found that membership on these advisory committees was not a financial benefit, id. at 183.
The fact that the Lanham Act derives from the Commerce Clause, not the Spending Clause, is further evidence that trademark registration is not a subsidy. The purpose of the Lanham Act is to regulate marks used in interstate commerce, prevent customer confusion, and protect the goodwill of markholders, 15 U.S.C. § 1127, not to subsidize markholders. Moreover, the government funding cases have thus far been limited to situations where the government has chosen to limit funding to individuals that are advancing the goals underlying the program the government seeks to fund. See generally Agency for Int’l Dev., 133 S.Ct. at 2324–25; Rust, 500 U.S. at 191, 111 S.Ct. 1759; cf. American Library Ass’n, 539 U.S. at 211, 123 S.Ct. 2297 (it is not unconstitutional for the government to insist that “public funds be spent for the purposes for which they were authorized”). The restriction on the registration of disparaging marks bears no relation to the objectives, goals, or purpose of the federal trademark registration program. Preventing disparaging marks does not protect trademark owners’ investments; in fact, because § 2(a) can be brought in cancellation proceedings decades after a mark is granted, this provision actually undermines this important purpose of the Lanham Act. And the disparagement proscription has never been alleged to prevent consumer confusion or deception. The government’s viewpoint—and content-based discrimination in this case is completely untethered to the purposes of the federal trademark registration program. It would be a radical extension of existing precedent to permit the government to rely upon its power to subsidize to justify its viewpoint discrimination, when that discrimination has nothing to do with the goals of the program in which it is occurring.
Were we to accept the government’s argument that trademark registration is a government subsidy and that therefore the government is free to restrict speech within the confines of the trademark program, it would expand the “subsidy” exception to swallow nearly all government regulation. In many ways, trademark registration resembles copyright registration. Under the logic of the government’s approach, it follows that the government could refuse to register copyrights without the oversight of the First Amendment. Congress could pass a law prohibiting the copyrighting of works containing “racial slurs,” “religious insults,” “ethnic caricatures,” and “misogynistic images.” Appellee’s En Banc Br. 2–3. It is difficult to imagine how trademark registration with its attendant benefits could be deemed a government subsidy but copyright registration with its attendant benefits would not amount to a government subsidy. And if both must be treated as government subsidies by virtue of their conference of benefits or advantages, though not public money, then the government has the right to make content—or viewpoint-based determinations over which works to grant registration. This idea—that the government can control speech by denying the benefits of copyright registration to disfavored speech—is anathema to the First Amendment. With this, the government agrees, arguing that copyright registration, unlike trademark registration, is protected by the First Amendment. Oral Arg. at 36:45–38:50. But the government has advanced no principled reason to treat trademark registration differently than copyright registration for present purposes. The government admits that *1355 any message-based regulation of copyrights would be subject to the First Amendment. We agree, and extend the government’s reasoning to § 2(a)’s message-based regulation of trademarks. These registration programs are prototypical examples of regulatory regimes. The government may not place unconstitutional conditions on trademark registration. We reject the government’s argument that it is free to restrict constitutional rights within the confines of its trademark registration program.
III. Section 2(a) Is Unconstitutional Even Under the Central Hudson Test for Commercial Speech
As discussed above, § 2(a) regulates expressive speech, not commercial speech, and therefore strict scrutiny is appropriate. Trademarks have at times been referred to as commercial speech. See, e.g., Friedman v. Rogers, 440 U.S. 1, 11, 99 S.Ct. 887, 59 L.Ed.2d 100 (1979) (holding that the trade name of an optometrist was commercial speech). They are, after all, commercial identifiers, the symbols and words by which companies distinguish and identify their brands. Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, Inc., 425 U.S. 748, 765, 96 S.Ct. 1817, 48 L.Ed.2d 346 (1976) (defining commercial speech as the “dissemination of information as to who is producing and selling what product, for what reason, and at what price”). It does not follow, however, that all government regulation of trademarks is properly reviewed under the Central Hudson intermediate scrutiny standard. Section 2(a) bars registration of disparaging marks. This regulation is squarely based on the expressive aspect of the speech, not its commercial-speech aspects. It should therefore be evaluated under the First Amendment standards applicable to the regulation of expressive speech. Discrimination against a mark by virtue of its offensive, disparaging nature discriminates against the mark’s political or social message. Section 2(a) should be subject to strict scrutiny, and be invalidated for its undisputed inability to survive such scrutiny.
Even if we were to treat § 2(a) as a regulation of commercial speech, it would fail to survive. In Central Hudson, the Supreme Court laid out the intermediate-scrutiny framework for determining the constitutionality of restrictions on commercial speech. 447 U.S. at 566, 100 S.Ct. 2343. First, commercial speech “must concern lawful activity and not be misleading.” Id. If this is the case, we ask whether “the asserted governmental interest is substantial,” id., and whether the regulation “directly and materially advanc[es]” the government’s asserted interest and is narrowly tailored to achieve that objective. Lorillard Tobacco Co. v. Reilly, 533 U.S. 525, 555–56, 121 S.Ct. 2404, 150 L.Ed.2d 532 (2001). “Under a commercial speech inquiry, it is the State’s burden to justify its content-based law as consistent with the First Amendment.” Sorrell, 131 S.Ct. at 2667.
First, we ask whether the regulated activity is lawful and not misleading. Cent. Hudson, 447 U.S. at 563–64, 100 S.Ct. 2343. Unlike many other provisions of § 2, the disparagement provision does not address misleading, deceptive, or unlawful marks. There is nothing illegal or misleading about a disparaging trademark like Mr. Tam’s mark.
Next, for speech that is lawful and not misleading, a substantial government interest must justify the regulation. Id. at 566, 100 S.Ct. 2343. But § 2(a) immediately fails at this step. The entire interest of the government in § 2(a) depends on disapproval of the message. That is an insufficient interest to pass the test of intermediate scrutiny, as the Supreme Court made clear in *1356 Sorrell. 131 S.Ct. at 2668 (law must not “seek to suppress a disfavored message”); id. at 2670 (rejecting message-based interest as “contrary to basic First Amendment principles”); see id. at 2667–68 (finding it unnecessary to rely on strict scrutiny; rejecting justification under Central Hudson); Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 69–72, 103 S.Ct. 2875, 77 L.Ed.2d 469 (1983); Carey v. Population Servs., Int’l, 431 U.S. 678, 701 & n. 28, 97 S.Ct. 2010, 52 L.Ed.2d 675 (1977).
The government proffers several interests to justify its bar on disparaging trademarks. It argues principally that the United States is “entitled to dissociate itself from speech it finds odious.” Appellee’s En Banc Br. 41. This core argument rests on intense disapproval of the disparaging marks. See, e.g., Appellee’s En Banc Br. 1 (“the most vile racial epithets and images”); id. at 2–3 (“racial slurs ... or religious insults, ethnic caricatures, misogynistic images, or any other disparaging terms or logos”); id. at 14 (“racial epithets”); id. at 21 (“racial slurs and similar disparagements”); id. at 22 (“including the most vile racial epithets”); id. at 41 (“speech [the government] finds odious”); id. at 44 (“racial slurs”). And that disapproval is not a legitimate government interest where, as here, for the reasons we have already discussed, there is no plausible basis for treating the speech as government speech or as reasonably attributed to the government by the public.
The government also argues that it has a legitimate interest in “declining to expend its resources to facilitate the use of racial slurs as source identifiers in interstate commerce.” Appellee’s En Banc Br. 43. The government’s interest in directing its resources does not warrant regulation of these marks. As discussed, trademark registration is user-funded, not taxpayer-funded. The government expends few resources registering these marks. See supra at 1353–54. Its costs are the same costs that would be incidental to any governmental registration: articles of incorporation, copyrights, patents, property deeds, etc. In fact, the government spends far more significant funds defending its refusal decisions under the statute. See McGinley, 660 F.2d at 487 (Rich, J., dissenting) (“More ‘public funds’ are being expended in the prosecution of this appeal than would ever result from the registration of the mark.”). Finally, labeling this sort of interest as substantial creates an end-run around the unconstitutional conditions doctrine, as virtually all government benefits involve the resources of the federal government in a similar sense. Nearly every government act could be justified under this ground, no matter how minimal. For example, the government could also claim an interest in declining to spend resources to issue permits to racist, sexist, or homophobic protests. The government cannot target speech on this basis, even if it must expend resources to grant parade permits or close down streets to facilitate such speech.
This holds true even though the government claims to have a “compelling interest in fostering racial tolerance.” Appellee’s En Banc Br. 43 (citing Bob Jones Univ. v. United States, 461 U.S. 574, 604, 103 S.Ct. 2017, 76 L.Ed.2d 157 (1983)). Bob Jones University does not stand for the broad proposition the government claims. Bob Jones University is a case about racially discriminatory conduct, not speech. The Court held that the government has an interest in combating “racial discrimination in education,” not a more general interest in fostering racial tolerance that would justify preventing disparaging speech. Id. at 595, 103 S.Ct. 2017.
The invocation of the general racial-tolerance interest to support speech *1357 regulation is a sharply different matter, as the Supreme Court explained in R.A.V.:
One must wholeheartedly agree with the Minnesota Supreme Court that “[i]t is the responsibility, even the obligation, of diverse communities to confront [virulent notions of racial supremacy] in whatever form they appear,” but the manner of that confrontation cannot consist of selective limitations upon speech. St. Paul’s brief asserts that a general “fighting words” law would not meet the city’s needs because only a content-specific measure can communicate to minority groups that the “group hatred” aspect of such speech “is not condoned by the majority.” The point of the First Amendment is that majority preferences must be expressed in some fashion other than silencing speech on the basis of its content.
505 U.S. at 392, 112 S.Ct. 2538 (first alteration in original; citations omitted). What is true of direct “silencing” is also true of the denial of important legal rights. “[I]n public debate we must tolerate insulting, and even outrageous, speech in order to provide adequate breathing space to the freedoms protected by the First Amendment.” Snyder, 562 U.S. at 458, 131 S.Ct. 1207 (quoting Boos v. Barry, 485 U.S. 312, 322, 108 S.Ct. 1157, 99 L.Ed.2d 333 (1988)) (alterations omitted). The case law does not recognize a substantial interest in discriminatorily regulating private speech to try to reduce racial intolerance.
Moreover, at the level of generality at which the government invokes “racial tolerance,” it is hard to see how one could find that § 2(a) “directly and materially advanc[es]” this interest and is narrowly tailored to achieve that objective. Lorillard Tobacco Co., 533 U.S. at 555–56, 121 S.Ct. 2404. Disparaging speech abounds on the Internet and in books and songs bearing government registered copyrights. And the PTO has granted trademark registrations of many marks with a racially charged character. Further, the connection to a broad goal of racial tolerance would be even weaker to the extent that the government suggests, contrary to our conclusion in II.A supra, that denial of registration has no meaningful effect on the actual adoption and use of particular marks in the marketplace.
Finally, the government argues that it has a legitimate interest in “allowing States to make their own determinations about whether trademarks should be unenforceable on grounds of public policy.” Appellee’s En Banc Br. 44. However, this interest cannot stand alone. If § 2(a) is otherwise unconstitutional, the government cannot render it constitutional by arguing that it is necessary so that states can partake in the same unconstitutional message-based regulation of trademarks. The government, in essence, argues that it has a legitimate interest in leaving the door open for states to violate the Constitution. This interest is certainly not legitimate, let alone substantial.
We conclude that the government has not presented us with a substantial government interest justifying the § 2(a) bar on disparaging marks. All of the government’s proffered interests boil down to permitting the government to burden speech it finds offensive. This is not a legitimate interest. With no substantial government interests, the disparagement provision of § 2(a) cannot satisfy the Central Hudson test. We hold the disparagement provision of § 2(a) unconstitutional under the First Amendment.
Although we find the disparagement provision of § 2(a) unconstitutional, nothing we say should be viewed as an endorsement of the mark at issue. We recognize that invalidating this provision may lead to the wider registration of marks *1358 that offend vulnerable communities. Even Mr. Tam, who seeks to reappropriate the term “slants,” may offend members of his community with his use of the mark. See Br. of Amici Curiae Nat’l Asian Pacific Am. Bar Ass’n 3, 5. But much the same can be (and has been) said of many decisions upholding First Amendment protection of speech that is hurtful or worse. Whatever our personal feelings about the mark at issue here, or other disparaging marks, the First Amendment forbids government regulators to deny registration because they find the speech likely to offend others. Even when speech “inflict[s] great pain,” our Constitution protects it “to ensure that we do not stifle public debate.” Snyder, 562 U.S. at 461, 131 S.Ct. 1207. The First Amendment protects Mr. Tam’s speech, and the speech of other trademark applicants.
We hold that the disparagement provision of § 2(a) is unconstitutional because it violates the First Amendment. We vacate the Board’s holding that Mr. Tam’s mark is unregistrable, and remand this case to the Board for further proceedings.
O’MALLEY, Circuit Judge, with whom WALLACH, Circuit Judge, joins, concurring.
I agree that the disparagement provision of 15 U.S.C. § 1052(a) (“§ 2(a)”) is unconstitutional on its face. I agree, moreover, that § 2(a) cannot survive the searching constitutional scrutiny to which the majority subjects it under the First Amendment to the United States Constitution. On this point, the majority rightly dispenses with this court’s precedent in In re McGinley, 660 F.2d 481 (C.C.P.A.1981) and its progeny. I write separately, however, because, I believe § 2(a) is also unconstitutionally vague, rendering it unconstitutional under the Fifth Amendment to the United States Constitution.
While the majority acknowledges the vague and uncertain application of § 2(a), Maj. Op. 1341-43, it finds that “[a]ll we need say about the uncertainty here, however, is that it contributes significantly to the chilling effect on speech,” id. at 1342-43. I agree with the majority’s concern about the uncertain nature of § 2(a), but believe those concerns should lead us to do more than note 2(a)’s undoubted chilling effect on speech. I find § 2(a)’s disparagement provision to be so vague that I would find it to be unconstitutional, whether or not it could survive Appellant’s First Amendment challenge.
Section 2(a) provides that the Trademark Trial and Appeal Board (“Board”) may refuse an application when the trademark “[c]onsists of or comprises ... matter which may disparage ... persons, living or dead, institutions, beliefs, or national symbols, or bring them into contempt, or disrepute.” (emphasis added). As the majority correctly notes, the language of the statute creates “uncertainty as to what might be deemed disparaging.” Maj. Op. 1341-42. Both would-be applicants and the Board are left to guess at what may have the potential to disparage a broad range of persons, institutions, symbols, and even undefined “beliefs.” And, they are left to guess at whether “may disparage” is the equivalent of bringing into contempt or disrepute, or is a distinct category of impropriety from these latter evils.
Where, as here, the language of a statute evades clarity, “[t]he area of proscribed conduct will be adequately defined and the deterrent effect of the statute contained within constitutional limits only by authoritative constructions sufficiently illuminating the contours of an otherwise vague prohibition.” Dombrowski v. Pfister, 380 U.S. 479, 490–91, 85 S.Ct. 1116, 14 L.Ed.2d 22 (1965). The Board has developed *1359 a two-step test to determine whether a mark is disparaging:
(1) What is 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; and
(2) If that meaning is found to refer to identifiable persons, institutions, beliefs or national symbols, whether that meaning may be disparaging to a substantial composite of the referenced group.
Trademark Manual of Exam. Proc. (“TMEP”) § 1203.03(b)(i) (Oct.2015 ed.) (citing, inter alia, In re Geller, 751 F.3d 1355, 1358 (Fed.Cir.2014); Harjo v. Pro–Football Inc., 50 U.S.P.Q.2d 1705, 1740–41 (T.T.A.B.1999)). Thus, the Board has concluded that a mark may disparage within the meaning of § 2(a) when a majority of the Board believes it “dishonor[s] by comparison with what is inferior, slight[s], deprecate[s], degrade[s], or affect[s] or injure [s] by unjust comparison.” Pro–Football, Inc. v. Harjo, 284 F.Supp.2d 96, 124 (D.D.C.2003) (internal quotation marks omitted) (quoting Harjo v. Pro–Football, Inc., 50 U.S.P.Q.2d 1705, 1737 n. 98 (T.T.A.B.1999)).
The two-step test does little to alleviate § 2(a)’s uncertainty. Indeed, by adding the caveat that a mark can be rejected whenever a mark’s meaning may be disparaging to “a substantial composite” of an “identifiable” group, (TMEP § 1203.03(b)(i)), the TMEP compounds the confusion the statute engenders. Thus a mark need only potentially disparage a subset of any group as long as that group can be “identifi[ed].”
One need only examine the disparate ways in which § 2(a) has been applied to see the confusion. While it is true that a “fertile legal ‘imagination can conjure up hypothetical cases in which the meaning of [disputed] terms will be in nice question,’ ” Grayned v. City of Rockford, 408 U.S. 104, 112 n. 15, 92 S.Ct. 2294, 33 L.Ed.2d 222 (1972) (alteration in original) (quoting Am. Commc’ns Ass’n v. Douds, 339 U.S. 382, 412, 70 S.Ct. 674, 94 L.Ed. 925 (1950)), the arbitrary application of § 2(a) is easily demonstrated. The majority discusses numerous examples of inconsistent registration decisions. Maj. Op. 1342 n. 7. These include examples where there is no conceivable difference between the applied-for marks, yet one is approved and the other rejected. Compare HAVE YOU HEARD SATAN IS A REPUBLICAN (Trademark Application Serial No. 85,077,647) (rejected because it disparaged the Republican Party), with THE DEVIL IS A DEMOCRAT, Registration No. 85,525,066 (accepted and later abandoned for other reasons). I agree with the majority that there appears to be “no rationale for the PTO’s seemingly arbitrary registration decisions, let alone one that would give applicants much guidance.” Maj. Op. 1342 n. 7.1
*1360 For § 2(a) to survive a vagueness challenge, the Supreme Court requires it “give the person of ordinary intelligence a reasonable opportunity to know what is prohibited, so that he may act accordingly.” Grayned, 408 U.S. at 108, 92 S.Ct. 2294. Further, “if arbitrary and discriminatory enforcement is to be prevented, laws must provide explicit standards for those who apply them.” Id. Given the subjective and hypothetical language of the statute and its well-documented, inconsistent application by the Board, § 2(a) is void for vagueness under even a lax test for vagueness. But the standard we should apply to § 2(a) is not lax.
“The degree of vagueness that the Constitution tolerates ... depends in part on the nature of the enactment.” Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489, 498, 102 S.Ct. 1186, 71 L.Ed.2d 362 (1982). “[P]erhaps the most important factor affecting the clarity that the Constitution demands of a law is whether it threatens to inhibit the exercise of constitutionally protected rights. If, for example, the law interferes with the right of free speech ..., a more stringent vagueness test should apply.” Id. at 499, 102 S.Ct. 1186. The First Amendment concerns articulated by the majority support application of a “more stringent vagueness test”—one that § 2(a) simply cannot pass.
a. First Amendment Concerns Require a Stringent Vagueness Test
As the majority points out, “[i]t is beyond dispute that § 2(a) discriminates on the basis of content.” Maj. Op. 1335. “[T]he test for disparagement—whether a substantial composite of the referenced group would find the mark disparaging—makes clear that it is the nature of the message conveyed by the speech which is being regulated. If the mark is found disparaging by the referenced group, it is denied registration.” Id. at 1335. Indeed, the problems with § 2(a) are more substantial than the majority even acknowledges—not only is a trademark’s registrability adjudged by the message it conveys, but the message conveyed is adjudged by the potential sensibilities of a broad range of potential listeners.
Under First Amendment principles, “content-based regulation of speech ... raises special First Amendment concerns because of its obvious chilling effect on free speech.” Reno v. ACLU, 521 U.S. 844, 872, 117 S.Ct. 2329, 138 L.Ed.2d 874 (1997). Indeed, “[b]road prophylactic rules in the area of free expression are suspect. Precision of regulation must be the touchstone in an area so closely touching our most precious freedoms.” Edenfield v. Fane, 507 U.S. 761, 777, 113 S.Ct. 1792, 123 L.Ed.2d 543 (1993) (internal quotation marks omitted) (quoting NAACP v. Button, 371 U.S. 415, 438, 83 S.Ct. 328, 9 L.Ed.2d 405 (1963)). The Supreme Court’s emphasis on precision for content-based regulations is premised on its understanding of
at least two connected but discrete due process concerns: first, that regulated parties should know what is required of them so they may act accordingly; second, precision and guidance are necessary so that those enforcing the law do not act in an arbitrary or discriminatory way. When speech is involved, rigorous adherence to those requirements is necessary to ensure that ambiguity does not chill protected speech.
*1361 FCC v. Fox Television Stations, Inc., ––– U.S. ––––, 132 S.Ct. 2307, 2317, 183 L.Ed.2d 234 (2012) (citing Grayned, 408 U.S. at 108–109, 92 S.Ct. 2294).
b. Section 2(a) is Void for Vagueness
Section 2(a)’s undeniable chilling effect on speech requires it to pass a “more stringent test” for vagueness in order to pass constitutional muster. Hoffman, 455 U.S. at 498, 102 S.Ct. 1186. Recognizing that due process vagueness challenges are more difficult to sustain where civil regulation—as distinct from criminal penalty provisions—are at issue, I believe § 2(a)’s inherent ambiguity makes it difficult for would-be applicants to discern its boundaries and leads to inconsistent and unreliable actions on the part of the government as it seeks to regulate on the basis of content.
First, the imprecise, content-based regulation of trademark registration affects the types of marks sought by would-be registrants. “Vague laws force potential speakers to “ ‘steer far wider of the unlawful zone” ... than if the boundaries of the forbidden areas were clearly marked.’ ” Brown v. Entm’t Merchants Ass’n, ––– U.S. ––––, 131 S.Ct. 2729, 2743, 180 L.Ed.2d 708 (2011) (quoting Baggett v. Bullitt, 377 U.S. 360, 372, 84 S.Ct. 1316, 12 L.Ed.2d 377 (1964)). The majority opinion rightly concludes that, given the Board’s inconsistency, “the public would have a hard time drawing much reliable guidance.” Maj. Op. 1342. The “uncertainty of speech-affecting standards has long been recognized as a First Amendment problem,” and the uncertainty inherent in § 2(a) “contributes significantly to the chilling effect on speech.” Maj. Op. 1342.2
Next, the absence of clear standards for the application of § 2(a) provides the government with virtually unlimited ability to pick and choose which marks to allow and which to deny. And neither § 2(a) itself nor the TMEP’s two-step test provides the PTO, the courts, or the public with any certainty as to what may disparage a given subset of any given population or group of believers. That is simply inadequate under the Fifth Amendment. See Nat’l Endowment for the Arts v. Finley, 524 U.S. 569, 588, 118 S.Ct. 2168, 141 L.Ed.2d 500 (1998) (“Under the First and Fifth Amendments, speakers are protected from arbitrary and discriminatory enforcement of vague standards.”); Grayned, 408 U.S. at 108–09, 92 S.Ct. 2294 (1972) (“[I]f arbitrary and discriminatory enforcement is to be prevented, laws must provide explicit standards for those who apply them. A vague law impermissibly delegates basic policy matters to policemen, judges, and juries for resolution on an ad hoc and subjective basis, with the attendant dangers of arbitrary and discriminatory application.”) (footnotes omitted). Cf. Kolender v. Lawson, 461 U.S. 352, 357–58, 103 S.Ct. 1855, 75 L.Ed.2d 903 (1983) (noting in the context of a criminal penalty scheme that, although the vagueness doctrine “focuses both on actual notice to citizens and arbitrary enforcement, we have recognized recently that the more important aspect of vagueness doctrine ‘is not actual notice, but the other principal element of the doctrine *1362 —the requirement that a legislature establish minimal guidelines to govern law enforcement.’ Where the legislature fails to provide such minimal guidelines, a criminal statute may permit ‘a standardless sweep [that] allows policemen, prosecutors, and juries to pursue their personal predilections.’ ” (quoting Smith v. Goguen, 415 U.S. 566, 574, 575, 94 S.Ct. 1242, 39 L.Ed.2d 605 (1974))).
Other circuits to have considered the use of the subjective terms connoting insult—like disparagement—have expressed similar concerns about the absence of objective standards governing their application.
In Dambrot v. Central Michigan University, 55 F.3d 1177 (6th Cir.1995), for example, the Sixth Circuit considered the discriminatory harassment policy of Central Michigan University (“CMU”). That policy defined racial and ethnic harassment as:
any intentional, unintentional, physical, verbal, or nonverbal behavior that subjects an individual to an intimidating, hostile or offensive educational, employment or living environment by ... (c) de-meaning or slurring individuals through ... written literature because of their racial or ethnic affiliation; or (d) using symbols, [epithets] or slogans that infer negative connotations about the individual’s racial or ethnic affiliation.
Id. at 1182 (emphases added). The court found the policy impermissibly vague because it required “one [to] make a subjective reference” and because “different people find different things offensive.” Id. at 1184. As such, the policy’s enforcement was too tied to subjective reference and, thus, both failed to “provide fair notice” and gave rise to an “unrestricted delegation of power” to university officials. Id. See also Wynn Oil Co. v. Purolator Chem. Corp., 536 F.2d 84, 86 (5th Cir.1976) (finding the subsection of an “injunction which restrains defendants from ‘slandering and disparaging the Wynn Oil Co. and its products’ [to be] impermissively vague”).
In Ridley v. Massachusetts Bay Transportation Authority, 390 F.3d 65 (1st Cir.2004), the First Circuit upheld the validity of the Massachusetts Bay Transportation Authority’s (“MBTA”) “guideline prohibiting demeaning or disparaging material,” id. at 93, because, in that case, “there [was] no serious concern about either notice or chilling effects[ ] where there [were] no consequences for submitting a non-conforming advertisement and having it rejected” id. at 94. But that court specifically distinguished the guidelines at issue—“given the nature of the MBTA’s advertising program and its chief purpose of raising revenue without losing ridership,” id. at 94—from “the concern over subjective decision making[, which has the] most effect in government licensing schemes” id. at 95. While the trademark registration scheme is not a traditional public forum making use of a licensing scheme to “maintain basic order,” it implicates the “[e]xcessive discretion and vagueness inquiries under the First Amendment” in much the same way. Id. at 94. As the majority notes, trademark registrants receive substantial benefits from the fact of registration, Maj. Op. 1328-29; denial of those benefits based on the subjective views of governmental employees about the potential subjective views of those who might be exposed to the proposed mark is an essentially standardless measure.
In McGinley, we found § 2(a)’s ban on scandalous subject matter, “sufficiently precise to enable the PTO and the courts to apply the law fairly and to notify a would-be registrant that the mark he adopts will not be granted a federal registration.” 660 F.2d at 484. While I agree that the PTO is capable of “notify[ing] a would-be registrant” of its decision to deny registration under § 2(a), the law is by no means precise enough to “enable the PTO and the courts to apply [it] fairly.” Id. As *1363 the majority points out, the Board has allowed use of a term by one trademark holder while disallowing use of precisely the same term by another based apparently on its view of how use of that term might be received by the audience the Board has chosen to “identify.” Maj. Op. 1336-38. This fact alone evidences the absence of explicit standards for the application of § 2(a).
As it turns out, the PTO’s Assistant Commissioner was correct in 1939 in expressing concern that “the word ‘disparage’ ... is going to cause a great deal of difficulty in the Patent Office, because ... it is always going to be just a matter of the personal opinion of the individual parties as to whether they think it is disparaging.” Hearing on H.R. 4744 Before the Subcomm. on Trademarks of the H. Comm. on Patents, 76th Cong. 21 (1939) (statement of Leslie Frazer). The Board has likewise commented on the vague and subjective nature of § 2(a). See, e.g., In re In Over Our Heads, 1990 WL 354546, at *1 (T.T.A.B.1990) (“[T]he guidelines for determining whether a mark is scandalous or disparaging are somewhat vague and the determination of whether a mark is scandalous or disparaging is necessarily a highly subjective one.”) (bracketing and quotation marks omitted); Harjo v. Pro–Football, Inc., 1999 WL 375907, at *35 (T.T.A.B.1999) (noting that whether a mark is disparaging “is highly subjective and, thus, general rules are difficult to postulate”).
“It is a basic principle of due process that an enactment is void for vagueness if its prohibitions are not clearly defined.” Grayned, 408 U.S. at 108, 92 S.Ct. 2294. The need for clarity is especially relevant when a law implicates First Amendment rights, as § 2(a) indisputably does. Section 2(a) does not provide a “person of ordinary intelligence a reasonable opportunity to know what is prohibited, so that he may act accordingly.” Id. And inconsistent, indeed seemingly rudderless, application of § 2(a) demonstrates the “arbitrary and discriminatory enforcement” that occurs when regulations do not “provide explicit standards for those who apply them.” Id.
While I agree with the majority’s thoughtful First Amendment analysis, I do not believe it is the only predicate to the conclusion that § 2(a) is unconstitutional.
For the above reasons, I concur in the majority’s conclusions and separately concur in the result.
DYK, Circuit Judge, concurring in part and dissenting in part, with whom Circuit Judges LOURIE and REYNA join with respect to parts I, II, III, and IV.
The majority is correct that the bar on registration of disparaging marks is unconstitutional as applied to Mr. Tam. But in my view the majority errs in going beyond the facts of this case and holding the statute facially unconstitutional as applied to purely commercial speech.
It is noteworthy that the majority seeks to justify its sweeping holding by describing § 2(a) as being something it is not. The provision bars the registration of marks that “disparage ... or bring into contempt, or disrepute.” 15 U.S.C. § 1052(a) (otherwise identified as § 2(a)). The majority repeatedly asserts that “[t]he government enacted § 2(a), and defends it today, because it is hostile to the messages conveyed by the refused marks.”1 Maj. *1364 Op. at 1337. In my view, there is nothing in the statute itself or the legislative history that supports this interpretation. On its face, and as interpreted by the Trademark Trial and Appeal Board (“the Board”), the statute is designed to preclude the use of government resources not when the government disagrees with a trademark’s message, but rather when its meaning “may be disparaging to a substantial composite of the referenced group.” In re Lebanese Arak Corp., 94 U.S.P.Q.2d 1215, 1217 (T.T.A.B.2010) (emphasis added). The PTO uses an objective test in making this determination, looking to dictionaries, the relationship of the matter to the other elements of 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. See id.2
Thus the purpose of the statute is to protect underrepresented groups in our society from being bombarded with demeaning messages in commercial advertising. The question is whether the statute so designed can survive First Amendment scrutiny. My answer is that the statute is constitutional as applied to purely commercial trademarks, but not as to core political speech, of which Mr. Tam’s mark is one example. Ultimately, unlike the majority, I do not think that the government must support, or society tolerate, disparaging trademarks in the name of commercial speech. The majority’s opinion not only invalidates the bar on disparaging marks in § 2(a) but may also effectively invalidate the bar on scandalous marks and the analogous provisions of the Model State Trademark Act. See 1964 Model State Trademark Act, § 2(b). The government need not support the inevitable consequence of this decision—“the wider registration of marks that offend vulnerable communities.” Maj. Op. at 1357-58.
As the majority notes, the Supreme Court has long recognized the protection of offensive speech that constitutes core political expression. “The right to free speech ... may not be curtailed simply because the speaker’s message may be offensive to his audience.” Hill v. Colorado, 530 U.S. 703, 716, 120 S.Ct. 2480, 147 L.Ed.2d 597 (2000). Underpinning the First Amendment’s protection of core speech that is disparaging is the fundamental constitutional value of preserving an “uninhibited marketplace of ideas in which truth will ultimately prevail,” a marketplace that provides “suitable access to social, political, esthetic, moral, and other ideas and experiences.” Red Lion, 395 U.S. at 390, 89 S.Ct. 1794. Integral to an “uninhibited marketplace of ideas” is the ability to incite debate. “[A] principal function of free speech under our system of government is to invite dispute. It may indeed best serve its high purpose when it induces a condition of unrest, creates dissatisfaction with conditions as they are, or even stirs people to anger.” Texas v. *1365 Johnson, 491 U.S. 397, 408–09, 109 S.Ct. 2533, 105 L.Ed.2d 342 (1989). Thus to maintain a “meaningful dialogue of ideas,” “we must tolerate insulting, and even outrageous, speech in order to provide adequate breathing space to the freedoms protected by the First Amendment.” Snyder v. Phelps, 562 U.S. 443, 452, 458, 131 S.Ct. 1207, 179 L.Ed.2d 172 (2011) (internal quotation marks, citations, and alterations omitted).3 At bottom, as Justice Holmes described, in the core speech area the First Amendment enshrines the “principle of free thought—not free thought for those who agree with us but freedom for the thought that we hate.” U.S. v. Schwimmer, 279 U.S. 644, 654–55, 49 S.Ct. 448, 73 L.Ed. 889 (1929) (Holmes, J., dissenting).
But this principle simply does not apply in the commercial context. For example, it is well established that racially or sexually disparaging speech in the workplace, when severe, may constitute a violation of Title VII, either as harassment or the creation of a hostile work environment. See, e.g., Faragher v. City of Boca Raton, 524 U.S. 775, 787–88, 118 S.Ct. 2275, 141 L.Ed.2d 662 (1998); Rodgers v. Western–Southern Life Ins. Co., 12 F.3d 668, 675 (7th Cir.1993). The same is necessarily true in the context of federal public accommodations law governing commercial establishments. No case of which I am aware suggests that imposing liability for disparaging speech in those commercial contexts, even when separated from conduct, violates the First Amendment.
So too in the area of commercial speech race or sex disparagement can claim no First Amendment protection. Unlike core political expression, the “extension of First Amendment protection to commercial speech is justified principally by the value to consumers of the information such speech provides.” Zauderer v. Office of Disciplinary Counsel of Supreme Court of Ohio, 471 U.S. 626, 651, 105 S.Ct. 2265, 85 L.Ed.2d 652 (1985). Its constitutional protection derives not from any dialogic function in the marketplace of ideas, but rather from its “informational function” in the marketplace of goods and services, Central Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n of N.Y., 447 U.S. 557, 563, 100 S.Ct. 2343, 65 L.Ed.2d 341 (1980), in other words, “who is producing and selling what product, for what reason, and at what price.” Va. State Bd. Of Pharmacy v. Va. Citizens Consumer Council, Inc., 425 U.S. 748, 765, 96 S.Ct. 1817, 48 L.Ed.2d 346 (1976); see also Sorrell v. IMS Health Inc., –––U.S. ––––, 131 S.Ct. 2653, 2673–74, 180 L.Ed.2d 544 (2011) (Breyer, J., dissenting). We protect the dissemination of this information to ensure that “private economic decisions” are “intelligent and well informed.” Va. State Bd. Of Pharmacy, 425 U.S. at 765, 96 S.Ct. 1817.
Speech proposing a commercial transaction is “an area traditionally subject to government regulation.” 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484, 499, 116 S.Ct. 1495, 134 L.Ed.2d 711 (1996) (citing and quoting Ohralik v. Ohio State Bar Ass’n, 436 U.S. 447, 456, 98 S.Ct. 1912, 56 L.Ed.2d 444 (1978)). The Court has “been careful to distinguish commercial speech from speech at the First Amendment’s core,” Florida Bar v. Went For It, Inc., 515 U.S. 618, 623, 115 S.Ct. 2371, 132 L.Ed.2d 541 (1995), recognizing the “commonsense distinctions that exist between commercial and noncommercial speech.” 44 Liquormart, 517 U.S. at 502, 116 S.Ct. 1495 (quoting *1366 Virginia Bd. of Pharmacy, 425 U.S. at 771 n. 24, 96 S.Ct. 1817). The “greater objectivity” and “greater hardiness” of commercial speech and the different constitutional values underlying its protection “likely diminish [ ] the chilling effect that may attend its regulation.” 44 Liquormart, 517 U.S. at 499, 116 S.Ct. 1495 (internal quotation marks and citations omitted). Accordingly, the Court has explained that “the State may regulate some types of commercial advertising more freely than other forms of protected speech,” id. at 498, 116 S.Ct. 1495 (internal quotations marks and citations omitted), and “the State may at times prescribe what shall be orthodox in commercial advertising,” Hurley v. Irish–American Gay, Lesbian & Bisexual Group of Boston, 515 U.S. 557, 573, 115 S.Ct. 2338, 132 L.Ed.2d 487 (1995) (internal quotation marks and citations omitted)—something it could never do with core political speech.
Recognizing the more limited protection of commercial speech, the Court has repeatedly upheld regulations “protect[ing] consumers from misleading, deceptive, or aggressive sales practices,” because such regulations are “consistent with the reasons for according constitutional protection to commercial speech” in the first place. 44 Liquormart, 517 U.S. at 501, 116 S.Ct. 1495; see also, e.g., Florida Bar, 515 U.S. 618, 115 S.Ct. 2371, 132 L.Ed.2d 541 (1995); Metromedia, Inc. v. City of San Diego, 453 U.S. 490, 101 S.Ct. 2882, 69 L.Ed.2d 800 (1981); Bates v. State Bar of Ariz., 433 U.S. 350, 97 S.Ct. 2691, 53 L.Ed.2d 810 (1977). “There can be no constitutional objection to the suppression of commercial messages that do not accurately inform the public about lawful activity.” Central Hudson, 447 U.S. at 563, 100 S.Ct. 2343.
This stands in stark contrast to core political speech, for which “constitutional protection does not turn upon ‘the truth ... of the ideas and beliefs which are offered.’ ” N.Y. Times Co. v. Sullivan, 376 U.S. 254, 271, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964) (quoting NAACP v. Button, 371 U.S. 415, 445, 83 S.Ct. 328, 9 L.Ed.2d 405 (1963)). “The erroneous statement is inevitable in free debate, and [ ] it must be protected [absent a showing of actual malice] if the freedoms of expression are to have the breathing space that they need to survive.” Id. at 271–72, 84 S.Ct. 710 (internal quotation marks, citations, and alterations omitted). “Authoritative interpretations of the First Amendment guarantees have consistently refused to recognize an exception for any test of truth.” N.Y. Times, 376 U.S. at 271, 84 S.Ct. 710. See also Hustler Magazine, Inc. v. Falwell, 485 U.S. 46, 52, 108 S.Ct. 876, 99 L.Ed.2d 41 (1988).
To be sure, the Court has held that commercial advertising cannot be restricted just because the product or service may be offensive to some members of the audience. See Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 71, 103 S.Ct. 2875, 77 L.Ed.2d 469 (1983); Carey v. Population Servs. Int’l, 431 U.S. 678, 701, 97 S.Ct. 2010, 52 L.Ed.2d 675 (1977). But, at the same time, the Court has explained that the manner of advertising itself may be restricted to protect the audience’s privacy interests. See Florida Bar v. Went For It, Inc., 515 U.S. 618, 630, 115 S.Ct. 2371, 132 L.Ed.2d 541 (1995). “[T]he existence of [First Amendment] protection does not deprive the State of all power to regulate such advertising in order to minimize its offensiveness.” Bolger, 463 U.S. at 84, 103 S.Ct. 2875 (1983) (Stevens, J., concurring) (citing and quoting from Carey, 431 U.S. at 716, 97 S.Ct. 2010 (Stevens, J., concurring)).
For example, in Florida Bar the Court upheld a ban on lawyer advertising targeted to recent accident victims and their families. *1367 515 U.S. at 634–35, 115 S.Ct. 2371. There the Court distinguished Bolger, which rejected a total ban on advertising related to contraceptives, because the government’s interest in Bolger had been only to shield citizens from generally “offensive” and “intrusive” products. See id. at 630–31, 115 S.Ct. 2371. That interest, the Court explained, was entirely different from the interest in “protecting the personal privacy and tranquility of [Florida’s] citizens from crass commercial intrusion by attorneys upon their personal grief in times of trauma.” Id. at 630, 115 S.Ct. 2371 (alterations omitted). The Court thus had “little trouble crediting the Bar’s” “privacy-based” interest as “substantial,” and held that it was sufficient to justify the advertising ban. Id. at 625, 629, 635, 115 S.Ct. 2371.
Disparagement as defined by the Board “is essentially a violation of one’s right of privacy—the right to be let alone from contempt or ridicule.” TMEP § 1203.03(b). While in the trademark context the dissemination of the disparaging material is not limited to the disparaged group, the disparaged group is nonetheless targeted in the sense that it is singled out for ridicule. Furthermore, the fact that the dissemination of the disparaging advertising is not limited to the disparaged group makes the government’s interest here all the greater—the effect on the disparaged group is amplified, not lessened, by disseminating the disparaging material to the public at large.
This well-recognized disparity in the types of restrictions that are permissible as applied to commercial as opposed to political speech derives from the very different constitutional values underlying their protection in the first place. The Court has recognized that the government has greater authority to “distinguish between the relative value of different categories of commercial speech” than of noncommercial speech. Metromedia, 453 U.S. at 514, 101 S.Ct. 2882. Specifically, the government has a distinct and substantial interest in “proscribing intrusive and unpleasant formats” for commercial expression. Members of City Council of L.A. v. Taxpayers for Vincent, 466 U.S. 789, 806, 104 S.Ct. 2118, 80 L.Ed.2d 772 (1984); see also Lehman v. City of Shaker Heights, 418 U.S. 298, 304, 94 S.Ct. 2714, 41 L.Ed.2d 770 (1974); Metromedia, 453 U.S. at 514, 101 S.Ct. 2882. Indeed, “it may not be the content of the speech, as much as the deliberate ‘verbal or visual assault,’ that justifies proscription.” Hill, 530 U.S. at 716, 120 S.Ct. 2480 (quoting Erznoznik v. Jacksonville, 422 U.S. 205, 210–11, n. 6, 95 S.Ct. 2268, 45 L.Ed.2d 125 (1975)).
Unlike core political speech, where offensiveness or disparagement has recognized value in its tendency to provoke debate, disparagement in commercial advertising furthers no First Amendment value. Indeed, neither counsel at oral argument nor the majority in its opinion has identified any First Amendment value served by disparaging speech in the commercial context. Thus even blanket bans on commercial speech may be the kind of consumer protective regulations that are consistent with the “informational function” of commercial advertising. See Central Hudson, 447 U.S. at 563, 100 S.Ct. 2343.
The majority, apparently recognizing that purely commercial speech is entitled to lesser protection, urges that all disparaging trademarks deserve heightened First Amendment protection because they have an expressive component. See Maj. Op. at 1337-38. While I agree that some marks, including Mr. Tam’s, have an expressive component, it would seem beyond debate that many do not, as is the case with respect to routine product identifiers. Indeed, the Supreme Court confirmed the lack of an expressive component in most *1368 trade names in Friedman v. Rogers, where it explicitly distinguished between advertisements that “editorialize on any subject, cultural, philosophical, or political,” which might be entitled to greater First Amendment protection, and the “mere solicitation of patronage implicit in a trade name,” which “is a form of commercial speech and nothing more.” 440 U.S. 1, 11, n. 10, 99 S.Ct. 887, 59 L.Ed.2d 100 (1979). The Court again recognized this distinction in S.F. Arts & Athletics Inc. v. U.S. Olympic Comm’n, 483 U.S. 522, 535, 107 S.Ct. 2971, 97 L.Ed.2d 427 (1987). “To the extent that [the statute] applies to uses for the purpose of trade [or] to induce the sale of any goods or services, its application is to commercial speech.” Id. (alterations omitted).
In short, many trademarks lack the kind of “expressive character” that would merit First Amendment protection for offensive content, and a regulation of the use of those marks could satisfy the Central Hudson test for commercial speech—a substantial government interest reflected in a narrowly tailored regulation. The majority’s contrary conclusion seems to me to be unsupported.
Even if disparaging commercial speech were protected from government ban or regulation, this case does not turn on the legitimacy of a regulation or a “blanket ban” on disparaging commercial speech. The refusal to register disparaging marks is not a regulation or “blanket ban” on anything. Rather, it involves the denial of a subsidy, and because it is a subsidy, it may be content based. It is “well established that the government can make content-based distinctions when it subsidizes speech.” Davenport v. Wash. Educ. Ass’n, 551 U.S. 177, 188–89, 127 S.Ct. 2372, 168 L.Ed.2d 71 (2007). The First Amendment “does not confer an affirmative right to use government [ ] mechanisms for the purpose of” expression, nor is the government “required to assist others in funding the expression of particular ideas, including political ones.” Ysursa v. Pocatello Educ. Ass’n, 555 U.S. 353, 355, 358, 129 S.Ct. 1093, 172 L.Ed.2d 770 (2009) (internal quotations and citations omitted). Significantly, every single Supreme Court decision upholding the protection of commercial speech has involved a prohibition or restriction of speech—not a subsidy.4
That trademark registration is a subsidy is not open to doubt. Contrary to the majority’s characterization, federal trademark registration is not a “regulatory regime.” Maj. Op. at 1353. Section 2(a) does not regulate any speech, much less impose a blanket ban. It merely deprives a benefit. The majority claims that federal trademark registration is not a subsidy because “the subsidy cases have all involved government funding or government property.” Maj. Op. at 1351. But this assertion is belied by the Court’s recent decisions in Davenport and Ysursa—neither involving government funding or property. Each made clear that the government can make content-based distinctions when it provides a benefit.
*1369 In Davenport, the Court considered a government benefit that gave unions “the power, in essence, to tax government employees,” by having the state collect fees from its employees on behalf of the unions. Davenport, 551 U.S. at 184, 127 S.Ct. 2372. The state limited this collection mechanism by refusing to collect nonmember fees for election-related purposes unless the nonmember affirmatively consented. Id. at 180, 127 S.Ct. 2372. The unions argued that this restriction was an unconstitutional content-based discrimination. Id. at 188, 127 S.Ct. 2372. The Court disagreed. The First Amendment’s usual aversion to content-based speech regulation is inapposite when “the government is acting in a capacity other than as regulator,” such as “when it subsidizes speech.” Id. at 188, 127 S.Ct. 2372. Because the collection of nonmember fees was a “state-bestowed entitlement,” “a matter of grace [that] [it] can, of course, disallow ... as it chooses,” Regan v. Taxation With Representation of Wash., 461 U.S. 540, 549, 103 S.Ct. 1997, 76 L.Ed.2d 129 (1983) (internal quotations and citations omitted), the content-based condition on that benefit did not raise a “realistic possibility that official suppression of ideas is afoot.” Davenport, 551 U.S. at 189–90, 127 S.Ct. 2372 (citations and quotation marks omitted). The unions remained “as free as any other entity to participate in the electoral process with all available funds other than the state-coerced agency fees.” Id. at 190, 127 S.Ct. 2372. Thus the Court declined to apply heightened scrutiny and upheld the restriction in light of the state’s “narrow” and legitimate interest in “protect[ing] the integrity of the election process.” See id. at 189–90, 127 S.Ct. 2372.
In Ysursa, the Court considered a similar benefit where the state collected dues on behalf of unions by providing payroll deductions. Ysursa, 555 U.S. at 355, 129 S.Ct. 1093. The state restricted that collection mechanism by preventing unions from using payroll deductions for any political purposes. Id. Again the unions argued that this restriction was an impermissible content-based speech restriction, and again the Court disagreed. The First Amendment “protects the right to be free from government abridgement of speech,” not the right to be “assist[ed] [ ] in funding the expression of particular ideas.” Id. at 358, 129 S.Ct. 1093. “While publicly administered payroll deductions for political purposes can enhance the unions’ exercise of First Amendment rights, Idaho is under no obligation to aid the unions in their political activities.” Id. at 359, 129 S.Ct. 1093. Because collecting payroll deductions was a government benefit, the State’s decision not to extend that benefit was “not an abridgement of the unions’ speech.” Id. As in Davenport, the unions remained “free to engage in such speech as they see fit. They simply are barred from enlisting the State in support of that endeavor.” Id. Thus the Court again declined to apply heightened scrutiny and upheld the regulation in light of the “government’s interest” in “avoiding the reality or appearance of government favoritism.” Id.
The same is true here. Federal trademark registration, like the state-bestowed collection mechanisms for unions in Davenport and Ysursa, is a government-bestowed collection mechanism for enforcing trademarks. It opens the federal courts to enforce trademark rights by providing, inter alia, original jurisdiction in federal courts for infringement claims, eligibility for treble damages for willful infringement, the ability to petition Customs to prevent the importation of infringing articles, and various enhanced protections for marks. See 15 U.S.C. §§ 1057(c), 1141, 1117, 1124. These benefits all “enlist” the government in support of the mark holder’s commercial identification, much like the collection of nonmember fees in Davenport *1370 and the payroll deductions in Ysursa enlisted the states in support of the unions’ political speech. See Ysursa, 555 U.S. at 359, 129 S.Ct. 1093. Just as the states were not obligated to enable labor unions to collect nonmember fees or take payroll deductions in the first place, the federal government is not obligated to provide these benefits of a trademark enforcement mechanism. And just as the unions remained free to speak for election-related purposes using all other funds, trademark holders remain free to use their marks—however disparaging—as far as the federal government is concerned.5 That states may deny state-law protection to these marks cannot make the denial of the federal subsidy any less constitutional.
Finally, the majority argues that § 2(a) should be treated as a regulatory provision because the denial of registration benefits will have a chilling effect on the use of disparaging marks and cause mark holders to abandon such marks. See Maj. Op. at 1342-43. But that is commonly the effect of the denial of subsidies, as the Supreme Court has recognized. See Regan, 461 U.S. at 550, 103 S.Ct. 1997 (“Although TWR does not have as much money as it wants, and thus cannot exercise its freedom of speech as much as it would like,” the decision not to subsidize its speech does not violate the First Amendment). A chilling effect does not turn a subsidy provision into a regulatory provision, so long as the subsidy is not designed to limit speech outside of the subsidized program. That is not the case here.
“[T]he relevant distinction that has emerged from our cases is between conditions that define the limits of the government spending program—those that specify the activities Congress wants to subsidize—and conditions that seek to leverage funding to regulate speech outside the contours of the program itself.” Agency for Int’l Dev. v. Alliance for Open Soc’y Int’l, Inc., –––U.S. ––––, 133 S.Ct. 2321, 2328, 186 L.Ed.2d 398 (2013) (“AID ”). An example of such impermissible leverage was found in FCC v. League of Women Voters, where federal funds were denied to public broadcasters if they engaged in editorializing. 468 U.S. 364, 399–401, 104 S.Ct. 3106, 82 L.Ed.2d 278 (1984). The restriction was invalidated because it affected editorializing engaged in without federal funds. Id. Section 2(a) is not designed to limit speech outside of the federal trademark program. Accordingly, it does not run afoul of the unconstitutional conditions doctrine.6 See id.
The majority’s contrary arguments are the very arguments rejected in the Supreme Court’s recent decision in AID. See 133 S.Ct. at 2328. AID explicitly disclaimed the majority’s assertion that the condition must be limited to “advancing the goals underlying the program the government seeks to fund.” Maj. Op. at 1354. The question is not whether “the condition is [ ] relevant to the objectives of the program,” but rather whether the condition “seek[s] to leverage funding to regulate speech outside the contours of the program *1371 itself,” which the restriction here does not. AID, 133 S.Ct. at 2328. Similarly, in Regan the Court upheld a requirement that nonprofit organizations seeking tax-exempt status under 26 U.S.C. § 501(c)(3) not engage in lobbying. 461 U.S. at 544, 103 S.Ct. 1997. The Court upheld that condition not because it was related in some way to the “goals” of 501(c)(3) tax exemption, but rather because “the condition did not prohibit that organization from lobbying Congress” with separate funds, i.e., it did not leverage funds outside of the nonprofit structure. Id. at 2329. The majority’s arguments fail to show a colorable violation of the unconstitutional conditions doctrine here.
The majority urges, however, that subsidies require viewpoint neutrality, and argues that the subsidy provided by § 2(a) discriminates based on viewpoint because favorable racial and other marks are allowed while disparaging ones are not. See Maj. Op. at 1336-38. Contrary to the majority, the Supreme Court has never held that this kind of subsidy must be viewpoint neutral. The question was raised, but not answered, in Davenport and Ysursa. See Davenport, 551 U.S. at 189, 127 S.Ct. 2372 (“Even if it be thought necessary that the content limitation be reasonable and viewpoint neutral ...”); Ysursa, 555 U.S. at 361, n. 3, 129 S.Ct. 1093. And the Court has upheld subsidies that were facially viewpoint discriminatory. See, e.g., Rust v. Sullivan, 500 U.S. 173, 111 S.Ct. 1759, 114 L.Ed.2d 233 (1991) (upholding a condition limiting Title X funding to clinics that do not advocate abortion as a method of family planning). The Court made an exception in a subsidy case involving the unique context of legal services, where “the traditional role of the [subsidized] attorneys” is to “speak[ ] on the behalf of his or her private, indigent client” and viewpoint discrimination undermined the very purpose of the subsidy. Legal Servs. Corp. v. Velazquez, 531 U.S. 533, 542, 544, 121 S.Ct. 1043, 149 L.Ed.2d 63 (2001). There is no tradition of unfettered advocacy in commercial advertising. Thus even if the regulation here could be deemed viewpoint discriminatory, it would not fail under the First Amendment. See Davenport, 551 U.S. at 189, 127 S.Ct. 2372.
But § 2(a) is in any event viewpoint neutral. In Boos v. Barry, 485 U.S. 312, 108 S.Ct. 1157, 99 L.Ed.2d 333 (1988), the Court addressed a nearly identical standard as applied to core political speech. The law there prohibited the display of any sign within 500 feet of a foreign embassy if the sign would tend to bring that foreign government into “public odium” or “disrepute.” Id. at 315, 108 S.Ct. 1157. Justice O’Connor’s plurality opinion confirmed that the restriction is “content-based,” but it specifically found that “the provision is not viewpoint based.” Id. at 319, 108 S.Ct. 1157 (emphasis added). “The display clause determines which viewpoint is acceptable in a neutral fashion by looking to the policies of foreign governments.” Id. (emphasis added). This “prevents the display clause from being directly viewpoint based, a label with potential First Amendment ramifications of its own.” Id. This aspect of the plurality opinion has since been cited with approval by a majority of the Court in Turner Broadcasting System, Inc. v. FCC, 512 U.S. 622, 645, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994). The same reasoning applies here. Just as the restriction in Boos operated in a “neutral fashion” by looking only to foreign governments, the bar on registration of disparaging marks operates in a “neutral fashion” by looking only to the views of the referenced group. Accordingly, just as the restriction in Boos was viewpoint neutral, so too is § 2(a). In Ridley v. Massachusetts Bay Transportation Authority, 390 F.3d 65 (1st Cir.2004), *1372 the First Circuit arrived at the same conclusion, holding that a regulation “prohibit[ing] demeaning or disparaging ads” was viewpoint neutral because “the state is not attempting to give one group an advantage over another in the marketplace of ideas.” Id. at 90–91.
Finding § 2(a) to be viewpoint neutral is consistent with the Court’s treatment of viewpoint discrimination in other areas. The Court has defined viewpoint discrimination as the government’s disagreement with the underlying “ideology,” “opinion” or “perspective of the speaker.” Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819, 829, 115 S.Ct. 2510, 132 L.Ed.2d 700 (1995). Here, as in Boos, the standard is not based on the government’s disagreement with anything. Rather, it is based on an objective, “neutral” assessment of a non-government perspective—in this case, a “substantial composite of the referenced group.” As in Davenport and Ysursa, there is no “realistic possibility that official suppression of ideas is afoot,” Davenport, 551 U.S. at 190, 127 S.Ct. 2372 and the content-based regulation here is not subject to heightened First Amendment scrutiny.
Even in subsidy cases, however, the government needs some interest sufficient to justify its regulation defined in terms of “reasonableness.” See Ysursa, 555 U.S. at 359, 129 S.Ct. 1093; Regan, 461 U.S. at 550, 103 S.Ct. 1997. In my view, the protection of disparaged groups is sufficient. As demonstrated on college campuses across the nation, members of some groups, whether or not justified, are particularly sensitive to disparaging material.7 There is significant social science evidence demonstrating the harmful psychological effects of holding a minority group up for ridicule on a national stage, particularly on children and young adults.8 In the case of core protected speech, as discussed above, the government has no legitimate interest in protecting disparaged groups. The groups must tolerate the disparagement in pursuit of the greater goal of a free marketplace of ideas. But, as discussed above, commercial speech is different. Disparagement as defined by the Board “is essentially a violation of one’s right of privacy—the right to be let alone from contempt or ridicule.” TMEP § 1203.03(c).
The government has an interest in “proscribing intrusive and unpleasant formats” for commercial expression. Taxpayers for Vincent, 466 U.S. at 806, 104 S.Ct. 2118; see also Lehman, 418 U.S. at 304, 94 S.Ct. 2714; Metromedia, 453 U.S. at 514, 101 S.Ct. 2882. The Supreme Court’s “precedents [ ] leave no room for doubt that the protection of potential clients’ privacy is a substantial state interest.” Florida Bar, 515 U.S. at 625, 115 S.Ct. 2371 (internal quotations marks omitted). We need not decide whether this interest is sufficiently compelling to justify a ban of disparaging commercial speech. It is more than sufficient to justify the government’s “decision not to assist” disparaging commercial expression. *1373 Ysursa, 555 U.S. at 360 n. 2, 129 S.Ct. 1093; Taxpayers for Vincent, 466 U.S. at 806, 104 S.Ct. 2118. At the same time, there is no countervailing First Amendment interest. It is certainly difficult to imagine, for example, how the disparaging elements of an advertisement such as “CHLORINOL SODA BLEACHING—we are going to use Chlorinol and be like de white nigger,”9 or “The Plucky Little Jap Shredded Wheat Biscuit,”10 or “Dr. Scott’s Electric Hair Brush—will not save an Indian’s scalp from his enemies but it will preserve yours from dandruff,”11 further any legitimate “informational function” associated with the relevant product.
Finally, contrary to the majority’s implication, it is quite feasible to distinguish between core and commercial speech. Congress has already determined that trademark law should distinguish between pure commercial speech and fully protected speech. Section 1125(c)(3) of title 15 excludes from liability for dilution parody, criticism, and any noncommercial use of a mark. And the noncommercial use of a mark, for parody, as an example, weighs against likelihood of confusion. See, e.g., Rogers v. Grimaldi, 875 F.2d 994 (2d Cir.1989); Davis v. Walt Disney Co., 430 F.3d 901 (8th Cir.2005); see also Cliffs Notes, Inc. v. Bantam Doubleday Dell Publ’g Group, Inc., 886 F.2d 490, 494–95 (2d Cir.1989) (“the expressive elements of titles require [ ] more protection than the labeling of ordinary commercial products ... so here the expressive element of parodies requires more protection than the labeling of ordinary products.”). Congress has made a similar judgment in the copyright context. See 17 U.S.C. § 107 (one of four fair use factors includes assessing whether the use is commercial). I see no reason why the Board would be unable to make such distinctions here.
Turning from the application of § 2(a) to commercial speech to the facts of this case, I agree with the majority that the bar on registration of disparaging marks is unconstitutional as applied to Mr. Tam. Here there can be no doubt that Mr. Tam’s speech is both political and commercial. Unlike Friedman, where the trade name proponent did “not wish to editorialize on any subject, cultural, philosophical, or political,” 440 U.S. at 11, 99 S.Ct. 887 Mr. Tam’s choice of mark reflects a clear desire to editorialize on cultural and political subjects. Mr. Tam chose THE SLANTS at least in part to reclaim the negative racial stereotype it embodies: “We want to take on these stereotypes that people have about us, like the slanted eyes, and own them. We’re very proud of being Asian—we’re not going to hide that fact.” In re Simon Shiao Tam, 108 U.S.P.Q.2d 1305, 2013 WL 5498164 at *2 (T.T.A.B.2013). See Maj. Op. at 1332 (Mr. Tam “selected the mark in order to ‘own’ the stereotype it represents.”).
Given the indisputably expressive character of Mr. Tam’s trademark in this case, the government’s recognized interests in protecting citizens from targeted, demeaning advertising and proscribing intrusive formats of commercial expression—interests that are sufficient to justify the provision *1374 as applied to commercial speech—are insufficient to justify application of the provision to Mr. Tam. As discussed, because of the fundamental values underlying the First Amendment’s robust protection of offensive speech that are unique to core political expression, the government cannot justify restricting disparaging trademarks when those marks, like Mr. Tam’s, actually consist of core expression. See, e.g., Snyder, 562 U.S. at 459–61, 131 S.Ct. 1207. Accordingly, because no government interest can justify restricting Mr. Tam’s core speech on the basis of its capacity to injure others, § 2(a) is invalid as applied. This also explains why the majority’s concern regarding copyright is misplaced. See, e.g., Maj. Op. at 1354–55. Copyrights, unlike trademarks, principally cover core protected expression. Thus, as for Mr. Tam, any government interest related to suppressing offensive speech would be insufficient to justify a comparable restriction as applied to copyright registration except for commercial advertising.
No case before the majority’s opinion today has imposed an obligation on the government to subsidize offensive, commercial speech. As Judge Lourie points out, the bar on registration of disparaging marks is longstanding, and we have previously upheld it in a number of decisions. I see no basis for invalidating it now as applied to commercial speech. I would adhere to those decisions in this respect, and I respectfully dissent.
LOURIE, Circuit Judge, dissenting.
I join Parts I–IV of Judge Dyk’s concurrence-in-part, dissent-in-part, but I respectfully dissent with respect to the result reached by the majority holding the disparagement provision of § 2(a) unconstitutional as violating the First Amendment. For the following additional reasons, I would affirm the USPTO’s decision refusing to register Mr. Tam’s trademark.
First, one wonders why a statute that dates back nearly seventy years—one that has been continuously applied—is suddenly unconstitutional as violating the First Amendment. Is there no such thing as settled law, normally referred to as stare decisis? Since the inception of the federal trademark registration program in 1905, the federal government has declined to issue registrations of disparaging marks. The Trademark Act of 1905 provided specific authority to refuse to register immoral or scandalous marks, see Act of Feb. 20, 1905, ch. 592, 33 Stat. 724; the USPTO refused to register disparaging marks on those grounds before the Lanham Act of 1946 was enacted, which explicitly incorporated a disparagement proscription, see Appellee’s En Banc Br. 6. The USPTO’s authority to refuse to issue trademark registrations with certain offensive content has thus existed in U.S. law for over one hundred years. As the majority notes, these are not prohibitions that have lain unused and latent for all of those years. The USPTO has been rejecting applications for trademark registrations on this basis throughout this period of time. By finding § 2(a) unconstitutional, we interfere with the long-standing Congressional policy of delegating authority to the USPTO to filter out certain undesirable marks from the federal trademark registration system. We should not further the degradation of civil discourse by overturning our precedent that holds that the First Amendment is not implicated by § 2(a)’s prohibition against disparaging trademarks.
In addition, the refusal of the USPTO to register a trademark is not a denial of an applicant’s right of free speech. The markholder may still generally use the mark as it wishes; without federal registration, it simply lacks access to certain federal statutory enforcement mechanisms for excluding others from confusingly similar *1375 uses of the mark. Mr. Tam may use his trademark as he likes, whether it be encouraging discussion on or taking ownership of racial slurs, or identifying goods and services with his band. In fact, it seems quite likely that Mr. Tam will continue to use his band name to make a statement regardless of federal registration—the expressive purpose of his mark undoubtedly overshadows the commercial considerations. The argument, therefore, that a trademark applicant’s right of free speech has been impaired by the failure of the USPTO to grant a federal registration is unconvincing.
Furthermore, it is not entirely clear that a trademark, even an expressive trademark, is protected commercial speech. The lack of a federal registration does not alter the informational function of a trademark: disparaging marks may still be used to identify the source of goods or services. The government’s decision to support certain choices and not others will invariably have some discouraging effect, but the government does not necessarily violate an individual’s constitutional rights merely by refusing to grant registration and thereby provide additional assistance in the enforcement of trademark rights.
Moreover, trademark rights, as amicus International Trademark Association informs us, are not limited to those marks deemed registrable by the USPTO. “Section 43(a) of the Lanham Act is available to protect all designations of origin, even—indeed, especially—those that cannot be registered under Section 2(a).” Br. of amicus curiae Int’l. Trademark Ass’n 4. The fear that mark-holders would be left with absolutely no recourse for trademark protection, once an application for federal registration is denied, appears unfounded. Rather, all that is at issue here is the government’s decision not to facilitate enforcement with the additional mechanisms attendant to federal registration. The denial of federal trademark registration thus does not deprive the mark-holder of trademark protection because of the content of its mark; the markholder still has trademark rights under the Act in addition to its common law rights.
Finally, it has been questioned whether federal registration imparts the “imprimatur” of the federal government on a mark, such that registration could be permissibly restricted as government speech. I believe that such action is justified. The USPTO does in fact “publish” trademarks, in the Trademark Official Gazette. Despite being in electronic form, it is still a form of government speech that is partially controlled or affected by government action. The USPTO may also require that a disclaimer of unregistrable components be included for publication. Moreover, a federally registered mark is usually “stamped” with some indication of government oversight, viz., the use of the ® symbol or a phrase that the mark is registered in the USPTO, giving proof to the public that the government has in some sense approved the mark. Without that designation, the markholder cannot take advantage of some of the benefits of federal registration, e.g., constructive notice for damages.
Similarly to specialty license plate designs, federally registered trademarks can be identified with two message contexts: one from the provider of goods or services, who has chosen to use a certain mark to link its product or services to itself, and one from the government, which has deemed the mark qualified for the federal registration program. The evaluation of disparagement is not based on the government’s moral judgment, despite any distaste expressed in its briefing for cancelled or applied-for marks; a mark is disqualified based only on evidence of its perception by the affected persons. The government action does not include a judgment *1376 on the worthiness or the effectiveness of the mark; if it did, it might—but not necessarily—venture into viewpoint-discrimination territory. And while a trademark alone, as a word placed on private property, is not government speech, once it claims that federally registered status, it becomes more than the private owner’s speech. It is not simply private speech as is the holding of a placard in a parade.
In my view, holding the disparagement provision of § 2(a) unconstitutional would be unsound, and the USPTO’s refusal to register Mr. Tam’s disparaging mark should therefore be affirmed.
Accordingly, I respectfully dissent.
REYNA, Circuit Judge, dissenting.
The Majority holds today that Mr. Tam’s speech, which disparages those of Asian descent, is valuable political speech that the government may not regulate except to ban its use in commerce by everyone but Mr. Tam. I believe the refusal to register disparaging marks under § 2(a) of the Lanham Act is an appropriate regulation that directly advances the government’s substantial interest in the orderly flow of commerce. Because I would uphold the constitutionality of § 2(a), I respectfully dissent.
Trademarks are commercial speech. And precisely because trademarks are commercial speech, the government’s decision to grant or deny registration must be reviewed under an intermediate standard of scrutiny. Intermediate scrutiny is satisfied whenever the decision is narrowly tailored to directly advance a substantial government interest. When the commercial or political content of a trademark threatens the government’s substantial interest in the orderly flow of commerce, appropriate regulation may be justified.
A. Intermediate Scrutiny Applies Because Trademarks Are Commercial Speech
The Supreme Court has held that trademarks are “a form of commercial speech and nothing more.” Friedman v. Rogers, 440 U.S. 1, 11, 99 S.Ct. 887, 59 L.Ed.2d 100 (1979); accord San Francisco Arts & Athletics, Inc. v. U.S. Olympic Comm., 483 U.S. 522, 563, 107 S.Ct. 2971, 97 L.Ed.2d 427 (1987). The purpose of a trademark is merely to “propos[e] a commercial transaction” by identifying the source of goods or services. Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n of New York, 447 U.S. 557, 562, 100 S.Ct. 2343, 65 L.Ed.2d 341 (1980).
Because “the Constitution accords less protection to commercial speech than to other constitutionally safeguarded forms of expression,” Bolger v. Youngs Drug Products Corp., 463 U.S. 60, 64–65, 103 S.Ct. 2875, 77 L.Ed.2d 469 (1983), the government may regulate the use of trademarks to ensure the orderly flow of commerce. For example, the government may disallow trade names that create “[t]he possibilities for deception,” even if the names are not untruthful. Friedman, 440 U.S. at 13, 99 S.Ct. 887. The government may similarly implement a trademark registration program, as it did through the Lanham Act, which provides certain speakers exclusive rights to their chosen marks in commerce. Such regulation is permissible under the First Amendment only because the speech being regulated is commercial and because the government has a substantial interest in facilitating commerce by “insuring that the stream of commercial information flows cleanly as well as freely.” Virginia State Bd. of Pharmacy v. Virginia Citizens Consumer Council, Inc., 425 U.S. 748, 771–72, 96 S.Ct. 1817, 48 L.Ed.2d 346 (1976).
*1377 The courts have long recognized that some trademarks can include expressive elements concerning matters of public interest, and that such trademarks nevertheless remain commercial speech. Historically, commercial speech received no First Amendment protection, see Valentine v. Chrestensen, 316 U.S. 52, 54, 62 S.Ct. 920, 86 L.Ed. 1262 (1942), and the seminal cases bringing commercial speech within the First Amendment’s purview did so, at least in part, because commercial speech often communicates on matters of public interest. Virginia State Bd., 425 U.S. at 764–65, 96 S.Ct. 1817. As the Supreme Court recognized in Virginia State Board, “not all commercial messages contain the same or even a very great public interest element,” but “[t]here are few to which such an element, however, could not be added.” Id.
The protections of commercial speech are therefore based, at least in part, on the recognition that commercial speech is not always entirely commercial, but that it may contain political messages that make the speech “ ‘commercial’ in widely varying degrees.” Bigelow v. Virginia, 421 U.S. 809, 826, 95 S.Ct. 2222, 44 L.Ed.2d 600 (1975). For this reason, the Supreme Court has routinely held that various examples of speech “constitute commercial speech notwithstanding the fact that they contain discussions of important public issues.” Bolger, 463 U.S. at 67, 103 S.Ct. 2875; see also Bd. of Trustees of State Univ. of New York v. Fox, 492 U.S. 469, 109 S.Ct. 3028, 106 L.Ed.2d 388 (1989). Put simply, commercial speech does not transform into core political speech with full First Amendment protections simply because it “links a product to a current public debate.” Cent. Hudson, 447 U.S. at 563, 100 S.Ct. 2343.
To determine whether speech is commercial, we consider “the nature of the speech taken as a whole.” Riley v. Nat’l Fed’n of the Blind, 487 U.S. 781, 796, 108 S.Ct. 2667, 101 L.Ed.2d 669 (1988). For example, in Bolger, the Supreme Court found that certain pamphlets were commercial speech, despite containing “discussions of important public issues,” because (1) the speaker conceded that the pamphlets were advertisements, (2) the pamphlets referenced a specific product, and (3) the speaker had an economic motivation for mailing the pamphlets. Bolger, 463 U.S. at 66–68, 103 S.Ct. 2875. The Court concluded that “[t]he combination of all these characteristics” supported the conclusion that “the informational pamphlets are properly characterized as commercial speech.” Id.
All three factors from Bolger are necessarily also present in trademarks. Trademarks are used to identify specific products and to advertise the sources of those products. Trademarks, and in particular those federally registered for exclusive use in interstate commerce, are necessarily tools of commerce used with an “economic motive.”1 A trademark is therefore commercial speech, and as such, it lacks full First Amendment protections, regardless of whether it also includes a political element.
The Majority reasons that because the commercial and political elements of trademarks are “inextricably intertwined,” the combined whole must be treated as expressive speech. Maj. Op. at 1339 (citing Riley, 487 U.S. at 796, 108 S.Ct. 2667). But as explained above, commercial speech *1378 is frequently intertwined with political elements, and this intertwining does not necessarily alter the essentially commercial character of the speech. Riley, on which the Majority relies, is not to the contrary. Riley only reiterates that “in deciding what level of scrutiny to apply” we must consider “the nature of the speech taken as a whole.” Riley, 487 U.S. at 796, 108 S.Ct. 2667. The nature of trademarks seeking federal registration for use in interstate commerce, when considered as a whole, is indisputably commercial, not political.
Judge Dyk concurs in the result today only because he believes the content of Mr. Tam’s mark is so “indisputably expressive” that it cannot be regulated under the lesser standards applied to commercial speech. Dyk, J., concurring at 1373-74. But if the expressive content of the mark precludes regulation, on what authority may the government grant Mr. Tam the exclusive right to use this mark in commerce? Whatever standard of scrutiny protects the content of Mr. Tam’s trademark from government regulation, that same standard must necessarily be overcome by the government’s substantial interest in the orderly flow of commerce, or no trademark could issue.
B. Intermediate Scrutiny Applies Because Section 2(a) is Content–Neutral
The Majority applies strict scrutiny not necessarily because of the expressive content of Mr. Tam’s mark, but because of the government’s supposed purpose of suppressing the political elements of the mark. Maj. Op. at 1337-39. The Majority thus invokes the modern test for content-neutrality, under which the “principal inquiry” is “whether the government has adopted a regulation of speech because of disagreement with the message it conveys.” Ward v. Rock Against Racism, 491 U.S. 781, 791, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989). Under Ward, “[t]he government’s purpose is the controlling consideration.” Id. The Supreme Court has endorsed the applicability of this test to commercial speech. Sorrell v. IMS Health Inc., ––– U.S. ––––, 131 S.Ct. 2653, 2664, 180 L.Ed.2d 544 (2011).
If this appeal turns on a content-neutrality analysis, we should be clear that the government has never stated that the purpose of § 2(a) is to suppress speech. Only the Majority has advanced this rationale, and it has done so only by default after eliminating all other interests of which it could conceive. I do not think we need to search so hard and so far. The purpose of § 2(a) is the same as the purpose of the Lanham Act as a whole—to promote the orderly flow of commerce.
The Lanham Act declares unequivocally that “[t]he intent of this chapter is to regulate commerce.” 15 U.S.C.A. § 1127. In analyzing content-neutrality, an apparently content-based law is nevertheless considered content-neutral if the government’s purpose is not to suppress speech, but to address the harmful secondary effects of that speech. See City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 106 S.Ct. 925, 89 L.Ed.2d 29 (1986); Young v. Am. Mini Theatres, 427 U.S. 50, 96 S.Ct. 2440, 49 L.Ed.2d 310 (1976). The Supreme Court has repeatedly applied this “Secondary Effects” doctrine to uphold not only time, place, and manner restrictions on particular types of speech, id. (upholding regulations on the locations of adult businesses), but also regulations on the content of expression itself, see, e.g., City of Erie v. Pap’s A.M., 529 U.S. 277, 120 S.Ct. 1382, 146 L.Ed.2d 265 (2000) (upholding ban on fully nude dancing); Barnes v. Glen Theatre, 501 U.S. 560, 111 S.Ct. 2456, 115 L.Ed.2d 504 (1991) (same). For example, applying Ward, the Supreme Court upheld a city’s ban on fully nude dancing *1379 because the ban was only a minimal burden on speech and was narrowly tailored to advance the “substantial government interest in protecting order and morality.” Barnes, 501 U.S. at 569, 111 S.Ct. 2456. In City of Erie, the Court upheld a nearly identical statute as content-neutral because it did “not attempt to regulate the primary effects of the expression” but rather, “the secondary effects, such as impacts on public health, safety, and welfare.” City of Erie, 529 U.S. at 291, 120 S.Ct. 1382.
The Supreme Court has also permitted regulation of speech based on the speech’s effect on commerce. For instance, it was under Ward that the Supreme Court upheld the FCC’s must-carry provisions as content-neutral, despite the provisions’ mandate that cable providers transmit particular types of content. Turner Broad. Sys., Inc. v. F.C.C., 512 U.S. 622, 647, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994). The Court upheld the must-carry regulations because they furthered the substantial government interest in “protecting non-cable households from loss of regular television broadcasting service.” Id. The Court has also upheld regulations on highly-protected private speech where the government sought to eliminate the secondary effects of that speech on the market for illegal goods. See Osborne v. Ohio, 495 U.S. 103, 110 S.Ct. 1691, 109 L.Ed.2d 98 (1990). Thus, when a regulation’s purpose is to address the secondary effects of certain speech, intermediate scrutiny is appropriate, even if the regulation implicates content.
Section 2(a) serves the same substantial government interest as the Lanham Act as a whole—the orderly flow of commerce. Commercial speech that insults groups of people, particularly based on their race, gender, religion, or other demographic identity, tends to disrupt commercial activity and to undermine the stability of the marketplace in much the same manner as discriminatory conduct. The government’s refusal to promote such speech in commerce is not an effort to suppress free expression, but to mitigate the disruptive secondary effects that a particular type of low-value speech may have when used in a commercial context. Because the government’s purpose is to mitigate these secondary effects on commerce rather than to suppress speech, the regulation is content-neutral and intermediate scrutiny applies.
C. Section 2(a) Advances the Substantial Government Interest in the Orderly Flow of Commerce
The government’s interest in the orderly flow of commerce is substantial. If it were not, the government would be powerless to implement a trademark registry because doing so necessarily requires a ban on infringing commercial speech. The government has a substantial interest in regulating “deceptive or misleading” commercial speech, even if that speech is not wholly false, because of the government’s substantial interest in “insuring that the stream of commercial information flow cleanly as well as freely.” Virginia State Bd. of Pharmacy, 425 U.S. at 771, 96 S.Ct. 1817. The Supreme Court has never held, however, that deceptive and misleading speech is the only type of commercial speech subject to regulation for its disruptive effect. See Cent. Hudson, 447 U.S. at 566, 100 S.Ct. 2343 (“For commercial speech to come within that provision, it at least must concern lawful activity and not be misleading.”) (emphasis added). Instead, any speech that substantially undermines the orderly flow of commerce may potentially be subject to at least some regulation.
The marketplace of ideas differs dramatically from the marketplace of goods and services. While the marketplace of ideas *1380 may tolerate or even benefit from the volatility that accompanies disparaging and insulting speech, the marketplace of goods and services is a wholly different animal. Commerce does not benefit from political volatility, nor from insults, discrimination, or bigotry. Commerce is a communal institution regulated for the mutual economic benefit of all. Commercial speech that discredits or brings reproach upon groups of Americans, particularly based on their race, has a discriminatory impact that undermines commercial activity and the stability of the marketplace in much the same manner as discriminatory conduct.
That discriminatory conduct disrupts commerce is long established. In upholding Title II of the Civil Rights Act, for example, the Supreme Court noted a record “replete with testimony of the burdens placed on interstate commerce by racial discrimination.” Katzenbach v. McClung, 379 U.S. 294, 299, 85 S.Ct. 377, 13 L.Ed.2d 290 (1964). The Court cited an “impressive array of testimony that discrimination in restaurants had a direct and highly restrictive effect upon interstate travel,” and that such discrimination therefore “obstructs interstate commerce.” Id. at 300, 85 S.Ct. 377. It cited “many references” to discrimination causing “a depressant effect on general business conditions in the respective communities” and it noted evidence that discrimination “deterred professional, as well as skilled, people from moving into areas where such practices occurred and thereby caused industry to be reluctant to establish there.” Id. The Court thus found “ample basis for the conclusion that established restaurants in such areas sold less interstate goods because of the discrimination, that interstate travel was obstructed directly by it, that business in general suffered and that many new businesses refrained from establishing there as a result of it.” Id.
Although these findings were specific to public accommodations, they are applicable to commerce generally. Commercial goods and services pervade all economic channels, including all public accommodations, such as stores, restaurants, hotels, theaters, and the like. Discriminatory messages within such commercial channels threaten the same disruptive effects as the discrimination itself. Although the Majority distinguishes between conduct and speech, Maj. Op. at 1356-57, the distinction is without a difference in this context. Whether a restaurant named “SPICS NOT WELCOME” would actually serve a Hispanic patron is hardly the point. The mere use of the demeaning mark in commerce communicates a discriminatory intent as harmful as the fruit produced by the discriminatory conduct.
Because even speech without accompanying conduct can have a discriminatory impact, other parts of the Civil Rights Act expressly regulate pure speech in commerce. For instance, Title VIII specifically bans advertising that indicates a discriminatory preference, even where discriminatory conduct is legal. See 42 U.S.C. § 3604(c); see also § 3603(b) (listing exemptions). Title VII places similar restrictions on job advertisements. See 42 U.S.C. § 2000e–3(b). Title VII also bans pure speech in the workplace when the speech is harassing, even when unaccompanied by any adverse employment action, because such speech creates a discriminatory impact. See Harris v. Forklift Sys., Inc., 510 U.S. 17, 114 S.Ct. 367, 126 L.Ed.2d 295 (1993); see also Burlington Indus., Inc. v. Ellerth, 524 U.S. 742, 118 S.Ct. 2257, 141 L.Ed.2d 633 (1998).
Nearly every disparaging mark identified in the voluminous briefing and opinions in this case has involved disparagement of race, gender, ethnicity, national origin, religion, sexual orientation, and *1381 similar demographic classification. The impact of advancing these bigoted messages through the ubiquitous channels of commerce may be discriminatory, and even if not discriminatory, at least disruptive to commerce. The only question is whether the government’s interest in avoiding this commercial disruption outweighs the modest “burden” that its refusal to register the offending marks places on the freedom of speech. I believe it does.
D. Section 2(a) Survives Intermediate Scrutiny
To be clear, I do not believe that the government may ban any speech it finds commercially undesirable, but only that when we are presented with a regulation, we must engage meaningfully in “the task of assessing the First Amendment interest at stake and weighing it against the public interest allegedly served by the regulation.” Bigelow, 421 U.S. at 826, 95 S.Ct. 2222. Here, the government’s substantial interest in the orderly flow of commerce is counterbalanced only by a minimal “burden” on a small subset of low-value commercial speech. Section 2(a) should survive intermediate scrutiny because it is only an “incidental restriction on First Amendment freedom [that] is no greater than is essential to the furtherance of the governmental interest” in the orderly flow of commerce. See Barnes, 501 U.S. at 561, 111 S.Ct. 2456.
Section 2(a) imposes only a modest “burden” on speech. First, the statute applies only in the commercial context, meaning that it does nothing to impact private speech. Mr. Tam remains free to spread his chosen message to all who would listen without fear of government intervention or reprisal. Second, § 2(a) does not strictly “burden” Mr. Tam’s speech, but only denies him a government-created benefit—the exclusive right to use that speech in commerce in connection with the sale of particular goods or services. At bottom, the only burden the application of § 2(a) imposes in this case is that Mr. Tam is free to communicate his chosen message within or without commerce, so long as he is willing to permit others to do the same.
Section 2(a) also implicates only a modest sliver of particularly low-value speech. Speech that disparages is a narrow subset of speech that offends, and it is a particularly low-value subset at that. See Am. Freedom Def. Initiative v. Mass. Bay Transp. Auth., 989 F.Supp.2d 182, 192 (D.Mass.2013) aff’d, 781 F.3d 571 (1st Cir.2015) (distinguishing speech that “crosses the line from being offensive or hurtful to being demeaning or disparaging”). To borrow a phrase from Justice Stevens, few of us would march our sons and daughters off to war to preserve the citizen’s right to be the exclusive purveyor of “OLD COON SMOKING TOBACCO.” See Young, 427 U.S. at 70, 96 S.Ct. 2440; McCann v. Anthony, 21 Mo.App. 83, 91–92 (1886).
The Supreme Court has routinely considered the relative value of burdened speech in its First Amendment analysis. See, e.g., Bethel Sch. Dist. No. 403 v. Fraser, 478 U.S. 675, 683, 106 S.Ct. 3159, 92 L.Ed.2d 549 (1986); Young, 427 U.S. at 70–71, 96 S.Ct. 2440; Tinker v. Des Moines Indep. Cmty. Sch. Dist., 393 U.S. 503, 510–11, 89 S.Ct. 733, 21 L.Ed.2d 731 (1969). For instance, the Court has held that a student’s interest in high-value political speech outweighed his school’s interest in avoiding a “substantial disruption,” Tinker, 393 U.S. at 510–11, 89 S.Ct. 733, but that a student’s interest in low-value “insulting” speech did not, Fraser, 478 U.S. at 683, 106 S.Ct. 3159. When low-value materials are concerned, “the State may legitimately use the content of these materials as the basis for placing them in a different classification” of First Amendment protection. Young, 427 U.S. at 71, 96 S.Ct. 2440.
*1382 At the extremes, disparaging speech enjoys no First Amendment protection. Chaplinsky v. New Hampshire, 315 U.S. 568, 62 S.Ct. 766, 86 L.Ed. 1031 (1942). “Insulting” words, which “by their very utterance inflict injury” are part of the “limited classes of speech, the prevention and punishment of which have never been thought to raise any Constitutional problem.” Id. at 571–72, 62 S.Ct. 766. To whatever extent “disparaging” speech differs from “insulting” speech, its value is not much greater.
Additionally, any minimal value disparaging speech might offer in the marketplace of ideas is far diminished in the marketplace of goods and services, which is the only context at issue in this appeal. One can hardly imagine what legitimate interest a vendor of goods or services may have in insulting potential customers. Whatever value disparaging speech might possess when used in private life, it loses when used in commerce.
When we balance the government’s substantial interest in the orderly flow of commerce against the modest imposition of § 2(a) on a narrowly tailored portion of particularly low-value speech, the standards of intermediate scrutiny are satisfied. Whatever modest imposition the statute makes on the free flow of public discourse, it is nothing more than an “incidental restriction on First Amendment freedom [that] is no greater than is essential to the furtherance of the governmental interest” in the orderly flow of commerce. See Barnes, 501 U.S. at 561, 111 S.Ct. 2456. For the foregoing reasons, I believe that § 2(a) is constitutional. I respectfully dissent.
We limit our holding in this case to the constitutionality of the § 2(a) disparagement provision. Recognizing, however, that other portions of § 2 may likewise constitute government regulation of expression based on message, such as the exclusions of immoral or scandalous marks, we leave to future panels the consideration of the § 2 provisions other than the disparagement provision at issue here. To be clear, we overrule In re McGinley, 660 F.2d 481 (C.C.P.A.1981), and other precedent insofar as they could be argued to prevent a future panel from considering the constitutionality of other portions of § 2 in light of the present decision.
This is Mr. Tam’s second application for the mark THE SLANTS. In 2010, Mr. Tam filed App. No. 77/952,263 seeking to register the mark for “Entertainment, namely, live performances by a musical band.” The examiner found the mark disparaging to people of Asian descent under § 2(a) and therefore refused to register it. Mr. Tam appealed that refusal to the Board, but the case was dismissed for failure to file a brief.
We reinstate the panel’s holding that Mr. Tam’s mark is disparaging.
See, e.g., Ritchie v. Simpson, 170 F.3d 1092, 1103 & n. 1 (Fed.Cir.1999) (Newman, J., dissenting); Pro–Football Inc. v. Harjo, No. 99–1385(CKK), 2000 WL 1923326, at *4 (D.D.C. Dec. 11, 2000); Stephen Baird, Moral Intervention in the Trademark Arena: Banning the Registration of Scandalous and Immoral Trademarks, 83 TRADEMARK REPORTER 661, 685–86 (1993); Justin G. Blankenship, The Cancellation of Redskins as a Disparaging Trademark: Is Federal Trademark Law an Appropriate Solution for Words That Offend?, 72 U. COLO. L. REV. 415, 443–44 (2001); Terence Dougherty, Group Rights to Cultural Survival: Intellectual Property Rights in Native American Cultural Symbols, 29 COLUM. HUM. RTS. L. REV. 355, 383 (1998); Bruce C. Kelber, ‘‘Scalping the Redskins:” Can Trademark Law Start Athletic Teams Bearing Native American Nicknames and Images on the Road to Racial Reform?, 17 HAMLINE L. REV. 533, 556 (1994); Paul Kuruk, Goading a Reluctant Dinosaur: Mutual Recognition Agreements as a Policy Response to the Misappropriation of Foreign Traditional Knowledge in the United States, 34 PEPP. L. REV. 629, 662 n. 209 (2007); Michelle B. Lee, Section 2(a) of the Lanham Act as a Restriction on Sports Team Names: Has Political Correctness Gone Too Far?, 4 SPORTS L.J. 65, 66–67 (1997); Jeffrey Lefstin, Does the First Amendment Bar Cancellation of Redskins?, 52 STAN. L. REV. 665, 676–77 (2000); Nell Jessup Newton, Memory and Misrepresentation: Representing Crazy Horse, 27 CONN. L. REV. 1003, 1030 n. 109 (1995); Ron Phillips, A Case for Scandal and Immorality: Proposing Thin Protection of Controversial Trademarks, 17 U. BALT. INTELL. PROP. L.J. 55, 67–68 (2008); Jendi Reiter, Redskins and Scarlet Letters: Why “Immoral” and “Scandalous” Trademarks Should Be Federally Registrable, 6 FED. CIR. BAR. J. 191, 197 (1996); Lilit Voskanyan, The Trademark Principal Register as a Nonpublic Forum, 75 U. CHI. L.REV. 1295, 1302 (2008).
Both parties agree that this appeal is appropriately viewed as involving a facial challenge. A law is facially invalid if “a substantial number of its applications are unconstitutional, judged in relation to the statute’s plainly legitimate sweep.” Wash. State Grange v. Wash. State Republican Party, 552 U.S. 442, 449 n. 6, 128 S.Ct. 1184, 170 L.Ed.2d 151 (2008) (internal quotation marks omitted). In other words, to succeed in his facial challenge, Mr. Tam must “demonstrate a substantial risk that application of the provision will lead to the suppression of speech.” Nat’l Endowment for the Arts v. Finley, 524 U.S. 569, 580, 118 S.Ct. 2168, 141 L.Ed.2d 500 (1998). The marks refused registration under the disparagement provision are protected speech. And the government refused to register all of these marks because it found they convey a disparaging message. More than a “substantial number” of § 2(a)’s applications of the disparagement provision rest on disapproval of the expressive message conveyed—every rejection under the disparagement provision is a message-based denial of otherwise-available legal rights. Thus, we conclude that § 2(a) is invalid on its face. That conclusion follows from the standards for First Amendment facial invalidation and also fits the rationale for those standards: it avoids maintaining on the books a rule that called for case-by-case litigation over particular marks, based on speakers’ intent and government interests or other factors, which would threaten to produce the very chilling effect that First Amendment facial-invalidity standards condemn.
In 1939, the Assistant Commissioner of Patents testified during congressional hearings on the Lanham Act that “it is always going to be just a matter of the personal opinion of the individual parties as to whether they think it is disparaging.” See Hearings on H.R. 4744 Before the Subcomm. on Trademarks of the House Comm. on Patents, 76th Cong., 1st Sess. 18–21 (1939) (statement of Leslie Frazer, Assistant Comm’r of Patents) (Mr. Frazer). And further interpretation has helped little. The definition of a disparaging mark—a mark that “dishonors by comparison with what is inferior, slights, deprecates, degrades, or affects or injures by unjust comparison”—provides little clarity. Geller, 751 F.3d at 1358 (alterations omitted). In In re In Over Our Heads, the PTO admitted that “[t]he guidelines for determining whether a mark is scandalous or disparaging are somewhat vague and the determination of whether a mark is scandalous or disparaging is necessarily a highly subjective one.” No. 755,278, 1990 WL 354546, at *1 (T.T.A.B.1990) (alterations and quotation marks omitted).
The PTO’s record of trademark registrations and denials often appears arbitrary and is rife with inconsistency. The PTO denied the mark HAVE YOU HEARD SATAN IS A REPUBLICAN because it disparaged the Republican Party, App. Ser. No. 85/077647, but did not find the mark THE DEVIL IS A DEMOCRAT disparaging, App. Ser. No. 85/525,066 (abandoned after publication for other reasons). The PTO registered the mark FAGDOG three times and refused it twice, at least once as disparaging. Compare Reg. Nos. 2,926,775; 2,828,396; and 3,174,475, with App. Ser. Nos. 76/454,927 and 75/950,535. The PTO refused to register the marks FAG FOREVER A GENIUS!, App. Ser. No. 86/089,512, and MARRIAGE IS FOR FAGS, App. Ser. No. 77/477,549, but allowed the mark F*A*G FABULOUS AND GAY, Reg. No. 2,997,761 (abandoned after publication for other reasons). And PTO examiners have registered DANGEROUS NEGRO, CELEBRETARDS, STINKY GRINGO, MIDGET–MAN, and OFF–WHITE TRASH—all marks that could be offensive to a substantial composite of the referenced group. We see no rationale for the PTO’s seemingly arbitrary registration decisions, let alone one that would give applicants much guidance.
A vague law that regulates speech on the basis of message “raises special First Amendment concerns because of its obvious chilling effect on free speech.” Reno v. ACLU, 521 U.S. 844, 871–72, 117 S.Ct. 2329, 138 L.Ed.2d 874 (1997). Thus, if a “law interferes with the right of free speech or of association, a more stringent vagueness test should apply.” Vill. of Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489, 499, 102 S.Ct. 1186, 71 L.Ed.2d 362 (1982). The Supreme Court reiterated these principles just three years ago:
And even if Mr. Tam could register his mark in a state, the benefits of state registration are limited by the boundaries of the individual state or the geographic scope of the actual use of the mark within the state. They are by no means the nationwide benefits afforded to federally registered trademarks.
Not surprisingly, holders of disparaging marks like Mr. Tam have not argued that they lack these common law rights on account of their marks not being registrable. They have little incentive to give this argument away.
The government also argues that Mr. Tam “may” have rights under 15 U.S.C. § 1125(a) (“Section 43(a)”). First, those rights would not include the benefits afforded to federally registered marks. Furthermore, it is not at all clear that Mr. Tam could bring a § 43(a) unfair competition claim. Section 43(a) allows for a federal suit to protect an unregistered trademark, much like state common law. But there is no authority extending § 43(a) to marks denied under § 2(a)’s disparagement provision. To the contrary, courts have suggested that § 43(a) is only available for marks that are registrable under § 2. See Two Pesos, 505 U.S. at 768, 112 S.Ct. 2753 (section 43(a) “protects qualifying unregistered trademarks and ... 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)”); Yarmuth–Dion, Inc. v. D’ion Furs, Inc., 835 F.2d 990, 992 (2d Cir.1987) (requiring a plaintiff to “demonstrate that his [unregistered] mark merits protection under the Lanham Act”); see also Renna v. Cty. of Union, 88 F.Supp.3d 310, 320 (D.N.J.2014) (“Section 2 declares certain marks to be unregistrable because they are inappropriate subjects for trademark protection. It follows that such unregistrable marks, not actionable as registered marks under Section 32, are not actionable under Section 43, either.”). And we have found no case allowing a § 43(a) action on a mark rejected or cancelled under § 2(a).
Counsel for the United States at oral argument disclaimed the notion that a government forum approach was appropriate in the context of trademark registration. See Oral Argument at 1:14:25–1:14:58; 1:16:20–1:17:15.
Amici also were easily able to uncover examples of inconsistencies in the application of the § 2(a). See Br. for American Civil Liberties Union, the American Civil Liberties Union of Oregon, and the American Civil Liberties Union of the Nation’s Capital as Amici Curiae 22–24 (discussing “a long line of arbitrary and contradictory decisions” as evidenced by the “countless examples of such irregularities,” including, but not limited to, examples where the same mark is rejected in one instance and accepted in another, even for the same use—for example compare MADONNA, In re Riverbank Canning Co., 25 CCPA 1028, 95 F.2d 327 (CCPA 1938) (affirming rejection of mark for use on wines as scandalous), with MADONNA, Registration No. 3,545,635 (accepted for use on wine) (Dec. 16, 2008); and MESSIAS, In re Sociedade Agricola E. Comerical Dos Vinhos Messias, S.A.R.L., 159 U.S.P.Q. 275 (T.T.A.B.1968) (rejected for use on wine and brandy), with IL MESSIA, Registration No. 4,093,035 (accepted for use on wine) (Jan. 31, 2012)). These examples further highlight the subjective nature of the registration standard under § 2(a): it is an unstable standard that apparently depends on shifting sensibilities over time.
Numerous amici have come to the same conclusion. See, e.g., Br. for First Amendment Lawyers Ass’n as Amicus Curiae 14 (“The multitude of Section 2(a) cases show that Section 2(a) does not convey ‘sufficiently definite warning as to the proscribed conduct when measured by common understanding and practices,’ as required by the Constitution.” (quoting Roth v. United States, 354 U.S. 476, 491, 77 S.Ct. 1304, 1 L.Ed.2d 1498 (1957))); Br. for Pro–Football, Inc. as Amicus Curiae 33 n. 13 (“Even if Section 2(a) sought to advance a legitimate state interest, its language is impermissibly vague to advance that interest. The statute provides no guidance as to which trademarks will be deemed disparaging, scandalous, or immoral.”).
The majority frequently characterizes the statute as “discriminat[ing] on the basis of message conveyed” and hence “viewpoint.” Maj. Op. at 1335. “It does so as a matter of avowed and undeniable purpose, and it does so on its face.” Id. “Denial of these benefits creates a serious disincentive to adopt a mark which the government may deem offensive or disparaging.” Id. at 1341. “The entire interest of the government in § 2(a) depends on disapproval of the message.” Id. at 1355. “All of the government’s proffered interests boil down to permitting the government to burden speech it finds offensive.” Id. at 1357.
To be sure, the Board may have rendered inconsistent results in some cases, but this has no bearing on the facial validity of § 2(a). See, e.g., Nat’l Endowment for the Arts v. Finley, 524 U.S. 569, 587, 118 S.Ct. 2168, 141 L.Ed.2d 500 (1998); Red Lion Broad. Co. v. F.C.C., 395 U.S. 367, 396, 89 S.Ct. 1794, 23 L.Ed.2d 371 (1969). In any event, when the government is not acting in its sovereign, regulatory capacity, “the consequences of imprecision are not constitutionally severe.” Finley, 524 U.S. at 589, 118 S.Ct. 2168.
See also, e.g., Cohen v. California, 403 U.S. 15, 25, 91 S.Ct. 1780, 29 L.Ed.2d 284 (1971); Hess v. Indiana, 414 U.S. 105, 107, 94 S.Ct. 326, 38 L.Ed.2d 303 (1973); Denver Area Educ. Telecomm. Consortium, Inc. v. F.C.C., 518 U.S. 727, 753–54, 116 S.Ct. 2374, 135 L.Ed.2d 888 (1996).
See, e.g., Linmark Assocs., Inc. v. Twp. of Willingboro, 431 U.S. 85, 97, 97 S.Ct. 1614, 52 L.Ed.2d 155 (1977) (striking down a ban on placing “For Sale” and “Sold” signs on residential property); Carey, 431 U.S. at 701–02, 97 S.Ct. 2010 (invalidating a ban on all advertising and display of contraceptives); Bolger, 463 U.S. at 71, 103 S.Ct. 2875 (invalidating a ban on unsolicited mailing of contraceptive advertisements); Va. State Bd. of Pharmacy, 425 U.S. at 773, 96 S.Ct. 1817 (invalidating a ban on advertising prescription drug prices); Sorrell v. IMS Health Inc., ––– U.S. ––––, 131 S.Ct. 2653, 2659, 180 L.Ed.2d 544 (2011) (invalidating a state law that prohibited the sale, disclosure, and use of pharmacy records without the prescriber’s consent and subject to limited exceptions).
That alternative federal enforcement under 15 U.S.C. § 1125(a) is potentially available to denied applicants only bolsters this point. See Maj. Op. at 1344-45 n. 11.
Bullfrog Films, Inc. v. Wick, 847 F.2d 502, 503 (9th Cir.1988), Dep’t of Tex., Veterans of Foreign Wars v. Tex. Lottery Comm’n, 760 F.3d 427, 430 (5th Cir.2014) (en banc), and Autor v. Pritzker, 740 F.3d 176, 177 (D.C.Cir.2014), relied on by the majority, Maj. Op. at 1351–53, are all inapposite. In all three cases, the government was attempting to leverage speech outside of the “contours” of its defined program, thus running afoul of the unconstitutional conditions doctrine. Here, on the other hand, no expression beyond the trademark is suppressed, and therefore no unconstitutional condition obtains.
See, e.g., Chuck Culpepper, How Missouri foot-ball’s boycott helped bridge a familiar campus divide, Wash. Post (Nov. 13, 2015), https://www.washingtonpost.com/sports/colleges/howmissouri-footballs-boycott-helped-unite-a-troubled-campus/2015/11/13/64fe68ea–8a0f–11e5–be8b–1ae2e4f50f76—story.html.
See, e.g., American Psychological Ass’n, APA Resolution Recommending the Immediate Retirement of American Indian Mascots, Symbols, Images, and Personalities by Schools, Colleges, Universities, Athletic Teams, and Organizations (2011), available at http://www.apa.org/about/policy/mascots.pdf (citing many studies finding psychological harm of exposure to negative stereotypes).
Julian Casablancas, 15 Shockingly Racist Vintage Ads, Business Pundit (Dec. 17, 2012), http://www.businesspundit.com/15–shockingly–racist–vintage–ads/?img=42884.
Dan Beard, 24 Recreation 1 (1905) available at https://books.google.com/books?id=LPQXAAAAYAAJ & pg= PA474–IA18# v=onepage & g & f=false.
Brian D. Behnken & Gregory D. Smithers, Racism in American Popular Media: From Aunt Jemima to the Frito Bandito 39 (2015).
The registration of a trademark confers a competitive advantage in the marketplace to the owner of the mark. Typically, in trademark disputes, opposition to the registration or use of a certain mark involves the commercial activities of a competitor. In such cases, the interests of both the owner and competitor are fundamentally commercial in nature.