4 Overlap: Functional Characteristics 4 Overlap: Functional Characteristics
4.1 Star Athletica, L.L.C. v. Varsity Brands, Inc. 4.1 Star Athletica, L.L.C. v. Varsity Brands, Inc.
Copyright
STAR ATHLETICA, L.L.C., Petitioner
v.
VARSITY BRANDS, INC., et al.
No. 15-866.
Supreme Court of the United States
Argued Oct. 31, 2016.
Decided March 22, 2017.
John J. Bursch, Grand Rapids, MI, for petitioner.
*1007William M. Jay, Washington, DC, for respondents.
Eric J. Feigin for the United States as amicus curiae, by special leave of the Court, supporting the respondents.
Steven M. Crosby, Stephen E. Feldman, The Feldman Law Group, New York, NY, Michael F. Rafferty, Emily Hamm Huseth, Harris Shelton Hanover, Walsh, P.L.L.C., Memphis, TN, John J. Bursch, Bursch Law PLLC, Caledonia, MI, Matthew T. Nelson, Conor B. Dugan, Warner Norcross & Judd LLP, Grand Rapids, MI, for petitioner.
Grady M. Garrison, Adam S. Baldridge, Baker, Donelson, Bearman, Caldwell & Berkowitz P.C., Memphis, TN, William M. Jay, Brian T. Burgess, Goodwin Procter LLP, Washington, DC, Thomas Kjellberg, Cowan, Liebowitz & Latman, P.C., New York, NY, Robert D. Carroll, Goodwin Procter LLP, Boston, MA, Charles T. Cox Jr., Goodwin Procter LLP, Washington, DC, for respondents.
Congress has provided copyright protection for original works of art, but not for industrial designs. The line between art and industrial design, however, is often difficult to draw. This is particularly true when an industrial design incorporates artistic elements. Congress has afforded limited protection for these artistic elements by providing that "pictorial, graphic, or sculptural features" of the "design of a useful article" are eligible for copyright protection as artistic works if those features "can be identified separately from, and are capable of existing independently of, the utilitarian aspects of the article." 17 U.S.C. § 101.
We granted certiorari to resolve widespread disagreement over the proper test for implementing § 101's separate-identification and independent-existence requirements. 578 U.S. ----, 136 S.Ct. 1823, 194 L.Ed.2d 829 (2016). We hold that a feature incorporated into the design of a useful article is eligible for copyright protection only if the feature (1) can be perceived as a two- or three-dimensional work of art separate from the useful article and (2) would qualify as a protectable pictorial, graphic, or sculptural work-either on its own or fixed in some other tangible medium of expression-if it were imagined separately from the useful article into which it is incorporated. Because that test is satisfied in this case, we affirm.
I
Respondents Varsity Brands, Inc., Varsity Spirit Corporation, and Varsity Spirit Fashions & Supplies, Inc., design, make, and sell cheerleading uniforms. Respondents have obtained or acquired more than 200 U.S. copyright registrations for two-dimensional designs appearing on the surface of their uniforms and other garments. These designs are primarily "combinations, positionings, and arrangements of elements" that include "chevrons ..., lines, curves, stripes, angles, diagonals, inverted [chevrons], coloring, and shapes." App. 237. At issue in this case are Designs 299A, 299B, 074, 078, and 0815. See Appendix, infra .
Petitioner Star Athletica, L.L.C., also markets and sells cheerleading uniforms. Respondents sued petitioner for infringing their copyrights in the five designs. The District Court entered summary judgment for petitioner on respondents' copyright claims on the ground that the designs did *1008not qualify as protectable pictorial, graphic, or sculptural works. It reasoned that the designs served the useful, or "utilitarian," function of identifying the garments as "cheerleading uniforms" and therefore could not be "physically or conceptually" separated under § 101"from the utilitarian function" of the uniform. 2014 WL 819422, *8-*9 (W.D.Tenn., Mar. 1, 2014).
The Court of Appeals for the Sixth Circuit reversed. 799 F.3d 468, 471 (2015). In its view, the "graphic designs" were "separately identifiable" because the designs "and a blank cheerleading uniform can appear 'side by side'-one as a graphic design, and one as a cheerleading uniform." Id., at 491 (quoting Compendium of U.S. Copyright Office Practices § 924.2(B) (3d ed. 2014) (Compendium)). And it determined that the designs were " 'capable of existing independently' " because they could be incorporated onto the surface of different types of garments, or hung on the wall and framed as art. 799 F.3d, at 491, 492.
Judge McKeague dissented. He would have held that, because "identifying the wearer as a cheerleader" is a utilitarian function of a cheerleading uniform and the surface designs were "integral to" achieving that function, the designs were inseparable from the uniforms. Id., at 495-496.
II
The first element of a copyright-infringement claim is "ownership of a valid copyright." Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340, 361, 111 S.Ct. 1282, 113 L.Ed.2d 358 (1991). A valid copyright extends only to copyrightable subject matter. See 4 M. Nimmer & D. Nimmer, Copyright § 13.01[A] (2010) (Nimmer). The Copyright Act of 1976 defines copyrightable subject matter as "original works of authorship fixed in any tangible medium of expression." 17 U.S.C. § 102(a).
"Works of authorship" include "pictorial, graphic, and sculptural works," § 102(a)(5), which the statute defines to include "two-dimensional and three-dimensional works of fine, graphic, and applied art, photographs, prints and art reproductions, maps, globes, charts, diagrams, models, and technical drawings, including architectural plans," § 101. And a work of authorship is " 'fixed' in a tangible medium of expression when it[ is] embodi[ed] in a" "material objec[t] ... from which the work can be perceived, reproduced, or otherwise communicated." Ibid. (definitions of "fixed" and "copies").
The Copyright Act also establishes a special rule for copyrighting a pictorial, graphic, or sculptural work incorporated into a "useful article," which is defined as "an article having an intrinsic utilitarian function that is not merely to portray the appearance of the article or to convey information." Ibid. The statute does not protect useful articles as such. Rather, "the design of a useful article" is "considered a pictorial, graphical, or sculptural work only if, and only to the extent that, such design incorporates pictorial, graphic, or sculptural features that can be identified separately from, and are capable of existing independently of, the utilitarian aspects of the article." Ibid .
Courts, the Copyright Office, and commentators have described the analysis undertaken to determine whether a feature can be separately identified from, and exist independently of, a useful article as "separability." In this case, our task is to determine whether the arrangements of lines, chevrons, and colorful shapes appearing on the surface of respondents' cheerleading uniforms are eligible for copyright protection as separable features *1009of the design of those cheerleading uniforms.
A
As an initial matter, we must address whether separability analysis is necessary in this case.
1
Respondents argue that "[s]eparability is only implicated when a [pictorial, graphic, or sculptural] work is the 'design of a useful article.' " Brief for Respondents 25. They contend that the surface decorations in this case are "two-dimensional graphic designs that appear on useful articles," but are not themselves designs of useful articles. Id., at 52. Consequently, the surface decorations are protected two-dimensional works of graphic art without regard to any separability analysis under § 101. Ibid. ; see 2 W. Patry, Copyright § 3:151, p. 3-485 (2016) (Patry) ("Courts looking at two-dimensional design claims should not apply the separability analysis regardless of the three-dimensional form that design is embodied in"). Under this theory, two-dimensional artistic features on the surface of useful articles are "inherently separable." Brief for Respondents 26.
This argument is inconsistent with the text of § 101. The statute requires separability analysis for any "pictorial, graphic, or sculptural features" incorporated into the "design of a useful article." "Design" refers here to "the combination" of "details" or "features" that "go to make up" the useful article. 3 Oxford English Dictionary 244 (def. 7, first listing) (1933) (OED). Furthermore, the words "pictorial" and "graphic" include, in this context, two-dimensional features such as pictures, paintings, or drawings. See 4 id., at 359 (defining "[g]raphic" to mean "[o]f or pertaining to drawing or painting"); 7 id., at 830 (defining "[p]ictorial" to mean "of or pertaining to painting or drawing"). And the statute expressly defines "[p]ictorial, graphical, and sculptural works" to include "two-dimensional ... works of ... art." § 101. The statute thus provides that the "design of a useful article" can include two-dimensional "pictorial" and "graphic" features, and separability analysis applies to those features just as it does to three-dimensional "sculptural" features.
2
The United States makes a related but distinct argument against applying separability analysis in this case, which respondents do not and have not advanced. As part of their copyright registrations for the designs in this case, respondents deposited with the Copyright Office drawings and photographs depicting the designs incorporated onto cheerleading uniforms. App. 213-219; Appendix, infra . The Government argues that, assuming the other statutory requirements were met, respondents obtained a copyright in the deposited drawings and photographs and have simply reproduced those copyrighted works on the surface of a useful article, as they would have the exclusive right to do under the Copyright Act. See Brief for United States as Amicus Curiae 14-15, 17-22. Accordingly, the Government urges, separability analysis is unnecessary on the record in this case. We generally do not entertain arguments that were not raised below and that are not advanced in this Court by any party, Burwell v. Hobby Lobby Stores, Inc., 573 U.S. ----, ----, 134 S.Ct. 2751, 189 L.Ed.2d 675 (2014), because "[i]t is not the Court's usual practice to adjudicate either legal or predicate factual questions in the first instance," CRST Van Expedited, Inc. v. EEOC, 578 U.S. ----, ----, 136 S.Ct. 1642, 1653, 194 L.Ed.2d 707 (2016). We decline to depart from our usual practice here.
We must now decide when a feature incorporated into a useful article "can be identified separately from" and is "capable of existing independently of" "the utilitarian aspects" of the article. This is not a free-ranging search for the best copyright policy, but rather "depends solely on statutory interpretation." Mazer v. Stein, 347 U.S. 201, 214, 74 S.Ct. 460, 98 L.Ed. 630 (1954). "The controlling principle in this case is the basic and unexceptional rule that courts must give effect to the clear meaning of statutes as written." Estate of Cowart v. Nicklos Drilling Co., 505 U.S. 469, 476, 112 S.Ct. 2589, 120 L.Ed.2d 379 (1992). We thus begin and end our inquiry with the text, giving each word its "ordinary, contemporary, common meaning." Walters v. Metropolitan Ed. Enterprises, Inc., 519 U.S. 202, 207, 117 S.Ct. 660, 136 L.Ed.2d 644 (1997) (internal quotation marks omitted). We do not, however, limit this inquiry to the text of § 101 in isolation. "[I]nterpretation of a phrase of uncertain reach is not confined to a single sentence when the text of the whole statute gives instruction as to its meaning." Maracich v. Spears, 570 U.S. ----, ----, 133 S.Ct. 2191, 2203, 186 L.Ed.2d 275 (2013). We thus "look to the provisions of the whole law" to determine § 101's meaning. United States v. Heirs of Boisdore, 8 How. 113, 122, 12 L.Ed. 1009 (1849).
1
The statute provides that a "pictorial, graphic, or sculptural featur [e]" incorporated into the "design of a useful article" is eligible for copyright protection if it (1) "can be identified separately from," and (2) is "capable of existing independently of, the utilitarian aspects of the article." § 101. The first requirement-separate identification-is not onerous. The decisionmaker need only be able to look at the useful article and spot some two- or three-dimensional element that appears to have pictorial, graphic, or sculptural qualities. See 2 Patry § 3:146, at 3-474 to 3-475.
The independent-existence requirement is ordinarily more difficult to satisfy. The decisionmaker must determine that the separately identified feature has the capacity to exist apart from the utilitarian aspects of the article. See 2 OED 88 (def. 5) (defining "[c]apable" of as "[h]aving the needful capacity, power, or fitness for"). In other words, the feature must be able to exist as its own pictorial, graphic, or sculptural work as defined in § 101 once it is imagined apart from the useful article. If the feature is not capable of existing as a pictorial, graphic, or sculptural work once separated from the useful article, then it was not a pictorial, graphic, or sculptural feature of that article, but rather one of its utilitarian aspects.
Of course, to qualify as a pictorial, graphic, or sculptural work on its own, the feature cannot itself be a useful article or "[a]n article that is normally a part of a useful article" (which is itself considered a useful article). § 101. Nor could someone claim a copyright in a useful article merely by creating a replica of that article in some other medium-for example, a cardboard model of a car. Although the replica could itself be copyrightable, it would not give rise to any rights in the useful article that inspired it.
2
The statute as a whole confirms our interpretation. The Copyright Act provides "the owner of [a] copyright" with the "exclusive righ[t] ... to reproduce the copyrighted work in copies." § 106(1). The statute clarifies that this right "includes the right to reproduce the [copyrighted] work in or on any kind of article, whether useful or otherwise." § 113(a).
*1011Section 101 is, in essence, the mirror image of § 113(a). Whereas § 113(a) protects a work of authorship first fixed in some tangible medium other than a useful article and subsequently applied to a useful article, § 101 protects art first fixed in the medium of a useful article. The two provisions make clear that copyright protection extends to pictorial, graphic, and sculptural works regardless of whether they were created as freestanding art or as features of useful articles. The ultimate separability question, then, is whether the feature for which copyright protection is claimed would have been eligible for copyright protection as a pictorial, graphic, or sculptural work had it originally been fixed in some tangible medium other than a useful article before being applied to a useful article.
3
This interpretation is also consistent with the history of the Copyright Act. In Mazer, a case decided under the 1909 Copyright Act, the respondents copyrighted a statuette depicting a dancer. The statuette was intended for use as a lamp base, "with electric wiring, sockets and lamp shades attached." 347 U.S., at 202, 74 S.Ct. 460. Copies of the statuette were sold both as lamp bases and separately as statuettes. Id., at 203, 74 S.Ct. 460. The petitioners copied the statuette and sold lamps with the statuette as the base. They defended against the respondents' infringement suit by arguing that the respondents did not have a copyright in a statuette intended for use as a lamp base. Id., at 204-205, 74 S.Ct. 460.
Two of Mazer 's holdings are relevant here. First, the Court held that the respondents owned a copyright in the statuette even though it was intended for use as a lamp base. See id., at 214, 74 S.Ct. 460. In doing so, the Court approved the Copyright Office's regulation extending copyright protection to works of art that might also serve a useful purpose. See ibid. (approving 37 C.F.R. § 202.8(a) (1949) (protecting "works of artistic craftsmanship, in so far as their form but not their mechanical or utilitarian aspects are concerned")).
Second, the Court held that it was irrelevant to the copyright inquiry whether the statuette was initially created as a freestanding sculpture or as a lamp base. 347 U.S., at 218-219, 74 S.Ct. 460 ("Nor do we think the subsequent registration of a work of art published as an element in a manufactured article, is a misuse of copyright. This is not different from the registration of a statuette and its later embodiment in an industrial article"). Mazer thus interpreted the 1909 Act consistently with the rule discussed above: If a design would have been copyrightable as a standalone pictorial, graphic, or sculptural work, it is copyrightable if created first as part of a useful article.
Shortly thereafter, the Copyright Office enacted a regulation implementing the holdings of Mazer . See 1 Nimmer § 2A.08[B][1][b] (2016). As amended, the regulation introduced the modern separability test to copyright law:
"If the sole intrinsic function of an article is its utility, the fact that the article is unique and attractively shaped will not qualify it as a work of art. However, if the shape of a utilitarian article incorporates features, such as artistic sculpture, carving, or pictorial representation, which can be identified separately and are capable of existing independently as a work of art, such features will be eligible for registration." 37 C.F.R. § 202.10(c) (1960) (punctuation altered).
Congress essentially lifted the language governing protection for the design of a useful article directly from the post-Mazer regulations and placed it into § 101 of the *10121976 Act. Consistent with Mazer, the approach we outline today interprets §§ 101 and 113 in a way that would afford copyright protection to the statuette in Mazer regardless of whether it was first created as a standalone sculptural work or as the base of the lamp. See 347 U.S., at 218-219, 74 S.Ct. 460.
C
In sum, a feature of the design of a useful article is eligible for copyright if, when identified and imagined apart from the useful article, it would qualify as a pictorial, graphic, or sculptural work either on its own or when fixed in some other tangible medium.
Applying this test to the surface decorations on the cheerleading uniforms is straightforward. First, one can identify the decorations as features having pictorial, graphic, or sculptural qualities. Second, if the arrangement of colors, shapes, stripes, and chevrons on the surface of the cheerleading uniforms were separated from the uniform and applied in another medium-for example, on a painter's canvas-they would qualify as "two-dimensional ... works of ... art," § 101. And imaginatively removing the surface decorations from the uniforms and applying them in another medium would not replicate the uniform itself. Indeed, respondents have applied the designs in this case to other media of expression-different types of clothing-without replicating the uniform. See App. 273-279. The decorations are therefore separable from the uniforms and eligible for copyright protection.1
The dissent argues that the designs are not separable because imaginatively removing them from the uniforms and placing them in some other medium of expression-a canvas, for example-would create "pictures of cheerleader uniforms." Post, at 1035 - 1036 (opinion of BREYER, J.). Petitioner similarly argues that the decorations cannot be copyrighted because, even when extracted from the useful article, they retain the outline of a cheerleading uniform. Brief for Petitioner 48-49.
This is not a bar to copyright. Just as two-dimensional fine art corresponds to the shape of the canvas on which it is painted, two-dimensional applied art correlates to the contours of the article on which it is applied. A fresco painted on a wall, ceiling panel, or dome would not lose copyright protection, for example, simply because it was designed to track the dimensions of the surface on which it was painted. Or consider, for example, a design etched or painted on the surface of a guitar. If that entire design is imaginatively removed from the guitar's surface and placed on an album cover, it would still resemble the shape of a guitar. But the image on the cover does not "replicate" the guitar as a useful article. Rather, the design is a two-dimensional work of art that corresponds to the shape of the useful article to which it was applied. The statute protects that work of art whether it is first drawn on the album cover and then applied to the guitar's surface, or vice versa. Failing to protect that art would create an anomaly: It would extend protection to two-dimensional designs that cover a part of a useful article but would not protect the same design if it covered the entire article. The statute does not support that distinction, nor can it be reconciled *1013with the dissent's recognition that "artwork printed on a t-shirt" could be protected. Post, at 1019 (internal quotation marks omitted).
To be clear, the only feature of the cheerleading uniform eligible for a copyright in this case is the two-dimensional work of art fixed in the tangible medium of the uniform fabric. Even if respondents ultimately succeed in establishing a valid copyright in the surface decorations at issue here, respondents have no right to prohibit any person from manufacturing a cheerleading uniform of identical shape, cut, and dimensions to the ones on which the decorations in this case appear. They may prohibit only the reproduction of the surface designs in any tangible medium of expression-a uniform or otherwise.2
D
Petitioner and the Government raise several objections to the approach we announce today. None is meritorious.
1
Petitioner first argues that our reading of the statute is missing an important step. It contends that a feature may exist independently only if it can stand alone as a copyrightable work and if the useful article from which it was extracted would remain equally useful. In other words, copyright extends only to "solely artistic" features of useful articles. Brief for Petitioner 33. According to petitioner, if a feature of a useful article "advance[s] the utility of the article," id., at 38, then it is categorically beyond the scope of copyright, id., at 33. The designs here are not protected, it argues, because they are necessary to two of the uniforms' "inherent, essential, or natural functions"-identifying the wearer as a cheerleader and enhancing the wearer's physical appearance. Id., at 38, 48; Reply Brief 2, 16. Because the uniforms would not be equally useful without the designs, petitioner contends that the designs are inseparable from the "utilitarian aspects" of the uniform. Brief for Petitioner 50.
The Government raises a similar argument, although it reaches a different result. It suggests that the appropriate test is whether the useful article with the artistic feature removed would "remai[n] similarly useful." Brief for United States as Amicus Curiae 29 (emphasis added). In the view of the United States, however, a plain white cheerleading uniform is "similarly useful" to uniforms with respondents' designs. Id., at 27-28.
The debate over the relative utility of a plain white cheerleading uniform is unnecessary. The focus of the separability inquiry is on the extracted feature and not on any aspects of the useful article that remain after the imaginary extraction. The statute does not require the decisionmaker to imagine a fully functioning useful article without the artistic feature. Instead, it requires that the separated feature qualify as a nonuseful pictorial, graphic, or sculptural work on its own.
Of course, because the removed feature may not be a useful article-as it would *1014then not qualify as a pictorial, graphic, or sculptural work-there necessarily would be some aspects of the original useful article "left behind" if the feature were conceptually removed. But the statute does not require the imagined remainder to be a fully functioning useful article at all, much less an equally useful one. Indeed, such a requirement would deprive the Mazer statuette of protection had it been created first as a lamp base rather than as a statuette. Without the base, the "lamp" would be just a shade, bulb, and wires. The statute does not require that we imagine a nonartistic replacement for the removed feature to determine whether that feature is capable of an independent existence.
Petitioner's argument follows from its flawed view that the statute protects only "solely artistic" features that have no effect whatsoever on a useful article's utilitarian function. This view is inconsistent with the statutory text. The statute expressly protects two- and three-dimensional "applied art." § 101. "Applied art" is art "employed in the decoration, design, or execution of useful objects," Webster's Third New International Dictionary 105 (1976) (emphasis added), or "those arts or crafts that have a primarily utilitarian function, or ... the designs and decorations used in these arts," Random House Dictionary 73 (1966) (emphasis added); see also 1 OED 576 (2d ed. 1989) (defining "applied" as "[p]ut to practical use"). An artistic feature that would be eligible for copyright protection on its own cannot lose that protection simply because it was first created as a feature of the design of a useful article, even if it makes that article more useful.
Indeed, this has been the rule since Mazer . In holding that the statuette was protected, the Court emphasized that the 1909 Act abandoned any "distinctions between purely aesthetic articles and useful works of art." 347 U.S., at 211, 74 S.Ct. 460. Congress did not enact such a distinction in the 1976 Act. Were we to accept petitioner's argument that the only protectable features are those that play absolutely no role in an article's function, we would effectively abrogate the rule of Mazer and read "applied art" out of the statute.
Because we reject the view that a useful article must remain after the artistic feature has been imaginatively separated from the article, we necessarily abandon the distinction between "physical" and "conceptual" separability, which some courts and commentators have adopted based on the Copyright Act's legislative history. See H.R. Rep. No. 94-1476, p. 55 (1976). According to this view, a feature is physically separable from the underlying useful article if it can "be physically separated from the article by ordinary means while leaving the utilitarian aspects of the article completely intact." Compendium § 924.2(A); see also Chosun Int'l, Inc. v. Chrisha Creations, Ltd., 413 F.3d 324, 329 (C.A.2 2005). Conceptual separability applies if the feature physically could not be removed from the useful article by ordinary means. See Compendium § 924.2(B); but see 1 P. Goldstein, Copyright § 2.5.3, p. 2:77 (3d ed. 2016) (explaining that the lower courts have been unable to agree on a single conceptual separability test); 2 Patry §§ 3:140-3:144.40 (surveying the various approaches in the lower courts).
The statutory text indicates that separability is a conceptual undertaking. Because separability does not require the underlying useful article to remain, the physical-conceptual distinction is unnecessary.
Petitioner next argues that we should incorporate two "objective" components, Reply Brief 9, into our test to provide guidance to the lower courts: (1) "whether the design elements can be identified as reflecting the designer's artistic judgment exercised independently of functional influence," Brief for Petitioner 34 (emphasis deleted and internal quotation marks omitted), and (2) whether "there is [a] substantial likelihood that the pictorial, graphic, or sculptural feature would still be marketable to some significant segment of the community without its utilitarian function," id., at 35 (emphasis deleted and internal quotation marks omitted).
We reject this argument because neither consideration is grounded in the text of the statute. The first would require the decisionmaker to consider evidence of the creator's design methods, purposes, and reasons. Id., at 48. The statute's text makes clear, however, that our inquiry is limited to how the article and feature are perceived, not how or why they were designed. See Brandir Int'l, Inc. v. Cascade Pacific Lumber Co., 834 F.2d 1142, 1152 (C.A.2 1987) (Winter, J., concurring in part and dissenting in part) (The statute "expressly states that the legal test is how the final article is perceived, not how it was developed through various stages").
The same is true of marketability. Nothing in the statute suggests that copyrightability depends on market surveys. Moreover, asking whether some segment of the market would be interested in a given work threatens to prize popular art over other forms, or to substitute judicial aesthetic preferences for the policy choices embodied in the Copyright Act. See Bleistein v. Donaldson Lithographing Co., 188 U.S. 239, 251, 23 S.Ct. 298, 47 L.Ed. 460 (1903) ("It would be a dangerous undertaking for persons trained only to the law to constitute themselves final judges of the worth of pictorial illustrations, outside of the narrowest and most obvious limits").
3
Finally, petitioner argues that allowing the surface decorations to qualify as a "work of authorship" is inconsistent with Congress' intent to entirely exclude industrial design from copyright. Petitioner notes that Congress refused to pass a provision that would have provided limited copyright protection for industrial designs, including clothing, when it enacted the 1976 Act, see id., at 9-11 (citing S. 22, Tit. II, 94th Cong., 2d Sess., 122 Cong. Rec. 3856-3859 (1976)), and that it has enacted laws protecting designs for specific useful articles-semiconductor chips and boat hulls, see 17 U.S.C. §§ 901 - 914, 1301 - 1332 -while declining to enact other industrial design statutes, Brief for Petitioner 29, 43. From this history of failed legislation petitioner reasons that Congress intends to channel intellectual property claims for industrial design into design patents. It therefore urges us to approach this question with a presumption against copyrightability. Id., at 27.
We do not share petitioner's concern. As an initial matter, "[c]ongressional inaction lacks persuasive significance" in most circumstances. Pension Benefit Guaranty Corporation v. LTV Corp., 496 U.S. 633, 650, 110 S.Ct. 2668, 110 L.Ed.2d 579 (1990) (internal quotation marks omitted). Moreover, we have long held that design patent and copyright are not mutually exclusive. See Mazer, 347 U.S., at 217, 74 S.Ct. 460. Congress has provided for limited copyright protection for certain features of industrial design, and approaching the statute with presumptive hostility toward protection for industrial design would undermine Congress' choice.
*1016In any event, as explained above, our test does not render the shape, cut, and physical dimensions of the cheerleading uniforms eligible for copyright protection.
III
We hold that an artistic feature of the design of a useful article is eligible for copyright protection if the feature (1) can be perceived as a two- or three-dimensional work of art separate from the useful article and (2) would qualify as a protectable pictorial, graphic, or sculptural work either on its own or in some other medium if imagined separately from the useful article. Because the designs on the surface of respondents' cheerleading uniforms in this case satisfy these requirements, the judgment of the Court of Appeals is affirmed.
It is so ordered.
APPENDIX TO OPINION OF THE COURT
*1018Justice GINSBURG, concurring in the judgment.
I concur in the Court's judgment but not in its opinion. Unlike the majority, I would not take up in this case the separability test appropriate under 17 U.S.C. § 101.1 Consideration of that test is unwarranted because the designs at issue are not designs of useful articles. Instead, the designs are themselves copyrightable pictorial or graphic works reproduced on useful articles.2
A pictorial, graphic, or sculptural work (PGS work) is copyrightable. § 102(a)(5). PGS works include "two-dimensional and three-dimensional works of fine, graphic, and applied art." § 101. Key to this case, a copyright in a standalone PGS work "includes the right to reproduce the work in or on any kind of article, whether useful or otherwise." § 113(a). Because the owner of a copyright in a pre-existing PGS work may exclude a would-be infringer from reproducing that work on a useful article, there is no need to engage in any separability inquiry to resolve the instant petition.
The designs here in controversy are standalone pictorial and graphic works that respondents Varsity Brands, Inc., et al. (Varsity) reproduce on cheerleading uniforms. Varsity's designs first appeared as pictorial and graphic works that Varsity's design team sketched on paper. App. 281. Varsity then sought copyright protection for those two-dimensional designs, not for cheerleading costumes; its registration statements claimed "2-Dimensional artwork" and "fabric design (artwork)." Appendix, infra, at 1020 - 1023, 1025 - 1026, 1028 - 1030. Varsity next reproduced its two-dimensional graphic designs on cheerleading uniforms, also on other garments, including T-shirts and jackets. See, e.g., App. 274, 276.3
*1019In short, Varsity's designs are not themselves useful articles meet for separability determination under § 101 ; they are standalone PGS works that may gain copyright protection as such, including the exclusive right to reproduce the designs on useful articles.4
Appendix to opinion of GINSBURG, J.
APPENDIX
EXHIBIT 15
Certificate of Registration Additional certificate (17 U.S.C. 706)
[Seal of the United States Copyright Office 1870]
This Certificate issued under the seal of the Copyright Office in accordance with title 17,United States Code, attests that registration has been made for the work identified below. The information on this certificate has been made a part of the Copyright Office records.
Form VA
For a Work of the Visual Arts
UNITED STATES COPYRIGHT OFFICE
RE VA 1-417-427
EFFECTIVE DATE OF REGISTRATION
5 21 07
Month Day Year
Maria A. Pallante
Acting Register of Copyrights, United States of America
DO NOT WRITE ABOVE THIS LINE. IF YOU NEED MORE SPACE, USE A SEPARATE CONTINUATION SHEET
1 Title of This Work NATURE OF THIS Design Number 078 WORK See instructions 2-dimensional artwork Preious or Alternative Titles __________________________________________________
Publication as a Contribution If this work was published as a contribution to a periodical, serial, or collection, give information about the collective work in which the contribution appeared.Title of Collective Work
________________________________________________ If published in a periodical or serial give: Volume Number Issue Date On Pages ______________________________________________
2 NOTE Under the law the "author" of a "work made for hire " is generally the employer, not the employee (see instructions). For any part of this work that was "made for hire" check "Yes" in the space provided, give the employer (or other person for whom the work was prepared) as "Author" of that part, and leave the space for dates of birth and death blank.
a NAME OF AUTHOR Varsity Brands, Inc. DATES OF BIRTH AND DEATH Year Born Year Died _____________ _____________ Was this contribution to the work a "work made for hire"? [X] Yes [ ] No Author's Nationality or Domicile Name of Country Citizen of _______________ or Domiciled in United States Was this Author's Contribution to the Work Anonymous? [ ] Yes [X] No Pseudonymous? [ ] Yes [X] No
If the answer to either of these questions is "Yes," see detailed instructions.
Nature of Authorship Check appropriate box(es) See Instructions [ ] 3 Dimensional sculpture [X] 2 Dimensional artwork
EXHIBIT 16
Certificate of Registration Additional certificate (17 U.S.C. 706)
[Seal of the United States Copyright Office 1870]
This Certificate issued under the seal of the Copyright Office in accordance with title 17,United States Code, attests that registration has been made for the work identified below. The information on this certificate has been made a part of the Copyright Office records.
Registration Number: VA 1-675-905
Effective date of registration: May 12, 2008
Maria A. Pallante Acting Register of Copyrights, United States of America
Title _________________________________________ Title of Work: 0815 Nature of Work: 2-dimensional artwork Completion/Publication _________________________ Year of Completion: 2007 Date of 1st Publication: January 2, 2008 Nation of 1st Publication: United States Author _________________________________________ Author: Varsity Brands, Inc. Author Created: 2-dimensional artwork Work made for hire: Yes Domiciled in: United States Anonymous: No Pseudonymous: No Copyright claimant _____________________________ Copyright Claimant: Varsity Brands, Inc.
EXHIBIT 17
Certificate of Registration Additional certificate (17 U.S.C. 706) [Seal of the United States Copyright Office 1870]
This Certificate issued under the "sealof the Copyright Office in accordance with title 17,United States Code, attests that registration has been made for the work identified below. The information on this certificate has been made a part of the Copyright Office records.
Form VA For a Work of the Visual Arts UNITED STATES COPYRIGHT OFFICE RE VA 1-319-228 EFFECTIVE DATE OF REGISTRATION April 29 2005 Month Day Year
Maria A. Pallante Acting Register of Copyrights, United States of America
DO NOT WRITE ABOVE THIS LINE. IF YOU NEED MORE SPACE, USE A SEPARATE CONTINUATION SHEET
____________________________________________________ 1 Title of This Work NATURE OF THIS 299A WORK See instructions FABRIC DESIGN (ARTWORK) Previous or Alternative Titles _____________________________________________________
Publication as a Contribution If this work was published as a contribution to a periodical, serial, or collection, give information about the collective work in which the contribution appeared.Title of Collective Work
______________________________________________________ If published in a periodical or serial give: Volume Number Issue Date On Pages ______________________________________________________
2 NOTE Under the law the "author" of a "work made for hire " is generally the employer not the employee (see instructions) For any part of this work that wasmade for hire check "Yes" in the space provided, give the employer (or other person for whom the work was prepared) as "Author" of that part and leave the space for dates of birth and death blank.
a NAME OF AUTHOR VARSITY SPIRIT FASHIONS & SUPPLIES INC DATES OF BIRTH AND DEATH Year Born Year Died _____________ _____________ Was this contribution to the work a "work made for hire"? [X] Yes [ ] No Author's Nationality or Domicile Name of Country Citizen of _______________ or Domiciled in United States Was this Author's Contribution to the Work Anonymous? [ ] Yes [X] No Pseudonymous? [ ] Yes [X] No
If the answer to either of these questions is "Yes," see detailed instructions.
Nature of Authorship Check appropriate box(es) See Instructions [ ] 3 Dimensional sculpture
[X] 2 Dimensional artwork [ ] Reproduction of work of art [ ] Map [ ] Photograph [ ] Jewelry design [ ] Technical drawing [ ] Text [ ] Architectural work b Name of Author ______________________ Dates of Birth and Death Year Born Year Died ______________ ______________ Was this contribution to the work a "work made for hire"? [ ] Yes [ ] No Author's Nationality or Domicile Name of Country Citizen of _______________ or Domiciled in United States Was this Author's Contribution to the Work Anonymous? [ ] Yes [ ] No Pseudonymous? [ ] Yes [ ] No
If the answer to either of these questions is "Yes," see detailed instructions.
Nature of Authorship Check appropriate box(es) See Instructions [ ] 3 Dimensional sculpture
EXHIBIT 18
Certificate of Registration Additional certificate (17 U.S.C. 706)
[Seal of the United States Copyright Office 1870]
This Certificate issued under the seal of the Copyright Office in accordance with title 17,United States Code, attests that registration has been made for the work identified below. The information on this certificate has been made a part of the Copyright Office records.
Form VA For a Work of the Visual Arts UNITED STATES COPYRIGHT OFFICE
RE VA 1-319-226 EFFECTIVE DATE OF REGISTRATION Month Day Year April 29 2005
Maria A. Pallante Acting Register of Copyrights, United. States of America
DO NOT WRITE ABOVE THIS LINE. IF YOU NEED MORE SPACE, USE A SEPARATE CONTINUATION SHEET
____________________________________________________ 1 Title of This Work NATURE OF THIS 299B WORK See instructions FABRIC DESIGN (ARTWORK) Previous or Alternative Titles _____________________________
Publication as a Contribution If this work was published as a contribution to a periodical, serial, or collection, give information about the collective work in which the contribution appeared.
Title of Collective Work ________________________________________________________ If published in a periodical or serial give: Volume Number Issue Date On Pages ________________________________________________________
2 NOTE Under the law the "author" of a "work made for hire " is generally the employer not the employee (see Instructions) For any part of this work that was made for hire, check Yes in the space provided, give the employer (or other person for whom the work was prepared) as "Author" of that part and leave the space for dates of birth and death blank.
a NAME OF AUTHOR VARSITY SPIRIT FASHIONS & SUPPLIES INC DATES OF BIRTH AND DEATH Year Born Year Died _____________ _____________ Was this contribution to the work a "work made for hire"? [X] Yes [ ] No Author's Nationality or Domicile Name of Country Citizen of _______________ or Domiciled in USA Was this Author's Contribution to the Work Anonymous? [ ] Yes [X] No Pseudonymous? [ ] Yes [X] No
If the answer to either of these questions is "Yes," see detailed instructions.
Nature of Authorship Check appropriate box(es) See Instructions
[ ] 3 Dimensional sculpture [X] 2 Dimensional artwork [ ] Reproduction of work of art [ ] Map [ ] Photograph [ ] Jewelry design [ ] Technical drawing [ ] Text [ ] Architectural work ______________________________________________ b Name of Author ______________________ Dates of Birth and Death Year Born Year Died _____________ _____________ Was this contribution to the work a "work made for hire"? [X] Yes [ ] No Author's Nationality or Domicile Name of Country Citizen of _______________ or Domiciled in _____________ Was this Author's Contribution to the Work Anonymous? [ ] Yes [X] No Pseudonymous? [ ] Yes [X] No
If the answer to either of these questions is "Yes," see detailed instructions.
Nature of Authorship Check appropriate box(es) See Instructions
AMENDED EXHIBIT 19
Certificate of Registration [Seal of the United States Copyright Office 1870]
This Certificate issued under the seal of the Copyright Office in accordance with title 17,United States Code, attests that registration has been made for the work identified below. The information on this certificate has been made a part of the Copyright Office records.
[Marybeth Peters] Register of Copyrights, United States of America
Form VA
For a Work of the Visual Arts
UNITED STATES COPYRIGHT OFFICE RE VA 1-411-535 [BARCODE]
EFFECTIVE DATE OF REGISTRATION
May 09 2007 Month Day Year
RATE CONTINUATION SHEET: _______________________________________________ 1 Title of This Work NATURE OF THIS Design Number 074 WORK See instructions 2-dimensional artwork Previous or Alternative Titles ________________________________________________
Publication as a Contribution If this work was published as a contribution to a periodical, serial, or collection, give information about the collective work in which the contribution appeared. Title of Collective Work _________________________________________________
If published in a periodical or serial give: Volume Number Issue Date On Pages ______________________________________________
2 NOTE Under the law the "author" of a "work made for hire" is generally the employer, not the employee (see instructions). For any part of this work that was "made for hire" check "Yes" in the space provided, give the employer (or other person for whom the "work" was prepared) as "Author" of that part and leave the space for dates of birth and death blank
a NAME OF AUTHOR Varsity Brands, Inc. DATES OF BIRTH AND DEATH Year Born Year Died _____________ _____________ Was this contribution to the work a "work made for hire"? [X] Yes [ ] No Author's Nationality or Domicile Name of Country Citizen of _______________ or Domiciled at United States Was this Author's Contribution to the Work Anonymous? [ ] Yes [X] No Pseudonymous? [ ] Yes [X] No
If the answer to either of these questions is "Yes," see detailed instructions.
Nature of Authorship Check appropriate box(es) See Instructions [ ] 3 Dimensional sculpture [X] 2 Dimensional artwork [ ] Reproduction of work of art
*1030Justice BREYER, with whom Justice KENNEDY joins, dissenting.
I agree with much in the Court's opinion. But I do not agree that the designs that Varsity Brands, Inc., submitted to the Copyright Office are eligible for copyright protection. Even applying the majority's test, the designs cannot "be perceived as ... two- or three-dimensional work[s] of art separate from the useful article." Ante, at 1007.
Look at the designs that Varsity submitted to the Copyright Office. See Appendix to opinion of the Court, ante . You will see *1031only pictures of cheerleader uniforms. And cheerleader uniforms are useful articles. A picture of the relevant design features, whether separately "perceived" on paper or in the imagination, is a picture of, and thereby "replicate[s]," the underlying useful article of which they are a part. Ante, at 1007, 1031. Hence the design features that Varsity seeks to protect are not "capable of existing independently o[f] the utilitarian aspects of the article." 17 U.S.C. § 101.
I
The relevant statutory provision says that the "design of a useful article" is copyrightable "only if, and only to the extent that, such design incorporates pictorial, graphic, or sculptural features that can be identified separately from, and are capable of existing independently of, the utilitarian aspects of the article." Ibid . But what, we must ask, do the words "identified separately" mean? Just when is a design separate from the "utilitarian aspect of the [useful] article?" The most direct, helpful aspect of the Court's opinion answers this question by stating:
"Nor could someone claim a copyright in a useful article merely by creating a replica of that article in some other medium-for example, a cardboard model of a car. Although the replica could itself be copyrightable, it would not give rise to any rights in the useful article that inspired it." Ante, at 1010.
Exactly so. These words help explain the Court's statement that a copyrightable work of art must be "perceived as a two- or three-dimensional work of art separate from the useful article." Ante, at 1007, 1015 - 1016. They help clarify the concept of separateness. Cf. 1 M. Nimmer & D. Nimmer, Nimmer on Copyright § 2A.08[A][1] (2016) (Nimmer) (describing courts' difficulty in applying that concept). They are consistent with Congress' own expressed intent. 17 U.S.C. § 101 ; H.R. Rep. No. 94-1476, pp. 55, 105 (1976) (H.R. Rep.). And they reflect long held views of the Copyright Office. See Compendium of U.S. Copyright Office Practices § 924.2(B) (3d ed. 2014), online at http://www.copyright.gov/comp3/docs/compendium.pdf (as last visited Mar. 7, 2017) (Compendium).
Consider, for example, the explanation that the House Report for the Copyright Act of 1976 provides. It says:
"Unless the shape of an automobile, airplane, ladies' dress, food processor, television set, or any other industrial product contains some element that, physically or conceptually, can be identified as separable from the utilitarian aspects of that article, the design would not be copyrighted...." H.R. Rep., at 55 (emphasis added).
These words suggest two exercises, one physical, one mental. Can the design features (the picture, the graphic, the sculpture) be physically removed from the article (and considered separately), all the while leaving the fully functioning utilitarian object in place? If not, can one nonetheless conceive of the design features separately without replicating a picture of the utilitarian object? If the answer to either of these questions is "yes," then the design is eligible for copyright protection. Otherwise, it is not. The abstract nature of these questions makes them sound difficult to apply. But with the Court's words in mind, the difficulty tends to disappear.
An example will help. Imagine a lamp with a circular marble base, a vertical 10-inch tall brass rod (containing wires) inserted off center on the base, a light bulb fixture emerging from the top of the brass rod, and a lampshade sitting on top. In front of the brass rod a porcelain Siamese cat sits on the base facing outward. Obviously, *1032the Siamese cat is physically separate from the lamp, as it could be easily removed while leaving both cat and lamp intact. And, assuming it otherwise qualifies, the designed cat is eligible for copyright protection.
Now suppose there is no long brass rod; instead the cat sits in the middle of the base and the wires run up through the cat to the bulbs. The cat is not physically separate from the lamp, as the reality of the lamp's construction is such that an effort to physically separate the cat and lamp will destroy both cat and lamp. The two are integrated into a single functional object, like the similar configuration of the ballet dancer statuettes that formed the lamp bases at issue in Mazer v. Stein, 347 U.S. 201, 74 S.Ct. 460, 98 L.Ed. 630 (1954). But we can easily imagine the cat on its own, as did Congress when conceptualizing the ballet dancer. See H.R. Rep., at 55 (the statuette in Mazer was "incorporated into a product without losing its ability to exist independently as a work of art"). In doing so, we do not create a mental picture of a lamp (or, in the Court's words, a "replica" of the lamp), which is a useful article. We simply perceive the cat separately, as a small cat figurine that could be a copyrightable design work standing alone that does not replicate the lamp. Hence the cat is conceptually separate from the utilitarian article that is the lamp. The pair of lamps pictured at Figures 1 and 2 in the Appendix to this opinion illustrate this principle.
Case law, particularly case law that Congress and the Copyright Office have considered, reflects the same approach. Congress cited examples of copyrightable design works, including "a carving on the back of a chair" and "a floral relief design on silver flatware." H.R. Rep., at 55. Copyright Office guidance on copyrightable designs in useful articles include "an engraving on a vase," "[a]rtwork printed on a t-shirt," "[a] colorful pattern decorating the surface of a shopping bag," "[a] drawing on the surface of wallpaper," and "[a] floral relief decorating the handle of a spoon." Compendium § 924.2(B). Courts have found copyrightable matter in a plaster ballet dancer statuette encasing the lamp's electric cords and forming its base, see Mazer, supra, as well as carvings engraved onto furniture, see Universal Furniture Int'l, Inc. v. Collezione Europa USA, Inc., 618 F.3d 417, 431-435 (C.A.4 2010) (per curiam ), and designs on laminated floor tiles, see Home Legend, LLC v. Mannington Mills, Inc., 784 F.3d 1404, 1412-1413 (C.A.11 2015). See generally Brief for Intellectual Property Professors as Amici Curiae .
By way of contrast, Van Gogh's painting of a pair of old shoes, though beautifully executed and copyrightable as a painting, would not qualify for a shoe design copyright. See Appendix, fig. 3, infra ; 17 U.S.C. §§ 113(a) - (b). Courts have similarly denied copyright protection to objects that begin as three-dimensional designs, such as measuring spoons shaped like heart-tipped arrows, Bonazoli v. R.S.V.P. Int'l, Inc., 353 F.Supp.2d 218, 226-227 (D.R.I.2005) ; candleholders shaped like sailboats, Design Ideas, Ltd. v. Yankee Candle Co., 889 F.Supp.2d 1119, 1128 (C.D.Ill.2012) ; and wire spokes on a wheel cover, Norris Industries, Inc. v. International Tel. & Tel. Corp., 696 F.2d 918, 922-924 (C.A.11 1983). None of these designs could qualify for copyright protection that would prevent others from selling spoons, candleholders, or wheel covers with the same design. Why not? Because in each case the design is not separable from the utilitarian aspects of the object to which it relates. The designs cannot be physically separated because they themselves make up the shape of the spoon, candleholders, or wheel covers of which they are a part.
*1033And spoons, candleholders, and wheel covers are useful objects, as are the old shoes depicted in Van Gogh's painting. More importantly, one cannot easily imagine or otherwise conceptualize the design of the spoons or the candleholders or the shoes without that picture, or image, or replica being a picture of spoons, or candleholders, or wheel covers, or shoes . The designs necessarily bring along the underlying utilitarian object. Hence each design is not conceptually separable from the physical useful object.
The upshot is that one could copyright the floral design on a soupspoon but one could not copyright the shape of the spoon itself, no matter how beautiful, artistic, or esthetically pleasing that shape might be: A picture of the shape of the spoon is also a picture of a spoon; the picture of a floral design is not. See Compendium § 924.2(B).
To repeat: A separable design feature must be "capable of existing independently" of the useful article as a separate artistic work that is not itself the useful article. If the claimed feature could be extracted without replicating the useful article of which it is a part, and the result would be a copyrightable artistic work standing alone, then there is a separable design. But if extracting the claimed features would necessarily bring along the underlying useful article, the design is not separable from the useful article. In many or most cases, to decide whether a design or artistic feature of a useful article is conceptually separate from the article itself, it is enough to imagine the feature on its own and ask, "Have I created a picture of a (useful part of a) useful article?" If so, the design is not separable from the useful article. If not, it is.
In referring to imagined pictures and the like, I am not speaking technically. I am simply trying to explain an intuitive idea of what separation is about, as well as how I understand the majority's opinion. So understood, the opinion puts design copyrights in their rightful place. The law has long recognized that drawings or photographs of real world objects are copyrightable as drawings or photographs, but the copyright does not give protection against others making the underlying useful objects. See, e.g., Burrow-Giles Lithographic Co. v. Sarony, 111 U.S. 53, 4 S.Ct. 279, 28 L.Ed. 349 (1884). That is why a copyright on Van Gogh's painting would prevent others from reproducing that painting, but it would not prevent others from reproducing and selling the comfortable old shoes that the painting depicts. Indeed, the purpose of § 113(b) was to ensure that " 'copyright in a pictorial, graphic, or sculptural work, portraying a useful article as such, does not extend to the manufacture of the useful article itself.' " H.R. Rep., at 105.
II
To ask this kind of simple question-does the design picture the useful article?-will not provide an answer in every case, for there will be cases where it is difficult to say whether a picture of the design is, or is not, also a picture of the useful article. But the question will avoid courts focusing primarily upon what I believe is an unhelpful feature of the inquiry, namely, whether the design can be imagined as a "two- or three-dimensional work of art." Ante, at 1007, 1015 - 1016. That is because virtually any industrial design can be thought of separately as a "work of art": Just imagine a frame surrounding the design, or its being placed in a gallery. Consider Marcel Duchamp's "readymades" series, the functional mass-produced objects he designated as art. See Appendix, fig. 4, infra . What is there in the world that, viewed through an esthetic lens, cannot *1034be seen as a good, bad, or indifferent work of art? What design features could not be imaginatively reproduced on a painter's canvas? Indeed, great industrial design may well include design that is inseparable from the useful article-where, as Frank Lloyd Wright put it, "form and function are one." F. Wright, An Autobiography 146 (1943) (reprint 2005). Where they are one, the designer may be able to obtain 15 years of protection through a design patent. 35 U.S.C. §§ 171, 173 ; see also McKenna & Strandburg, Progress and Competition in Design, 17 Stan. Tech. L. Rev. 1, 48-51 (2013). But, if they are one, Congress did not intend a century or more of copyright protection.
III
The conceptual approach that I have described reflects Congress' answer to a problem that is primarily practical and economic. Years ago Lord Macaulay drew attention to the problem when he described copyright in books as a "tax on readers for the purpose of giving a bounty to writers." 56 Parl. Deb. (3d Ser.) (1841) 341, 350. He called attention to the main benefit of copyright protection, which is to provide an incentive to produce copyrightable works and thereby "promote the Progress of Science and useful Arts." U.S. Const. Art. I, § 8, cl. 8. But Macaulay also made clear that copyright protection imposes costs. Those costs include the higher prices that can accompany the grant of a copyright monopoly. They also can include (for those wishing to display, sell, or perform a design, film, work of art, or piece of music, for example) the costs of discovering whether there are previous copyrights, of contacting copyright holders, and of securing permission to copy. Eldred v. Ashcroft, 537 U.S. 186, 248-252, 123 S.Ct. 769, 154 L.Ed.2d 683 (2003) (BREYER, J., dissenting). Sometimes, as Thomas Jefferson wrote to James Madison, costs can outweigh "the benefit even of limited monopolies." Letter from Thomas Jefferson to James Madison (July 31, 1788), in 13 Papers of Thomas Jefferson 443 (J. Boyd ed. 1956) (Jefferson Letter). And that is particularly true in light of the fact that Congress has extended the "limited Times" of protection, U.S. Const. Art. I, § 8, cl. 8, from the "14 years" of Jefferson's day to potentially more than a century today. Jefferson Letter 443; see also Eldred, supra, at 246-252, 123 S.Ct. 769 (opinion of BREYER, J.).
The Constitution grants Congress primary responsibility for assessing comparative costs and benefits and drawing copyright's statutory lines. Courts must respect those lines and not grant copyright protection where Congress has decided not to do so. And it is clear that Congress has not extended broad copyright protection to the fashion design industry. See, e.g., 1 Nimmer § 2A.08[H][3][c] (describing how Congress rejected proposals for fashion design protection within the 1976 Act and has rejected every proposed bill to this effect since then); Esquire, Inc. v. Ringer, 591 F.2d 796, 800, n. 12 (C.A.D.C.1978) (observing that at the time of the 1976 Copyright Act, Congress had rejected every one of the approximately 70 design protection bills that had been introduced since 1914); e.g., H.R. 5055, 109th Cong., 2d Sess.: "To Amend title 17, United States Code, to provide protection for fashion design" (introduced Mar. 30, 2006; unenacted). Congress has left "statutory ... protection ... largely unavailable for dress designs." 1 Nimmer § 2A.08[H][3] [a]; Raustiala & Sprigman, The Piracy Paradox: Innovation and Intellectual Property in Fashion Design, 92 Va. L.Rev. 1687, 1698-1705 (2006).
*1035Congress' decision not to grant full copyright protection to the fashion industry has not left the industry without protection. Patent design protection is available. 35 U.S.C. §§ 171, 173. A maker of clothing can obtain trademark protection under the Lanham Act for signature features of the clothing. 15 U.S.C. § 1051 et seq. And a designer who creates an original textile design can receive copyright protection for that pattern as placed, for example, on a bolt of cloth, or anything made with that cloth. E.g., Compendium § 924.3(A)(1). "[T]his [type of] claim ... is generally made by the fabric producer rather than the garment or costume designer," and is "ordinarily made when the two-dimensional design is applied to the textile fabric and before the garment is cut from the fabric." 56 Fed.Reg. 56531 (1991).
The fashion industry has thrived against this backdrop, and designers have contributed immeasurably to artistic and personal self-expression through clothing. But a decision by this Court to grant protection to the design of a garment would grant the designer protection that Congress refused to provide. It would risk increased prices and unforeseeable disruption in the clothing industry, which in the United States alone encompasses nearly $370 billion in annual spending and 1.8 million jobs. Brief for Council of Fashion Designers of America, Inc., as Amicus Curiae 3-4 (citing U.S. Congress, Joint Economic Committee, The New Economy of Fashion 1 (2016)). That is why I believe it important to emphasize those parts of the Court's opinion that limit the scope of its interpretation. That language, as I have said, makes clear that one may not "claim a copyright in a useful article merely by creating a replica of that article in some other medium," which "would not give rise to any rights in the useful article that inspired it." Ante, at 1010.
IV
If we ask the "separateness" question correctly, the answer here is not difficult to find. The majority's opinion, in its appendix, depicts the cheerleader dress designs that Varsity submitted to the Copyright Office. Can the design features in Varsity's pictures exist separately from the utilitarian aspects of a dress? Can we extract those features as copyrightable design works standing alone, without bringing along, via picture or design, the dresses of which they constitute a part?
Consider designs 074, 078, and 0815. They certainly look like cheerleader uniforms. That is to say, they look like pictures of cheerleader uniforms, just like Van Gogh's old shoes look like shoes. I do not see how one could see them otherwise. Designs 299A and 2999B present slightly closer questions. They omit some of the dresslike context that the other designs possess. But the necklines, the sleeves, and the cut of the skirt suggest that they too are pictures of dresses. Looking at all five of Varsity's pictures, I do not see how one could conceptualize the design features in a way that does not picture, not just artistic designs, but dresses as well.
Were I to accept the majority's invitation to "imaginatively remov[e]" the chevrons and stripes as they are arranged on the neckline, waistline, sleeves, and skirt of each uniform, and apply them on a "painter's canvas," ante, at 1011 - 1012, that painting would be of a cheerleader's dress. The esthetic elements on which Varsity seeks protection exist only as part of the uniform design-there is nothing to separate out but for dress-shaped lines that replicate the cut and style of the uniforms. Hence, each design is not physically separate, nor is it conceptually separate, from the useful article it depicts, namely, a *1036cheerleader's dress. They cannot be copyrighted.
Varsity, of course, could have sought a design patent for its designs. Or, it could have sought a copyright on a textile design, even one with a similar theme of chevrons and lines.
But that is not the nature of Varsity's copyright claim. It has instead claimed ownership of the particular " 'treatment and arrangement' " of the chevrons and lines of the design as they appear at the neckline, waist, skirt, sleeves, and overall cut of each uniform. Brief for Respondents 50. The majority imagines that Varsity submitted something different-that is, only the surface decorations of chevrons and stripes, as in a textile design. As the majority sees it, Varsity's copyright claim would be the same had it submitted a plain rectangular space depicting chevrons and stripes, like swaths from a bolt of fabric. But considered on their own, the simple stripes are plainly unoriginal. Varsity, then, seeks to do indirectly what it cannot do directly: bring along the design and cut of the dresses by seeking to protect surface decorations whose "treatment and arrangement" are coextensive with that design and cut . As Varsity would have it, it would prevent its competitors from making useful three-dimensional cheerleader uniforms by submitting plainly unoriginal chevrons and stripes as cut and arranged on a useful article. But with that cut and arrangement, the resulting pictures on which Varsity seeks protection do not simply depict designs. They depict clothing. They depict the useful articles of which the designs are inextricable parts. And Varsity cannot obtain copyright protection that would give them the power to prevent others from making those useful uniforms, any more than Van Gogh can copyright comfortable old shoes by painting their likeness.
I fear that, in looking past the three-dimensional design inherent in Varsity's claim by treating it as if it were no more than a design for a bolt of cloth, the majority has lost sight of its own important limiting principle. One may not "claim a copyright in a useful article merely by creating a replica of that article in some other medium," such as in a picture. Ante, at 1010. That is to say, one cannot obtain a copyright that would give its holder "any rights in the useful article that inspired it." Ante, at 1010.
With respect, I dissent.
APPENDIX TO OPINION OF BREYER, J.
4.2 Brenner v. Manson 4.2 Brenner v. Manson
Patents
BRENNER, COMMISSIONER OF PATENTS v. MANSON.
No. 58.
Argued November 17, 1965.
Decided March 21, 1966.
*520 Paul Bender argued the cause for petitioner, pro hac vice, by special leave of Court. With him on the brief were Solicitor General Marshall, Assistant Attorney General Douglas, Sherman L. Cohn and Edward Berlin.
Dean Laurence argued the cause for respondent. With him on the brief were Herbert I. Sherman and John L. White.
W. Brown Morton, Jr., and Ellsworth H. Mosher filed a brief for the American Patent Law Association, as amicus curiae, urging affirmance.
delivered the opinion of the Court.
This case presents two questions of importance to the administration of the patent laws: First, whether this Court has certiorari jurisdiction, upon petition of the Commissioner of Patents, to review decisions of the Court of Customs and Patent Appeals; and second, whether the practical utility of the compound produced by a chemical process is an essential element in establishing a prima facie case for the patentability of the process. The facts are as follows:
In December 1957, Howard Ringold and George Rosen-kranz applied for a patent on an allegedly novel process for making certain known steroids.1 They claimed *521priority as of December 17, 1956, the date on which they had filed for a Mexican patent. United States Patent No. 2,908,693 issued late in 1959.
In January 1960, respondent Manson, a chemist engaged in steroid research, filed an application to patent precisely the same process described by Ringold and Rosenkranz. He asserted that it was he who had discovered the process, and that he had done so before December 17, 1956. Accordingly, he requested that an “interference” be declared in order to try out the issue of priority between his claim and that of Ringold and Rosenkranz.2
A Patent Office examiner denied Manson’S application, and the denial was affirmed by the Board of Appeals within the Patent Office. The ground for rejection was the failure “to disclose any utility for” the chemical compound produced by the process. Letter of Examiner, dated May 24, 1960. This omission was not *522cured, in the opinion of the Patent Office, by Manson’s reference to an article in the November 1956 issue of the Journal of Organic Chemistry, 21 J. Org. Chem. 1333-1335, which revealed that steroids of a class which included the compound in question were undergoing screening for possible tumor-inhibiting effects in mice, and that a homologue3 adjacent to Manson’s steroid had proven effective in that role. Said the Board of Appeals, “It is our view that the statutory requirement of usefulness of a product cannot be presumed merely because it happens to be closely related to another compound which is known to be useful.”
The Court of Customs and Patent Appeals (hereinafter CCPA) reversed, Chief Judge Worley dissenting. 52 C. C. P. A. (Pat.) 739, 745, 333 F. 2d 234, 237-238. The court held that Manson was entitled to a declaration of interference since “where a claimed process produces a known product it is not necessary to show utility for the product,” so long as the product “is not alleged to be detrimental to the public interest.” Certiorari was granted, 380 U. S. 971, to resolve this running dispute over what constitutes “utility” in chemical process claims,4 as well as to answer the question concerning our certiorari jurisdiction.
*523I.
Section 1256 of Title 28 U. S. C. (1964 ed.), enacted in 1948, provides that “Cases in the Court of Customs and Patent Appeals may be reviewed by the Supreme Court by writ of certiorari.” This unqualified language would seem to foreclose any challenge to our jurisdiction in the present case. Both the Government5 and the respondent urge that we have certiorari jurisdiction over patent decisions of the CCPA, although the latter would confine our jurisdiction to those petitions filed by dissatisfied applicants and would deny the Commissioner of Patents the right to seek certiorari.6 This concert of opinion, does not settle the basic question because jurisdiction cannot be conferred by consent of the parties. The doubt that does exist stems from a decision of this *524Court, rendered in January 1927, in Postum Cereal Co. v. California Fig Nut Co., 272 U. S. 693, which has been widely interpreted as precluding certiorari jurisdiction over patent and trademark decisions of the CCPA.
Postum, however, was based upon a statutory scheme materially different from the present one. Postum involved a proceeding in the Patent Office to cancel a trademark. The Commissioner of Patents rejected the application. An appeal was taken to the then Court of Appeals for the District of Columbia, which in 1927 exercised the jurisdiction later transferred to the CCPA. Under the statutory arrangement in effect at the time, the judgment of the Court of Appeals was not definitive because it was not an order to the Patent Office determinative of the controversy. A subsequent bill in equity could be brought in the District Court and it was possible that a conflicting adjudication could thus be obtained. On this basis, the Court held that it could not review the decision of the Court of Appeals. It held that the conclusion of the Court of Appeals was an “administrative decision” rather than a “judicial judgment”: “merely an instruction to the Commissioner of Patents by a court which is made part of the machinery of the Patent Office for administrative purposes.” 272 U. S., at 698-699. Therefore, this Court concluded, the proceeding in the Court of Appeals — essentially administrative in nature — was neither case nor controversy within the meaning of Article III of the Constitution. Congress might confer such “administrative” tasks upon the courts of the District of Columbia, wrote Chief Justice Taft, but it could not empower this Court to participate therein.
Congress soon amended the statutory scheme. In March of 1927 it provided that an action in the District Court was to be alternative and not cumulative to appellate review, that it could not be maintained to overcome
*525an adjudication in the Court of Appeals.7 In 1929 Congress transferred appellate jurisdiction over the Commissioner’s decisions from the Court of Appeals to what had been the Court of Customs Appeals and was now styled the Court of Customs and Patent Appeals.8 Whereas the Court of Appeals had been empowered to take additional evidence and to substitute its judgment for that of the Commissioner, the CCPA was confined to the record made in the Patent Office.9 Compare Federal Communications Comm’n v. Pottsville Broadcasting Co., 309 U. S. 134, 144-145. Despite these changes, however, Postum had acquired a life of its own. It continued to stand in the way of attempts to secure review here of CCPA decisions respecting the Commissioner of Patents. See, e. g., McBride v. Teeple, 311 U. S. 649, denying certiorari for “want of jurisdiction” on the authority of Postum. 10
This was the background against which Congress, in its 1948 codification of statutes pertaining to the judiciary, enacted § 1256, blandly providing in unqualified language for review on certiorari of “[cjases in the Court of Customs and Patent Appeals.” Nothing in the legislative materials relating to the statute, except its language, is of assistance to us in the resolution of the present problem: Did the statutory changes which followed *526 Postura mean that a patent decision by the CCPA was a “judicial” determination reviewable by this Court under Article III? And, if so, was § 1256 intended to create such jurisdiction?
Assistance came with the 1958 revision of the Judicial Code. Congress there declared the CCPA “a court established under article III . . . ,” that is, a constitutional court exercising judicial rather than administrative power. 28 U. S. C. § 211 (1964 ed.). In 1962 this Court addressed itself to the nature and status of the CCPA. Olidden Co. v. Zdanok, 370 U. S. 530, raised the question whether a judge of the CCPA was an Article III judge, capable of exercising federal judicial power. In answering that question in the affirmative, MR. Justice Harlan's opinion, for three of the seven Justices participating, expressly left open the question whether § 1256 conferred certiorari jurisdiction over patent and trademark cases decided in the CCPA, 370 U. S., at 578 n. 49. It suggested, however, that Postura might be nothing-more than a museum piece. The opinion noted that Postura “must be taken to be limited to the statutory scheme in existence before” 1929. 370 U. S., at 579. The concurring opinion of Mr. Justice Clark, in which The Chief Justice joined, did not reflect any difference on this point.
Thus, the decision sought to be reviewed is that of an Article III court. It is “judicial” in character. It is not merely an instruction to the Commissioner or part of the “administrative machinery” of the Patent Office. It is final and binding in the usual sense.11 In sum, Postura *527has no vitality in the present setting, and there remains no constitutional bar to our jurisdiction.
Having arrived at this conclusion, we have no difficulty in giving full force and effect to the generality of the language in § 1256. It would be entirely arbitrary for us to assume, despite the statutory language, that Congress in 1948 intended to enshrine Postum — dependent as it was upon a statutory scheme fundamentally altered in 1927 and 1929 — as a hidden exception to the sweep of § 1256. The contrary is more plausible: that by using broad and unqualified language, Congress intended our certiorari jurisdiction over CCPA cases to be as broad as the Constitution permits.
This conclusion is reinforced by reference to the anomalous consequences which would result were we to adopt a contrary view of § 1256. Determinations of the Patent Office may be challenged either by appeal to the CCPA or by suit instituted in the United States District Court for the District of Columbia. 35 U. S. C. § 145, 28 U. S. C. § 1542 (1964 ed.). Where the latter route is elected, the decision obtained may be reviewed in the Court of Appeals for the District of Columbia Circuit, and ultimately in this Court upon writ of certiorari. Hoover Co. v. Coe, 325 U. S. 79. It would be strange indeed if corresponding certiorari jurisdiction did not exist where the alternative route was elected. Were that so, in the event of conflict between the CCPA and the courts of the District of Columbia, resolution by this Court would be achievable only if the litigants chose to proceed through the latter. Obviously, the orderly administration both of our certiorari jurisdiction and of the patent laws requires that ultimate review be available in this Court, regardless of the route chosen by the litigants.
*528We therefore conclude that § 1256 authorizes the grant of certiorari in the present case. We now turn to the merits.12
II.
Our starting point is the proposition, neither disputed nor disputable, that one may patent only that which is *529“useful.” In Graham v. John Deere Co., ante, p.' 1, at 5-10, we have reviewed the history of the requisites of patentability, and it need not be repeated here. Suffice it to say that the concept of utility has maintained a central place in all of our patent legislation, beginning with the first patent law in 1790 13 and culminating in the present law’s provision that
“Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.” 14
As is so often the case, however, a simple, everyday word can be pregnant with ambiguity when applied to the facts of life. That this is so is demonstrated by the present conflict between the Patent Office and the CCPA over how the test is to be applied to a chemical process which yields an already known product whose utility— other than as a possible object of scientific inquiry — has not yet been evidenced. It was not long ago that agency and court seemed of one mind on the question. In Application of Bremner, 37 C. C. P. A. (Pat.) 1032, 1034, 182 F. 2d 216, 217, the court affirmed rejection by the Patent Office of both process and product claims. It noted that “no use for the products claimed to be developed by the processes had been shown in the specification.” It held that “It was never intended that a patent be granted upon a product, or a process producing a product, unless such product be useful.” Nor was this new doctrine in the court. See Thomas v. Michael, 35 C. C. P. A. (Pat.) 1036, 1038-1039, 166 F. 2d 944, 946-947.
*530The Patent Office has remained steadfast in this view. The CCPA, however, has moved sharply away from Bremner. The trend began in Application of Nelson, 47 C. C. P. A. (Pat.) 1031, 280 F. 2d 172. There, the court reversed the Patent Office’s rejection of a claim on a process yielding chemical intermediates “useful to chemists doing research on steroids,” despite the absence of evidence that any of the steroids thus ultimately produced were themselves “useful.” The trend has accelerated,15 culminating in the present case where the court held it sufficient that a process produces the result intended and is not “detrimental to the public interest.” 52 C. C. P. A. (Pat.), at 745, 333 F. 2d, at 238.
It is not remarkable that differences arise as to how the test of usefulness is to be applied to chemical processes. Even if we knew precisely what Congress meant in 1790 when it devised the “new and useful” phraseology and in subsequent re-enactments of the test, we should have difficulty in applying it in the context of contemporary chemistry where research is as comprehensive as man’s grasp and where little or nothing is wholly beyond the pale of “utility” — if that word is given its broadest reach.
Respondent does not — -at least in the first instance— rest upon the extreme proposition, advanced by the court below, that a novel chemical process is patentable so long *531as it yields the intended product16 and so long as the product is not itself “detrimental.” Nor does he commit the outcome of his claim to the slightly more conventional proposition that any process is “useful” within the meaning of § 101 if it produces a compound whose potential usefulness is under investigation by serious scientific researchers, although he urges this position, too, as an alternative basis for affirming the decision of the CCPA. Rather, he begins with the much more orthodox argument that his process has a specific utility which would entitle him to a declaration of interference even under the Patent Office’s reading of § 101. The claim is that the supporting affidavits filed pursuant to Rule 204 (b), by reference to Ringold’s 1956 article, reveal that an adjacent homologue of the steroid yielded by his process has been demonstrated to have tumor-inhibiting effects in mice, and that this discloses the requisite utility. We do not accept any of these theories as an adequate basis for overriding the determination of the Patent Office that the “utility” requirement has not been met.
Even on the assumption that the process would be patentable were respondent to show that the steroid produced had a tumor-inhibiting effect in mice,17 we would *532not overrule the Patent Office finding that respondent has not made such a showing. The Patent Office held that, despite the reference to the adjacent homologue, respondent’s papers did not disclose a sufficient likelihood that the steroid yielded by his process would have similar tumor-inhibiting characteristics. Indeed, respondent himself recognized that the presumption that adjacent homologues have the same utility18 has been challenged in the steroid field because of “a greater known unpredictability of compounds in that field.” 19 In these circumstances and in this technical area, we would not overturn the finding of the Primary Examiner, affirmed by the Board of Appeals and not challenged by the CCPA.
The second and third points of respondent’s argument present issues of much importance. Is a chemical process “useful” within the meaning of § 101 either (1) because it works — i. e., produces the intended product? or (2) because the compound yielded belongs to a class of compounds now the subject of serious scientific investigation? These contentions present the basic problem for our adjudication. Since we find no specific assistance in the legislative materials underlying § 101, we are remitted to an analysis of the problem in light of the general intent of Congress, the purposes of the patent system, and the implications of a decision one way or the other.
In support of his plea that we attenuate the requirement of “utility,” respondent relies upon Justice Story’s *533well-known statement that a “useful” invention is one “which may be applied to a beneficial use in society, in contradistinction to an invention injurious to the morals, health, or good order of society, or frivolous and insignificant” 20 — and upon the assertion that to do so would encourage inventors of new processes to publicize the event for the benefit of the entire scientific community, thus widening the search for uses and increasing the fund of scientific knowledge. Justice Story’s language sheds little light on our subject. Narrowly read, it does no more than compel us to decide whether the invention in question is “frivolous and insignificant” — a query no easier of application than the one built into the statute. Read more broadly, so as to allow the patenting of any invention not positively harmful to society, it places such a special meaning on the word “useful” that we cannot accept it in the absence of evidence that Congress so intended. There are, after all, many things in this world which may not be considered “useful” but which, nevertheless, are totally without a capacity for harm.
It is true, of course, that one of the purposes of the patent system is to encourage dissemination of information concerning discoveries and inventions.21 And it may be that inability to patent a process to some extent discourages disclosure and leads to greater secrecy than would otherwise be the case. The inventor of the process, or the corporate organization by which he is employed, has some incentive to keep the invention *534secret while uses for the product are searched out. However, in light of the highly developed art of drafting patent claims so that they disclose as little useful information as possible — while broadening the scope of the claim as widely as possible — the argument based upon the virtue of disclosure must be warily evaluated. Moreover, the pressure for secrecy is easily exaggerated, for if the inventor of a process cannot himself ascertain a “use” for that which his process yields, he has every incentive to make his invention known to those able to do so. Finally, how likely is disclosure of a patented process to spur research by others into the uses to which the product may be put? To the extent that the patentee has power to enforce his patent, there is little incentive for others to undertake a search for uses.
Whatever weight is attached to the value of encouraging disclosure and of inhibiting secrecy, we believe a more compelling consideration is that a process patent in the chemical field, which has not been developed and pointed to the degree of specific utility, creates a monopoly of knowledge which should be granted only if clearly commanded by the statute. Until the process claim has been reduced to production of a product shown to be useful, the metes and bounds of that monopoly are not capable of precise delineation. It may engross a vast, unknown, and perhaps unknowable area. Such a patent may confer power to block off whole areas of scientific development,22 without compensating benefit to the public. The basic quid pro quo contemplated by the Constitution and the Congress for granting a patent monopoly is the benefit derived by the public from an invention with substantial utility. Unless and until a process is refined and developed to this point — where specific bene*535fit exists in currently available form — there is insufficient justification for permitting an applicant to engross what may prove to be a broad field.
These arguments for and against the patentability of a process which either has no known use or is useful only in the sense that it may be an object of scientific research would apply equally to the patenting of the product produced by the process. Respondent appears to concede that with respect to a product, as opposed to a process, Congress has struck the balance on the side of non-patentability unless “utility” is shown. Indeed, the decisions of the CCPA are in accord with the view that a product may not be patented absent a showing of utility greater than any adduced in the present case.23 We find absolutely no warrant for the proposition that although Congress intended that no patent be granted on a chemical compound whose sole “'utility” consists of its potential role as an object of use-testing, a different set of rules was meant to apply to the process which yielded the unpatentable product.24 That proposition seems to us little more than an attempt to evade the impact of the rules which concededly govern patentability of the product itself.
This is not to say that we mean to disparage the importance of contributions to the fund of scientific infor*536mation short of the invention of something “useful,” or that we are blind to the prospect that what now seems without “use” may tomorrow command the grateful attention of the public. But a patent is not a hunting license. It is not a reward for the search, but compensation for its successful conclusion. “[A] patent system must be related to the world of commerce rather than to the realm of philosophy. . . .” 25
The judgment of the CCPA is Reversed.
Mr. Justice Douglas, while acquiescing in Part I of the Court’s opinion, dissents on the merits of the controversy for substantially the reasons stated by Mr. Justice Harlan.
concurring in part and dissenting in part.
While I join the Court’s opinion on the issue of cer-tiorari jurisdiction, I cannot agree with its resolution of the important question of patentability.
Respondent has contended that a workable chemical process, which is both new and sufficiently nonobvious to satisfy the patent statute, is by its existence alone a contribution to chemistry and “useful” as the statute employs that term.1 Certainly this reading of “useful” in the statute is within the scope of the constitutional grant, which states only that “[t]o promote the Progress of Science and useful Arts,” the exclusive right to “Writings and Discoveries” may be secured for limited times to those who produce them. Art. I, § 8. Yet the patent statute is somewhat differently worded and is on *537its face open both to respondent’s construction and to the contrary reading given it by the Court. In the absence of legislative history on this issue, we are thrown back on policy and practice. Because I believe that the Court’s policy arguments are not convincing and that past practice favors the respondent, I would reject the narrow definition of “useful” and uphold the judgment of the Court of Customs and Patent Appeals (hereafter CCPA).
The Court’s opinion sets out about half a dozen reasons in support of its interpretation. Several of these arguments seem to me to have almost no force. For instance, it is suggested that “[ujntil the process claim has been reduced to production of a product shown to be useful, the metes and bounds of that monopoly are not capable of precise delineation” (p. 534, ante) and “[i]t may engross a vast, unknown, and perhaps unknowable area” (p. 534, ante). I fail to see the relevance of these assertions; process claims are not disallowed because the products they produce may be of “vast” importance nor, in any event, does advance knowledge of a specific product use provide much safeguard on this score or fix “metes and bounds” precisely since a hundred more uses may be found after a patent is granted and greatly enhance its value.
The further argument that an established product use is part of “[t]he basic quid pro quo” (p. 534, ante) for the patent or is the requisite “successful conclusion” (p. 536, ante) of the inventor’s search appears to beg the very question whether the process is “useful” simply because it facilitates further research into possible product uses. The same infirmity seems to inhere in the Court’s argument that chemical products lacking immediate utility cannot be distinguished for present purposes from the processes which create them, that respondent appears to concede and the CCPA holds that *538the products are nonpatentable, and that therefore the processes are nonpatentable. Assuming that the two classes cannot be distinguished, a point not adequately considered in the briefs, and assuming further that the CCPA has firmly held such products non-patentable,2 this permits us to conclude only that the CCPA is wrong either as to the products or as to the processes and affords no basis for deciding whether both or neither should be patentable absent a specific product use.
More to the point, I think, are the Court’s remaining, prudential arguments against patentability: namely, that disclosure induced by allowing a patent is partly undercut by patent-application drafting techniques, that disclosure may occur without granting a patent, and that a patent will discourage others from inventing uses for the product. How far opaque drafting may lessen the public benefits resulting from the issuance of a patent is not shown by any evidence in this case but, more important, the argument operates against all patents and gives no reason for singling out the class involved here. The thought that these inventions may be more likely than most to be disclosed even if patents are not allowed may have more force; but while empirical study of the industry might reveal that chemical researchers would behave in this fashion, the abstractly logical choice for them seems to me to maintain secrecy until a product use can be discovered. As to discouraging the search by *539others for product uses, there is no doubt this risk exists but the price paid for any patent is that research on other uses or improvements may be hampered because the original patentee will reap much of the reward. From the standpoint of the public interest the Constitution seems to have resolved that choice in favor of patentability.
What I find most troubling about the result reached by the Court is the impact it may have on chemical research. Chemistry is a highly interrelated field and a tangible benefit for society may be the outcome of a number of different discoveries, one discovery building upon the next. To encourage one chemist or research facility to invent and disseminate new processes and products may be vital to progress, although the product or process be without “utility” as the Court defines the term, because that discovery permits someone else to take a further but perhaps less difficult step leading to a commercially useful item. In my view, our awareness in this age of the importance of achieving and publicizing basic research should lead this Court to resolve uncertainties in its favor and uphold the respondent’s position in this case.
This position is strengthened, I think, by what appears to have been the practice of the Patent Office during most of this century. While available proof is not conclusive, the commentators seem to be in agreement that until Application of Bremner, 37 C. C. P. A. (Pat.) 1032, 182 F. 2d 216, in 1950, chemical patent applications were commonly granted although no resulting end use was stated or the statement was in extremely broad terms.3 Taking this to be true, Bremner represented *540a deviation from established practice which the CCPA has now sought to remedy in part only to find that the Patent Office does not want to return to the beaten track. If usefulness was typically regarded as inherent during a long and prolific period of chemical research and development in this country, surely this is added reason why the Court’s result should not be adopted until Congress expressly mandates it, presumably on the basis of empirical data which this Court does not possess.
Fully recognizing that there is ample room for disagreement on this problem when, as here, it is reviewed in the abstract, I believe the decision below should be affirmed.
4.3 Norris Industries, Inc. v. International Telephone & Telegraph Corp. 4.3 Norris Industries, Inc. v. International Telephone & Telegraph Corp.
Copyright
NORRIS INDUSTRIES, INC., Plaintiff-Appellant, v. INTERNATIONAL TELEPHONE AND TELEGRAPH CORPORATION and David L. Ladd, Register of Copyrights, Defendants-Appellees.
No. 81-5915.
United States Court of Appeals, Eleventh Circuit.
Jan. 24, 1983.
Rehearing and Rehearing En Banc Denied March 18, 1983.
*919Ervin, Varn, Jacobs, Odom & Kitchen, Wilfred C. Varn, Tallahassee, Fla., Harness, Dickey & Pierce, Joseph R. Papp, Charles H. Blair, Sohn A. Sinclair, Birmingham, Mich., for plaintiff-appellant.
Pennie & Edmonds, Robert M. Kunstadt, John E. Kidd, Joseph J.C. Ranalli, New York City, Robert J. Kelly, Tallahassee, Fla., Ronald W. Alice, New York City, for defendants-appellees.
Before RONEY and HATCHETT, Circuit Judges, and WISDOM,* Senior Circuit Judge.
This case involves whether wire-spoked wheel covers for automobiles are entitled to copyright protection as opposed to industrial design or some other trade protection. The district court in this copyright infringement suit concluded on summary judgment that the wheel covers are not entitled to copyright protection because they are useful articles without separable features. We affirm.
There are essentially two issues in the case: (1) whether the wheel covers are useful or ornamental and (2) whether, even if useful, there is separable art work that deserves copyright protection. The difference between the copyright protection sought by the plaintiff and protection as an industrial design primarily lies in the length of time and scope of protection given.1 That the wheel covers might be protected under some other law is not before this panel. The sole issue is whether the wheel covers as presented for registration are copyrightable.
Background
In 1975 Norris Industries, Inc. created a design for a wheel cover that simulates a wire wheel and applied to the Copyright Office for copyright registration. The first two applications were rejected. Norris reapplied and pointed out a district court decision that found the design of a lighting fixture to be copyrightable. Esquire, Inc. v. Ringer, 414 F.Supp. 939 (D.D.C.1976). The Copyright Office “in line with [its] policy of resolving doubtful cases in favor of registration” then granted copyright registration. In 1977 Norris created a second design. Four applications for copyright registration were submitted and rejected. *920During this process the D.C.Circuit Court of Appeals reversed Esquire. Esquire, Inc. v. Ringer, 192 U.S.App.D.C. 187, 591 F.2d 796 (1978), cert. denied, 440 U.S. 908, 99 S.Ct. 1217, 59 L.Ed.2d 456 (1979). The second wheel cover was rejected by the Copyright Office as a utilitarian article without separable copyrightable aspects on the authority of Esquire. 2
Late in 1980 Norris instituted suit in the Northern District of Florida against International Telephone and Telegraph Corporation (ITT), alleging copyright and patent infringement. Pursuant to 17 U.S.C.A. § 411(a), the Register of Copyrights joined in the suit as a party defendant for the purpose of contesting registrability.3 ITT then filed suit against Norris in a federal court in California for declaratory judgment as to the validity of Norris’ copyrights and patents. On Norris’ motion, the California suit was transferred to Florida,4 and the two suits were consolidated.5 Both ITT and the Register moved for partial summary judgment on the copyright counts. The district court granted the two motions and declared both the registered and the unregistered copyrights invalid as a matter of law.
Applicable Copyright Law
As an initial skirmish the parties disagree on whether the Copyright Act of 1909, ch. 320, 35 Stat. 1075, as amended, or the Copyright Act of 1976, 17 U.S.C.A. §§ 101-810, is applicable to the instant case although we are not sure it makes any difference. Norris’ first registered copyright was granted before the January 1, 1978 effective date of the 1976 Act and therefore the 1909 Act applied. Norris also applied for registration of the second wheel cover before January 1, 1978. The transitional and supplementary provisions of the 1976 Act provide that registration shall be made in accordance with the Act as it existed on December 31,1977, where the deposit application and fee were received before January 1, 1978. Copyright Act of 1976, P.L. No. 94-553, § 109, 90 Stat. 2600 (1976). Therefore, we agree with Norris that the Copyright Act of 1909 applies to questions of registrability.
In any event, the legislative history of the Act of 1976 indicates that Congress endorsed the practice of the Copyright Office under the Act of 1909 as it related to registrability of utilitarian articles. The express congressional intent was merely to clarify the statutory distinction between works of applied art eligible for copyright protection and industrial designs ineligible for protection. H.R.Rep. No. 94-1476, 94th Cong. 2d Sess. 54-55, reprinted in [1976] U.S.Code Cong. & Ad.News 5659, 5667-68. See also Esquire, Inc. v. Ringer, 591 F.2d at 803 (1976 Act and its legislative history express congressional understanding of scope of protection for utilitarian articles). *921The language from existing guidelines and regulations of the Copyright Office was incorporated into the 1976 Act. Under this Act the definition of useful article is “an article having an intrinsic utilitarian function that is not merely to portray the appearance of the article or to convey information” and includes articles that are normally a part of a useful article. 17 U.S. C.A. § 101. The design of a useful article qualifies as a copyrightable pictorial, graphic, or sculptural work only to the extent that the design incorporates features that “can be identified separately from, and are capable of existing independently of, the utilitarian aspects of the article.” Id.
Because Congress intended to codify the existing practice of the Copyright Office with respect to the dichotomy between ornamental and useful articles, the 1976 Act and its legislative history are useful aids in interpreting the 1909 Act and regulations adopted pursuant to it. A further reason. for noting the identity of treatment of this matter under the 1909 Act and the 1976 Act is to explain why cases arising under the 1976 Act are relevant to this case arising under the 1909 Act.
Important to this litigation is the intended scope of the copyright laws. Congressional authority to provide authors copyright protection for their writings is found in Article I, Section 8 of the Constitution. Without extending copyright protection to the limits of the Constitution, Congress has sought a flexible definition of writings. The 1909 Act gave content to the phrase “writings of an author,” Copyright Act, ch. 391, § 4, 61 Stat. 654 (1947), by setting out classes of works eligible for registration including “works of art.” Id. § 5(g), 61 Stat. 654 (1947).6
Under guidelines established by the Copyright Office, a distinction was made between ornamental and useful articles.7 Or*922namental articles are “all decorative articles designed primarily to beautify, embellish, or adorn,” Compendium of Copyright Office Practices, 2.8.1.I.b.1., and if the requirements of originality and creativity are satisfied, are entitled to registration. Useful articles are “designed primarily to serve a utilitarian function,” id. 2.8.1.I.b.2., and are not entitled to registration, except for features “which can be identified separately and are capable of existing independently as a work of art.” 37 C.F.R. § 202.10(c) (1977) [superseded]. As indicated in the above discussion, these regulations were effectively codified in the 1976 Act. 17 U.S. C.A. § 101.
Discussion
Norris contends first that its wheel covers are ornamental articles, not useful, designed to beautify, embellish, and adorn the wheels of automobiles. The district court held that the wheel covers are utilitarian articles serving as hubcaps to protect lug-nuts, brakes, wheels, and axles from damage and corrosion, as had been determined by the Register in rejecting the copyright application.8 The court added that the wheel covers were designed to be part of an automobile, which is a useful article.
The district court properly gave some deference to the expertise of the Register in its decision. The Supreme Court has accorded deference to the Register’s drawing of the line between ornamental and useful articles. In Mazer v. Stein, 347 U.S. 201, 74 S.Ct. 460, 98 L.Ed. 630 (1954), the seminal case on the copyrightability of works of art used in industry, the Court referred to the agency’s construction of the statute in its regulation permitting registration of works of artistic craftsmanship only as to form and not as to mechanical or utilitarian aspects. Id. at 212-13, 74 S.Ct. at 467-68. The District of Columbia Circuit has noted the considerable expertise of the Register in defining the boundaries between copyrightable works of art and noncopyrightable industrial designs and has accorded considerable weight to the refusal of registration. See Esquire, Inc. v. Ringer, 591 F.2d at 801. The Fourth Circuit has also expressly deferred to the Register’s view on the registrability of utilitarian articles. See Eltra Corp. v. Ringer, 579 F.2d 294, 297-98 (4th Cir.1978). The expertise relied on is not technical expertise in the use of the article submitted for registration but expertise in the interpretation of the law and its application to the facts presented by the copyright application. The Copyright Office has been concerned with the distinction between copyrightable and noncopyrightable works of art since the Copyright Act of 1870 characterized copyrightable subject matter as works of fine art. See Mazer v. Stein, 347 U.S. at 209-13, 74 S.Ct. at 466-68.
These determinations are routinely made by the Register and are unquestionably related to the substantive area of the agency’s business. Therefore, we find no error in the district court’s reliance on the Register’s opinion that Norris’ wheel covers are useful articles as that term is used in the Act. This is not to say that the court should simply accept the Register’s decision without question. But our view of the uncontradicted evidence and the graphic description of the wheel cover reveals no abuse of administrative discretion by the Register. See Esquire, Inc. v. Ringer, 591 F.2d at 805-06 & n. 28. The court’s statement that it took judicial notice of the history of the automobile hubcap, a statement faulted by Norris, was nothing more than an assessment of the evidence in light of common knowledge and does not require a reversal of the court’s decision.
We therefore hold that the district court was correct in deciding that the wirespoked wheel covers are useful within the meaning of the copyright law.
*923As useful articles any features that can be identified separately and exist independently of the utilitarian aspects of the article may still be copyrighted as a work of art. 37 C.F.R. § 202.10(c) (1977) [superseded]; 17 U.S.C.A. § 101. Both case law and legislative history indicate that separability encompasses works of art that are either physically severable from the utilitarian article or conceptually severable. 1 M. Nimmer, Nimmer on Copyright, § 2.08[B][3], at 2-94 — 2-96.1 (1982). In a noted case involving physical separability, a simulated antique telephone which encased a pencil sharpener was found to be copyrightable. Ted Arnold, Ltd. v. Silvercraft Co., 259 F.Supp. 733 (S.D.N.Y.1966). Although the pencil sharpener was a useful article, the telephone casing was physically severable and capable of existing independently as a work of art. The court stressed that “the copyrighted article is the simulation of an antique telephone, not the pencil sharpener inside, and not the combination of the two.” Id. at 734. Similarly, in Mazer v. Stein, 347 U.S. 201, 74 S.Ct. 460, 98 L.Ed. 630 (1954), the statuettes used as lamp bases were undeniably capable of existing as independent works of art. In fact, the dancing figures without any lamp components were first registered as works of art. Id. at 202-03, 74 S.Ct. at 462. The Supreme Court held that works of art do not lose their copyrightability when incorporated into a useful article. Id. at 218, 74 S.Ct. at 471.
The district court considered the ornamental and utilitarian aspects of Norris’ wheel covers to be inseparable. The court noted that the appearance of a wire wheel was conveyed by the entire wheel cover taken as a whole. The record indicates that Norris sought copyright registration for the entire wheel cover. Even if the arrangement of spokes could be identified separately, however, they are incapable of existing independently. As the Register and the district court observed, the pattern resulting from the placement of spokes is an inseparable component of the wheel cover. The spokes are attached to the rim and to the hub, and once the spokes are removed from their position the pattern ceases to exist. See SCOA Industries, Inc. v. Famolare, Inc., 192 U.S.P.Q. 216, 218 (S.D.N.Y. 1976) (sculptured design of wavy lines, ridges, and troughs in shoe soles cannot be identified nor exist independently as works of art). Nimmer concludes that “[p]hysical separability would presumably mean that after removal of those features which are necessary for the utilitarian function of the article, the artistic features would nevertheless remain intact.” 1 M. Nimmer, supra, § 2.08[B][3], at 2-96. Norris emphasizes that the spokes and hub can be removed from the backing plate, but the hub and spokes do not constitute the simulated wire wheel for which copyright protection was sought. See Esquire, Inc. v. Ringer, 591 F.2d at 806. Nor do the hub and spokes constitute a feature capable of existing independently as a work of art.
Norris argues that, even if not physically separable, the aesthetically pleasing pattern formed by the spokes is nevertheless conceptually separable. Attempting to draw a line between “copyrightable works of applied art and uncopyrighted works of industrial design,” the Committee commented on the conceptually separable elements of useful articles such as a carving on the back of a chair and a floral relief design on silver flatware. H.R.Rep. No. 94-1476, 94th Cong., 2d Sess. 55, reprinted in [1976] U.S. Code Cong. & Ad.News 5659, 5668.9 These *924items are ornamental, superfluous designs contained within useful objects. See Scarves by Vera, Inc. v. United Merchants and Manufacturers, Inc., 173 F.Supp. 625 (S.D.N.Y.1959) (designs printed on blouses copyrightable); Syracuse China Corp. v. Stanley Roberts, Inc., 180 F.Supp. 527 (S.D.N.Y.1960) (designs on dinnerware copyrightable).
On the other hand, functional components of useful articles, no matter how artistically designed, have generally been denied copyright protection unless they are physically separable from the useful article. See Esquire, Inc. v. Ringer, 192 U.S.App.D.C. 187, 591 F.2d 796 (1978), cert. denied, 440 U.S. 908, 99 S.Ct. 1217, 59 L.Ed.2d 456 (1979) (overall design of lighting fixture not copyrightable); SCOA Industries, Inc. v. Famolare, Inc., 192 U.S.P.Q. 216 (S.D.N.Y.1976) (wavy lines on shoe soles not copyrightable); Vacheron & Constantin — Le Coultre Watches, Inc. v. Benrus Watch Co., 155 F.Supp. 932 (S.D.N.Y.1957), aff’d in part, reversed on other grounds, 260 F.2d 637 (2d Cir.1958) (artistically designed watch face not copyrightable). The wire wheel covers in this case are not superfluous ornamental designs, but functional components of utilitarian articles.
In Kieselstein-Cord v. Accessories By Pearl, Inc., 632 F.2d 989 (2d Cir.1980), a case “on a razor’s edge of copyright law,” 632 F.2d at 990, the Second Circuit upheld the Copyright Office’s determination that sculptured belt buckles were eligible for copyright protection. The court viewed their primary sculptural elements as “conceptually separable” from their subsidiary utilitarian function. Id. at 993. The belt buckles in Kieselstein-Cord had been registered by the Copyright Office as jewelry. Unlike the shape of useful articles, the form of jewelry has traditionally been considered copyrightable. See Boucher v. DuBoyes, Inc., 253 F.2d 948, 949 (2d Cir.), cert. denied, 357 U.S. 936, 78 S.Ct. 1384, 2 L.Ed.2d 1550 (1958). Noting this traditional copyright protection of jewelry, the court stated that its conclusion that the sculptured belt buckles were copyrightable was “not at variance with the expressed congressional intent to distinguish copyrightable applied art and uncopyrighted industrial design.” Kieselstein-Cord, 632 F.2d at 993.
Wire wheel covers do not fall within any of the categories of fine or applied art which have been traditionally considered copyrightable. The wheel covers do not contain a superfluous sculptured design, serving no function, that can be identified apart from the wheel covers themselves. As we have suggested to be appropriate, the court in Kieselstein-Cord gave some deference to the Copyright Office’s determination that the belt buckles were copyrightable works of art, noting that “[t]he Copyright Office continually engages in the drawing of lines between that which may be and that which may not be copyrighted.” Id. at 994. The Copyright Register here refused to register the wheel covers.
The district court did not err in concluding that the wire wheel covers, as useful articles containing no physically or conceptually separable works of art, are not entitled to copyright protection.
AFFIRMED.
4.4 Traffix Devices, Inc. v. Marketing Displays, Inc. 4.4 Traffix Devices, Inc. v. Marketing Displays, Inc.
Trademark
532 U.S. 23 (2001)
TRAFFIX DEVICES, INC.
v.
MARKETING DISPLAYS, INC.
No. 99-1571.
United States Supreme Court.
Argued November 29, 2000.
Decided March 20, 2001.
CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT
[25] [25] Kennedy, J., delivered the opinion for a unanimous Court.
John G. Roberts, Jr., argued the cause for petitioner. With him on the briefs were Gregory G. Garre and JeanneMarie Marshall.
Deputy Solicitor General Wallace argued the cause for the United States as amicus curiae urging reversal. With him on the brief were Solicitor General Waxman, Assistant Attorney General Ogden, Jeffrey A. Lamken, Anthony J. Steinmeyer, and Mark S. Davies.
John A. Artz argued the cause for respondent. With him on the brief were John S. Artz, Robert P. Renke, and Lisa A. Sarkisian.[1]
Justice Kennedy, delivered the opinion of the Court.
Temporary road signs with warnings like "Road Work Ahead" or "Left Shoulder Closed" must withstand strong gusts of wind. An inventor named Robert Sarkisian obtained two utility patents for a mechanism built upon two springs (the dual-spring design) to keep these and other outdoor signs upright despite adverse wind conditions. The holder of the now-expired Sarkisian patents, respondent Marketing Displays, Inc. (MDI), established a successful business in the manufacture and sale of sign stands incorporating the patented feature. MDI's stands for road signs were recognizable to buyers and users (it says) because the dual-spring design was visible near the base of the sign.
[26] This litigation followed after the patents expired and a competitor, TrafFix Devices, Inc.,sold sign stands with a visible spring mechanism that looked like MDI's. MDI and TrafFix products looked alike because they were. When TrafFix started in business, it sent an MDI product abroad to have it reverse engineered, that is to say copied. Complicating matters, TrafFix marketed its sign stands under a name similar to MDI's. MDI used the name "WindMaster," while TrafFix, its new competitor, used "WindBuster."
MDI brought suit under the Trademark Act of 1946 (Lanham Act), 60 Stat. 427, as amended, 15 U. S. C. § 1051 et seq., against TrafFix for trademark infringement (based on the similar names), trade dress infringement (based on the copied dual-spring design), and unfair competition. TrafFix counterclaimed on antitrust theories. After the United States District Court for the Eastern District of Michigan considered cross-motions for summary judgment, MDI prevailed on its trademark claim for the confusing similarity of names and was held not liable on the antitrust counterclaim; and those two rulings, affirmed by the Court of Appeals, are not before us.
I
We are concerned with the trade dress question. The District Court ruled against MDI on its trade dress claim. 971 F. Supp. 262 (ED Mich. 1997). After determining that the one element of MDI's trade dress at issue was the dualspring design, id., at 265, it held that "no reasonable trier of fact could determine that MDI has established secondary meaning" in its alleged trade dress, id., at 269. In other words, consumers did not associate the look of the dualspring design with MDI. As a second, independent reason to grant summary judgment in favor of TrafFix, the District Court determined the dual-spring design was functional. On this rationale secondary meaning is irrelevant because there can be no trade dress protection in any event. In ruling on the functional aspect of the design, the District Court [27] noted that Sixth Circuit precedent indicated that the burden was on MDI to prove that its trade dress was nonfunctional, and not on TrafFix to show that it was functional (a rule since adopted by Congress, see 15 U. S. C. § 1125(a)(3) (1994 ed., Supp. V)), and then went on to consider MDI's arguments that the dual-spring design was subject to trade dress protection. Finding none of MDI's contentions persuasive, the District Court concluded MDI had not "proffered sufficient evidence which would enable a reasonable trier of fact to find that MDI's vertical dual-spring design is non functional." 971 F. Supp., at 276. Summary judgment was entered against MDI on its trade dress claims.
The Court of Appeals for the Sixth Circuit reversed the trade dress ruling. 200 F. 3d 929 (1999). The Court of Appeals held the District Court had erred in ruling MDI failed to show a genuine issue of material fact regarding whether it had secondary meaning in its alleged trade dress, id., at 938, and had erred further in determining that MDI could not prevail in any event because the alleged trade dress was in fact a functional product configuration, id., at 940. The Court of Appeals suggested the District Court committed legal error by looking only to the dual-spring design when evaluating MDI's trade dress. Basic to its reasoning was the Court of Appeals' observation that it took "little imagination to conceive of a hidden dual-spring mechanism or a tri or quad-spring mechanism that might avoid infringing [MDI's] trade dress." Ibid. The Court of Appeals explained that "[i]f TrafFix or another competitor chooses to use [MDI's] dual-spring design, then it will have to find some other way to set its sign apart to avoid infringing [MDI's] trade dress." Ibid. It was not sufficient, according to the Court of Appeals, that allowing exclusive use of a particular feature such as the dual-spring design in the guise of trade dress would "hinde[r] competition somewhat." Rather, "[e]xclusive use of a feature must `put competitors at a significant non-reputation-related disadvantage' before trade [28] dress protection is denied on functionality grounds." Ibid. (quoting Qualitex Co. v. Jacobson Products Co., 514 U. S. 159, 165 (1995)). In its criticism of the District Court's ruling on the trade dress question, the Court of Appeals took note of a split among Courts of Appeals in various other Circuits on the issue whether the existence of an expired utility patent forecloses the possibility of the patentee's claiming trade dress protection in the product's design. 200 F. 3d, at 939. Compare Sunbeam Products, Inc. v. West Bend Co., 123 F. 3d 246 (CA5 1997) (holding that trade dress protection is not foreclosed), Thomas & Betts Corp. v. Panduit Corp., 138 F. 3d 277 (CA7 1998) (same), and Midwest Industries, Inc. v. Karavan Trailers, Inc., 175 F. 3d 1356 (CA Fed 1999) (same), with Vornado Air Circulation Systems, Inc. v. Duracraft Corp., 58 F. 3d 1498, 1500 (CA10 1995) ("Where a product configuration isa significant inventive component of an invention covered by a utility patent . . . it cannot receive trade dress protection"). To resolve the conflict, we granted certiorari. 530 U. S. 1260 (2000).
II
It is well established that trade dress can be protected under federal law. The design or packaging of a product may acquire a distinctiveness which serves to identify the product with its manufacturer or source; and a design or package which acquires this secondary meaning, assuming other requisites are met, is a trade dress which may not be used in a manner likely to cause confusion as to the origin, sponsorship, or approval of the goods. In these respects protection for trade dress exists to promote competition. As we explained just last Term, see Wal-Mart Stores, Inc. v. Samara Brothers, Inc., 529 U. S. 205 (2000), various Courts of Appeals have allowed claims of trade dress infringement relying on the general provision of the Lanham Act which provides a cause of action to one who is injured when a person uses "any word, term name, symbol, or device, or any [29] combination thereof . . . which is likely to cause confusion . . . as to the origin, sponsorship, or approval of his or her goods." 15 U. S. C. § 1125(a)(1)(A). Congress confirmed this statutory protection for trade dress by amending the Lanham Act to recognize the concept. Title 15 U. S. C. § 1125(a)(3) (1994 ed., Supp. V) provides: "In a civil action for trade dress infringement under this chapter for trade dress not registered on the principal register, the person who asserts trade dress protection has the burden of proving that the matter sought to be protected is not functional." This burden of proof gives force to the well-established rule that trade dress protection may not be claimed for product features that are functional. Qualitex, supra, at 164-165; Two Pesos, Inc. v. Taco Cabana, Inc., 505 U. S. 763, 775 (1992). And in Wal-Mart, supra, we were careful to caution against misuse or over extension of trade dress. We noted that "product design almost invariably serves purposes other than source identification." Id., at 213.
Trade dress protection must subsist with the recognition that in many instances there is no prohibition against copying goods and products. In general, unless an intellectual property right such as a patent or copyright protects an item, it will be subject to copying. As the Court has explained, copying is not always discouraged or disfavored by the laws which preserve our competitive economy. Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U. S. 141, 160 (1989). Allowing competitors to copy will have salutary effects in many instances. "Reverse engineering of chemical and mechanical articles in the public domain often leads to significant advances in technology." Ibid.
The principal question in this case is the effect of an expired patent on a claim of trade dress infringement. A prior patent, we conclude, has vital significance in resolving the trade dress claim. A utility patent is strong evidence that the features therein claimed are functional. If trade dress protection is sought for those features the strong evidence [30] of functionality based on the previous patent adds great weight to the statutory presumption that features are deemed functional until proved otherwise by the party seeking trade dress protection. Where the expired patent claimed the features in question, one who seeks to establish trade dress protection must carry the heavy burden of showing that the feature is not functional, for instance by showing that it is merely an ornamental, incidental, or arbitrary aspect of the device.
In the case before us, the central advance claimed in the expired utility patents (the Sarkisian patents) is the dualspring design; and the dual-spring design is the essential feature of the trade dress MDI now seeks to establish and to protect. The rule we have explained bars the trade dress claim, for MDI did not, and cannot, carry the burden of overcoming the strong evidentiary inference of functionality based on the disclosure of the dual-spring design in the claims of the expired patents.
The dual springs shown in the Sarkisian patents were well apart (at either end of a frame for holding a rectangular sign when one full side is the base) while the dual springs at issue here are close together (in a frame designed to hold a sign by one of its corners). As the District Court recognized, this makes little difference. The point is that the springs are necessary to the operation of the device. The fact that the springs in this very different-looking device fall within the claims of the patents is illustrated by MDI's own position in earlier litigation. In the late 1970's, MDI engaged in a long-running intellectual property battle with a company known as Winn-Proof. Although the precise claims of the Sarkisian patents cover sign stands with springs "spaced apart," U. S. Patent No. 3,646,696, col. 4; U. S. Patent No. 3,662,482, col. 4, the Winn-Proof sign stands (with springs much like the sign stands at issue here) were found to infringe the patents by the United States District Court for the District of Oregon, and the Court of Appeals for the [31] Ninth Circuit affirmed the judgment. Sarkisian v. WinnProof Corp., 697 F. 2d 1313 (1983). Although the WinnProof traffic sign stand (with dual springs close together) did not appear, then, to infringe the literal terms of the patent claims (which called for "spaced apart" springs), the WinnProof sign stand was found to infringe the patents under the doctrine of equivalents, which allows a finding of patent infringement even when the accused product does not fall within the literal terms of the claims. Id., at 1321-1322; see generally Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U. S. 17 (1997). In light of this past ruling—a ruling procured at MDI's own insistence—it must be concluded the products here at issue would have been covered by the claims of the expired patents.
The rationale for the rule that the disclosure of a feature in the claims of a utility patent constitutes strong evidence of functionality is well illustrated in this case. The dual-spring design serves the important purpose of keeping the sign upright even in heavy wind conditions; and, as confirmed by the statements in the expired patents, it does so in a unique and useful manner. As the specification of one of the patents recites, prior art "devices, in practice, will topple under the force of a strong wind." U. S. Patent No. 3,662,482, col. 1. The dual-spring design allows sign stands to resist toppling in strong winds. Using a dual-spring design rather than a single spring achieves important operational advantages. For example, the specifications of the patents note that the "use of a pair of springs . . . as opposed to the use of a single spring to support the frame structure prevents canting or twisting of the sign around a vertical axis," and that, if not prevented, twisting "may cause damage to the spring structure and may result in tipping of the device." U. S. Patent No. 3,646,696, col. 3. In the course of patent prosecution, it was said that "[t]he use of a pair of spring connections as opposed to a single spring connection . . . forms an important part of this combination" because it [32] "forc[es] the sign frame to tip along the longitudinal axis of the elongated ground-engaging members." App. 218. The dual-spring design affects the cost of the device as well; it was acknowledged that the device "could use three springs but this would unnecessarily increase the cost of the device." Id., at 217. These statements made in the patent applications and in the course of procuring the patents demonstrate the functionality of the design. MDI does not assert that any of these representations are mistaken or inaccurate, and this is further strong evidence of the functionality of the dual-spring design.
III
In finding for MDI on the trade dress issue the Court of Appeals gave insufficient recognition to the importance of the expired utility patents, and their evidentiary significance, in establishing the functionality of the device. The error likely was caused by its misinterpretation of trade dress principles in other respects. As we have noted, even if there has been no previous utility patent the party asserting trade dress has the burden to establish the nonfunctionality of alleged trade dress features. MDI could not meet this burden. Discussing trademarks, we have said "`[i]n general terms, a product feature is functional,' and cannot serve as a trademark, `if it is essential to the use or purpose of the article or if it affects the cost or quality of the article.' " Qualitex, 514 U. S., at 165 (quoting Inwood Laboratories, Inc. v. Ives Laboratories, Inc., 456 U. S. 844, 850, n. 10 (1982)). Expanding upon the meaning of this phrase, we have observed that a functional feature is one the "exclusive use of [which] would put competitors at a significant nonreputation-related disadvantage." 514 U. S., at 165. The Court of Appeals in the instant case seemed to interpret this language to mean that a necessary test for functionality is "whether the particular product configuration is a competitive necessity." 200 F. 3d, at 940. See also Vornado, 58 F. 3d, at 1507 ("Functionality, by contrast, has been defined [33] both by our circuit, and more recently by the Supreme Court, in terms of competitive need"). This was incorrect as a comprehensive definition. As explained in Qualitex, supra, and Inwood, supra, a feature is also functional when it is essential to the use or purpose of the device or when it affects the cost or quality of the device. The Qualitex decision did not purport to displace this traditional rule. Instead, it quoted the rule as Inwood had set it forth. It is proper to inquire into a "significant non-reputation-related disadvantage" in cases of esthetic functionality, the question involved in Qualitex. Where the design is functional under the Inwood formulation there is no need to proceed further to consider if there is a competitive necessity for the feature. In Qualitex, by contrast, esthetic functionality was the central question, there having been no indication that the green-gold color of the laundry press pad had any bearing on the use or purpose of the product or its cost or quality.
The Court has allowed trade dress protection to certain product features that are inherently distinctive. Two Pesos, 505 U. S., at 774. In Two Pesos, however, the Court at the outset made the explicit analytic assumption that the trade dress features in question (decorations and other features to evoke a Mexican theme in a restaurant) were not functional. Id., at 767, n. 6. The trade dress in those cases did not bar competitors from copying functional product design features. In the instant case, beyond serving the purpose of informing consumers that the sign stands are made by MDI (assuming it does so), the dual-spring design provides a unique and useful mechanism to resist the force of the wind. Functionality having been established, whether MDI's dual-spring design has acquired secondary meaning need not be considered.
There is no need, furthermore, to engage, as did the Court of Appeals, in speculation about other design possibilities, such as using three or four springs which might serve the same purpose. 200 F. 3d, at 940. Here, the functionality of the spring design means that competitors need not explore [34] whether other spring juxtapositions might be used. The dual-spring design is not an arbitrary flourish in the configuration of MDI's product; it is the reason the device works. Other designs need not be attempted.
Because the dual-spring design is functional, it is unnecessary for competitors to explore designs to hide the springs, say, by using a box or framework to cover them, as suggested by the Court of Appeals. Ibid. The dual-spring design assures the user the device will work. If buyers are assured the product serves its purpose by seeing the operative mechanism that in itself serves an important market need. It would be at cross-purposes to those objectives, and something of a paradox, were we to require the manufacturer to conceal the very item the user seeks.
In a case where a manufacturer seeks to protect arbitrary, incidental, or ornamental aspects of features of a product found in the patent claims, such as arbitrary curves in the legs or an ornamental pattern painted on the springs, a different result might obtain. There the manufacturer could perhaps prove that those aspects do not serve a purpose within the terms of the utility patent. The inquiry into whether such features, asserted to be trade dress, are functional by reason of their inclusion in the claims of an expired utility patent could be aided by going beyond the claims and examining the patent and its prosecution history to see if the feature in question is shown as a useful part of the invention. No such claim is made here, however. MDI in essence seeks protection for the dual-spring design alone. The asserted trade dress consists simply of the dual-spring design, four legs, a base, an upright, and a sign. MDI has pointed to nothing arbitrary about the components of its device or the way they are assembled. The Lanham Act does not exist to reward manufacturers for their innovation in creating a particular device; that is the purpose of the patent law and its period of exclusivity. The Lanham Act, furthermore, does not protect trade dress in a functional design simply [35] because an investment has been made to encourage the public to associate a particular functional feature with a single manufacturer or seller. The Court of Appeals erred in viewing MDI as possessing the right to exclude competitors from using a design identical to MDI's and to require those competitors to adopt a different design simply to avoid copying it. MDI cannot gain the exclusive right to produce sign stands using the dual-spring design by asserting that consumers associate it with the look of the invention itself. Whether a utility patent has expired or there has been no utility patent at all, a product design which has a particular appearance may be functional because it is "essential to the use or purpose of the article" or "affects the cost or quality of the article." Inwood, 456 U. S., at 850, n. 10.
TrafFix and some of its amici argue that the Patent Clause of the Constitution, Art. I, § 8, cl. 8, of its own force, prohibits the holder of an expired utility patent from claiming trade dress protection. Brief for Petitioner 33-36; Brief for Panduit Corp. as Amicus Curiae 3; Brief for Malla Pollack as Amicus Curiae 2. We need not resolve this question. If, despite the rule that functional features may not be the subject of trade dress protection, a case arises in which trade dress becomes the practical equivalent of an expired utility patent, that will be time enough to consider the matter. The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion.
It is so ordered.
[1] Briefs of amici curiae urging reversal were filed for the Holmes Group, Inc.,by James W. Dabney; for Panduit Corp. by Roy E. Hofer, Jerome Gilson, Cynthia A. Homan, and Philip A. Jones; and for Malla Pollack, pro se.
Briefs of amici curiae urging affirmance were filed for the American Intellectual Property Association by Louis T. Pirkey; and for Thomas & Betts Corp. by Sidney David and Roy H. Wepner.
Theodore H. Davis, Jr., Marie V. Driscoll, and Helen Hill Minsker filed a brief for the International Trademark Association as amicus curiae.
4.5 Jay Franco & Sons, Inc. v. Franek 4.5 Jay Franco & Sons, Inc. v. Franek
Trademark
JAY FRANCO & SONS, INC., Plaintiff-Appellee, v. Clemens FRANEK, Defendant-Appellant.
No. 09-2155.
United States Court of Appeals, Seventh Circuit.
Argued May 24, 2010.
Decided Aug. 11, 2010.
*856Ezra Sutton (argued), Attorney, Wood-bridge, NJ, for Plaintiff-Appellee.
Mark D. Roth (argued), Attorney, Orum & Roth, Chicago, IL, for Defendant-Appellant.
Before EASTERBROOK, Chief judge, and POSNER and EVANS, Circuit Judges.
The same year Huey Lewis and the News informed America that it’s “Hip To Be Square”, Clemens Franek sought to trademark the circular beach towel. His company, CLM Design, Inc., pitched the towel as a fashion statement — “the most radical beach fashion item since the bikini,” declared one advertisement. “Bound to be round! Don’t be square!” proclaimed another. CLM also targeted lazy sunbathers: “The round shape eliminates the need to constantly get up and move your towel as the sun moves across the sky. Instead merely reposition yourself.”
The product enjoyed some initial success. Buoyed by an investment and promotional help from the actor Woody Harrelson (then a bartender on the TV show Cheers), CLM had sold more than 30,000 round beach towels in 32 states by the end of 1987. To secure its status as the premier circular-towel maker, the company in 1986 applied for a trademark on the towel’s round design. The Patent and Trademark Office registered the “configuration of a round beach towel” as trademark No. 1,502,261 in 1988. But this was not enough to save CLM: Six years later it dissolved. The mark was assigned to Franek, who continues to sell circular towels.
In 2006 Franek discovered that Jay Franco & Sons, a distributor of bath, bedding, and beach accessories, was selling round beach towels. After settlement negotiations failed, Franek sued two of Jay Franco’s customers, Target and Walmart, for unauthorized use of his registered trademark in violation of § 32 of the Lanham Act, 15 U.S.C. § 1114. Jay Franco had agreed to indemnify and defend its customers in such suits, so it sued Franek to invalidate his mark. (The pending suits against Target and Walmart made the claim ripe, just as insurers can bring declaratory-judgment suits to resolve disputes about a policy’s scope once an insured has been sued and asserts that the policy applies.) The district judge consolidated the two cases, granted summary judgment in Jay Franco’s favor, and dismissed the remaining claims and counterclaims. 2009 U.S. Dist. Lexis 20361 (N.D.I11. Mar. 13, 2009). Franek appeals from that judgment; Target and Walmart are not part of the appeal.
One way to void a trademark is to challenge its distinctiveness. A valid trademark identifies the source of the good *857it marks. Designs do not inherently communicate that information, so to be valid a product-design mark must have acquired a “secondary meaning” — a link in the minds of consumers between the marked item and its source. See Wal-Mart Stores, Inc. v. Samara Brothers, Inc., 529 U.S. 205, 120 S.Ct. 1339, 146 L.Ed.2d 182 (2000); Bretford Manufacturing, Inc. v. Smith System Manufacturing Corp., 419 F.3d 576, 578-80 (7th Cir.2005). Cf. Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 112 S.Ct. 2753, 120 L.Ed.2d 615 (1992). But this type of invalidation is unavailable to Jay Franco. Franek (and before him CLM) has continuously used the round-towel mark since its 1988 registration. That makes the mark “incontestable,” 15 U.S.C. § 1065, a status that eliminates the need for a mark’s owner in an infringement suit to show that his mark is distinctive. See 15 U.S.C. § 1115(b); Park ‘N Fly, Inc. v. Dollar Park and Fly, Inc., 469 U.S. 189, 105 S.Ct. 658, 83 L.Ed.2d 582 (1985).
Unfortunately for Franek, incontestable marks are not invincible. The Lanham Act lists a number of affirmative defenses an alleged infringer can parry with; one is a showing that the mark is “functional.” See § 1115(b)(8); Specialized Seating, Inc. v. Greenwich Industries, L.P., 616 F.3d 722, 724, 2010 WL 3155922, *1 (7th Cir.2010) (discussing functionality and other ways to defeat incontestable marks). As our companion opinion in Specialized Seating explains, patent law alone protects useful designs from mimicry; the functionality doctrine polices the division of responsibilities between patent and trademark law by invalidating marks on useful designs. This was the route Jay Franco pursued. The district judge agreed, finding Franek’s mark “functional” under the definition the Supreme Court gave that concept in TrafFix Devices, Inc. v. Marketing Displays, Inc., 532 U.S. 23, 32-35, 121 S.Ct. 1255, 149 L.Ed.2d 164 (2001). The judge got it right.
TrafFix says that a design is functional when it is “essential to the use or purpose of the device or when it affects the cost or quality of the device,” 532 U.S. at 33, 121 S.Ct. 1255, a definition cribbed from Inwood Laboratories, Inc. v. Ives Laboratories, Inc., 456 U.S. 844, 850 n. 10, 102 S.Ct. 2182, 72 L.Ed.2d 606 (1982). So if a design enables a product to operate, or improves on a substitute design in some way (such as by making the product cheaper, faster, lighter, or stronger), then the design cannot be trademarked; it is functional because consumers would pay to have it rather than be indifferent toward or pay to avoid it. A qualification is that any pleasure a customer derives from the design’s identification of the product’s source — the joy of buying a marked good over an identical generic version because the consumer prefers the status conferred by the mark — doesn’t count. That broad a theory of functionality would penalize companies for developing brands with cachet to distinguish themselves from competitors, which is the very purpose of trademark law. In short, a design that produces a benefit other than source identification is functional.
Figuring out which designs meet this criterion can be tricky. Utility patents serve as excellent cheat sheets because any design claimed in a patent is supposed to be useful. See 35 U.S.C. § 101; Brenner v. Manson, 383 U.S. 519, 528-36, 86 S.Ct. 1033, 16 L.Ed.2d 69 (1966). For this reason, TrafFix held that expired utility patents provide “strong evidence that the features therein claimed are functional.” 532 U.S. at 29, 121 S.Ct. 1255. The parties in this case wrangle over the relevance of a handful of utility patents that claim circu*858lar towels. We need discuss only one (No. 4,794,029), which describes a round beach towel laced with drawstrings that can be pulled to turn the towel into a satchel. This patent’s first two claims are:
1. A towel-bag construction comprising: a non-rectangular towel;
a casing formed at the perimeter of said towel;
a cord threaded through said casing; and
a section of relatively non-stretehable fabric of a shape geometrically similar to that of said towel attached with its edges equidistant from the edges of said towel.
2. A towel-bag construction as set forth in claim 1 wherein said towel is circular in shape, whereby a user while sunbathing may reposition his or her body towards the changing angle of the sun while the towel remains stationary.
Claim 2 sounds like Franek’s advertisements, which we quoted above. The patent’s specification also reiterates, in both the summary and the detailed description, that a circular towel is central to the invention because of its benefit to lazy sunbathers.
Franek argues that claim 2 does not trigger the TrajFix presumption of functionality because his towel does not infringe the '029 patent. He notes that claim 2 incorporates claim 1 (in patent parlance, claim 1 is “independent” and claim 2 “dependent,” see 35 U.S.C. § 112) with the added condition that the towel be circular. An item can infringe a dependent claim only if it also violates the independent claim incorporated by the dependent claim. See Monsanto Co. v. Syngenta Seeds, Inc., 503 F.3d 1352, 1359 (Fed.Cir.2007). Franek reasons that because his towel lacks a perimeter casing, drawstring, and non-stretchable section of fabric, it does not infringe claim 1, and thus cannot infringe claim 2. Even if his towel could infringe claim 2, Franek maintains that the claim, is invalid because the towel-to-bag patent was sought in 1987, two years after Franek started selling a round beach towel, and thus too late to claim its invention. See 35 U.S.C. § 102(b); Pfaff v. Wells Electronics, Inc., 525 U.S. 55, 119 S.Ct. 304, 142 L.Ed.2d 261 (1998).
Proving patent infringement can be sufficient to show that a trademarked design is useful, as it means that the infringing design is quite similar to a useful invention. See Raytheon Co. v. Roper Corp., 724 F.2d 951, 959 (Fed.Cir.1983). But such proof is unnecessary. Functionality, is determined by a feature’s usefulness, not its patentability or its infringement of a patent. TrafFix’s ruling that an expired patent (which by definition can no longer be infringed) may evince a design’s functionality demonstrates that proof of infringement is unnecessary. If an invention is too useless to be patentable, or too dissimilar to a design to shed light on its functions, then the lack of proof of patent infringement is meaningful. Otherwise it is irrelevant. A design may not infringe a patented invention because the invention is obvious or taught by prior art, see 35 U.S.C. §§ 102(a), 103(a), but those and other disqualifers do not mean that the design is not useful. Just so here: Franek’s towel may lack some of the components in claim 1 necessary to infringe claim 2, but claim 2’s coverage of a circular beach towel for sunbathing is enough to signal that a round-towel design is useful for sunbathers. Each claim in a patent is evaluated individually, see Altoona Publix Theatres, Inc. v. American Tri-Ergon Corp., 294 U.S. 477, 487, 55 S.Ct. 455, 79 L.Ed. 1005 (1935), each must be substantially different, see 37 C.F.R. § 1.75(b); Manual of Patent Examining Procedure *859§ 2173.05(n)(b) (8th ed., July 2010 rev.), and each is presumed valid, 35 U.S.C. § 282. We must therefore presume that the unique component in claim 2 — the round shape of the towel — is useful.
Nor does it matter that the '029 patent application was filed two years after Franek began selling round towels. As we’ve explained, a patent’s invalidity for a reason other than uselessness says nothing about the claimed design’s functionality. And a design patented yesterday can be as good evidence of a mark’s functionality as a design patented 50 years ago. Indeed, more recent patents are often better evidence because technological change can render designs that were functional years ago no longer so. See Eco Manufacturing LLC v. Honeywell International Inc., 357 F.3d 649, 653 (7th Cir.2003). The Court in TrafFix may have dealt only with expired utility patents, but the logic it employed is not limited to them.
To put things another way, a trademark holder cannot block innovation by appropriating designs that under-gird further improvements. Patent holders can do this, but a patent’s life is short; trademarks can last forever, so granting trademark holders this power could permanently stifle product development. If we found Franek’s trademark nonfunctional, then inventors seeking to build an improved round beach towel would be out of luck. They’d have to license Franek’s mark or quell their inventiveness. That result does not jibe with the purposes of patent or trademark law.
This “strong evidence” of the round towel’s functionality is bolstered by Franek’s own advertisements, which highlight two functional aspects of the round beach towel’s design. One, also discussed in the '029 patent, is that roundness enables heliotropic sunbathers — tanners who swivel their bodies in unison with the sun’s apparent motion in order to maintain an even tan — to remain on their towels as they rotate rather than exert the energy to stand up and reposition their towels every so often, as conventional rectangular towels require.
Franek responds that whatever its shape (golden-ratio rectangle, square, nonagon) any towel can satisfy a heliotropic tanner if it has enough surface area — the issue is size, not shape. That’s true, and it is enough to keep the roundness of his towel from being functional under the first prong of TrafFix’s definition (“essential to the use or purpose of the device”) but not the second. For heliotropic sunbathers, a circle surpasses other shapes because it provides the most rotational space without waste. Any non-circle polygon will either limit full rotations (spinning on a normal beach towel leads to sandy hair and feet) or not use all the surface area (a 6' tall person swiveling on a 6' by 6' square towel won’t touch the corners). Compared to other shapes that permit full rotations, the round towel requires less material, which makes it easier to fold and carry. That’s evidence that the towel’s circularity “affects the ... quality of the device.” (The reduction in needed material also suggests that round towels are cheaper to produce than other-shaped towels, though Franek contends that cutting and hemming expenses make them costlier. We express no view on the matter.)
But let us suppose with Franek — who opposed summary judgment and who is thus entitled to all reasonable inferences— that round towels are not measurably better for spinning with the sun. After all, other shapes (squircles, regular icosagons) are similar enough to circles that any qualitative difference may be lost on tanners. Plus, the ability to rotate 180 degrees may be an undesired luxury. Few lie out from dawn ‘til dusk (if only to avoid skin cancer) *860and the daily change in the sun’s declination means it will rise due east and set due west just twice a year, during the vernal and autumnal equinoxes. A towel shaped like a curved hourglass that allows only 150 or 120 degrees of rotation (or even fewer) may be all a heliotropic tanner wants. No matter. Franek’s mark still is functional.
Franek’s advertisements declare that the round towel is a fashion statement. Fashion is a form of function. A design’s aesthetic appeal can be as functional as its tangible characteristics. See Qualitex Co. v. Jacobson Products Co., 514 U.S. 159, 169-70, 115 S.Ct. 1300, 131 L.Ed.2d 248 (1995); Wal-Mart, 529 U.S. at 214, 120 S.Ct. 1339; TrafFix, 532 U.S. at 33, 121 S.Ct. 1255; W.T. Rogers Co. v. Keene, 778 F.2d 334 (7th Cir.1985); Publications International, Ltd. v. Landoll, Inc., 164 F.3d 337 (7th Cir.1998); Abercrombie & Fitch Stores, Inc. v. American Eagle Outfitters, Inc., 280 F.3d 619 (6th Cir.2002). And many cases say that fashionable designs can be freely copied unless protected by patent law. See, e.g., Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 109 S.Ct. 971, 103 L.Ed.2d 118 (1989); Sears, Roebuck & Co. v. Stiffel Co., 376 U.S. 225, 84 S.Ct. 784, 11 L.Ed.2d 661 (1964); Compco Corp. v. Day-Brite Lighting, Inc., 376 U.S. 234, 84 S.Ct. 779, 11 L.Ed.2d 669 (1964); Kellogg Co. v. National Biscuit Co., 305 U.S. 111, 59 S.Ct. 109, 83 L.Ed. 73 (1938); Singer Manufacturing Co. v. June Manufacturing Co., 163 U.S. 169, 16 S.Ct. 1002, 41 L.Ed. 118 (1896).
The chief difficulty is distinguishing between designs that are fashionable enough to be functional and those that are merely pleasing. Only the latter group can be protected, because trademark law would be a cruel joke if it limited companies to tepid or repugnant brands that discourage customers from buying the marked wares. We discussed this problem at length in Keene. See also Eco Manufacturing, 357 F.3d at 654; Schwinn Bicycle Co. v. Ross Bicycles, Inc., 870 F.2d 1176, 1188-91 (7th Cir.1989). The Supreme Court broached the subject in Qualitex when it discussed the functionality of the green-gold color of a dry cleaning pad. Unwilling to say that the pad required a green-gold hue or was improved by it, the Court still thought that the color would be functional if its exclusive use by a single designer “would put competitors at a significant non-reputation-related disadvantage.” 514 U.S. at 165, 115 S.Ct. 1300. This is a problem for Franek’s round-towel mark.
Franek wants a trademark on the circle. Granting a producer the exclusive use of a basic element of design (shape, material, color, and so forth) impoverishes other designers’ palettes. See, e.g., Brunswick Corp., v. British Seagull Ltd., 35 F.3d 1527 (Fed.Cir.1994) (black color of boat engines is functional because it is compatible with boats of many different colors). Qualitex’s determination that “color alone, at least sometimes, can meet the basic legal requirements for use as a trademark” (514 U.S. at 166, 115 S.Ct. 1300), means that there is no per se rule against this practice. See also Thomas & Betts Corp. v. Panduit Corp., 138 F.3d 277, 299 (7th Cir.1998). The composition of the relevant market matters. But the more rudimentary and general the element — all six-sided shapes rather than an irregular, perforated hexagon; all labels made from tin rather than a specific tin label; all shades of the color purple rather than a single shade — the more likely it is that restricting its use will significantly impair competition. See, e.g., Keene, 778 F.2d at 343; Lorillard v. Pride, 28 F. 434 (C.C.N.D.Ill.1886); Saint-Gobain Corp. v. 3M Co., 90 U.S.P.Q.2d 1425 (T.T.A.B.2007). Franek’s towel is of this ilk. He has trademarked the “configuration of a round beach towel.” *861Every other beach towel manufacturer is barred from using the entire shape as well as any other design similar enough that consumers are likely to confuse it with Franek’s circle (most regular polygons, for example).
Contrast Franek’s mark with the irregular hexagon at issue in Keene or the green-gold hue in Qmlitex. Those marks restrict few design options for competitors. Indeed, they are so distinctive that competitors’ only reason to copy them would be to trade on the goodwill of the original designer. Cf. Service Ideas, Inc. v. Traex Corp., 846 F.2d 1118, 1123-24 (7th Cir.1988) (purposeful copying of a beverage server’s arbitrary design indicated a lack of aesthetic functionality). That’s not so here. A circle is the kind of basic design that a producer like Jay Franco adopts because alternatives are scarce and some consumers want the shape regardless of who manufactures it. There are only so many geometric shapes; few are both attractive and simple enough to fabricate cheaply. Cf. Qualitex, 514 U.S. at 168-69, 115 S.Ct. 1300 (functionality doctrine invalidates marks that would create color scarcity in a particular market). And some consumers crave round towels — beachgoers who prefer curved edges to sharp corners, those who don’t want to be “square,” and those who relish the circle’s simplicity. A producer barred from selling such towels loses a profitable portion of the market. The record does not divulge much on these matters, but any holes in the evidence are filled by the TrafFix presumption that Franek’s mark is functional, a presumption he has failed to rebut.
Franek chose to pursue a trademark, not a design patent, to protect the stylish circularity of his beach towel. Cf. Kohler Co. v. Moen Inc., 12 F.3d 632, 647 (7th Cir.1993) (Cudahy, J., dissenting) (calling Franek’s mark a “horrible example[ ]” of a registered trademark that should have been a design patent). He must live -with that choice. We cannot permit him to keep the indefinite competitive advantage in producing beach towels this trademark creates.
If Franek is worried that consumers will confuse Jay Franco’s round beach towels with his, he can imprint a distinctive verbal or pictorial mark on his towels. See Publications International, 164 F.3d at 343; Keene, 778 F.2d at 347-48. That will enable him to reap the benefits of his brand while still permitting healthy competition in the beach towel market.
Affirmed