834 F.2d 1142 (1987)2
No. 828, Docket 86-6260.
United States Court of Appeals, Second Circuit.
Argued March 2, 1987.
Decided December 2, 1987.
Blum Kaplan, New York City (Lawrence Rosenthal, Laura E. Goldbard, Anita K. Yeung, New York City, of counsel), for plaintiff-appellant.5
Fish & Neave, New York City (Donald E. Degling, Susan Progoff, Eric M. Lee, New York City, of counsel), for defendant-appellee.6
Before OAKES and WINTER, Circuit Judges, and ZAMPANO, District Judge.7
In passing the Copyright Act of 1976 Congress attempted to distinguish between  protectable "works of applied art" and "industrial designs not subject to copyright protection." See H.R.Rep. No. 1476, 94th Cong., 2d Sess. 54, reprinted in 1976 U.S.Code Cong. & Admin.News 5659, 5667 (hereinafter H.R.Rep. No. 1476). The courts, however, have had difficulty framing tests by which the fine line establishing what is and what is not copyrightable can be drawn. Once again we are called upon to draw such a line, this time in a case involving the "RIBBON Rack," a bicycle rack made of bent tubing that is said to have originated from a wire sculpture. (A photograph of the rack is contained in the appendix to this opinion.) We are also called upon to determine whether there is any trademark protection available to the manufacturer of the bicycle rack, appellant Brandir International, Inc. The Register of Copyright, named as a third-party defendant under the statute, 17 U.S.C. § 411, but electing not to appear, denied copyrightability. In the subsequent suit brought in the United States District Court for the Southern District of New York, Charles S. Haight, Jr., Judge, the district court granted summary judgment on both the copyright and trademark claims to defendant Cascade Pacific Lumber Co., d/b/a Columbia Cascade Co., manufacturer of a similar bicycle rack. We affirm as to the copyright claim, but reverse and remand as to the trademark claim.9
Against the history of copyright protection well set out in the majority opinion in Carol Barnhart Inc. v. Economy Cover Corp., 773 F.2d 411, 415-18 (2d Cir.1985), and in Denicola, Applied Art and Industrial Design: A Suggested Approach to Copyright in Useful Articles, 67 Minn.L.Rev. 707, 709-17 (1983), Congress adopted the Copyright Act of 1976. The "works of art" classification of the Copyright Act of 1909 was omitted and replaced by reference to "pictorial, graphic, and sculptural works," 17 U.S.C. § 102(a)(5). According to the House Report, the new category was intended to supply "as clear a line as possible between copyrightable works of applied art and uncopyrighted works of industrial design." H.R.Rep. No. 1476, at 55, U.S.Code Cong. & Admin.News 1976, p. 5668. The statutory definition of "pictorial, graphic, and sculptural works" states that "the design of a useful article, as defined in this section, shall be considered a pictorial, graphic, 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." 17 U.S.C. § 101. The legislative history added gloss on the criteria of separate identity and independent existence in saying:10
On the other hand, although the shape of an industrial product may be aesthetically satisfying and valuable, the Committee's intention is not to offer it copyright protection under the bill. 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 under the bill.
H.R.Rep. No. 1476, at 55, U.S.Code Cong. & Admin.News 1976, p. 5668.12
As courts and commentators have come to realize, however, the line Congress attempted to draw between copyrightable art and noncopyrightable design "was neither clear nor new." Denicola, supra, 67 Minn.L.Rev. at 720. One aspect of the distinction that has drawn considerable attention is the reference in the House Report to "physically or conceptually" (emphasis added) separable elements. The District of Columbia Circuit in Esquire, Inc. v. Ringer, 591 F.2d 796, 803-04 (D.C.Cir.1978) (holding outdoor lighting fixtures ineligible for copyright), cert. denied, 440 U.S. 908, 99 S.Ct. 1217, 59 L.Ed.2d 456 (1979), called this an "isolated reference" and gave it no significance. Professor Nimmer, however,  seemed to favor the observations of Judge Harold Leventhal in his concurrence in Esquire, who stated that "the overall legislative policy ... sustains the Copyright Office in its effort to distinguish between the instances where the aesthetic element is conceptually severable and the instances where the aesthetic element is inextricably interwoven with the utilitarian aspect of the article." 591 F.2d at 807; see 1 Nimmer on Copyright § 2.08[B] at 2-93 to 2-96.2 (1986). But see Gerber, Book Review, 26 U.C.L.A.L.Rev. 925, 938-43 (1979) (criticizing Professor Nimmer's view on conceptual separability). Looking to the section 101 definition of works of artistic craftsmanship requiring that artistic features be "capable of existing independently of the utilitarian aspects," Professor Nimmer queries whether that requires physical as distinguished from conceptual separability, but answers his query by saying "[t]here is reason to conclude that it does not." See 1 Nimmer on Copyright § 2.08[B] at 2-96.1. In any event, in Kieselstein-Cord v. Accessories by Pearl, Inc., 632 F.2d 989, 993 (2d Cir.1980), this court accepted the idea that copyrightability can adhere in the "conceptual" separation of an artistic element. Indeed, the court went on to find such conceptual separation in reference to ornate belt buckles that could be and were worn separately as jewelry. Kieselstein-Cord was followed in Norris Industries, Inc. v. International Telephone & Telegraph Corp., 696 F.2d 918, 923-24 (11th Cir.), cert. denied, 464 U.S. 818, 104 S.Ct. 78, 78 L.Ed.2d 89 (1983), although there the court upheld the Register's refusal to register automobile wire wheel covers, finding no "conceptually separable" work of art. See also Transworld Mfg. Corp. v. Al Nyman & Sons, Inc., 95 F.R.D. 95 (D.Del.1982) (finding conceptual separability sufficient to support copyright in denying summary judgment on copyrightability of eyeglass display cases).13
In Carol Barnhart Inc. v. Economy Cover Corp., 773 F.2d 411 (2d Cir.1985), a divided panel of this circuit affirmed a district court grant of summary judgment of noncopyrightability of four life-sized, anatomically correct human torso forms. Carol Barnhart distinguished Kieselstein-Cord, but it surely did not overrule it. The distinction made was that the ornamented surfaces of the Kieselstein-Cord belt buckles "were not in any respect required by their utilitarian functions," but the features claimed to be aesthetic or artistic in the Carol Barnhart forms were "inextricably intertwined with the utilitarian feature, the display of clothes." 773 F.2d at 419. But cf. Animal Fair, Inc. v. Amfesco Indus., Inc., 620 F.Supp. 175, 186-88 (D.Minn.1985) (holding bear-paw design conceptually separable from the utilitarian features of a slipper), aff'd mem., 794 F.2d 678 (8th Cir.1986). As Judge Newman's dissent made clear, the Carol Barnhart majority did not dispute "that `conceptual separability' is distinct from `physical separability' and, when present, entitles the creator of a useful article to a copyright on its design." 773 F.2d at 420.14
"Conceptual separability" is thus alive and well, at least in this circuit. The problem, however, is determining exactly what it is and how it is to be applied. Judge Newman's illuminating discussion in dissent in Carol Barnhart, see 773 F.2d at 419-24, proposed a test that aesthetic features are conceptually separable if "the article ... stimulate[s] in the mind of the beholder a concept that is separate from the concept evoked by its utilitarian function." Id. at 422. This approach has received favorable endorsement by at least one commentator, W. Patry, Latman's The Copyright Law 43-45 (6th ed. 1986), who calls Judge Newman's test the "temporal displacement" test. It is to be distinguished from other possible ways in which conceptual separability can be tested, including whether the primary use is as a utilitarian article as opposed to an artistic work, whether the aesthetic aspects of the work can be said to be "primary," and whether the article is marketable as art, none of which is very satisfactory. But Judge Newman's test was rejected outright by the majority as "a standard so ethereal as to amount to a `nontest' that would be extremely difficult, if not impossible, to administer or apply." 773 F.2d at 419 n. 5.15
 Perhaps the differences between the majority and the dissent in Carol Barnhart might have been resolved had they had before them the Denicola article on Applied Art and Industrial Design: A Suggested Approach to Copyright in Useful Articles, supra. There, Professor Denicola points out that although the Copyright Act of 1976 was an effort "`to draw as clear a line as possible,'" in truth "there is no line, but merely a spectrum of forms and shapes responsive in varying degrees to utilitarian concerns." 67 Minn.L.Rev. at 741. Denicola argues that "the statutory directive requires a distinction between works of industrial design and works whose origins lie outside the design process, despite the utilitarian environment in which they appear." He views the statutory limitation of copyrightability as "an attempt to identify elements whose form and appearance reflect the unconstrained perspective of the artist," such features not being the product of industrial design. Id. at 742. "Copyrightability, therefore, should turn on the relationship between the proffered work and the process of industrial design." Id. at 741. He suggests that "the dominant characteristic of industrial design is the influence of nonaesthetic, utilitarian concerns" and hence concludes that copyrightability "ultimately should depend on the extent to which the work reflects artistic expression uninhibited by functional considerations." Id. To state the Denicola test in the language of conceptual separability, if design elements reflect a merger of aesthetic and functional considerations, the artistic aspects of a work cannot be said to be conceptually separable from the utilitarian elements. Conversely, where design elements can be identified as reflecting the designer's artistic judgment exercised independently of functional influences, conceptual separability exists.16
We believe that Professor Denicola's approach provides the best test for conceptual separability and, accordingly, adopt it here for several reasons. First, the approach is consistent with the holdings of our previous cases. In Kieselstein-Cord, for example, the artistic aspects of the belt buckles reflected purely aesthetic choices, independent of the buckles' function, while in Carol Barnhart the distinctive features of the torsos — the accurate anatomical design and the sculpted shirts and collars — showed clearly the influence of functional concerns. Though the torsos bore artistic features, it was evident that the designer incorporated those features to further the usefulness of the torsos as mannequins. Second, the test's emphasis on the influence of utilitarian concerns in the design process may help, as Denicola notes, to "alleviate the de facto discrimination against nonrepresentational art that has regrettably accompanied much of the current analysis." Id. at 745. Finally, and perhaps most importantly, we think Denicola's test will not be too difficult to administer in practice. The work itself will continue to give "mute testimony" of its origins. In addition, the parties will be required to present evidence relating to the design process and the nature of the work, with the trier of fact making the determination  whether the aesthetic design elements are significantly influenced by functional considerations.17
Turning now to the facts of this case, we note first that Brandir contends, and its chief owner David Levine testified, that the original design of the RIBBON Rack stemmed from wire sculptures that Levine had created, each formed from one continuous undulating piece of wire. These sculptures were, he said, created and displayed in his home as a means of personal expression, but apparently were never sold or displayed elsewhere. He also created a wire sculpture in the shape of a bicycle and states that he did not give any thought to the utilitarian application of any of his sculptures until he accidentally juxtaposed the bicycle sculpture with one of the self-standing wire sculptures. It was not until November 1978 that Levine seriously began pursuing the utilitarian application of his sculptures, when a friend, G. Duff Bailey, a bicycle buff and author of numerous articles about urban cycling, was at Levine's home and informed him that the sculptures would make excellent bicycle racks, permitting bicycles to be parked under the overloops as well as on top of the underloops. Following this meeting, Levine met several times with Bailey and others, completing the designs for the RIBBON Rack by the use of a vacuum cleaner hose, and submitting his drawings to a fabricator complete with dimensions. The Brandir RIBBON Rack began being nationally advertised and promoted for sale in September 1979.18
In November 1982 Levine discovered that another company, Cascade Pacific Lumber Co., was selling a similar product. Thereafter, beginning in December 1982, a copyright notice was placed on all RIBBON Racks before shipment and on December 10, 1982, five copyright applications for registration were submitted to the Copyright Office. The Copyright Office refused registration by letter, stating that the RIBBON Rack did not contain any element that was "capable of independent existence as a copyrightable pictorial, graphic or sculptural work apart from the shape of the useful article." An appeal to the Copyright Office was denied by letter dated March 23, 1983, refusing registration on the above ground and alternatively on the ground that the design lacked originality, consisting of "nothing more than a familiar public domain symbol." In February 1984, after the denial of the second appeal of the examiner's decision, Brandir sent letters to customers enclosing copyright notices to be placed on racks sold prior to December 1982.19
Between September 1979 and August 1982 Brandir spent some $38,500 for advertising and promoting the RIBBON Rack, including some 85,000 pieces of promotional literature to architects and landscape architects. Additionally, since October 1982 Brandir has spent some $66,000, including full-, half-, and quarter-page advertisements in architectural magazines such as Landscape Architecture, Progressive Architecture, and Architectural Record, indeed winning an advertising award from Progressive Architecture in January 1983. The RIBBON Rack has been featured in Popular Science, Art and Architecture, and Design 384 magazines, and it won an Industrial Designers Society of America design award in the spring of 1980. In the spring of 1984 the RIBBON Rack was selected from 200 designs to be included among 77 of the designs exhibited at the Katonah Gallery in an exhibition entitled "The Product of Design: An Exploration of the Industrial Design Process," an exhibition that was written up in the New York Times.20
Sales of the RIBBON Rack from September 1979 through January 1985 were in excess of $1,367,000. Prior to the time Cascade Pacific began offering for sale its bicycle rack in August 1982, Brandir's sales were $436,000. The price of the RIBBON Rack ranges from $395 up to $2,025 for a stainless steel model and generally depends on the size of the rack, one of the most popular being the RB-7, selling for $485.21
Applying Professor Denicola's test to the RIBBON Rack, we find that the rack is not copyrightable. It seems clear that  the form of the rack is influenced in significant measure by utilitarian concerns and thus any aesthetic elements cannot be said to be conceptually separable from the utilitarian elements. This is true even though the sculptures which inspired the RIBBON Rack may well have been — the issue of originality aside — copyrightable.22
Brandir argues correctly that a copyrighted work of art does not lose its protected status merely because it subsequently is put to a functional use. The Supreme Court so held in Mazer v. Stein, 347 U.S. 201, 74 S.Ct. 460, 98 L.Ed. 630 (1954), and Congress specifically intended to accept and codify Mazer in section 101 of the Copyright Act of 1976. See H.R.Rep. No. 1476 at 54-55. The district court thus erred in ruling that, whatever the RIBBON Rack's origins, Brandir's commercialization of the rack disposed of the issue of its copyrightability.23
Had Brandir merely adopted one of the existing sculptures as a bicycle rack, neither the application to a utilitarian end nor commercialization of that use would have caused the object to forfeit its copyrighted status. Comparison of the RIBBON Rack with the earlier sculptures, however, reveals that while the rack may have been derived in part from one of more "works of art," it is in its final form essentially a product of industrial design. In creating the RIBBON Rack, the designer has clearly adapted the original aesthetic elements to accommodate and further a utilitarian purpose. These altered design features of the RIBBON Rack, including the spacesaving, open design achieved by widening the upper loops to permit parking under as well as over the rack's curves, the straightened vertical elements that allow in- and above-ground installation of the rack, the ability to fit all types of bicycles and mopeds, and the heavy-gauged tubular construction of rustproof galvanized steel, are all features that combine to make for a safe, secure, and maintenance-free system of parking bicycles and mopeds. Its undulating shape is said in Progressive Architecture, January 1982, to permit double the storage of conventional bicycle racks. Moreover, the rack is manufactured from 2 3/8-inch standard steam pipe that is bent into form, the six-inch radius of the bends evidently resulting from bending the pipe according to a standard formula that yields bends having a radius equal to three times the nominal internal diameter of the pipe.24
Brandir argues that its RIBBON Rack can and should be characterized as a sculptural work of art within the minimalist art movement. Minimalist sculpture's most outstanding feature is said to be its clarity and simplicity, in that it often takes the form of geometric shapes, lines, and forms that are pure and free of ornamentation and void of association. As Brandir's expert put it, "The meaning is to be found in, within, around and outside the work of art, allowing the artistic sensation to be experienced as well as intellectualized." People who use Foley Square in New York City see in the form of minimalist art the "Tilted Arc," which is on the plaza at 26 Federal Plaza. Numerous museums have had exhibitions of such art, and the school of minimalist art has many admirers.25
It is unnecessary to determine whether to the art world the RIBBON Rack properly would be considered an example of minimalist sculpture. The result under the copyright statute is not changed. Using the test we have adopted, it is not enough that, to paraphrase Judge Newman, the rack may stimulate in the mind of the reasonable observer a concept separate from the bicycle rack concept. While the RIBBON Rack may be worthy of admiration for its aesthetic qualities alone, it remains nonetheless the product of industrial design. Form and function are inextricably intertwined in the rack, its ultimate design being as much the result of utilitarian pressures as aesthetic choices. Indeed, the visually pleasing proportions and symmetricality of the rack represent design changes made in response to functional concerns. Judging from the awards the rack has received, it would seem in fact that Brandir has achieved with the RIBBON Rack the highest goal of modern industrial design, that is, the harmonious fusion of function and aesthetics. Thus there remains no artistic element of the RIBBON  Rack that can be identified as separate and "capable of existing independently, of, the utilitarian aspects of the article." Accordingly, we must affirm on the copyright claim.26
As to whether the configuration of Brandir's bicycle rack can be protected under either section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a), or New York State unfair competition law, we are reminded that the design of a product itself may function as its packaging or protectable trade dress. See LeSportsac, Inc. v. K mart Corp., 754 F.2d 71, 75 (2d Cir.1985). The district court dismissed Brandir's claims, saying that its analysis of the copyright issue was sufficient to dispose of the Lanham Act and common law claims. The court stated "the design feature of the Ribbon Racks is clearly dictated by the function to be performed, namely, holding up bicycles. If the steam pipes were not bent into the design, but instead remained flat, the bicycles would not stand up, they would fall down." But as Judge Newman noted in his dissent in Carol Barnhart, 773 F.2d at 420 n. 1, the principle of conceptual separability of functional design elements in copyright law is different from the somewhat similar principle of functionality as developed in trademark law. For trademark purposes, he pointed out, a design feature "has been said to be functional if it is `essential to the use or purpose of the article' or `affects the cost or quality of the article.'" Id. (quoting Inwood Laboratories, Inc. v. Ives Laboratories, Inc., 456 U.S. 844, 850 n. 10, 102 S.Ct. 2182, 2187 n. 10, 72 L.Ed.2d 606 (1982)); see LeSportsac, Inc. v. K mart Corp., 754 F.2d at 75-76 (trade dress of a product is eligible for protection if it has acquired a secondary meaning and is nonfunctional).27
Here, the district court limited its inquiry to determining whether portions of the RIBBON Rack performed the function of a bicycle rack. But the fact that a design feature performs a function does not make it essential to the performance of that function; it is instead the absence of alternative constructions performing the same function that renders the feature functional. Thus, the true test of functionality is not whether the feature in question performs a function, but whether the feature "is dictated by the functions to be performed," Warner Bros. Inc. v. Gay Toys, Inc., 724 F.2d 327, 331 (2d Cir.1983) (quoted in LeSportsac, Inc. v. K mart Corp., 754 F.2d at 76), as evidenced by available alternative constructions. See Metro Kane Imports, Ltd. v. Rowoco, Inc., 618 F.Supp. 273, 275-76 (S.D.N.Y.1985), aff'd mem., 800 F.2d 1128 (2d Cir.1986) (finding high-tech design of orange juice squeezer not dictated by function to be performed as there was no evidence that design permitted juicer to be manufactured at lower price or with altered performance). There are numerous alternative bicycle rack constructions. The nature, price, and utility of these constructions are material issues of fact not suitable for determination by summary judgment. For example, while it is true that the materials used by Brandir are standard-size pipes, we have no way of knowing whether the particular size and weight of the pipes used is the best, the most economical, or the only available size and weight pipe in the marketplace. We would rather think the opposite might be the case. So, too, with the dimension of the bends being dictated by a standard formula corresponding to the pipe size; it could be that there are many standard radii and that the particular radius of Brandir's RIBBON Rack actually required new tooling. This issue of functionality on remand should be viewed in terms of bicycle racks generally and not one-piece undulating bicycle racks specifically. See id. at 330-32; see also In re DC Comics, Inc., 689 F.2d 1042, 1045 (C.C.P.A.1982) (dolls generally and not Superman dolls are the class by  which functionality is determined). We reverse and remand as to the trademark and unfair competition claims.28
Judgment affirmed as to the copyright claim; reversed and remanded as to the trademark and unfair competition claims.29
Although I concur in the reversal of the district court's grant of summary judgment on the trademark and unfair competition claims, I respectfully dissent from the majority's discussion and disposition of the copyright claim.35
My colleagues, applying an adaptation of Professor Denicola's test, hold that the aesthetic elements of the design of a useful article are not conceptually separable from  its utilitarian aspects if "[f]orm and function are inextricably intertwined" in the article, and "its ultimate design [is] as much the result of utilitarian pressures as aesthetic choices." Applying that test to the instant matter, they observe that the dispositive fact is that "in creating the Ribbon Rack, [Levine] has clearly adapted the original aesthetic elements to accommodate and further a utilitarian purpose." (emphasis added). The grounds of my disagreement are that: (1) my colleagues' adaptation of Professor Denicola's test diminishes the statutory concept of "conceptual separability" to the vanishing point; and (2) their focus on the process or sequence followed by the particular designer makes copyright protection depend upon largely fortuitous circumstances concerning the creation of the design in issue.36
With regard to "conceptual separability," my colleagues deserve considerable credit for their efforts to reconcile Carol Barnhart Inc. v. Economy Cover Corp., 773 F.2d 411 (2d Cir.1985) with Kieselstein-Cord v. Accessories by Pearl, Inc., 632 F.2d 989 (2d Cir.1980). In my view, these cases are not reconcilable. Carol Barnhart paid only lip service to the fact that the "conceptual separability" of an article's aesthetic utilitarian aspects may render the design of a "useful article" a copyrightable "sculptural work." 17 U.S.C. § 101 (1982). Actually, the Carol Barnhart majority applied a test of physical separability. They thus stated:37
What distinguishes [the Kieselstein Cord] buckles from the Barnhart forms is that the ornamented surfaces of the buckles were not in any respect required by their utilitarian functions; the artistic and aesthetic features could thus be conceived of as having been added to, or superimposed upon, an otherwise utilitarian article. The unique artistic design was wholly unnecessary to performance of the utilitarian function. In the case of the Barnhart forms, on the other hand, the features claimed to be aesthetic or artistic, e.g., the life-size configuration of the breasts and the width of the shoulders are inextricably intertwined with the utilitarian feature, the display of clothes.
773 F.2d at 419 (emphasis added). In contrast, Kieselstein-Cord focused on the fact that the belt buckles at issue could be perceived as objects other than belt buckles:39
We see in appellant's belt buckles conceptually separable sculptural elements, as apparently have the buckles' wearers who have used them as ornamentation for parts of the body other than the waist.
632 F.2d at 993.41
My colleagues' adaptation of the Denicola test tracks the Carol Barnhart approach, whereas I would adopt that taken in Kieselstein-Cord, which allows for the copyrightability of the aesthetic elements of useful articles even if those elements simultaneously perform utilitarian functions. The latter approach received its fullest elaboration in Judge Newman's dissent in Carol Barnhart, where he explained that "[f]or the [artistic] design features to be `conceptually separate' from the utilitarian aspects of the useful article that embodies the design, the article must stimulate in the mind of the beholder a concept that is separate from the concept evoked by its utilitarian function." 773 F.2d at 422 (Newman, J., dissenting).42
In other words, the relevant question is whether the design of a useful article, however intertwined with the article's utilitarian aspects, causes an ordinary reasonable observer to perceive an aesthetic concept not related to the article's use. The answer to this question is clear in the instant case because any reasonable observer would easily view the Ribbon Rack as an  ornamental sculpture. Indeed, there is evidence of actual confusion over whether it is strictly ornamental in the refusal of a building manager to accept delivery until assured by the buyer that the Ribbon Rack was in fact a bicycle rack. Moreover, Brandir has received a request to use the Ribbon Rack as environmental sculpture, and has offered testimony of art experts who claim that the Ribbon Rack may be valued solely for its artistic features. As one of those experts observed: "If one were to place a Ribbon Rack on an island without access, or in a park and surround the work with a barrier, ... its status as a work of art would be beyond dispute."43
My colleagues also allow too much to turn upon the process or sequence of design followed by the designer of the Ribbon Rack. They thus suggest that copyright protection would have been accorded "had Brandir merely adopted ... as a bicycle rack" an enlarged version of one of David Levine's original sculptures rather than one that had wider upper loops and straightened vertical elements. I cannot agree that copyright protection for the Ribbon Rack turns on whether Levine serendipitously chose the final design of the Ribbon Rack during his initial sculptural musings or whether the original design had to be slightly modified to accommodate bicycles. Copyright protection, which is intended to generate incentives for designers by according property rights in their creations, should not turn on purely fortuitous events. For that reason, the Copyright Act expressly states that the legal test is how the final article is perceived, not how it was developed through various stages. It thus states in pertinent part:44
the design of a useful article ... shall be considered a ... sculptural work only if, and only to the extent that, such design incorporates ... sculptural features that can be identified separately from, and are capable of existing independently of, the utilitarian aspects of the article.
17 U.S.C. § 101 (1982) (emphasis added).46
I therefore dissent from the decision so far as it relates to copyrightability but concur in its discussion and holding as to the trademark and unfair competition claims.47
 Of the United States District Court for the District of Connecticut, sitting by designation.48
 The statute also defines "useful article" as one "having an intrinsic utilitarian function that is not merely to portray the appearance of the article or to convey information. An article that is normally a part of a useful article is considered a `useful article.'" 17 U.S.C. § 101.49
 Professor Denicola rejects the exclusion of all works created with some utilitarian application in view, for that would not only overturn Mazer v. Stein, 347 U.S. 201, 74 S.Ct. 460, 98 L.Ed. 630 (1954), on which much of the legislation is based, but also "a host of other eminently sensible decisions, in favor of an intractable factual inquiry of questionable relevance." 67 Minn.L.Rev. at 741. He adds that "[a]ny such categorical approach would also undermine the legislative determination to preserve an artist's ability to exploit utilitarian markets." Id. (citing 17 U.S.C. § 113(a) (1976)).50
 We are reminded not only by Judge Gesell in the district court in Esquire, 414 F.Supp. 939, 941 (D.D.C.1976), but by Holmes in Bleistein v. Donaldson Lithographing Co., 188 U.S. 239, 251-52, 23 S.Ct. 298, 300-01, 47 L.Ed. 460 (1903), by Mazer v. Stein, 347 U.S. at 214, 74 S.Ct. at 468, and by numerous other opinions, that we judges should not let our own view of styles of art interfere with the decisionmaking process in this area. Denicola suggests that the shape of a Mickey Mouse telephone is copyrightable because its form is independent of function, and "[a] telephone shape owing more to Arp, Brancusi, or Moore than Disney may be equally divorced from utilitarian influence." 67 Minn.L.Rev. at 746. This is true, of course, of the artist Christo's "Running Fence," approved (following Professor Nimmer) as an example of conceptual separability in Keiselstein-Cord, 632 F.2d at 993.51
 Because the district court viewed the rack as entirely functional, it therefore did not reach the next step of determining whether Brandir's RIBBON Rack had acquired secondary meaning by the time Cascade started to manufacture its bicycle rack.52
 Indeed, in addition to the numerous bicycle racks on the market, one may observe trees, awning supports, parking meters, signs, fire plugs, and many other objects used as bicycle racks.53
 Indeed, Kieselstein-Cord approved Professor Nimmer's example of Christo's "Running Fence" as an object whose sculptural features were conceptually, but not physically, separable from its utilitarian aspects. 632 F.2d at 993; see 1 Nimmer on Copyright § 2.08[B] at 2-96.1 & n. 112.2 (1987). The fact that the Running Fence's aesthetic features were "inextricably intertwined" with its functional aspects, however, creates doubt as to whether it is a copyrightable "sculptural work" under Carol Barnhart or the instant decision.54
 The reasonable observer may be forgiven, however, if he or she does not recognize the Ribbon Rack as an example of minimalist art.55
 The Copyright Office held that the Ribbon Rack was not copyrightable because it lacked originality. There may be some merit in that view in light of the Ribbon Rack's use of standard radii. This issue, however, was not raised in defendant's motion for summary judgment, was not addressed by the district court, and is not implicated here.