Central to this success has been a multipronged strategy to secure legal protection for the fruits of these investments. For more than a century, the bottle has been a pillar of that strategy, alongside Coca-Cola's secret formula and the Coca-Cola name itself. Today, the bottle is arguably the world's most famous example of "trade dress," a form of federal trademark law that protects product packaging that serves as a designator of source. 1 No one doubts the existence of so-called "secondary meaning," that consumer association between the bottle and the Company as a product source. When Supreme Court Justices ranging from Stephen Breyer to Antonin Scalia have written about trade dress, the bottle has served as a primary point of reference. 2 Despite the bottle's importance, the story of how it reached this exalted position is surprisingly neglected. This essay is an effort to fill that gap. We seek to recover the early history of the bottle with a view to understanding the interplay of law and marketing that enabled its present success. We draw upon a variety of sources, including the prosecution history of relevant patents, the proceedings of an early infringement suit, and the numerous settlements of litigation initiated by the Company against other soda producers. The latter were collected in a Company-sponsored publication called Opinions, Orders, Injunctions, and Decrees Relating to Unfair Competition and Infringement of Trade-Mark. 3 Opinions and Orders eventually grew to three volumes and was styled to resemble an authoritative bound compilation of case law such as the Federal Reporter.
As we explain, IP protection in the early years was fragile and contingent. 4 The story begins in 1914, when the Company's top lawyer spearheaded the effort to develop a new bottle. We examine the two main forms of legal protection available to the Company during this period. Design patents could protect the new bottle as a novel ornamental design. If successful, the Company would have exclusive rights to the design. Unfair competition law, a precursor to modern protection of trade dress, could protect the bottle as an identifier of source. If successful, rivals would be prohibited from using a confusingly similar bottle to "pass off" their colas as the real thing. Design patent law offered only short-term protection, given that a design patent expired 14 years from issuance. 5 Unfair competition offered long-term, effectively permanent protection once secondary meaning developed.
The Company was not first to employ these legal tools. Others had used design patents to protect packaging or had asserted unfair competition claims against rivals. The Company's efforts stood out, however, in their aggressiveness and sophistication. For example, it obtained not one but three design patents, which assertedly protected the bottle from imitators for about 36 years. It avidly sued rivals and published the results in Opinions and Orders. A quotation from Justice Oliver Wendell Holmes adorned the spine as a kind of warning to rivals. Writing for a unanimous Supreme Court, Justice Holmes had declared that the Coca-Cola name "means a single thing coming from a single source, and well known to the community." 6 By thus establishing that the name had achieved secondary meaning, the quotation simultaneously celebrated the cultural importance of Coca-Cola and handed the Company an important legal victory.
However, the Company's early efforts to protect the bottle were haunted by problems and risks. Its problems began with the first design patent, which covered a prototype that was significantly altered en route to production. That difference left the Company vulnerable, because the scope of the patent would not necessarily stretch to cover an alleged infringer that adopted the same shape as the production bottle. Worse, the Company revealed its weak position in an ill-advised patent suit against another soft drink maker. The court construed the patent narrowly, providing a road map for avoidance by others. The Company's later patents were no help either. The second design patent essentially covered the production bottle. However, the Company waited more than five years after production began to file an application. The resulting patent might well have been invalid given its close resemblance to the earlier production bottle. In any event, the patent could not have been applied retroactively to prevent copying of the production bottle. The surprising implication of these patent missteps is that other firms were probably legally free to use the hourglass shape as soon as the first bottle came off the production line in 1916. 7 The unfair competition approach faced uncertainties too. The patents helpfully bought time for the bottle to develop secondary meaning, a necessary part of any unfair competition claim. However, the patents' flaws put the development of secondary meaning at risk. Had rivals exploited the flaws, the unfair competition claim might have failed as well. The use of design patents to facilitate the development of secondary meaning also gave rise to an additional source of uncertainty. Early case law treated design patents as a bargain between the patentee and the public, in which the design was dedicated to the public once the patent expired. Courts rejected efforts to convert short-term patent exclusivity into long-term unfair competition protection. This hostility posed a threat to the viability of unfair competition as a means to protect the bottle from imitators.
Our discussion proceeds in four parts. Part I describes the severe challenge posed by copycat colas and the Company's development of a novel, distinctive bottle as a response. Part II examines the Company's design patent strategy and its shortcomings. Part III explains unfair competition law as a means to exclude rivals, including the central role played by secondary meaning. Part IV explores the interaction between design patents and unfair competition, particularly the risk that short-term protection using design patents might undermine the Company's long-term interesting in protection under unfair competition principles. Along the way, we identify significant modern echoes of the Company's strategies, including the pursuit of multiple patents to extend the duration of protection and the use of design patents as a way to buy time for the development of secondary meaning.
In the 1880s, the Company faced a major challenge. Copycat colas with similar names and bottle labels were everywhere. Noka-Cola and Coke-Ola, among many others, openly free-rode on the popularity of Coca-Cola. The Company responded with a flurry of lawsuits. Pursuing competitors one by one, however, was very difficult, given the sheer number of knockoffs. Moreover, as soon as one copycat was taken off the market, another one popped up elsewhere.
In response, the Company devised a second tool to deter knockoffs: the distinctive product delivery system and packaging that we know today as the Coca-Cola bottle. The idea of a distinctive bottle was not itself new. For decades, special bottles had been used to identify the drugs contained therein, and entire glassworks were devoted to manufacturing distinctive bottles for exactly this purpose. 8 The Company held a design contest with a $500 reward. 9 The idea was to devise a bottle that, in the words of a leading bottler, was identifiable "even when feeling it in the dark" and "so shaped that, even if broken, a person could tell at a glance what it was." 10 Eight glass manufacturers accepted the challenge. A team from the Root Glass Company visited the local library in Terre Haute, Indiana, to find images of coca and kola-ingredients in the original formulation of Coca-Cola-to somehow incorporate into their design. The team found nothing suitable; instead, legend has it, they found a picture of a cocoa pod on a nearby page of the Encyclopedia Britannica. Cocoa is used to make chocolate; it has nothing to do with coca. Nevertheless, the distinctive bulge of the cocoa pod was incorporated into the bottle design, resulting in the hourglass shape that we know today.
The new bottle was a potentially useful tool against copycats. Consumers would come to recognize the bottle by sight and by touch and to associate it with Coca-Cola. Armed with this association, consumers could successfully distinguish the real thing, no matter what the knockoffs named their beverage or how similar the soda itself looked. For this strategy to work, however, the Company had to solve two challenges: adoption and exclusivity. In other words, it was important that Coca-Cola be sold in the new bottle rather than a different bottle, and that competitors must be prevented from selling their cola in a bottle with this shape. Neither outcome was a foregone conclusion.
Adoption was not automatic because the Company did not do the bottling. Fountain Coca-Cola was the Company's initial focus, bottles an afterthought. Company founder Asa Candler thought bottles were low-class and left the bottling task to others, even going so far as to enter into a perpetual contract for syrup at a very low price because he was so dubious of the enterprise. Syrup was sold to bottling franchisees all over the country. Candler miscalculated, as bottle sales soon outpaced fountain sales. Even Americans who would never find themselves seated at the soda fountain could buy a bottle of Coca-Cola for a nickel. 11 While Candler's decision left huge profits on the table, the profitability of bottling had the happy side effect of encouraging entrepreneurs to spread the Coca-Cola gospel. Bottling turned out be one of the main driving forces behind consumer diversification and mass consumption. In hindsight, this misjudgment may well have been a driving force toward ubiquity of the product.
The Company needed the bottlers' cooperation to make the switch to a new bottle. Early bottles could be any shape or color, required by contract merely to have diamondshaped paper labels bearing the Company's name in capital letters. 12 Making the switch was costly. Moreover, as agents of the Company, some bottlers were faithless in the early 12 Id. at 20. days, furtively adulterating the syrup with saccharine, 13 suggesting that they might be a reluctant partner in the quest to stamp out free-riders. Harold Hirsch, the Company's general counsel, exhorted the bottlers to accept a new design "that we can adopt and call our own child." 14 In appealing to the bottlers' ambition, he also revealed his own: "We are not building Coca-Cola alone for today. We are building Coca-Cola forever, and it is our hope that Coca-Cola will remain the National drink to the end of time." 15 What must have seemed bombastic at the time also proved prophetic. In the end, the bottlers went along.
Exclusivity was important because otherwise, competitors could simply sell their soda in a bottle of identical shape. Conceivably, the higher cost of a specially manufactured glass bottle raised the costs of rivals seeking to market a knockoff. But at heart, exclusivity depended crucially upon intellectual property protection. When the Company introduced the new bottle, it stated its goal plainly in a newspaper ad that it furnished to its bottlers. "We've Bottled Up the Pirates of Business," the headline ran; "they cannot imitate the new Coca-Cola bottle-it is patented." 16 Legal exclusivity, once secured, facilitated the Company's heavy investments in marketing-investments that would have made free-riding even more tempting if it were permitted. By 1927, the Company claimed to have spent more than $20 million marketing the bottle, 17 and bottlers spent millions more. 18 These marketing efforts went far beyond advertising. For example, the Company worked with a range of artists to fashion the Coca-Cola image and infuse it into popular culture. Norman Rockwell's well-known "Out Fishin" (1935) portrayed a young boy fishing from his stoop on a tree stump, with his pole, his dog, and a bottle of Coca-Cola. Haddon Sundblum forever stamped his (and Coke's) mark on American cultural consciousness with his soon-to-be iconic Coca-Cola Santaplump, jolly, dressed in Coca-Cola red and white, and drinking from the bottle.
The investments paid off. Today, the bottle is famous, ubiquitous, and nearly synonymous with the product itself. Heavy marketing has inspired brand loyalty and, with it, a willingness to pay a premium for Coca-Cola. For some consumers, the taste of a Coke and its packaging might be said to have merged. 19 We now turn to the intellectual property protection that made this success possible.
Design patents were the Company's first line of defense, particularly important in the short run. A design patent protects the visual appearance of a product, including its overall shape and surface ornamentation. 20 Design patents are similar in important respects to the utility patents that cover useful inventions. Whereas a utility patent "protects the way an article is used and works," a design patent "protects the way an article looks." 21 To receive patent protection, a design must satisfy not only specific rules that are unique to design patents, but also rules initially developed for utility patents. 22 To be patentable, a design must be novel, nonobvious, and ornamental. 23 One modern formulation of the novelty test is that an ordinary observer, familiar with existing designs-in legal jargon, "prior art"-would not regard the new design to be "substantially the same" as any existing design. 24 Non-obviousness requires the existence of a single preexisting "primary reference" whose "design characteristics . . . are basically the same" as the new design. 25 Only if a primary reference can be identified may secondary references be used, in conjunction with the primary reference, to identify a design with "the same overall visual appearance" as the new design. 26 These tests are lenient, making it difficult for the Patent Office to reject design patent claims. 27 Coca-Cola was hardly the first firm to seek a design patent for an ornamental bottle. An early example is the 1890 patent for a "peculiar square-shaped, bulging-necked bottl[e]," used for rye whiskey by Cook & Bernheimer Co. 28 Novel bottle designs had also been patented for perfume and salad dressing, among other products. Thus, seeking design patent protection for the Coca-Cola bottle was very much part of a convention at the time. The Company's efforts were notable, however, in their aggressiveness. The Company secured a series of three patents, granted in 1915, 1923, and 1937, that collectively spanned nearly 36 years-more than double the 14-year duration of protection for a single patent. Patent #1, granted in 1915, covered the Root Glass prototype that won the design contest. 29 As shown in Figure 1, the design had a large bulge in the middle-so large that the bottle was unstable when placed on a surface. The production bottle, manufactured starting in 1916, was significantly slimmed down compared to the prototype. Patent #2, granted in 1923, was nearly identical to the production bottle, with a slight change in the "ribs." Patent #3, granted in 1937, was a further slight variant on the design in Patent #2. Patents in hand, the Company sued rivals for infringement. Patent #1 was the primary and typically the sole focus of these cases. The successful results were collected in the pages of Opinions and Orders. 32 A typical injunction prevented the use of any bottle "covered by" or "embodying the invention" of Patent #1. Often, the injunction also prohibited use of the Coca-Cola mark, likely reflecting an additional assertion of trademark infringement of the Coca-Cola name. Some also prohibited the use of the Company's actual bottles, presumably because the defendant was reusing empties. 33 In some cases, the defendant was also required to pay costs, which likely served as an additional deterrent to would-be infringers. The Company's patent strategy faced its most significant test in 1927, when it sued the Whistle Company, a seller of brightly colored soda, for infringement of Patent #1. 34 The Company identified, as the "characteristic feature" of the patent, that "line known to art and architecture as the 'line of beauty,' or ogee curve. It consists of a double or reverse curve, convex and concave." 35 Whistle's bottle also used a double curve but was different from the patented design in other respects, and indeed was covered by a design patent of its own. As shown in Figure 2, the Whistle design was tall and slender, where Patent #1 was short and bulging, and it lacked the ripples of the Coca-Cola design.
Viewed in hindsight, the Company's decision to sue Whistle is puzzling. Whistle was not a copycat or obvious free-rider on Coca-Cola's success. The Company did not even attempt an unfair competition claim. So far as appears, Whistle posed no real threat to Coca-Cola's business. Moreover, the success of the suit was far from assured, given the substantial differences in Whistle's design. There was also real potential downside, if the court invalidated the Company's flagship patent or construed it narrowly.
Whistle went to trial rather than settling. At trial, the parties sparred over the meaning of the Supreme Court's test for infringement, announced in Gorham Co. v. White:
[I]f, in the eye of an ordinary observer, giving such attention as a purchaser usually gives, two designs are substantially the same, if the resemblance is such as to deceive such an observer, inducing him to purchase one supposing it to be the other, the first one patented is infringed by the other. 33 At least as to patent law, a suit complaining of the reuse of empties would likely fail on exhaustion grounds. See Mitchell v. Hawley, 83 U.S. 544 (1872) ("Where patentee has constructed a machine and sold it without any condition . . . , patentee must be understood to have parted to such extent with all his exclusive rights, and that he ceases to have any interest whatever in the patented machine . . . ."). Such a suit might have a better chance under trademark law or unfair competition principles. 34 Coca-Cola Co. v. Whistle Co., 20 F.2d 955 (D. Del. 1927) (entering judgment for Whistle after a trial). In its complaint, the Company also asserted Patent #2 and an additional patent that it had acquired as part of a settlement with an alleged infringer. These claims were dropped before trial. 35 Id. at 956. 36 The Company tried to establish the necessary resemblance to an ordinary observer along dimensions other than the visual appearance of the two designs. Counsel argued, for example, that the designs would be difficult to distinguish by touch: an "ordinary dealer digging down into a tub of ice water at night" would have trouble distinguishing the two. 37 Counsel also arranged for Coca-Cola to be sold to consumers in Whistle bottles, to show that consumers didn't notice the switch, prompting the judge to muse that the same was likely true for a wide variety of bottle designs that didn't arguably infringe. The Company's overarching theme at trial was that an adverse ruling would devastate its business and render the design patent law a dead letter.
Whistle won the case. The court relied heavily on the fact that the prior art was full of other designs with double curves, including the designs shown in Figure 3. Thus, assuming that Patent #1 was valid despite the prior art, it must be narrow. The court's approach was in tune with an understanding of the ordinary observer test in which the comparison is made "in the context of the prior art." 38 In light of the prior art, the court concluded that the use of a double curve was not a basis for finding infringement. 39 Whistle revealed a significant flaw in the Company's patent strategy. Suppose another firm had decided in 1916 to sell a bottle that was identical in shape to the production bottle. Would this bottle infringe Patent #1? The answer is far from clear. A savvy defendant could make the same arguments that carried the day in Whistle: that the patent (if valid) was narrow in light of the prior art; that Gorham did not require the court to ignore the prior art. The use of a double curve-what the Company described as the "outstanding characteristic" of the patent-would not be enough to infringe. The defendant would also emphasize important differences compared to Patent #1. Just as in Whistle, the dimensions of defendant's bottle would be recognized as substantially different: tall and slender, rather than short and bulging. 40 The defendant would likely emphasize a further difference. The defendant's design, in adopting the shape of the production bottle, would contain a smooth panel on which the name of the product could be blown. Patent #1 is missing such a panel. 41 To be sure, this hypothetical case is more difficult than Whistle in some respects. The differences between Patent #1 and the Whistle bottle are larger than the differences between Patent #1 and the shape of the production bottle. If another firm used not only the same overall shape but also the ripples present in Patent #1 (and in the actual production bottle), the case would be closer still. Either way, the relevant question is whether the shape of the production bottle is different enough, such that the designs are not "substantially the same." Here, the fact that the production bottle effectively received its own patent (i.e., Patent #2), despite the existence of Patent #1 in the prior art, might tend to reinforce a conclusion of substantial difference. 42 Moreover, the Company's pursuit of Patent #2 might reflect an awareness that Patent #1 did not extend far enough. 43 Taken as a whole, a strong case can be made that Patent #1 did not stretch to cover the shape of the production bottle. 44 Our conclusion is reinforced by a candid moment toward the end of the Whistle trial. The Company's counsel argued that if the patent was so narrow that Whistle did not infringe, then neither would a copy of the production bottle. 45 The point of the argument was to present a parade of horribles to the court-that a narrow construal would open the door to imitation of the bottle. We believe that the Company counsel's argument is essentially correct. If we are right about that, then others could have used the same design as the production bottle-or, at minimum, the same shape-immediately upon its introduction in 1916.
If another company had taken this approach, Patent #2 could not rescue the Company from its predicament due to a basic principle of patent law: a patent cannot remove prior art from the public domain. That is, Patent #2 (granted in 1923, applied for the previous year) could not apply retroactively to remove the pre-existing (1916) design of the production bottle from the public domain. Thus, competitors were free to use the design, provided that no earlier patent covered it. If Patent #1 did not stand in the way, as 40 This difference was similarly emphasized in the application process for another design patent, in order to overcome the objection that it was too similar to Patent #1 to merit a separate patent. See Prosecution History, U.S. Patent No. D54,241 (Feb. 10, 1919) ("The shape of Samuelson's bottle [i.e., the design in Patent #1] is chunky, thick set and wholly devoid of the dainty slenderness and simple grace of outline which marks the applicant's design. . . . It appeals to the sense of the bizarre, the grotesque and the ridiculous. It is fat and funny."). D54,241 was not a Company design, but was eventually acquired by the Company as part of a settlement of patent infringement and initially asserted in the litigation against Whistle. 41 This point was made in the prosecution of U.S. Patent No. D54,241. See id. 42 See International Seaway Trading Corp. v. Walgreens Corp., 589 F.3d 1233, 1239-40 (Fed. Cir. 2009) (holding that the same standard applies to anticipation and infringement). 43 Dean points to Patent #2 as a possible effort to cure the defect. See DEAN, supra note 10, at 132. As we discuss below, the later patent could not cover the production bottle, which was part of the prior art. 44 Dean similarly suggests that the production bottle might be too different from Patent #1 to infringe. See id. 45 Whistle Transcript, supra note 17, at 166-67.
we discuss above, then the production bottle could have been used immediately. If (contrary to our argument above) Patent #1 was infringed, then the production bottle could have been used, at the latest, when Patent #1 expired in 1929.
Patent #2 had the additional problem that it might well have been invalid, given its near identity to the prior art production bottle. 46 Near identity is reflected in Coca-Cola's own public history of the bottle, which mistakenly describes Patent #2 as a "renew[al]" rather than a new patent. 47 Indeed, the designer reportedly was given instructions to make only small tweaks to the production bottle in developing the design that was the subject of Patent #2. 48 The Patent Office took nearly two years to issue the patent, initially rejecting the application on the ground that "the slight differences over [the prior art] are for either old features" or insufficiently novel to warrant a grant. 49 The prior art in question was yet another bottle patent; so far as appears, the production bottle was never discussed during the application process. 50 We recognize that even small differences from the prior art may be sufficient to support the grant of a fresh design patent. Nevertheless, the near identity of Patent #2 to the prior art raises serious doubts about validity. The patent's invalidity, if established, would be an independent reason why it would not block other bottles with the same shape. Invalidity could have a further consequence if the Company was aware of it, stemming from the Company's decision to mark its bottles with Patent #2 ("PAT'D DEC. 25, 1923") once Patent #1 expired. 51 Marking a bottle in this fashion when the bottle was in fact unpatented, "for the purpose of deceiving the public," would violate the patent marking statute, subjecting the Company to civil damages. 52 The Company's aggressive use of multiple patents to extend protection over time has an important modern echo. In the pharmaceutical industry, branded drug makers frequently secure multiple patents on a drug in an effort to extend the duration of patent protection and keep lower priced generic drugs out of the market. The later expiring patents are frequently challenged successfully by generic entrants, on the grounds (among others) that the patent is too narrow to cover the allegedly infringing product or else invalid in light 46 A similar issue arose in Kellogg Co. v. National Biscuit Co., 305 U.S. 111 (1938), a famous Supreme Court case about shredded wheat discussed in Part IV. See id. at 119 n.4 (1938) (noting that a court had previously held a pertinent design patent to be invalid because the design was in public use more than two years before a patent application was filed). 47 Coca-Cola History, supra note 9 ("In 1923, the patent for the bottle was renewed"). 48 DEAN, supra note 10, at 148, describes the work of his father, a designer at Root Glass: "Dean was instructed to make the change so minor that it would not be noticeable to preserve the integrity of the highly successful design." 49 Prosecution History, U.S. Pat. No. D63,657 (Sept. 2, 1922). 50 of the prior art. 53 Fortunately for the Company, apart from the Whistle case, the bottle patents appear not to have been subjected to much scrutiny, preserving them as a tool to stop and deter would-be copyists.
Design patents were at best only a short-term solution to the problem of copycats. The long-term source of protection for the bottle was unfair competition law (and ultimately federal trade dress protection). As noted in the introduction, the bottle has been repeatedly invoked as the paradigmatic example of federal trade dress. How did we get here?
Much as a trademark and trade dress today is protected from misappropriation by a counterfeiter, product packaging and design were similarly protected from copycats under the common law. Protection was administered by a mix of state and federal courts. A private party could sue in federal court, claiming that a competitor was creating consumer confusion by selling a product with the same design or packaging as the plaintiff. Often, the particular claim was that the defendant had engaged in "product simulation." Such unfair competition cases were a precursor to and form of modern trademark law.
Prior to the introduction of the Coca-Cola bottle, firms relied upon unfair competition law to enjoin competitors from using confusingly similar bottles. For example, in the late 1890s, Hires' Root Beer was known by its "peculiar" bottle: cylindrical with "high shoulders and short neck." 54 The distinctive shape was the purchaser's "principal method of identification." 55 Hires secured an injunction preventing a competitor from selling soda in a similarly shaped bottle. As the court of appeals explained, "[t]he defendant is not deprived of the right to market its root beer, but, at its peril, must see to it that its product is not dressed in the clothes of another." 56 The Hires court considered allowing the defendant to use the same shape but with a different label, and rejected this alternative as insufficient. 57 In another case around the same time, a federal court granted similar relief for another "peculiar" bottle, this one "square-shaped [and] bulging-necked" for rye whisky. 58 Other cases involving a diverse array of products frequently, though not always, resulted in an injunction. 59 53 55 Id. at 810. 56 Id. at 813-14. 57 Id. at 813 ("Any restraint . . . that did not include the form of bottle . . . would be ineffective to stay the wrong."). 58 Cook & Bernheimer Co. v. Ross, 73 F. 203, 205-06 (C.C.S.D.N.Y. 1896). There was no imitation of the label, notwithstanding some similarity: "Complainant's case rests solely on the form of package, which it claims has been so imitated as to make out a case of unfair competition." Id. at 204. 59 Secondary meaning was also important in a later period, as a basis for modern trade dress protection. As a technical matter, it might not be strictly necessary for the Company to establish secondary meaning for the bottle. Some marks and some trade dress are "inherently distinctive": their "intrinsic nature serves to identify a particular source." 67 KODAK for film is a famous example of an inherently distinctive mark. Certain forms of trade dress, including product packaging, can be inherently distinctive as well. Product design, by contrast, cannot. Whether the Coca-Cola bottle is product packaging, as opposed to product design, is unclear. The Supreme Court opinion that held that product designs cannot be inherently distinctive pointed to the Coca-Cola bottle as the paradigmatic "hard cas[e] at the margin" between the two: a classic glass Coca-Cola bottle, for instance, may constitute packaging for those consumers who drink the Coke and then discard the bottle, but may constitute the product itself for those consumers who are bottle collectors, or part of the product itself for those consumers who buy Coke in the classic glass bottle, rather than a can, because they think it more stylish to drink from the former. 68 In any event, even if secondary meaning were not strictly necessary for protectability, it would be helpful, since a stronger association translates to broader scope of protection for the trade dress. 69 The Company was also highly familiar with efforts to demonstrate secondary meaning from its trademark cases against rivals. As noted in Part I, it sued numerous competitors with similar names, including a copycat called Koke. The case eventually reached the Supreme Court, and the central issue addressed by the Court was whether the Coca-Cola name was understood by consumers as a signifier of source, as opposed to a description of the products' ingredients. This was important because as a description, the name was a failure. Originally, the name was descriptively accurate. Coca leaf gave the product its original cocaine kick, and the kola nut was known as a source of caffeine. The Company played up the connection with illustrations of coca leaves and kola nuts on bottle labels and advertisements. But times changed and the product was reformulated. Cocaine was removed from the "soft" drink, 70 and the kola nut was only used in trace amounts.
The inaccuracy gave Koke an opening. Koke argued that the Company had "unclean hands" for using a misleading mark. Unclean hands is an equitable doctrine that prevents a plaintiff from successfully invoking the equitable powers of the court if the plaintiff itself is engaged in certain forms of legal wrongdoing. Because the relief sought in all of these trademark suits was injunctive-an order for the defendant to stop marketing 67 Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 768 (1992). 68 Wal-Mart Stores, Inc. v. Samara Bros., 529 U.S. 205, 215 (2000). 69 its product with the confusing or deceptive name-the Company's wrongdoing could prevent legal remedy entirely.
The potency of such an accusation of misbranding was brought home by another (non-IP) case, quaintly named United States v. Forty Barrels & Twenty Kegs of Coca-Cola. 71 The Food and Drug Administration complained that because Coca-Cola contained "no coca and little if any cola," it was misbranded, in violation of the federal pure food law of 1906. The Company was in a no-win legal situation. If Coca-Cola had contained cocaine, the company would have been in trouble for the cocaine, which became illegal to distribute without a doctor's prescription in 1914. Yet, absent cocaine and kola, its mark was misleading and arguably its product misbranded. 72 The FDA's complaint, which ultimately settled, showed the Company's vulnerability to a misbranding claim. The Company's trademark suits eventually escaped the specter of misdescription, in large part because of the success of their marketing campaigns and the ubiquity of the product in the popular imagination. In 1920, the Supreme Court decided that it didn't matter that the Coca-Cola name was once misleading, because consumers had come to understand the name as a signifier of source. Thus, the Holmes quotation in the introduction: "The name now characterizes a beverage to be had at almost any soda fountain. It means a single thing coming from a single source, and well known to the community." 73 Put differently, the name Coca-Cola, although descriptively misleading (arguably to the point of unlawfulness), came to signify the source of the soda rather than a claim about the nature of the soda. Rather than running afoul of the law, by being a widely successful product, the name came to be protected as a trademark.
The Company's dual approach to bottle intellectual property-design patent protection in the short run, unfair competition (and later federal trade dress) in the long run-had some distinctive benefits and risks. First, and most obviously, every little bit helps. IP rights are generally probabilistic and uncertain; one never knows when a suit might fail. A design patent might be deemed too narrow to exclude a rival; a particular unfair competition suit might result not in an injunction but in a weaker remedy, such as requiring the defendant merely to affix a clarifying label. Perhaps better, then, to take a belt-and-suspenders approach.
The design patents also reinforced the unfair competition claims in an important way. Secondary meaning takes time to develop. If a design is copied quickly, before secondary meaning emerges, secondary meaning might never be achieved. This is not a merely theoretical concern. Today, designs that are copied before secondary meaning has 71 241 U.S. 265 (1916). 72 The litigation also included an allegation of adulteration, which was prohibited under the Pure Food and Drug Act of 1906 and the subsequent Food, Drug, and Cosmetic Act of 1938. 73 Coca-Cola Co. v. Koke Co. of America, 254 U.S. 143, 146 (1920). a chance to develop are out of luck, insofar as trade dress protection is concerned. 74 Here, the Company's design patents provided the Company with more than three decades of protection during which it was able to invest heavily in marketing. Without patent protection, it might not have been possible to convince imitators to stay out of the market for long enough to establish second meaning. If a rival had shown that Patent #1 did not cover the production bottle, much as it did not cover Whistle's, secondary meaning might never have developed.
At the same time, the design patents posed a significant threat to the Company's unfair competition claims. Intellectual property doctrine and policy reflects a longstanding concern with bootstrapping: that one type of claim might be used to compensate for a weakness or deficiency in another, and thereby extend the protection beyond its proper bounds. A particularly potent version of this concern involves patent law. Patent law provides powerful legal protection, but only for a limited time. The bargain between the patentee and the public is that after expiration, the patented invention or design is dedicated to the public. If time-limited patent protection could be converted to indefinite protection as a matter of unfair competition or trademark, the public would not receive the full use of the invention or design. This concern with bootstrapping was reflected in the case law well before the Company launched its new bottle. In the late 1800s, Singer, the dominant manufacturer of sewing machines, excluded competitors by means of utility patents on various aspects of its design. After the patents expired, Singer sued a competitor that made sewing machines with an identical "size, shape, ornamentation, and general external appearance" as Singer's. 75 Singer argued that the competitor had done so in order to "induc[e] the belief" that the machines were actually made by Singer. This was an unfair competition claim that effectively sought to preserve exclusivity beyond the point of patent expiration.
The Supreme Court accepted the unfair competition concern in principle, but rejected the implication that a competitor could be enjoined from selling a product with the identical form. Such an injunction would violate the bargain the patentee had made with the public: It is self-evident that on the expiration of a patent the monopoly created by it ceases to exist, and the right to make the thing formerly covered by the patent becomes public property. It is upon this condition that the patent is granted. It follows, as a matter of course, that on the termination of the patent there passes to the public the right to make the machine in the form in which it was constructed during the patent. We may therefore dismiss without further comment the complaint as to the form in which the defendant made his machines. 76 The Court accommodated the unfair competition concern to a limited degree by requiring the competitor to take steps to clarify to consumers that it, and not Singer, made the machines.
Singer signaled a potential problem for the Company's use of both patent law and unfair competition law to protect the bottle from competitors. The patents would create space for secondary meaning to develop, but at the potential cost of threatening the Company's ability to enjoin competitors' use of the bottle. A court might order the rival to merely label its bottles to clarify that the product was not Coca-Cola. Labeling was an imperfect solution that would presumably defeat the purpose of having distinctive bottles in the first place. To be sure, it fell short of the Company's aspiration to have a bottle that was identifiably associated with Coca-Cola, even by feel or broken into pieces. Implementation of a remedy short of injunction, moreover, was highly uncertain in practice. Paradoxically, to the extent these risks were recognized, it might have created additional pressure on the Company to pursue additional design patents. The Company's strategy differed from Singer's, in that the bottle was protected by design patents, rather than utility patents. However, arguably this was a distinction without a difference. A design patentee equally reaches a bargain made with the public. So the Second Circuit appeared to conclude in a 1918 case, not long after the first bottle design patent was granted and the bottle went into production. Shredded Wheat Co. v. Humphrey Cornell Co. 77 was a case about the shape of shredded wheat cereal biscuits. Henry Perky had developed the biscuits after being prescribed a diet of mushed wheat that he found inedible unless he baked it to a crisp. The product was protected not only by a utility patent on the biscuit, but also a design patent on the pillow shape. After the patents expired, a rival made a biscuit "identical in substance and appearance with the plaintiff's." 78 The plaintiff sued in unfair competition, alleging that secondary meaning had been acquired in the appearance of the biscuit.
The court applied Singer, concluding that upon patent expiration, the form of the product-i.e., the pillow shape-had been dedicated to the public and could therefore be copied freely. Moreover, Judge Learned Hand's opinion for the court explicitly connected this dedication to the public with the expiration of the design patent. 79 As in Singer, the court then proceeded to assess to what extent the plaintiff's secondary meaning might be protected by requiring the defendant to alter its biscuit in other respects, such as color or size.
Twenty years later, the Supreme Court came to a similar conclusion in another shredded wheat case. Nabisco, having acquired Perky's interests, sued Kellogg, which was manufacturing biscuits with the same shape. Nabisco argued, once again, that copying the pillow shape amounted to unfair competition-that Kellogg was thereby "passing off" 77 250 F. 960 (2d Cir. 1918). 78 Id. at 962. 79 "As to form, the plaintiff appears to us finally concluded by its own design patent. . . . [T]he plaintiff's formal dedication of the design is conclusive reason against any injunction based upon the exclusive right to that form, however necessary the plaintiff may find it for its protection." Id. at 964. Kellogg biscuits as Nabisco's. 80 The Court once again sided with the entrant, pointing to the expiration of utility patents (on the product and on machinery) and a design patent:
The plaintiff has not the exclusive right to sell shredded wheat in the form of a pillow-shaped biscuit-the form in which the article became known to the public. That is the form in which shredded wheat was made under the basic [utility] patent. The patented machines used were designed to produce only the pillow-shaped biscuits. And a design patent was taken out to cover the pillow-shaped form. Hence, upon expiration of the patents the form . . . was dedicated to the public. 81 The Kellogg Court then recited Singer's discussion about the patent bargain. As with Singer and Cornell, the Court once again recognized the unfair competition claim in part, insisting that the entrant market the pillow-shaped biscuit "in a manner which reasonably distinguishes its product from that of plaintiff" and analyzing at length the scope of its responsibility. 82 Taken together, these cases suggested that the Company would be unable to assert an unfair competition claim that blocked others' use of the bottle once the design patents had expired. The logic of the bargain theory applied with equal force to utility and design patents, and had been explicitly applied to design patents in Cornell and Kellogg.
Ultimately, however, doctrine went in a different direction. Alongside the bargain approach, a second line of thinking emerged that more tightly connected the "no bootstrapping" principle solely to utility patents, thereby leaving space for design patents to receive different treatment. The argument was that an incumbent should not be allowed the advantage of a functional feature-that is, a product feature that results in lower costs or higher quality-on a permanent basis. This argument appeared in Kellogg, which rested in part on the functional concern that a rival would be placed at a permanent practical advantage if it were denied access to the pillow shape. The Court opined that the "cost of the biscuit would be increased and its high quality lessened if some other form were substituted for the pillow shape." 83 The same concern also could be found in Cornell, which similarly evaluated possible changes to defendant's product in light of higher cost or lower quality. 84 This same point was hinted at in Hires, which noted that plaintiff's distinctive bottle served no functional purpose, but rather sought to identify the source of the product, which suggested in turn that defendant's use of the distinctive bottle shape could serve no other purpose but fraudulent imitation. 85 Today, this concern about functionality is central to modern trademark doctrine, which denies trade dress protection to functional features. 80 As in Singer, the broader case included allegations about use of the product name. 81 Kellogg Co. v. National Biscuit Co., 305 U. S. 111, 119-20 (1938). 82 See also id. at 120 (describing "obligation to identify its product lest it be mistaken for that of the plaintiff"). 83 Id. at 122. 84 Shredded Wheat Co. v. Humphrey Cornell Co., 250 F. 960, 965 (2d Cir. 1918) ("The question is always commercial; we ought not to impose any burdens which, either by changing the appearance of the article itself, or by imposing expense upon its production, will operate to give the plaintiff such advantage in the market as will substantially handicap his competitors."). 85 See Charles E. Hires Co. v. Consumers' Co., 100 F. 809, 812 (7th Cir. 1900) (noting that any functional justification for the imitation was "pretentious").
A concern with functionality provides a basis for distinguishing design patents from utility patents. Utility patents protect functional features, whereas design patents protect ornamental features. Thus, an unfair competition claim directed to a particular shape that had been the subject of a utility patent is likely to implicate functional concerns. Indeed, in modern trade dress doctrine, the existence of an expired utility patent is "strong evidence" of functionality. 86 Such concerns about functionality are unlikely to be present for design patents.
Ultimately, this second approach carried the day. Functionality, and a version of functionality confined to utility patents, is the principal doctrine that mediates the interaction between patent and trade dress. Today, an expired design patent is generally held to pose no barrier to continued trade dress protection or injunctive relief. 87 It is common for counsel to advise clients to follow the path blazed by the Coca-Cola Company: to secure a design patent and thereby facilitate the acquisition of secondary meaning, all with a view to enjoying trade dress protection down the line. 88 In this respect, as in so many others, Coca-Cola was ahead of its time.
The Coca-Cola bottle has become a ubiquitous signifier of the product and, by providing uniformity of quality, price, and access, an equalizing force in society. The bottle embodies part of why Warhol celebrated Coca-Cola as an American cultural icon: What's great about this country is that America started the tradition where the richest consumers buy essentially the same things as the poorest. You can be watching TV and see Coca-Cola, and you know that the President drinks Coke, Liz Taylor drinks Coke, and just think, you can drink Coke, too. A Coke is a Coke and no amount of money can get you a better Coke than the one the bum on the corner is drinking. All the Cokes are the same and all the Cokes are good. Liz Taylor knows it, the President knows it, the bum knows it, and you know it. 89 Today, the bottle has retained its powerful hold on the public imagination, even its actual use as a container for the product has waned. The Company eventually shifted from glass packaging to aluminum cans and plastic bottles, while continuing to feature the bottle in advertising. 90 In the 1990s, the bottle made a big return in plastic bottles that adopted a version of the hourglass shape. In the 2000s, the Company introduced a new aluminum bottle. As the Company noted in advertisements touting the bottle's return, sounding an almost apologetic note: "Certain things belong in certain packages. Anything else just doesn't seem right." 91 The silhouette appears pervasively in advertising and as an image printed on bottles and cans, often near the curvy "dynamic ribbon device," which is a further visual echo of the bottle's shape. The bottle's success has been built on the legal ability-or at least the perceived ability-to exclude and thereby deter knockoffs. Legal protection provides a way to secure a return on enormous investments in marketing and branding. And yet, as we have shown, the basis for excluding rivals was surprisingly thin and historically contingent in the early days. The design patents' flaws meant that the shape might have been available to others from the start. The Company got lucky that the defects were apparently not recognized. Had the flaws been discovered and probed at the time, secondary meaning might never have developed. Similarly, had the public bargain approach to patent expiration carried the day, the design might have been freely used by others. Thus, there is a good chance that, with scores of competitors using an identical bottle to deliver their beverages, the bottle would have been no more associated with Coca-Cola than a standard wine bottle is with Chateau Lafite Rothschild. 90 DEAN, supra note 10, at 110. 91 Id. at 113.