Industrial Design: Reinterpreting the Useful Article Doctrine

By: Chris Piazza

Industrial design is design that focuses not only “on the appearance of a product, but also on how it functions, is manufactured and ultimately the value and experience it provides for users.” As creations of the mind, industrial designs constitute protectable intellectual property; however, they fall uncomfortably into the three existing divisions of intellectual property with the potential to protect them—copyright, trade dress, and design patent. Some scholars contend that Congress ought to enact a distinct statutory scheme, apart from existing divisions of intellectual property, to protect industrial designs. This blog post examines each of these alternatives and concludes that slightly modifying copyright law’s useful article doctrine has the potential to improve industrial design protection.

Trade dress

Trade dress protects a product’s configuration, or “design and shape.” For a product’s configuration to qualify for trade dress protection, it cannot be functional and it must have acquired a secondary meaning. A configuration acquires secondary meaning when a significant number of consumers perceive it as denoting the product of a particular manufacturer (rather than being merely arbitrary or decorative) after it has been used in the market.

These qualifications render trade dress barely viable for protecting industrial design. The non-functional requirement disqualifies from protection any design that makes its product easier to manufacture, more convenient to ship, or work better. And industrial design frequently does just this. Furthermore, the secondary meaning requirement leaves many designs out in the cold: most do not have secondary meaning and those of upstarts, by definitions, cannot have secondary meaning.

Design patent

Design patents protect “new, original and ornamental design[s] for an article of manufacture.” While design patents lack a utility requirement, they are typically only granted for the appearance of a functional product. To qualify for a design patent, a design must be “primarily ornamental” and cannot have been obvious to a designer of ordinary skill of that type of product.

Intellectual property attorneys Richard Sybert and Lindsay Hulley argue that design patent is the best available mode of industrial design protection, primarily because it circumvents some of the problems that copyright and trade dress face. Unlike proving copyright infringement, proving design patent infringement does not require a showing that the infringed object was in fact copied. Unlike proving trade dress infringement, proving design patent infringement does not require a showing of existing customer confusion or acquired secondary meaning.

Improvement relative to flawed modes of industrial design protection does not, however, mean that design patent is always suited to the task. Design patent, because of its ornamentality requirement, shares with copyright and trade dress the problem that the more complete a design’s integration of function and form, the less likely it is to obtain protection. Additionally, obtaining a design patent is a costly and time-consuming process—which is particularly problematic as designs frequently live ephemeral vogues. Finally, requiring novelty and nonobviousness forecloses much design from design patent eligibility. As Magliocca explains:

[M]ost designs simply involve rearranging basic artistic elements (e.g. colors, shapes, and materials) into a new pattern. The result of this reshuffling process can almost always be considered obvious—and hence unpatentable—because any designer could have come up with the combination at issue.

Copyright

Copyright protects “original works of authorship.” Although “pictorial, graphic, and sculptural works” are among the enumerated classes of works that the Act protects, the Act further specifies, in what is known as the useful article doctrine, that the design of a useful article:

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.

Additionally, the Act defines a useful article as a work “having intrinsic utilitarian function that is not merely to portray the appearance of the article or to convey information.”

In Star Athletica v. Varsity Brands, the Supreme Court elaborated on the statutory language of the useful article doctrine, articulating a two-prong “separability” test. Under the separability test, a feature incorporated into the design of a useful article is copyrightable if it: (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 when fixed in some other tangible medium.

Critics contend that the useful article doctrine cripplingly limits copyright’s potential to protect industrial designs, which aim for total confluence of form and function. This concern is justified given the assumption implicit in Athletica’s separability test, namely that a feature of a design either is or is not, in binary fashion, separable from its useful article. This blog post discusses why this assumption is both doubtful and legally fraught in the Where to Improve section below.

Sui Generis

Scholars have proposed sui generis protection as an alternative to copyright, trade dress, and design patent protections. Latin for “of its own kind,” sui generis refers in intellectual property to “a regime designed to protect rights that fall outside the traditional patent, trademark, copyright, and trade secret.” American intellectual property law currently provides sui generis protection that incorporates elements of both copyright and patent law for vessel hulls and semiconductor chips. Unlike the United States, the European Union, Japan, and others have extended sui generis protection to industrial designs. Scholars contend that well-crafted sui generis legislation would better serve the idiosyncratic needs of industrial designs than the intellectual property divisions currently in place.

With good reason, Congress is reticent to grant matters sui generis protection. Granting such protection too widely would expand intellectual property’s domain, perhaps at the expense of intellectual property’s overarching purpose: incentivizing creation. Intellectual property’s primary method of incentivizing creation, after all, is limiting the ability of others to participate in the marketplace.

The ambit of intellectual property protection is a matter of purposeful legislative choice. For example, Congress has chosen to exclude from patent protection any design that would have been obvious to a skilled artisan. The Supreme Court has explained that this rule follows from the judgment that “[w]ere it otherwise patents might stifle, rather than promote, the progress of useful arts.” When sui generis protections lower the protectability threshold and award intellectual property rights over things that are not quite patentable, copyrightable, or the like, the potential effect is a diminishment of the public domain and interference with future innovation and competition.

Where to Improve: Reinterpreting the Useful Article Doctrine

The useful article doctrine is copyright’s primary limitation in protecting industrial designs. For a feature of a design to be copyrightable, the doctrine requires that feature be separable from its useful article. However, the Supreme Court’s binary distinction between features that can and cannot be separated is a false dichotomy. A court can find any design feature separable from its useful article by viewing it at a sufficiently specific level of abstraction. Likewise, a court can find any design feature inseparable by viewing it at a sufficiently general level of abstraction. Indeed, though the Supreme Court agreed in Star Athletica on the separability test, it was split, when applying the test, as to whether the design feature in question was separable. Regarding design features, all that one can fairly say is that some are separable at a more or less general levels of abstraction than others.

For an illustration, consider the bicycle rack in the image above. Using the separability test, one could successfully argue in court that the rack’s curving metal rod is uncopyrightable. To do so, one would contend, at a general level of abstraction, that the curving metal rod is inseparable from the rack as a whole, because both the curving form and the functional capacity of the rack reside in it alike.

One could also conceivably succeed in court on the claim that the rack’s curving form is copyrightable. To do so, one would argue, at a more specific level of abstraction, that, though that curving metal rod is an aspect of the bike rack, the rod itself has sub-aspects. One functional (and hence uncopyrightable) sub-aspect of the rod is its steel material, which lends it strength. Another aspect, this time aesthetic, is the rod’s particular curvature, on which the functionality of the rack, arguably at least, does not depend. To demonstrate the curvature’s separability, one might draw it out with a pencil on paper.

The separability test’s current binary approach dooms opposing parties to talk past one another, each at a different level of abstraction, with courts only clumsily able to direct the proceedings. Courts could more productively channel discourse if they considered separability across a spectrum. By applying a spectrum, courts would be able to determine the sorts of designs that are truly worthy of copyright protection with greater precision and sophistication. The resulting body of case law would make guessing whether an aspect of a design is separable less of a shot in the dark. Consequently, copyright would become a viable alternative to design patent for protecting industrial designs, the importance of which follows from the numerous situations where design patent, useful though it may be, falls short.