Patent

Inventions Made for Hire

Inventions Made for Hire
Joshua L. Simmons* A pdf version of this article may be downloaded here.  

Introduction

Despite the persistent notion that modern invention is performed by individual inventors in their garages, few would disagree that today most patentable inventive activity occurs in corporate and university settings and that most individuals who would be labeled “inventors” in the twenty-first century are employees of corporate entities.[1] In the corporate setting, if an employee creates a copyrightable work within the scope of his or her employment, the Copyright Act not only grants ownership of the work to the employer but actually considers the employer the “author” of the work.[2] By contrast, under the Patent Act, one who creates an invention is its inventor, and ownership will only pass to another, including an employer, through a written assignment.[3] In other words, unless there is an agreement to the contrary, an employer does not have any rights in an invention “which is the original conception of the employee alone.”[4] Given the close relationship between copyright law and patent law,[5] it is puzzling that employees’ works and inventions would be treated so differently under the two disciplines. Yet, when one considers their historical foundation, it becomes clear that there were pivotal differences between the needs of the two disciplines in the nineteenth century that led to their modern formulations.[6] In particular, whereas the type of copyrightable works created in the nineteenth century transitioned from individual labors to collaborative labors among multiple individuals working together, which necessitated the collecting of rights in order to make use of the resulting copyrightable work, patentable inventions continued to be perceived during the nineteenth century as the work of individuals. Moreover, patent law developed other, more limited doctrines that provided some rights to inventors’ employers. Today, however, most patentable inventions are invented by multiple inventors in a collaborative environment.[7] In fact between 1885 and 1950, the percentage of U.S. patents issued to corporations grew from 12% to at least 75%.[8] These numbers have only continued to increase in the past decade.[9] Moreover, the recently passed Leahy-Smith America Invents Act partially recognizes this change by permitting an “applicant for patent” to file a “substitute statement” instead of an inventor’s oath or declaration in certain circumstances.[10] Nevertheless, patent law remains stuck in the past. The failure to modernize how patent law handles employee invention has led to a string of significant court opinions, including from the Supreme Court and the Federal Circuit, holding that employers had not received adequate assignments to their employees’ patented inventions and thus could not bring suit based thereon.[11] In order to resolve these issues and bring patent law into the twenty-first century, the Patent Act should be amended to borrow from the Copyright Act and adopt a principle similar to the work made for hire doctrine that would grant employers the rights to their employees’ inventions made within the scope of their employment.[12] This article proceeds in four parts. Part I discusses the current state of copyright and patent law vis-à-vis employers’ rights in their employees’ intellectual labor. This part will compare copyright’s work made for hire doctrine to three patent law doctrines: the shop rights doctrine; the hired-to-invent doctrine, and the employee improvements doctrine. Part II describes the evolution of copyright law during the nineteenth century from the former regime, which vested employees with presumptive ownership of their work, to the current regime, which grants employers presumptive rights to their employees’ efforts. Part III describes patent law during the time that copyright law was changing so dramatically—including the development of the patent law doctrines described in Part I—and posits potential reasons that patent law did not make a similar move. In particular, (a) patent law’s development of the doctrines described in Part I provided similar benefits, although in the form of different rights, to those provided by copyright law’s work made for hire doctrine, and (b) the perceived nature of invention in the nineteenth century did not call for a unification and codification of those doctrines the way copyright law’s did. Finally, Part IV argues that patent law should modernize and develop an “inventions made for hire” doctrine.

I. IP and the Employer-Employee Relationship

It is well settled in the United States that copyright law and patent law treat legal title in the intellectual labor of employees differently.[13] Section A discusses the copyright law doctrine of work made for hire and its benefits for employers. Section B discusses doctrines that provide some of those benefits in patent law.

A. Copyright Law

Under the Copyright Act, title in a copyrightable work initially vests in the author or authors of a work.[14] However, “author” is a term of art with meaning beyond the creator of or “person who originates or gives existence” to a work.[15] In particular, where a “work made for hire”—another term of art—is concerned, the employer is “considered the author” unless the parties agree otherwise.[16] A work may be considered made for hire if (1) it was prepared within the scope of an employee’s employment, or (2) it is a certain type of commissioned work and the parties have so agreed previously.[17] Whether the creator of a copyrightable work is an employee for purposes of the work made for hire doctrine is determined by looking to the common law doctrine of agency.[18] In Community for Creative Non-Violence v. Reid, the Supreme Court identified thirteen factors that it considered relevant to whether an individual was considered an employee as a matter of agency law.[19] The Supreme Court declared that “[n]o one of these factors [was] determinative,”[20] but at least one circuit has held that not all factors are created equal and some will be important in “virtually every situation” while others “will often have little or no significance.”[21] The more important factors are to be given “more weight in the analysis.”[22] Once a work is considered made for hire, the employing or commissioning party enjoys several benefits over a work that is not considered made for hire, but rather is transferred to the employing party. First, unlike in patent law, it immediately becomes the owner of a legal right to the work.[23] In patent law, an inventor must first file an application with the United States Patent and Trademark Office (“PTO”), which is then examined by a PTO employee to determine if the alleged new invention is entitled to a patent.[24] Copyrights, on the other hand, vest immediately once an original work of authorship is fixed in a tangible medium of expression.[25] For a company to gain an ownership interest in a copyrighted work not made for hire, the interest must be negotiated for and transferred in a signed written document.[26] When a work qualifying as a work made for hire is created, however, no negotiation or written instrument is required for that specific work because, for copyright purposes, as long as agency would hold that the work’s creator was an employee working within the scope of his employment, his employer is considered the work’s author.[27] There are two subsidiary benefits to the immediate grant to the employer of a work’s copyrights. First, no court intervention is required to transfer the rights. As will be described below, in certain cases, patent law will grant employers rights in their employee’s patents, but to vest those rights the employee must still sign a document transferring ownership to his employer. If he refuses, a court order may substitute for the transfer. Second, the work made for hire doctrine allows a grant of rights without needing to define what is being granted. Whereas when a grant of rights is negotiated, the transferor may come back to the transferee to argue about which rights were actually transferred and the scope of the intended transfer.[28] Second, an employer holding the rights in a work made for hire may register the work in the Copyright Office in its own name.[29] This is an optional but important step, because without registration a copyright infringement action may not be initiated,[30] and generally, statutory damages are not available for infringements prior to registration.[31] Third, the work made for hire doctrine grants employers all of the rights associated with copyright ownership. Thus, they can exclude others from using the work, and leverage that right to extract rents in exchange for a license authorizing someone else to use the work.[32] Further, employers using works made for hire are authorized to then transfer their ownership of the works’ copyrights to another party, either in whole or in part.[33] They can also, of course, exercise any of the rights authorized under the Copyright Act itself, including creating derivative works.[34] One set of rights an employer is not granted under the Copyright Act, however, is the right to attribution and integrity granted for works of visual art.[35] Fourth, the work made for hire doctrine also protects employers from future rights granted to authors. For example, the Copyright Act of 1976 granted authors the right to terminate transfers, but those new rights did not apply to creators of works made for hire because their employers were considered the author.[36] In addition, after the 1976 revisions, employers remained shielded from the exercise of the transfer termination provisions because no transfer is considered to have taken place under the work made for hire doctrine;[37] the employer was the work’s original “author.” Fifth, the work made for hire doctrine grants employers certainty with regard to the duration of their copyrights. Works not made for hire subsist for the life of the author plus 70 years.[38] This means that those to whom the original author transfers must know the duration of the original author’s lifetime in order to calculate the duration of the copyrighted work’s term. Works made for hire, however, are merely granted a term of 95 years from publication or 120 years from creation, whichever expires first.[39] This time frame serves two functions. First, it simplifies the requirements of an employer to keep track of the duration of its copyrights. Second—grim though it may be—if the employee were to die within 25 years of publication or 50 years of creation, the employer would benefit from a longer copyright term than if the work were considered made for hire.[40] Finally, the work made for hire doctrine applies to all works universally, regardless of the type of work or the employee’s level of contribution. In other words, a work created and conceptualized entirely by only one employee is treated the same as a work created by hundreds of employees, each of whom contributed minor improvements and expression to the work.

B. Patent Law

Some of the benefits described in the previous section are mirrored by certain patent law doctrines—or are irrelevant with regard to patents—but others have no patent law equivalent. Section A described eleven benefits associated with copyright law’s work made for hire doctrine: (1) grant of title without negotiation;[41] (2) grant of title without transfer/assignment; (3) grant of title without court intervention; (4) registrability in the employer’s own name; (5) exclusion of others from using the work and licensing rights in the work to others; (6) use of the work; (7) transfer of title to others; (8) protection from future rights granted to work’s creator; (9) shield from transfer termination; (10) benefits with regard to copyright term duration; and (11) universal application regardless of level of contribution.[42]
Table 1: Work Made For Hire Benefits
No Negotiation
No Assignment from Employee
No Court Intervention
Registrability
Exclusion and Licensing
Use
Transfer to Others
Protection from Future Rights
Termination Shield
Duration
Universal Application
It is immediately apparent that the benefits of the termination shield and duration are not relevant to patent law because patents are granted for a set term of years—whether the base of 20 years or an extended term due to regulatory or PTO review[43]—not tied to the inventor’s lifetime. Therefore, there would be no special duration benefits should an inventor’s employer be considered an “inventor” for purposes of the Patent Act. In addition, unlike copyright transfers, which as discussed above may be terminated upon notice, patent assignments are not subject to termination.[44] The other benefits, however, require additional consideration. The three following sections describe modern patent law doctrines that give employers some, but not all, of the benefits of copyright law’s work made for hire doctrine.
1. Employer Use and Shop Rights
One of the benefits of the work made for hire doctrine is that an employer is guaranteed the right to make use of an employee’s creative work. Employers may receive a similar benefit from patent law’s shop rights doctrine.[45] However, because patent law describes negative rights, it is a protection against infringement actions by the employee patent holder and her assigns, and not an affirmative grant of use of the patented invention. Shop rights arise when an employee, “during his hours of employment, [and] working with his [employer’s] materials and appliances, conceives and perfects an invention for which he obtains a patent.”[46] The courts have held that in such circumstances, the employee by force of law must give his employer a nonexclusive, royalty-free right to practice the invention.[47] The doctrinal basis behind the shop right remains fuzzy,[48] as courts have described it being based on (a) a license implied in fact,[49] (b) estoppel,[50] and (c) equity and fairness.[51] Shop rights do not have the same scope as work made for hire, however. A shop right is not an ownership interest in the patent, which means that an employer does not have either the right to exclude others by threatening to sue for infringement or the right to file a patent application.[52] Similarly, a shop right is non-exclusive, which means that the employee patent holder is free to assign rights to others by granting them a patent license. In addition, a shop right is not transferable other than with the sale of the entire appurtenant business.[53]
Table2: Work Made For Hire v. Shop Rights
No Negotiation
No Assignment from Employee
No Court Intervention
Registrability
Exclusion and Licensing
Use
Transfer to Others
Protection from Future Rights
Termination Shield n/a
Duration n/a
Universal Application
2. Commissioned Invention and the Hired-to-Invent Doctrine
As mentioned in section A, under copyright law a work will be considered for hire if it was made by an employee within the scope of his or her employment. Similarly, patent law’s hired-to-invent doctrine grants an employer rights to the inventions of its employees if the employee was hired to invent them. Under this doctrine, an employee hired to solve a particular problem or to invent in a certain field will forfeit his patent rights even without a written contract.[54] Just as in the work made for hire context, courts have established a number of factors to indicate whether an inventor has been hired to invent.[55] However, unlike work made for hire, the hired-to-invent doctrine requires the employee to assign any patent obtained; it does not vest title immediately in the employer upon invention.[56] Although, failure to assign after a court order to do so may result in the order having the same effect as an assignment.[57] At first blush, this patent law doctrine would appear to accomplish much of what the copyright law work made for hire doctrine does. However, there are differences. First, work made for hire vests title in an employer immediately. The hired-to-invent doctrine merely obligates the inventor to assign the invention to his or her employer. True, no negotiation is required after the inventive activity occurs, but until an assignment is signed, the invention’s creator continues to hold the patent on the invention—albeit in trust for his employer. Furthermore, if the invention’s creator refuses to assign his patent rights, then a court order is required. Second, the doctrine is only effective against employees who are actually hired to invent. Employees who create an invention within the scope of their employment but who were not specifically directed to do so retain their patent rights.[58] Third, unlike the work made for hire doctrine, which permits an employer to file a copyright registration in its own name, the hired-to-invent doctrine still requires the patent application to be filed in the inventor’s name, even if the employer, as assignee, files a patent application on his behalf. Finally, as the employee is still considered the “inventor” under the Patent Act, it is possible for future rights to be given to the “inventor” that would not be automatically vested in the inventor’s employer. For example, in the copyright context, when the right to terminate transfers of ownership was introduced as a right of “authors,” it did not apply to those whose works were made for hire, as their employers were considered the “authors.” Similar rights could be granted to “inventors” that would affect the assignees of the inventors’ patent rights.
Table3: Work Made For Hire v. Hired-to-Invent
No Negotiation
No Assignment from Employee
No Court Intervention
Registrability
Exclusion and Licensing
Use
Transfer to Others
Protection from Future Rights
Termination Shield n/a
Duration n/a
Universal Application
3. Employer Inventions and Employee Improvements
In addition to the situation where an employer hires an employee to invent, an employer will also receive the benefit of insignificant improvements by his employees on inventions the employer has conceived of himself.[59] Under this doctrine, when an employer conceives of an invention, but receives ancillary suggestions or improvements from his employee, he retains all rights in the invention, including any employee improvements.[60] To qualify under the employee improvements doctrine, the employer must have had a “plan and preconceived design,”[61] and the improver must be an employee.[62] The employee improvements doctrine provides a minimal step toward the work made for hire doctrine, but does not come anywhere close to providing a substitute. First, the doctrine still requires there to be an inventor. All the doctrine does is subsume improvements by another person into the inventive exercise of the employing inventor; it does not operate to grant an employing corporation rights. In addition, the improvements that are granted must be ancillary, so any benefits from the doctrine will be minimal. Thus, if the employee creates a non-ancillary invention or improvement, the employer still needs to qualify under one of the previous doctrines or negotiate an assignment. At base, the employee improvements doctrine determines whether an employee should be considered a joint inventor or not, which does not come close to achieving the same benefits of the work made for hire doctrine.
Table4: Work Made For Hire v. Employee Improvements
No Negotiation
No Assignment from Employee
No Court Intervention
Registrability
Exclusion and Licensing
Use
Transfer to Others
Protection from Future Rights
Termination Shield n/a
Duration n/a
Universal Application
Given that patent law never developed a doctrine that defaulted patent rights to an employer the way copyright law did, one is left to wonder why. Why is it that copyright law in the nineteenth century began to apply an employer presumption? And why did the courts not feel the need to apply a similar presumption in patent cases? Part II discusses the development of the work made for hire doctrine in the nineteenth century, and Part III discusses potential reasons that a similar doctrine was not developed in patent law.

II. Development of the Work Made for Hire Doctrine

The work made for hire doctrine predated the Copyright Act of 1976. In fact, although it was first codified in the Copyright Act of 1909,[63] Professor Catherine L. Fisk has claimed that the doctrine “neither was invented by the drafters of the 1909 Act, nor was it well recognized in the cases before 1909.”[64] Instead, between the American Civil War and the passage of the Copyright Act of 1909 the default rule in copyright cases quietly switched from employee ownership to employer ownership. Professor Fisk argues that this move was justified by “cloaking” artists in an “aura of . . . romantic genius” to argue for greater legal protection, while simultaneously “downplaying or ignoring individual creative genius so as to assert corporate ownership over those copyrighted works.”[65]

A. Antebellum

Prior to the American Civil War, the default—and essentially the rule—was that the employee author was the owner of his works.[66] In the early days of copyright, protection was of limited application. The first copyright act—the Copyright Act of 1790—only protected maps, charts, and books.[67] Among the books that received copyright protection were the books in which court decisions were reported. Thus, it was only natural for the first copyright cases concerning employment to involve books containing those decisions and the authors that authored them. In the first case involving an employed author, the author of twelve volumes reporting the caselaw of the United States Supreme Court assigned his copyrights to a publisher. The publisher then sued a subsequent case reporter for copyright infringement because he had produced a volume titled “Condensed Reports of Cases in the Supreme Court of the United States,” which included the work of the prior reporter.[68] The Supreme Court, noting that the original reporter had failed to satisfy the formality requirements, held that it could not confer common law copyright on the initial reporter. However, what is telling about the opinion is that it appears to assume that the original reporter—and not the publisher—would be entitled to the copyrights in his work, even though his status as an employee was uncertain at best.[69] Later cases in the antebellum era involved school books,[70] theatrical works,[71] and cartography.[72] What is particularly interesting about this period of time is the courts’ clear pro-employee stance. In later decades, courts’ rhetoric would be inconsistent with the holdings of their decisions. While courts would claim to celebrate the genius and romanticism of the independent author, they would simultaneously grant first publishing houses and later employers the rights to that labor. During the antebellum nineteenth century, however, courts’ rhetoric praising the individual was consistent with the holdings of their decisions, which largely gave employees ownership of the copyrights in their works. The only cases where employers were held to have been granted copyrights in their employees’ works involved expressed contracts and cartography which, by its very nature, requires coordination among various individuals. True, smaller scale maps could be created by one person on their own, but for anything of a greater scale, coordination among various parties would be required.

B. The Nineteenth Century Postbellum

In the period during and after the American Civil War, courts began to hold that employers had been granted copyrights in their employees’ works, not by operation of contract, but based on the employment relationship. Particularly important to the discussion in Part III is that while cases originally held that the employment relationship granted an employer rights in his employees’ work by virtue of the involvement of a corporate president, later cases created the legal fiction of the company as author.[73] In the postbellum period, like the antebellum period, the relevant cases generally involved legal publications,[74] as well as theatrical works.[75] One of the theatrical work cases is particularly important for the purposes of this article. In Keene v. Wheatley, the court ruled against the plaintiff-employer on many of her arguments, but with regard to ownership of one of her employee’s works, the court stated:
Here, [the employee], while in the general theatrical employment of the [employer], engaged in the particular office of assisting in the adaptation of this play; and made the additions in question in the course of his willing performance of this duty. [The employer] consequently became the proprietor of them as products of his intellectual exertion in a particular service in her employment.[76]
For this proposition, the court did not rely on copyright law. Instead, the court looked to patent law, and determined that since “[w]here an inventor, in the course of his experimental essays, employs an assistant who suggests, and adapts, a subordinate improvement, it is, in law, an incident, or part, of the employer’s main invention,” the plaintiff in Keene was entitled to the “literary proprietorship” of her employee’s work.[77] In the eyes of the Keene court, the employee was merely adding minor enhancements akin to the patent doctrine of employee improvement. Yet, over time the doctrine of work made for hire found increasing judicial support and eventually became a default rule in favor of employers.[78] Over the next several years, this new doctrine was mentioned in dicta, but in no case did a court rule that an employer possessed rights over its employee by virtue of the default rule.[79] Instead, the presumption was switched surreptitiously. In Callaghan v. Myers, for example, while holding that a court reporter-employee held the copyrights in his work instead of his government employer, the Supreme Court determined that this was the case only due to “a tacit assent by the government to his exercising such privilege,”[80] This shift in emphasis made employer ownership a presumption that was only rebutted by tacit agreements and expressed contracts. Why did the courts make this shift? Professor Fisk posits three potential factors:
[1] Courts might have felt that a default rule of employer ownership was more likely to reflect the intent of most parties and wanted to save the parties the trouble of negotiating for employer ownership. . . . [2] [C]ourts might have begun to see employers as possessing a stronger moral claim and believed that any employee who planned to assert copyright ownership ought to be forced to disclose that intent and negotiate for it. . . . [3] [A]s changing assumptions about the nature of authorship strengthened the rhetorical force of the employer’s claim, a default rule of employer ownership might have seemed more intrinsically appealing, irrespective of whether the parties might negotiate around it.[81]
In response to the shifting sands of copyright law doctrine of the nineteenth century, one might expect that parties would have begun contracting around the default rules, but two factors may have made contracting prohibitively difficult:
First, the costs of transacting might be high when the parties have to discuss something as touchy as authorship. Employers might have been afraid to alienate employees by demanding assignment of the copyright, preferring to run the risk of litigation later. Employees may have lacked legal sophistication to realize that it was necessary to contract for copyright ownership. Second, the instability of the law may have made enforcement of any contract they did reach highly uncertain.[82]
In any case, by the nineteenth century, the work made for hire doctrine was in the wind, whether because of changes in the parties themselves, or merely the impressions of those sitting on the bench.

C. Corporate Ownership and Codification

The work made for hire doctrine was codified by Congress in 1909, but its foundations had existed prior to that time. As described in section B, the concept of authorship and ownership of copyright were shifting over the last half of the nineteenth century. However, it was not until 1885 that a court held that a corporation, and not an individual employer, held the copyrights in a work.[83] It was at this time that the legal fiction arose that a corporation could be an author. This idea makes sense when viewed one way: corporations are merely collections of individuals working toward, generally, hierarchically determined joint goals. However, it flies in the face of the entire notion of the individual romantic author. The concept behind the expression embodied in a copyrightable work can be created by any number of different people as it moves through the corporate process. This is particularly clear today when one looks at the large corporations behind music, television, and film.[84] These corporations are made up of hundreds of thousands of employees, and even more independent contractors. To produce one blockbuster film may require hiring and coordinating thousands of individuals, each of whom may contribute to a portion of the film. Without the work made for hire doctrine, these companies would have to rely on a messy web of contracts and the employee improvements doctrine, and even then, it is unlikely that they would be assured that all the relevant rights had been secured. By 1899, however, the courts had come to realize the need for a doctrine that allowed corporate employers to control the copyrights in the works of their employees by default. In Collier Engineer Co. v. United Correspondence Schools, a salaried employee hired “to compile, prepare, and revise . . . instruction and question papers” had moved to a new employer and prepared similar materials.[85] His former employer made a motion for a preliminary injunction, which the court denied, but in so doing determined that the employer was entitled to any copyright in the original materials.[86] This idea of protecting employers when their employees move to other employers will resurface in Part III as a justification for employer ownership of patent rights. Later cases embraced the idea that when an employer hired individuals to produce copyrightable works, the copyrights in those works were granted to the employer.[87] In this way, by the turn of the century, copyright law and patent law had developed fairly parallel doctrines with regard to employee works and inventions. In 1909, however, the Copyright Act was revised to codify the work made for hire doctrine by adding the following language:
[T]he word ‘author’ shall include an employer in the case of works made for hire.[88]
By including employers of those that create works made for hire within the definition of the term “author,” Congress continued in the direction the courts had been heading, but went farther by creating an explicit employer presumption. Professor Fisk identifies three reasons for this change: (1) an ease in drafting the Act, (2) avoiding constitutionality challenges, and (3) ensuring that copyright ownership would vest initially in employers so they could benefit from copyright renewals.[89] And that, ladies and gentlemen, was history. Since 1909, work made for hire has remained a mainstay of American copyright law. Yet, despite over one hundred years of the work made for hire doctrine in copyright law, patent law remains based on the rule that only the individuals that create an invention can be considered inventors. Moreover, for an employer to claim patent rights, he must either negotiate an assignment, or rely on one of the doctrines discussed in Part I.B.: shop rights, hired-to-invent, or employee improvements. Barring one of these options, an employer holds no rights in a patent absent a written assignment. Part III will explore the patent cases of the late nineteenth century, and posit potential reasons that patent law did not take a similar route.

III. Employment in Patent Law

As previously described, patent law never developed a doctrine that provided all of the benefits of copyright law’s now well-established work made for hire doctrine. This divergence between the two disciplines can be explained by two factors. First, inventive activity in the nineteenth century was perceived to be individualized and not to require significant coordination among individuals. Second, patent law developed the three doctrines described in Part I.B., which provided satisfactory protection to employers who wanted to continue using their employees’ inventions.

A. Antebellum

In order to understand the nature of inventive activity during the nineteenth century, one must understand that fundamental shifts were occurring in society at the time that changed both who the inventors were and what they were inventing. The United States has always been a country of inventors. In fact, both George Washington and Thomas Jefferson are credited with numerous inventions, including plows, adjustable writing desks and swivel chairs.[90] However, at the end of eighteenth century, there were “probably no more than 300 people who we would now class as scientists in the entire world,”[91] By 1800, there were about a thousand, and by 1900 there were approximately 100,000.[92] Due to the growth in the number of scientists, science began to shift from “a gentlemanly hobby, where the interests and abilities of a single individual [could] have a profound impact, to a well-populated profession, where progress depend[ed] on the work of many individuals who [were], to some extent, interchangeable.”[93] Similarly, the number of non-scientist inventors blossomed throughout the nineteenth century leading to an increased ability to appreciate the “interplay between science and technology, particularly in the fields of electricity and engine building,” which “led to a host of new practical machines that changed communications, transportation, the home, the workplace, and the farm,”[94] For example, at the beginning of the nineteenth century, America remained a nation of farmers who used the same hand implements that had been used for centuries to harvest their crops and who would travel primarily by horse or boat.[95] In the time before the Civil War, however, non-scientist inventors made many advances, including the invention of the steamboat, farming machinery, the telegraph, and synthetic materials.[96]  Such advances were only possible due to the increased interplay between science and technology that continued to grow throughout the nineteenth century.[97] These developments also led to societal changes, including increased prosperity and the end of slavery as new equipment was developed that could perform the work better and more cheaply.[98] Despite the democratization of science and invention in the nineteenth century, however, inventive activity continued to be considered the work of individuals. In fact, there are “no reported cases before 1843 in which an employer claimed, as against an employee, ownership of a patent because the inventor had been working for him at the time of the invention,”[99] Thus, in the nineteenth century, the general perception was that inventorship was the work of certain individuals, who were considered the great men of the time. One visual example of this mentality is Christian Schussele’s 1857 painting titled Men of Progress, which depicts Benjamin Franklin overlooking nineteen nineteenth century inventors:[100]
Figure 2
Figure 2: Men of Progress
Despite this public perception, however, the reality was that even these great men had assistance in creating their famous inventions. Among those depicted in Schussele’s painting is Charles Goodyear, who is remembered for his invention of vulcanized rubber in 1839.[101] He received a patent on his process in 1844,[102] and almost immediately became embroiled in patent litigation.[103] Not only was Goodyear forced to litigate against his competitors, however, but also against his own employees, one of whom attempted to patent one of Goodyear’s inventions.[104] Fortunately for Goodyear, despite recognizing a presumption in favor of the employee—who actually made the machine—the courts considered Goodyear’s employee’s activities to have been at Goodyear’s direction, and thus, they determined that it was Goodyear, and not his employee, that was the true inventor.[105] Despite the favorable result for Goodyear, the case only underscored the presumption that employees owned their inventions, as the court continued to articulate employee-ownership as the general presumption in such cases. Thus, time and time again, courts were required to find exceptions in order to protect employers from the presumption’s consequences. For example, in 1843 the Supreme Court, in McClurg v. Kingsland, determined that a license or grant from the employee to the employer may be presumed by virtue of the fact that the employee:
was employed by the defendants . . . receiving wages from them by the week; while so employed, he claimed to have invented the improvement patented, and after unsuccessful experiments made a successful one . . . the experiments were made in the [employer’s] foundry, and wholly at their expense, while [the employee] was receiving his wages, which were increased on account of the useful result.[106]
This implied license, or shop right, prevented the assignees of the employee’s patent from bringing suit against the former employer for infringement when all the employer was doing was continuing to use the process it had been using prior to the employee’s departure.[107] The Court reached this result without requiring an assignment of the patent. Rather, the Court held that the plaintiffs took their assignment from the employee subject to his license to his former employer.[108] As such, McClurg is generally thought of as the first shop rights case.[109] It can also be seen as a fairly pro-employee case, for the employer received no rights to the invention other than the right to use it despite the fact that the employee would not have been able to make his improvement but for his employer’s time, money and equipment.[110] During this same time period, other inventors were creating major innovations in both science and industry. For example, in 1844 Samuel Morse—also depicted in Men of Progress—connected the first intercity line between Washington and Baltimore and sent the message, “What hath God wrought!”[111] Moreover, scientists were making significant advances in electric theory, evolution, and thermodynamics.[112] Similarly, in the decade that followed, solo inventors continued to be credited with major breakthroughs in both science and industry. For example, during the 1850s, Heinrich Geissler is credited with the leaps forward that resulted in the invention of the vacuum tube.[113] Further, the 1850s saw huge advances in telecommunications technology, including Morse’s testing of the first transatlantic telegraph cable.[114] Even Abraham Lincoln took the time to consider the discoveries, inventions and improvements that resulted from America’s willingness to observe, reflect and experiment.[115] Despite the attribution of these major advances to individual inventors, the reality was that even in the 1840s, teams of workers were working together to develop new technologies, which “somewhat undermine[s] the heroic mythology surrounding the great individualistic and entrepreneurial American inventors,”[116] The dichotomy between the myth of the sole inventor and the reality that inventive activity was being carried out by groups of individuals in many ways mirrors the shift that was occurring in copyright law at that time. However, instead of acknowledging that multiple individuals were involved in the inventive process, courts turned a blind eye to the significant contributions employees were making to their employers’ inventions and dismissed those contributions as small modifications.[117] In King v. Gedney, for example, the court disregarded the following evidence as “vague and equivocal,” and instead determined that the employer “might have” given the employee—a draftsmen and general foreman—general directions about the improvement:
[The employee] was the foreman, and directed the making the alterations in the factory. [The employer] gave him no directions in relation to those alterations; [another employee] asked him for directions during the time that he was making alternations, and his answer was, ‘I do not know anything about it; go to [the employee].’ He said it was [the employee’s] improvement, and he knew nothing about it; if it did not work it was [the employee’s] fault; [the other employee] must go to him for all instructions. [The other employee] did not hear [the employer] give directions to any of the workmen in the shop. He has seen [the employer] and [the employee] conversing together at the factory, when chalk marks were made upon the floor by [the employee] to convey to [the employer] the manner in which he was going to prosecute the work.[118]
The court determined that if the general idea was the employing inventor’s, then the employee “could only be considered as acting as [his] servant,” and held the employer to be the original and sole inventor.[119] Thus, by crediting employers with being sole inventors, courts were able to continue to articulate the presumption that those who do the actual inventing will own the rights to their inventions. However, as employees became more and more involved with the inventive process, courts would find it increasingly difficult to rely on the employee improvements doctrine to rebut the presumption of employee ownership.[120]

B. The Nineteenth Century Postbellum

American invention did not stop during the American Civil War. In fact, the war spurred increased invention, particularly in agriculture. For example, Cyrus McCormick—another famed inventor depicted in Men of Progress—developed a reaper in 1861, which allowed northerners to increase food production “to the highest levels of surplus ever known” even though they had fewer farm hands due to the war.[121] Similarly, in the mid-1860s, scientists were making major advances in chemistry and thermodynamics.[122] Then, in 1865, George Westinghouse received his first patent for a rotary steam engine, followed by patents for railroad frogs and air brakes.[123] A few years later, a “self-taught practical innovator” by the name of Thomas Alva Edison would apply for his first patent.[124] Edison is a particularly interesting example of a nineteenth century inventor, because it is well documented that he “regularly employed not only mechanics and craftsmen but also college-trained chemists and engineers in his invention workshop,”[125] In some ways, then, it is not surprising that the Supreme Court first recognized the employee improvements doctrine in its 1868 Agawam Co. v. Jordan opinion. In Agawam, an inventor named John Goulding hired a blacksmith, Edward Winslow, to help in his machine shop.[126] Goulding had been working on his invention and was at the point of experimentation when Winslow, having visited another factory, made a suggestion of replacing a piece of Goulding’s invention with a spool and drum.[127]  Goulding went back and forth about the value of the suggestion, but after making another modification to make the change practicable, he adopted Winslow’s suggestion.[128] The new invention was patented and eventually assigned to the plaintiff who then sued the Agawam Woolen Company for infringement.[129] Agawam defended, in part, by arguing that it was Winslow, and not Goulding, that invented the improvements for which the patent was granted.[130] The Court held that Goulding’s claim to invention of the patent was sustained, because all that Agawam could prove was that Winslow suggested the spool and drum—an “auxiliary” part of the invention.[131] In handing another victory to an employer, however, the Court continued to recite that the general presumption is that employees hold rights to their inventions.[132] Similarly, as discussed above, during this time employers in copyright cases began to see the presumption switch from employee ownership to employer ownership. For example, five years prior to Agawam, the influential case of Kenne v. Wheatley was decided in favor of the employer relying on an employee improvements-type argument to grant a theater proprietor the copyrights to a play adapted by her and her theatrical company.[133] Whereas in theater it was undeniable that multiple individuals were required to develop new materials and produce it for the stage, invention in the 1870s continued to be perceived as the work of individuals with minimal help from employees instructed to carryout manual labor. Significant changes, however, were on the horizon for those industries that employed workers to design and construct their products. The “United States began to develop a single price and market system,” which led to capital intensive industries that “employed workers who owned no tools and had few skills, leading to social inequalities that created new political crises,”[134] Those inequities led to the formation of labor unions, which employed a variety of tactics to effect the status of workers in this country. “The decades from the 1870s through the early 20th century saw one disruptive strike after another and the beginning of the appeal of ideologies and reform strategies that would characterize the labor movement—agrarian reform, populism, socialism, anarchism, and progressive reformism.”[135] Despite these changes, inventive activity in the United States soldiered on. For example, Edison—who, as discussed above, employed a variety of workers in his invention workshop[136]—had filed for patents on 21 inventions by 1871.[137] As inventive activity began to involve more complex technology, the proportion of inventions devised by solo inventors continued to decline.[138] Perhaps in recognition of these changes, the Supreme Court began to hint at a willingness to recognize employers’ ownership of patented inventions that they hired employees to create.[139] Nevertheless, there were limits to the Court’s willingness to grant patent rights to one who employed another to invent. In Collar Co. v. Van Dusen, for example, the Supreme Court determined that the plaintiff, despite employing a set of paper-makers to create a paper collar that would be perceived as a real starched linen collar, was not entitled to a patent because only the paper-makers had invented something patentable.[140] The plaintiff had argued that the paper-makers had received inspiration and direction, but the Court determined that the paper-markers had provided the whole improvement and therefore the plaintiff had no rights to their invention under Agawam’s ancillary discoveries rule.[141] Despite its use of the terms “employer” and “employee,” however, the Collar case is better understood as concerning, using our modern parlance, independent contractors, who, even if Collar was a modern copyright case, would be unlikely to fall within the work made for hire doctrine. Rather than inspiring and directing the paper-makers’ improvement, the patentee derived all of the improvements he attempted to patent from the paper-makers. Similarly, other courts in the 1870s held that if an employee’s invention was outside the employee’s inventive arena, the employer would receive no more than a shop right to the invention.[142] By the 1880s, “farmers and industrial workers represented a force for political insurgency that began to coalesce into new movements that threatened to change the nature of society,”[143] These movements reflected a general anti-monopoly and big business sentiment at the turn of the century, as they began to lead to discussions of new government regulations “to control railroad rates and other large enterprises . . . to address the growing power of bankers and railroad magnates,”[144] At the same time, entrepreneur inventors, like Westinghouse, were beginning to build empires of their own from their own companies and inventions, as well as those of others.[145] These companies employed various individuals who developed patentable inventions during their employment. During the remainder of the nineteenth century, disputes over employees’ inventions became more frequent and led to cases that further developed the shop right[146] and employee improvements doctrines.[147] The courts, however, failed to develop a mechanism for dealing with employees that were explicitly hired to invent,[148] until the hired-to-invent doctrine was formally recognized by the Supreme Court in 1924.[149] Also, the rise of the corporate form introduced additional complications,[150] such as whether a shop right could be assigned.[151] Despite these developments, however, in not one of these cases was an employing company considered an “inventor” by virtue of its employees’ activity. Rather, for a company to receive patent ownership, an employee needed to assign it his or her rights. By the turn of the century, each and every person in the United States had been affected by the inventions of the nineteenth century. In particular, the poor were being put to work in factories using newly invented machinery and tools. Even the middle classes indirectly benefited from the benefits of the day:
For the middle classes, such as physicians, attorneys, engineers, academics, journalists, and managers, whose employment was less dependent on machinery and tools than on their education, the new technologies provided new sources of entertainment, comfort, and opportunities. By 1890, a person who worked in such a profession could come home, switch on the electric light, make a phone call, go to the refrigerator and take out a cold beer, light a factory-made cigarette with a safety match, then sit down to listen to a phonograph record while reading a newspaper with news of the day from all over the planet. It was a new world, for none of those experiences had been possible 40 years earlier, even to the wealthy.[152]

IV. Is Patent Law Stuck in the Past?

For inventors in the twentieth century, more and more inventive activity would occur within corporate settings. However, patentable inventive activity would continue to be credited to individuals, not groups, until late into the century. In fact, the average number of inventors per patent would not begin to change dramatically until the 1970s, when a “dramatic increase in the proportion of ‘highly collaborative’ inventions” would drive the average up:[153]
Chart 1
  Yet, despite the fact that today the “ordinary inventor is . . . a joint inventor who invents as part of a team,”[154] patent law continues to require that corporate employers receive assignments from their employees in order to effect a transfer of the ownership of the employees’ patent rights. In Part II, several reasons behind copyright law’s move toward the work made for hire doctrine were posited: (1) courts were attempting to reflect the intent of the parties; (2) courts felt that employees were in the best position to determine ex ante whether they intended to assert copyright ownership; (3) changing notions of the nature of authorship; (4) ease of statutory drafting; (5) avoiding constitutional concerns; and (6) concern about renewal terms. Three of these reasons simply do not apply to patent law. First, patents in the United States do not have renewal terms. Second, the copyright statute was being revised in 1909 for reasons beyond the codification of the work made for hire doctrine. The patent statute, by contrast, was not significantly revised until 1952. Without an independent push to redraft the Patent Act, the need to ease statutory drafting was absent, and constitutional concerns in granting patent rights to non-“inventors” would not need to be addressed. As for the remaining reasons for the shift in copyright law, patent law simply did not share those factors. First, the perceived nature of inventive activity in the nineteenth century did not lend itself to an argument for the need of a doctrine like work made for hire. As discussed in Part II, copyrighted works grew in complexity over the course of the nineteenth century. While copyrighted works were originally only embodied in single author books, they steadily grew to include complicated works involving the input of the multiple creative individuals. In particular, the creation of maps and theatrical works required coordination among multiple parties to insure that works could be exploited. Once the 1909 Act was passed, that complication grew exponentially as additional works were brought under the copyright statute. The perception of inventive activity in the nineteenth century, on the other hand, did not appear to require significant coordination among parties. Rather, credit for advances was given to individual inventors, whether or not they, like Edison, employed others in their invention workshops. Moreover, even in the nineteenth century, most patents were on incremental improvements of existing technologies, which made it even more likely that the number of individuals involved in each inventive push forward would be small. Further, even if an invention required multiple individuals to reduce it to practice, the individual credited with the invention could subsume any other inventive activity into his invention under the employee improvements doctrine.  In fact, in 1916, three-quarters of U.S. patents were issued to individuals.[155] Thu­­s, the need to coordinate among multiple individuals that existed in copyright law in the nineteenth century did not exist in patent law.[156] Second, the existing doctrines that had developed, including the hired-to-invent doctrine, provided sufficient ex ante protection to employers. The shop rights doctrine, which was well established in the early part of the century, provided employers with a way of ensuring that their expectations were met. Unlike today where many organizations that own patent rights merely license those rights to others to use, in the nineteenth century, most companies used patent rights merely to exclude competitors while they themselves developed products to sell to consumers. Without the need to license, companies merely needed the right to use, which the shop rights doctrine provided. Moreover, as described above, when an inventor did use assistance in the process of developing his invention, the employee improvements doctrine protected him from needing to secure rights from each assistant in his workshop. As long as the inventor couldn’t be seen as deriving his invention from someone else, he could be fairly content that his rights in the invention would be secure. Finally, by the end of the nineteenth century, employers were securing contracts with their employees that included provisions requiring the assignment of any patented inventions later developed.[157] In 1885, only 12% of U.S. patents were issued to corporations.[158] By 1950, it had grown to 75%, and by the 1980s, 84% of U.S. patents were assigned to corporations, “usually the employer of the actual inventor,”[159] In the past decade, those efforts have become increasingly successful.[160]
Chart 2
  In fact, it will come as no surprise that the Intellectual Property Owners Association’s list of top patent owners does not include the name of a single individual.[161] Further, by the turn of the century, when an employer hired an inventor, the hired-to-invent doctrine would require the inventor to assign their rights to the employer. Despite the protection of the existing patent law doctrines that give employers some of the benefits of the work made for hire doctrine, it is clear from Table 5 that not one doctrine captures all of the benefits from that doctrine:[162]
Table 5: Work Made For Hire Benefits v. Patent Doctrines
Shop Rights Hired-to-Invent Employee Improv.
No Negotiation
No Assignment from Employee
No Court Intervention
Registrability
Exclusion and Licensing
Use
Transfer to Others
Protection from Future Rights
Termination Shield n/a n/a n/a
Duration n/a n/a n/a
Universal Application
Moreover, as recent cases make clear, companies cannot rely on the patent law doctrines and contracts with their employees to ensure ownership of the rights to their employees’ inventions.[163] For example, the Supreme Court recently held that although an inventor working as a research fellow at Stanford University signed a Copyright and Patent Agreement that stated that he “‘agree[d] to assign’ to Stanford his ‘right, title and interest in’ inventions resulting from his employment at the University” and written assignments of rights, which Stanford used to file several patent applications, Stanford did not receive the inventor’s patent rights because he signed a Visitor’s Confidentiality Agreement that provided that he “‘will assign and do[es] hereby assign’ to Cetus his ‘right, title and interest in each of the ideas, inventions and improvements’ made ‘as a consequence of [his] access’ to Cetus” between signing the CPA and the assignments.[164] Further, many modern employers intend to license, rather than use, their employees’ inventions, which a shop right would not permit. Similarly, in Abbott Point of Care Inc. v. Epocal Inc., Abbott filed a patent infringement suit against Epocal, a competitor in the diagnostic field, alleging infringement of patents that “cover systems and devices for testing blood samples,”[165] Epocal, however, was founded by Dr. Imants Lauks, the named inventor of the patents at issue and a former employee of Abbott’s predecessors.[166] Despite the fact that Dr. Lauks had entered into two employment agreements, one of which contained a provision requiring him to assign all of his rights to any inventions that he might conceive, the Federal Circuit determined that Abbot did not hold rights to the patents and, thus, lacked standing to bring the lawsuit.[167] The court determined that Dr. Lauks had resigned from his previous employment in 1999, and that while the consulting agreement that he entered into after his resignation stated that various provisions of Dr. Lauks’ employment agreement remained in effect, it did not address invention assignments or obligations.[168] Thus, the two patent applications Dr. Lauks filed after entering into the consulting agreement did not belong to Abbot.[169] Additionally, despite the high assignment rate discussed above, when a patent assignment is found to be ineffective, the consequences can be extreme. This is particularly true where the employer acts on its belief that an assignment is effective and incurs costs associated with the patented technology, only to learn that they do not, in fact, hold any patent rights. As such, it is time for patent law to come to terms with the fact that it is stuck in the past and adopt an inventions made for hire doctrine. To do so, Congress need only pass an act amending two sections of the Patent Act.  First, Congress would add a subsection (c) to 35 U.S.C. § 111 as follows:
(c) In the case of an invention made for hire, the employer is considered the inventor for purposes of this title, and, unless the parties have expressly agreed otherwise in a written instrument signed by them, owns all patents rights to the invention.
Second, Congress would add a subsection (f) to 35 U.S.C. § 100 as follows:
(f) An “invention made for hire” is an invention invented or discovered by an employee within the scope of his or her employment.
Whether an individual is an employee for purposes of the invention made for hire doctrine will be determined by looking to the same common law doctrine of agency that the Supreme Court identified in Community for Creative Non-Violence v. Reid.[170] Moreover, there is over two decades of caselaw interpreting agency doctrine in light of the Copyright Act and over a century of caselaw interpreting the work made for hire doctrine for courts to rely on in implementing this new provision.[171]

Conclusion

As this article has suggested, patent law remains stuck in the past by continuing to rely on the rhetorical device that modern invention is performed by individual inventors in their garages, when, in fact, most patentable inventive activity occurs in corporate and university settings. Moreover, contrary to popular perception, most individuals who would be labeled “inventors” in the twenty-first century are employees of a corporate entity and assign their patent rights to their employer. Nevertheless, patent law continues to require written assignments to give employers ownership of their employees’ inventions. While that formulation may have made sense when invention was credited to individuals and required little coordination among inventors, it no longer does. Rather, employers should be able to feel secure in the knowledge that any inventions invented or discovered by their employees within the scope of their employment are owned by the party that financed and supported them (i.e., their employers). As such, it is time for the Patent Act to be amended to borrow from the Copyright Act and adopt an inventions made for hire doctrine.
* Mr. Simmons is an attorney in Kirkland & Ellis LLP’s New York office, specializing in intellectual property litigation, including copyright, patent and trademark. He received a B.A. from Brandeis University in 2006 and a J.D. from Columbia Law School in 2010, where he was awarded the Caroll G. Harper Prize for achievement in intellectual property. The author thanks Professor Harold Edgar, Henry Lebowitz and David Harris for their helpful suggestions for and reactions to this piece. The views, opinions, statements, analysis and information contained in this article are those of Mr. Simmons and do not reflect the views of Kirkland & Ellis LLP or any of its past, present or future clients.
[1] See infra Part IV.
[2] See Copyright Act of 1976, 17 U.S.C. § 201(b) (2006) (“In the case of a work made for hire, the employer or other person for whom the work was prepared is considered the author for purposes of this title, and, unless the parties have expressly agreed otherwise in a written instrument signed by them, owns all of the rights comprised in the copyright.”).
[3] See United States v. Dubilier Condenser Corp., 289 U.S. 178, 188 (1933) (noting that invention is “the product of original thought”); Solomons v. United States, 137 U.S. 342, 346 (1890) (“[W]hatever invention [an inventor] may thus conceive and perfect is his individual property.”); Gayler v. Wilder, 51 U.S. 477, 493 (1851) (“[T]he discoverer of a new and useful improvement is vested by law with an inchoate right to its exclusive use, which he may perfect and make absolute by proceeding in the manner which the law requires.”); 8 Chisum on Patents § 22.01 (2012) (“The presumptive owner of the property right in a patentable invention is the single human inventor . . . .”); see also Patent Act of 1952, 35 U.S.C. § 152 (2006) (“Patents may be granted to the assignee of the inventor of record in the Patent and Trademark Office, upon the application made and the specification sworn to by the inventor, except as otherwise provided in this title.”); id. § 261 (“Applications for patent, patents, or any interest therein, shall be assignable in law by an instrument in writing.”); Bd. of Trustees Trs. of Leland Stanford Junior Univ. v. Roche Molecular Sys., Inc., 131 S. Ct. 2188, 2192 (2011) (“Since 1790, the patent law has operated on the premise that rights in an invention belong to the inventor.”); Dubilier Condenser, 289 U.S. at 187 (“A patent is property and title to it can pass only by assignment.”); 8 Chisum on Patents § 22.01 (2012) (“The inventor . . . [may] transfer ownership interests by written assignment to anyone . . . .”). In fact, only the actual inventor is entitled to a patent, and only the inventor or someone he has assigned his patent rights to in writing may file a patent application, which in either case must be made on the inventor’s behalf. See 35 U.S.C. § 111(a)(2)(C) (“Such application shall include . . . an oath by the applicant as prescribed by section 115 of this title.”); id. § 115 (“The applicant shall make oath that he believes himself to be the original and first inventor of the process, machine, manufacture, or composition of matter, or improvement thereof, for which he solicits a patent . . . .”) (emphasis added); id. § 118 (“Whenever an inventor refuses to execute an application for patent, or cannot be found or reached after diligent effort, a person to whom the inventor has assigned or agreed in writing to assign the invention or who otherwise shows sufficient proprietary interest in the matter justifying such action, may make application for patent on behalf of and as agent for the inventor . . . and the Director may grant a patent to such inventor . . . .”) (emphasis added).
[4] Dubilier Condenser, 289 U.S. at 189; see also Stanford, 131 S. Ct. at 2196 (“We have rejected the idea that mere employment is sufficient to vest title to an employee’s invention in the employer.”).
[5] Certainly copyrights and patents share an affinity due to their finding support in the same clause of the Constitution. U.S. Const. art. I, § 8, cl. 8 (“The Congress shall have Power . . . [t]o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries . . . .”).
[6] As described below, in the nineteenth century, copyright law had a strong presumption of employee ownership, which slowly moved to an employer presumption that was finally codified in the twentieth century. See infra Part II. Patent law, on the other hand, developed various doctrines that permitted employers to be considered the owners of patent rights in certain circumstances, but not in others. See infra Part III.
[7] See infra Part IV.
[8] David F. Noble, America by Design: Science, Technology, and the Rise of Corporate Capitalism 87 (Alfred A. Knopf, Inc. 1977).
[9] Dennis Crouch & Jason Rantanen, Assignment of US Patents, Patent Law Blog (Patently-O) (Dec. 18, 2011, 9:07 PM), http://www.patentlyo.com/patent/2011/12/assignment-of-us-patents.html.
[10] Leahy-Smith American Invents Act, Pub. L. No. 112-29, § 4, 125 Stat. 284, 294 (2011) (“A substitute statement . . . is permitted with respect to any individual who—(A) is unable to file the oath or declaration . . . because the individual—(i) is deceased; (ii) is under legal incapacity; or (iii) cannot be found or reached after diligent effort; or (B) is under an obligation to assign the invention but has refused to make the oath or declaration required. . . .”) (emphasis added); see also Dennis Crouch, AIA Shifts USPTO Focus from Inventors to Patent owners, Patent Law Blog (Patently-O) (Aug. 14, 2012, 9:35 AM), http://www.patentlyo.com/patent/2012/08/aia-shifts-usptos-focus-from-inventors-to-patent-owners.html (describing the shift).
[11] See Stanford, 131 S. Ct. 2188 (2011) (holding that Stanford did not hold the rights to a patent that was allegedly infringed by Roche’s HIV test kits); Abbott Point of Care Inc. v. Epocal Inc., 666 F.3d 1299 (Fed. Cir. 2012) (affirming the dismissal of Abbott complaint for lack of standing to bring suit based on blood sample testing patents).
[12] Patent law and copyright law borrowing from each other is nothing new. The Supreme Court borrowed patent law’s construction of secondary liability to inform its copyright decision in Sony Corp. of America v. Universal City Studios. 464 U.S. 417, 439 (1984) (“There is no precedent in the law of copyright for the imposition of vicarious liability on such a theory. The closest analogy is provided by the patent law cases to which it is appropriate to refer because of the historic kinship between patent law and copyright law.”). That analysis was then expounded on in the copyright case of Metro-Goldwyn-Mayer Studios, Inc. v. Grokster, Ltd., 545 U.S. 913 (2005), and then the copyright focused construction was used as precedent in later patent cases.  See, e.g., Ricoh Co. v. Quanta Computer Inc., 550 F.3d 1325, 1336–40 (Fed. Cir. 2008) (citing Sony and Grokster to determine the standard for contributory infringement); DSU Med. Corp. v. JMS Co., 471 F.3d 1293, 1303–04 (Fed. Cir. 2006) (same).
[13] 3 R. Carl Moy, Moy’s Walker on Patents § 10:17 (4th ed. 2012).
[14] Copyright Act of 1976, 17 U.S.C. § 201(a) (2006) (“Copyright in a work protected under this title vests initially in the author or authors of the work. The authors of a joint work are coowners of copyright in the work.”).
[15] Oxford English Dictionary (2d ed. 1989).
[16] 17 U.S.C. § 201(b) (“In the case of a work made for hire, the employer or other person for whom the work was prepared is considered the author for purposes of this title, and, unless the parties have expressly agreed otherwise in a written instrument signed by them, owns all of the rights comprised in the copyright.”).
[17] Id. § 101 (“A ‘work made for hire’ is—
(1) a work prepared by an employee within the scope of his or her employment; or (2) a work specially ordered or commissioned for use as a contribution to a collective work, as a part of a motion picture or other audiovisual work, as a translation, as a supplementary work, as a compilation, as an instructional text, as a test, as answer material for a test, or as an atlas, if the parties expressly agree in a written instrument signed by them that the work shall be considered a work made for hire.”).
For example, comic book authors generally are required to sign agreements establishing that their work is commissioned as a work made for hire. See Joshua L. Simmons, Catwoman or the Kingpin: Potential Reasons Comic Book Publishers Do Not Enforce Their Copyrights Against Comic Book Infringers, 33 Colum. J.L. & Arts 267, 272 (2010) (describing comic book work made for hire agreements and caselaw).
[18] Cmty. for Creative Non-Violence v. Reid, 490 U.S. 730, 750–51 (1989) (holding the sculpture at issue was not a work made for hire, because the sculptor was an independent contractor and CCNV could not satisfy Section 101(2)).
[19] Id. at 751–52 (“In determining whether a hired party is an employee under the general common law of agency, we consider [1] the hiring party’s right to control the manner and means by which the product is accomplished.  Among the other factors relevant to this inquiry are [2] the skill required; [3] the source of the instrumentalities and tools; [4] the location of the work; [5] the duration of the relationship between the parties; [6] whether the hiring party has the right to assign additional projects to the hired party; [7] the extent of the hired party’s discretion over when and how long to work; [8] the method of payment; [9] the hired party’s role in hiring and paying assistants; [10] whether the work is part of the regular business of the hiring party; [11] whether the hiring party is in business; [12] the provision of employee benefits; [13] and the tax treatment of the hired party.”) (citations omitted).
[20] Id. at 752 (citing Ward v. Atl. Coast Line R.R., 362 US 396, 400 (1960); Hilton Int’l Co. v. NLRB, 690 F.2d 318, 321 (2d Cir. 1982)).
[21] Aymes v. Bonelli, 980 F.2d 857, 861 (2d Cir. 1992) (“Some factors, therefore, will often have little or no significance in determining whether a party is an independent contractor or an employee. In contrast, there are some factors that will be significant in virtually every situation. These include: (1) the hiring party’s right to control the manner and means of creation; (2) the skill required; (3) the provision of employee benefits; (4) the tax treatment of the hired party; and (5) whether the hiring party has the right to assign additional projects to the hired party.”).
[22] Id. (“These factors will almost always be relevant and should be given more weight in the analysis, because they will usually be highly probative of the true nature of the employment relationship.”).
[23] Under the Copyright Act of 1976, copyright vests “initially in the author or authors of the work,” 17 U.S.C. § 201(a) (2006). As “the employer or other person for whom the work was prepared is considered the author” when a work made for hire is involved, however, a work made for hire initially invests in the employer or commissioning party. Id. § 201(b).
[24] See Patent Act of 1952, 35 U.S.C. § 131 (2006).
[25] See Copyright Act of 1976, 17 U.S.C. § 102(a) (2006).
[26] See id. § 204(a).
[27] As described above, even works that are not created within the scope of an employee’s employment may be considered works made for hire. See supra note 17. However, works that are specially ordered or commissioned and considered works made for hire under Subsection 2 are outside the scope of this article.
[28] For example, in 2009, the founders of Skype, after selling the company to eBay, sued their former company and eBay for copyright infringement because, although eBay had purchased the good will in Skype, they had only licensed the source code to the program. See Complaint, Joltid Ltd. v. Skype Techs. S.A., 2009 WL 2958651 (N.D. Cal. Sept. 16, 2009) (No. 09 Civ. 4299); see also Brad Stone, Skype Founders File Copyright Suit Against eBay, N.Y. Times, Sept. 17, 2009, at B3, available at http://www.nytimes.com/2009/09/17/technology/ companies/17skype.html.
[29] See 17 U.S.C. § 408(a) (2006) (“[T]he owner of copyright or of any exclusive right in the work may obtain registration of the copyright claim . . . .”); id. at § 409 (“The application for copyright registration . . . shall include — (1) the name and address of the copyright claimant . . . (4) in the case of a work made for hire, a statement to this effect . . . .”).
[30] See id. § 411(a).
[31] See id. § 412.
[32] See id. § 106.
[33] See id. § 201(d).
[34] Id.
[35] Section 106A grants certain rights to “the author of a work of visual art,” See id. § 106A(a). However, section 101 defines a “work of visual art” as not including “any work made for hire,” Id. at § 101. In addition, in Dastar Corp. v. Twentieth Century Fox Film Corp., the Supreme Court held that trademark law cannot be used to require attribution “to the author of any idea, concept, or communication embodied in [tangible] goods,” 539 U.S. 23, 37 (2003).
[36] See 17 U.S.C. § 203(a) (stating that exclusive or nonexclusive transfers or licenses of copyrights to works created on or after January 1, 1978 and not works made for hire may be terminated); id. § 304(c) (noting that exclusive or nonexclusive transfers or licenses of the renewal copyrights to works within its first or renewal term on January 1, 1978 and not works made for hire may be terminated).
[37] Id. § 203; see Simmons, supra note 17, at 274–75 (describing cases of termination involving comic book talent); cf. James Grimmelmann, The Worst Part of Copyright: Termination of Transfers, The Laboratorium (Feb. 15, 2012, 10:23 AM), http://laboratorium.net/archive/2012/ 02/15/the_worst_part_of_copyright_termination_of_transfe (arguing that the inalienability created by the termination transfer provisions of the Copyright Act is unsupported by moral intuitions or logic).
[38] 17 U.S.C. § 302(a). A potential infringer may seek a certified report from the Copyright Office after 95 years for a published work or 120 years for an unpublished work. If the certified report does not disclose that the author of a work is alive or died less than 70 years prior to the report, the infringer is entitled to a presumption that shields her from infringement liability. Id. § 302(e).
[39] Id. § 302(c).
[40] On the other hand, a long-lived author could result in a significantly longer period of protection.
[41] By avoiding negotiation, parties also avoid the possibility that a court would interpret contract terms later in a way that the original parties to the agreement would not have anticipated. See, e.g., Marvel Characters, Inc. v. Simon, 310 F.3d 280 (2d Cir. 2002) (holding that although both parties had signed a settlement agreement stipulating that a work was made for hire, the provision was ineffective because it was signed subsequent to the work’s creation with purported retroactive effect).
[42] An additional societal benefit might include the idea that the work made for hire doctrine may permit and even incentivize the creation of works that would be impossible to create if the employing company had to negotiate with each employed author.
[43] Patent Act of 1952, 35 U.S.C. § 154(a)(2) (2006) (“[S]uch grant shall be for a term beginning on the date on which the patent issues and ending 20 years from the date on which the application for the patent was filed in the United States . . . .”). The Patent Act contains provisions providing for extensions to the patent term in certain circumstances. See id. § 154(b) (describing patent term adjustment for PTO delays); id. § 155 (describing patent term extension for Food and Drug Administration approval); id. § 156 (describing patent term extensions for other regulatory review).
[44] See id. § 261 (“Applications for patent, patents, or any interest therein, shall be assignable in law by an instrument in writing. The applicant, patentee, or his assigns or legal representatives may in like manner grant and convey an exclusive right under his application for patent, or patents, to the whole or any specified part of the United States.”).
[45] Kurtzon v. Sterling Indus., Inc., 228 F. Supp. 696, 697 (E.D. Pa. 1964) (“[A] ‘shop right’ or license to use, manufacture and sell.”).
[46] United States v. Dubilier Condenser Corp., 289 U.S. 178, 188 (1933) (holding that two employees of the radio section of the Bureau of Standards, who had received patents based on a “power amplifier” invention, were not required to assign their patents to the United States—their employer—instead the United States was entitled to a shop right, if anything); McClurg v. Kingsland, 42 U.S. 202 (1843) (affirming the Circuit Court’s jury instruction that a license between the patent holder and the employer-defendants could be presumed allowing the employer to continue to use the patented invention without liability, even if the patent was valid); Lariscey v. United States, 949 F.2d 1137, 1144 (Fed. Cir. 1991) (“A common law doctrine founded on equitable principles, the shop right rule allows an employer to use, without payment to the employee, an employee’s invention that was made using the employer’s time and/or materials, facilities, or equipment.”); see also Terry B. McDaniel, Shop Rights, Rights in Copyrights, Supersession of Prior Agreements, Modification of Agreement, Right of Assignment and Other Contracts, 14 AIPLA Q.J. 35, 38 (1986) (“The concept of shop right developed as a form of equitable compensation for situations where the employer has financed the employee’s invention, normally where the employee has used his employer’s time, facilities, and materials.”) (citation omitted).
[47] See McDaniel, supra note 46, at 36.
[48] Lariscey, 949 F. 2d at 1144 (“The shop right has variously been described as a form of implied license, founded on estoppel or acquiescence.”).
[49] See McClurg, 42 U.S. 202; see also Chisum,  supra note 3, at § 22.03[3][a] (“The actions of both the employee and the employer may warrant the assumption that the employee had consented to allow limited use of his invention in return for assistance in the development of the idea.”).
[50] See Gill v. United States, 160 U.S. 426 (1896); see also Chisum, supra note 3, at § 22.03[3][a] (“By participating in the installation of an embodiment of the invention in the employer’s business without demanding a royalty or other compensation, the employee is estopped from later demanding compensation or charging infringement after issuance of the patent.”).
[51] See Dubilier Condenser, 289 U.S. at 188; Beecroft & Blackman, Inc. v. Rooney, 268 F. 545 (S.D.N.Y. 1920); see also Chisum, supra note 3, at § 22.03[3][a] (“Because the employee has used employer resources to develop the invention, it is only fair and equitable that the employer have a limited right to use the invention. The party who invests in a project and risks the cost of failure should enjoy a return on his/her investment and should share in the benefits of success.”).
[52] See Dubilier Condenser, 289 U.S. at 188 (“[W]here a servant, during his hours of employment, working with his master’s materials and appliances, conceives and perfects an invention for which he obtains a patent, he must accord his master a non-exclusive right to practice the invention.”); Kurtzon, 228 F. Supp. at 697 (“[The employer] has the right to use the plaintiff’s invention but pays no royalties to the patentee. It has no property or title interest in the invention or the patent.  [It] has no exclusive contract with the [employee] that others shall not practice the invention. At most [it] has a ‘shop right’ or license to use, manufacture and sell the [product] which is not an exclusive contract and amounts to a bare license protecting [the employer] from a claim of infringement by [the employee]. . . . Such a licensee may neither sue alone nor join with the licensor patentee in an infringement action.”) (citations omitted).
[53] See Lane & Bodley Co. v. Locke, 150 U.S. 193 (1893) (holding that the Hapgood doctrine would not be extended to a company that “was organized upon the same basis as the [predecessor company]; that the business of the company was to be the same as that carried on by [the predecessor], and to be carried on in the same premises; that the entire property and assets of the firm and its liabilities and obligations were devolved upon the [successor] company”); Hapgood v. Hewitt, 119 U.S. 226 (1886) (holding that although after the original corporation entitled to a shop right dissolved, the stockholders had created a new corporation—its successor in interest—the shop right did not cover the new corporation).
[54] See Standard Parts Co. v. Peck, 264 U.S. 52 (1924) (holding that the invention’s creator held the legal title to the patented invention in trust for his former employer, and ordering him to assign the title in the patent to him employer’s successor); see also Houghton v. United States, 23 F.2d 386 (4th Cir. 1928) (“An employ[ee], performing all the duties assigned to him in his department of service, may exercise his inventive faculties in any direction he chooses, with the assurance that whatever invention he may thus conceive and perfect is his individual property. . . . But this general rule is subject to these limitations. If one is employed to devise or perfect an instrument, or a means for accomplishing a prescribed result, he cannot, after successfully accomplishing the work for which he was employed, plead title thereto as against his employer.  That which he has been employed and paid to accomplish becomes, when accomplished, the property of his employer.  Whatever rights as an individual he may have had in and to his inventive powers, and that which they are able to accomplish, he has sold in advance to his employer.” (citing Solomons v. United States, 137 U.S. 342, 346 (1890)); Nat’l Dev. Co. v. Gray, 55 N.E.2d 783, 787 (Mass. 1944) (“If the employee fails to reach his goal the loss falls upon the employer, but if he succeeds in accomplishing the prescribed result then the invention belongs to the employer even though the terms of employment contain no express provision dealing with the ownership of whatever inventions may be developed.”).
[55] Chisum, supra note 3, at § 22.03[2] (“The primary factor . . . is the specificity of the task assigned to the employee. . . . Evidentiary factors that carry some weight include (1) previous assignments of patents on other inventions by the employee; (2) a customary practice within the company for other similarly situated employees to assign; (3) whether the invention was conceived during the period of employment; (4) who originally posed the problem solved by the invention; (5) the employee’s authority within the company to determine to whom to give a problem for solution; (6) the relative importance of the idea to the employer’s business; (7) a previous inconsistent position on inventorship by the employer; (8) an agreement by the employer to pay royalties to the employee; (9) payment of patent procurement expenses by the employer or employee; and (10) the absence of initial interest by the employer when the employee first exposed the idea.”).
[56] See Standard Parts, 264 U.S. at 56–57 (Upholding the district court’s formal decree that the patentee, “within ten days from the date of the decree, assign and transfer to the company the legal title to the letters patent, and also transfer to it all other patents or pending applications for patents for inventions made by him, in connection with the processes and machinery developed in the performance of the agreement with the [company],” Furthermore, “if [the patentee] failed to perform the decree, ‘then and in that event’ the ‘decree [would] have the same force and effect as such assignments and transfers would have had if made.’”).
[57] Id. But see Leon Jaroff, Intellectual Chain Gang, Time Vol. 149, 64 (Feb. 10, 1997), available at http://www.time.com/time/magazine/ article/0,9171,985892,00.html (describing employee that chose to go to jail rather than to comply with a court order to assign his invention to his employer).
[58] See 1-5 Milgrim on Trade Secrets § 5.02[4][b] (“The general rule of law is that inventions made by an employee, although made during the hours of employment and with the use of his employer’s materials, facilities and personnel, are the employee’s property unless by the terms of his employment, or otherwise, he agreed to transfer the ownership (as distinguished from the use) of such inventions.”); see also United States v. Dubilier Condenser Corp., 289 U.S. 178, 187 (1933) (stating that inventions created by employees not hired to invent—even if the invention is within the same field as their employment—follow the general rule that title may only pass through negotiated assignments).
[59] Agawam Co. v. Jordan, 74 U.S. 583, 602 (1868) (“No one is entitled to a patent for that which he did not invent unless he can show a legal title to the same from the inventor or by operation of law; but where a person has discovered an improved principle in a machine, manufacture, or composition of matter, and employs other persons to assist him in carrying out that principle, and they, in the course of experiments arising from that employment, make valuable discoveries ancillary to the plan and preconceived design of the employer, such suggested improvements are in general to be regarded as the property of the party who discovered the original improved principle, and may be embodied in his patent as a part of his invention.”).
[60] Id. at 602 (“[W]here a person has discovered an improved principle in a machine, manufacture, or composition of matter, and employs other persons to assist him in carrying out that principle, and they, in the course of experiments arising from that employment, make valuable discoveries ancillary to the plan and preconceived design of the employer, such suggested improvements are in general to be regarded as the property of the party who discovered the original improved principle, and may be embodied in his patent as a part of his invention.”).
[61] Id.; see also Int’l Carrier Call & Television Corp. v. Radio Corp. of Am., 142 F.2d 493, 496 (2d Cir. 1944) (“An employer who seeks to patent the fruits of his employees’ labors must go further than merely to express a purpose to be realized.”).
[62] Agawam, 74 U.S. at 603 (“Persons employed, as much as employers, are entitled to their own independent inventions, but where the employer has conceived the plan of an invention and is engaged in experiments to perfect it, no suggestions from an employee, not amounting to a new method or arrangement, which, in itself is a complete invention, is sufficient to deprive the employer of the exclusive property in the perfected improvement.”); Larson v. Crowther, 26 F.2d 780, 790 (8th Cir. 1928) (“Where a party has discovered an improved principle in a manufacture, and employs another to assist him in carrying out that principle, and the employee makes valuable additions to the preconceived design of the employer, such improvements are, in general, to be regarded as the property of the employer, and may be embodied in his patent as part of his invention,” (citations omitted)).
[63] See Copyright Act of 1909, ch. 320, § 62, 35 Stat. 1075 (amended 1976) (“[T]he word “author” shall include an employer in the case of works made for hire.”).
[64] Catherine L. Fisk, Authors At Work: The Origins of the Work-For-Hire Doctrine, 15 Yale J.L. & Hum. 1, 5 (2003).
[65] Id. at 7 (“[S]ome courts did focus on the employer’s contribution or the employer’s right to control. But over time many came to rely simply on the legal fiction that employment renders the employer the author. The ‘crude legal fiction’ was not that the employer’s right to control made it the author; it was that the employer was the author.”).
[66] But see id. at 15 (discussing that despite courts’ statements to the contrary, studies have shown that “between forty-four and forty-nine percent of copyright registration between 1790 and 1800 were by a person other than the actual author.”) (citation omitted).
[67] Copyright Act of 1790, ch. 15 §1, 1 Stat. 124 (amended 1976). The first thing you notice in this group of works is that while books can certainly be created by an individual author, the most useful maps require coordinated effort by multiple individuals.
[68] See Wheaton v. Peters, 33 U.S. 591 (1834) (holding that the author of twelve volumes reporting the cases of the United States Supreme Court held copyright in his original elements added to the reports, although not the work of the justices themselves).
[69] See Fisk, supra note 64, at 16–18 (discussing early American legal publishing and the status of legal reporters, who were hired by the courts as independent contractors—although not called that—and encouraged to find publishers for their work by selling their copyrights). Fisk believes that the fact that early cases involved case reporters who, like Wheaton vis-à-vis the United States Supreme Court, had relationships with the judges likely deciding these cases, likely influenced their outcome. See id. at 21. The reporters were not traditional employees, but rather important and well-educated men who had agreed to perform a function that that was essential to the profession. See id.
[70] See Pierpont v. Fowle, 19 F. Cas. 652 (C.C.D. Mass. 1846) (holding that the renewal rights in a copyrighted work are not included in an assignment of the initial copyrights).
[71] See Atwill v. Ferrett, 2 F. Cas. 195 (1846) (holding that an author commissioned to create an opera remained the author of the work, but that an involved theater manager retained copyright in the version performed in his theater).
[72] See Pennsylvania v. Desilver, 3 Phila. 31 (C.P. 1858) (holding that the state employing a coordinating cartographer held the copyrights in his work); Heine v. Appleton, 11 F. Cas. 1031 (C.C.S.D.N.Y. 1857) (holding that an artist that had participated in a government funded expedition could not enjoin publication of books containing his illustrations, because he had expressly agreed otherwise and he had been paid to alter the images for that very purpose).
[73] See Fisk, supra note 64, at 32 (“[C]ourts came to understand that a corporation—the quintessential “corporate” (as in collective) author—should own the rights to the work created by all the persons who worked for the corporation.”); see also infra Part II.C.
[74] See Lawrence v. Dana, 15 F. Cas. 26 (C.C.D. Mass. 1869) (holding that an employer’s use of an employee’s work on a legal treatise, which she had contracted not to make use of in future editions, is an infringement). In Lawrence, the court in dicta noted that the employee had not acquired “any right to demand a copyright” in his work, but rather his employer held that right. Id. at 51.
[75] In both Roberts v. Myers and Boucicault v. Fox, the author of a commissioned play sued the theater manager who commissioned the piece for having it performed after the author, who was also acting in the piece, had quit. Roberts v. Myers, 20 F. Cas. 898 (C.C.D. Mass. 1860) (finding the author was entitled to the copyright in the work, despite being commissioned to write the play); Boucicault v. Fox, 3 F. Cas. 977 (C.C.S.D.N.Y. 1862) (finding that only an express agreement to assign copyright in the work would be sufficient to transfer copyright ownership). In Keene v. Wheatley, the proprietor of a New York theater had been given the rights to perform Our American Cousin, and in the process, she and her company adapted the play. 14 F. Cas. 180 (C.C.E.D. Pa. 1861).
Figure 1
Figure 1: Our American Cousin Playbill
 
[76] Keene, 14 F. Cas. at 187.
[77] Id. Although the court ultimately held that the employee’s work was not copyrightable, the court did determine that equitable principles permitted the plaintiff to recover based on the employee’s giving his work to the defendants to add to their production. Id. at 188.
[78] Professor Fisk makes much of the fact that the doctrine “slipped into the cases without the usual adversary process,” because in both Keene and Lawrence, the employees did not lose the rights to their work. Fisk, supra note 64, at 43–44. In Keene, the employee had already assigned whatever he had to the theater producer defendants; if anyone was going to lose out it would have been them. Keene v. Wheatley, 14 F. Cas. 180 (C.C.E.D. Pa. 1861). In Lawrence, despite the court acknowledging a copyright interest in the employer, the parties’ express contract protected the employee. Lawrence v. Dana, 15 F. Cas. 26 (C.C.D. Mass. 1869).
[79] See Callaghan v. Myers, 128 U.S. 617 (1888); Sarony v. Burrow-Giles Lithographic Co., 17 F. 591 (C.C.S.D.N.Y. 1883); In re Gould & Co., 2 A. 886 (Conn. 1885).
[80] Callaghan, 128 U.S. at 647.
[81] Fisk, supra note 64, at 45 (emphasis added).
[82] Id. at 50 (citing Rochelle Cooper Dreyfuss, Collaborative Research: Conflicts on Authorship, Ownership, and Accountability, 53 Vand. L. Rev. 1161, 1172–79 (2000)). But see Press Publishing Co. v. Monroe, 73 F. 196 (2d Cir. 1896) (holding that a contract provided copyright to the author); Mallory v. Mackaye, 86 F. 122 (C.C.N.Y. 1898) (holding that a contract allocated copyright in a theater manager over the playwright); Carte v. Evans, 27 F. 861 (C.C.D. Mass. 1886).
[83] Schumacher v. Schwencke, 25 F. 466 (C.C.S.D.N.Y. 1885). But see Edward Thompson Co. v. Am. Law Book Co., 119 F. 217, 219 (C.C.S.D.N.Y. 1902) (“It is conceded that the question whether a corporation can be an author within the meaning of the copyright laws has never been decided.”).
[84] In Aalmuhammed v. Lee, the plaintiff was hired to work on, without a written agreement, and made “very extensive” contributions to the film Malcolm X.  202 F.3d 1227, 1229 (9th Cir. 2000). However, the Ninth Circuit held that in the same way that the person who controlled the film’s hue was not a joint author of the film, the plaintiff would not be considered one either. Id. at 1233. Instead, the “superintended” or “master mind” of the whole work—in this case Spike Lee, who had signed a work for hire agreement—would be considered the work’s author. Id. at 1233, 1235. To have held otherwise would have opened authors to claims by “research assistants, editors, and former spouses, lovers and friends,” not to mention dramaturgies, actors, cinematographers, film editors, art directors, and costume, makeup, set and sound designers. See id. at 1235–36.
[85] 94 F. 152, 153 (C.C.S.D.N.Y. 1899).
[86] Id. at 153 (“[U]nder his contract . . . the literary product of [the employee’s] work became the property of the [employer], which it was entitled to copyright, and which, when copyrighted, [the employee] would have no more right than any stranger to copy or reproduce.”). The court denied the motion because it was unclear whether the newly created pamphlets infringed because both sets of pamphlets were compilations. Id.
[87] See Edward Thompson, 119 F. at 219 (In response to the defendant’s demurrer that the plaintiff corporation was not the copyright holder, the court stated, “It sufficiently appears that complainant’s publication is the result of the intellectual labor of the editors and compilers employed by complainant.  It is unnecessary, as it might be impracticable, to set forth the names of the persons engaged in the preparation of the work.”);
[88] Copyright Act of 1909, ch. 320, § 62, 35 Stat. 1075 (amended 1976).
[89] Fisk, supra note 64, at 62 (“First, it was a matter of ease in statutory drafting (“author” is a term of art used throughout the statute). Second, it avoided constitutional doubts about a default rule of employer ownership stemming from the constitutional provision that Congress may give “authors” a copyright.  Third, and most importantly, the drafters of the revision wanted to be sure that the employer would be the initial copyright owner rather than an assignee, because only the initial owner is entitled to obtain a renewal.”) (citations omitted).
[90] Jon Noonan, Nineteenth-Century Inventors, at ix (Facts on File 1992).
[91] John Gribbin, The Scientists: A History of Science Told Through the Lives of Its Greatest Inventors 359–61 (Random House 2002) (citing Frank Greenaway, John Dalton and the Atom (Heinermann, London, 1966)).
[92] Id. “But remember that the whole population of Europe doubled, from about 100 million to about 200 million, between 1750 and 1850, and the population of Britain alone doubled between 1800 and 1850, from roughly 9 million to roughly 18 million. The number of scientists did increase as a proportion of the population, but not as dramatically as the figures for scientists alone suggested at first sight,” Id.
[93] Id. at 359.
[94] Rodney P. Carlisle, Inventions and Discoveries, Scientific American 224 (John Wiley & Sons, Inc. 2004).
[95] Noonan, supra note 90, at xi.
[96] Id. (“In 1830, there were only 23 miles of usable track in the United States, but that was soon to change. By 1850, there were 5,000 miles of track and 20 years later, 75,000 miles. Steamboats increased in size and number; new farming equipment filled the country with food; the telegraph lines grew alongside the expanding railroads; and a thousand uses were found for vulcanized rubber and other new materials.”).
[97] Carlisle, supra note 94, at 223 (“The machines were built first; then the principles that governed them were thought through. With the newly discovered principles in place, it was possible to build better machines.”).
[98] Noonan, supra note 90, at x–xi.
[99] Catherine L. Fisk, Removing the ‘Fuel of Interest’ from the ‘Fire of Genius’: Law and the Employee-Inventor, 1830–1930, 65 U. Chi. L. Rev. 1127, 1138 (1998); see also id. 1138 n.33 (explaining that it is “implausible that there were no such cases before 1843,” but there were fewer opinions written and reported at that time).
[100] Depicted left to right: William Morton; James Bogardus; Samuel Colt; Cyrus Hall McCormick; Joseph Saxton; Charles Goodyear; Peter Cooper; Jordan Mott; Joseph Henry; Eliphalet Nott; John Ericsson; Frederick Sickels; Samuel F. B. Morse; Henry Burden; Richard Hoe; Erastus Bigelow; Isaiah Jennings; Thomas Blanchard; Elias Howe.
[101] Noonan, supra note 90, at 33–34, 39 (“A sample of his rubber, composed of his most recent chemical formula, came in contact with a hot stove. The natural gum always softened in summer heat. With the extreme heat of the stove, the gum should have melted. Instead of softening, however, Goodyear’s newest rubber charred like leather. It could still stretch but it wasn’t soft or sticky! Goodyear noticed that the edge of his sample was perfectly cured and not charred. If he could find the correct temperature, he could make fully cured rubber.”).
[102] Id. at 36.
[103] Id.  (“Several companies honored licenses from Goodyear to make rubber products using his methods. Some did not. These other companies copied Goodyear’s process without paying royalty fees. Goodyear had to get involved in lawsuits to protect his rights.”).
[104] In Warner v. Goodyear, Charles Goodyear patented an invention and then Solomon C. Warner, who had been employed as a machinist for Goodyear, attempted to patent the same invention. 29 F. Cas. 256, 256 (C.C.D.C. 1846). The invention at issue was a combination for “manufacturing corrugated or shirred India-rubber goods” using “metallic calendar rollers an elastic endless apron and a stretching-frame,” Id.
[105] The patent commissioner refused Warner’s patent, and when he appealed, the court found that although an inference could be drawn for Warner because he made the machine, it was rebutted by his employment and that while still employed he “stood by and saw Mr. Goodyear apply for and obtain a patent for it without objection,” Id. at 257. Moreover, the court found a presumption in Goodyear’s favor:
from the fact that Warner made the machine for Goodyear at his request, for his benefit, and at his expense, [the presumption] is that it was made according to his directions; and the burden of proof is then on Warner to show that the machine was not according to his directions.
Id. The court went on to find that Goodyear had conducted a number of experiments and that Warner had merely reduced to practice Goodyear’s combination. Id. Furthermore, the court may have been swayed by some evidence of bad faith, as Wilson did not apply for a patent for 18 months, six months after Goodyear had obtained his patent, and had reached an agreement with another company to pay for the costs in getting the patent. Id. Cf. Collier Eng’r Co. v. United Correspondence Sch., 94 F. 152 (C.C.S.D.N.Y. 1899), discussed supra notes 85–86 and accompanying text.
[106] 42 U.S. 202, 205 (1843). Furthermore, the employee continued to work at the foundry making rollers, and although his employer declined his proposal to apply for a patent and have the employer purchase his rights,
he made no demand on theme for any compensation for using his improvement, nor gave them any notice not to use it, till, on some misunderstanding on another subject, he gave them such notice, about the time of his leaving their foundry, and after making the agreement with the plaintiffs, who owned [another] foundry . . . for an assignment to them of his right.
Id.
[107] Id. at 204. The patent had been granted on an “improvement in the mode of casting chilled rollers and other metallic cylinders and cones . . . [so that] iron rollers or cylinders could be so cast that when the metal was introduced into the mould it should cause a sw[i]rl or rotary motion, by which the flog or dross would be thrown into the centre instead of the surface of the cylinder,” Id. The improvement was to change the direction of a tube exiting a furnace and entering the cylindrical mold from a perpendicular position to an angular one. Id. at 204–05 (“The tube or tubes, or passages called gates, through which the metal to be conveyed into the moulds shall not enter the mould perpendicularly at the bottom, but slanting, or in a direction approaching to a tangent of the cylinder, or if the gates enter the moulds horizontally or nearly so, shall not enter in the direction of the axis of the cylinder, but in a tangent form, or inclining towards a tangent of the cylinder.”).
[108] Id. at 206.
[109] Interestingly, McClurg has reasserted itself in modern copyright law to establish the idea that legislative expansion of intellectual property rights does not conflict with the constitution’s requirement that copyrights and patents be only for “limited times,” See Golan v. Holder, 132 S. Ct. 873, 886–87 (2012). (“This Court again upheld Congress’ restoration of an invention to protected status in McClurg v. Kingsland . . . . There we enforced an 1839 amendment that recognized a patent on an invention despite its prior use by the inventor’s employer. Absent such dispensation, the employer’s use would have rendered the invention unpatentable, and therefore open to exploitation without the inventor’s leave.”) (citations omitted); Eldred v. Ashcroft, 537 U.S. 186, 187–88 (2003); Golan v. Ashcroft, 310 F. Supp. 2d 1215, 1219 (D. Colo. 2004) (“As explained in Eldred, the McClurg court determined that [the employee] was ‘unprotected under the law in force when the patent issued because he had allowed his [ex-]employer briefly to practice the invention before he obtained the patent. Only upon enactment, two years later, of an exemption for such allowances did the patent become valid, retroactive to the time it issued.’ The McClurg Court upheld retroactive application of the new amended laws ‘for though they may be retrospective in their operation, that is not a sound objection to their validity.’”) (citations omitted).
[110] Cf. Fuller & Johnson Mfg. Co. v. Bartlett, 31 N.W. 747, 752 (Wis. 1887) (“The mere fact that, in making the invention, an employe [sic] uses the materials of his employer, and is aided by the services and suggestions of his co-employes [sic] and employer in perfecting and bringing the same into successful use, is insufficient to preclude him from all right thereto as an inventor.”).
[111] Noonan, supra note 90, at 26.
[112] In 1841, James Joule produced papers on the relationship between electricity and head for the Philosophical Magazine and the Manchester Literary and Philosophical Society. Gribbin, supra note 91, at 384. Between 1839 and 1845, Darwin made significant advances in developing his theory of evolution. Id. at 349. In 1848 and 1849 respectively, Julius Robert von Mayer developed ideas about heat and energy into a discussion of the age of the Earth and the Sun, and William Thomson coined the term thermodynamics. Id. at 381, 383.
[113] First, Geissler invented a better kind of vacuum tube that used mercury to make airtight contact. Id. at 487. Then, he pushed the invention further and “sealed two electrodes into the evacuated glass vessel, creating a tube in which there was a permanent vacuum,” Id. at 488.
[114] Noonan, supra note 90, at 14–15, 26. In 1851, over 50 telegraph companies “started stretching wires,” and within 15 years, “almost half a million miles of lines in America, Europe, and Asia would be connected,” Id. at 23–24.
[115] Abraham Lincoln, Sole Hope of the Future (Feb. 11, 1859), in Mario M. Cuomo & Harold Holzer, Lincoln on Democracy 148, 150 (Fordham Univ. Press 2004).
[116] Gribbin, supra note 91, at 224.
[117] For example, in Wellman v. Blood, an employer attempted to patent an invention, and an interference party argued that he was not the inventor, but rather the inventor was a man who had created a drawing of the invention with a modification of his own. 29 F. Cas. 628, 629–31 (C.C.D.C. 1856) (involving improved machinery for “cleaning top cards of carding machines”). The court determined, however, that
[i]f the employer conceives the result embraced in the invention, or the general idea of a machine upon a particular principle, and in order to carry his conception into effect, it is necessary to employ manual dexterity, or even inventive skill, in the mechanical details and arrangements requisite for carrying out the original conception; in such cases the employer will be the inventor, and the servant will be a mere instrument through which he realizes his idea.
Id. at 631. It went on to decide that Wellman had conceived of the improvement first and “practically reduced it,” even though the final invention included the employee’s small modification. Id.
[118] King v. Gedney, 14 F. Cas. 526, 530 (C.C.D.C 1856) (involving improvement on washing machinery that included a four-way valve at the breach end of the tub). In King, the employer sought a patent on an improvement to a previously patented machine, his employee also sought a patent, and an interference was declared. Id. at 526. The patent commissioner decided invention priority in favor of the employee, and the employer appealed. Id. at 527.
[119] Id. at 530–31; see also Dental Vulcanite Co. v. Wetherbee, 7 F. Cas. 498, 502–03 (C.C.D. Mass. 1866); Goodyear v. Day, 10 F. Cas. 677, 677 (C.C.D.N.J. 1852); Sparkman v. Higgins, 22 F. Cas. 879, 880 (C.C.S.D.N.Y. 1846); Alden v. Dewey, 1 F. Cas. 329, 330 (C.C.D. Mass. 1840); Dixon v. Moyer, 7 F. Cas. 758, 759 (C.C.D. Pa. 1821). The court also repeated the rule stated in Wellman, supra note 117, but attributed it to Curtis on Patents. King, 14 F. Cas. at 530.
[120] Dwight B. Cheever, The Rights of Employer and Employee to Inventions Made by Either During the Relationship, 1 Mich. L. Rev. 384, 385 (1902–1903) (“[T]here is . . . no question for the courts where the employer does all the inventing and the employee does only work requiring ‘mechanical skill,” As soon, however, as the employee begins to think that his ‘mechanical skill’ has turned to inventive skill the trouble begins . . . .”).
[121] Noonan, supra note 90, at 48–49.
[122] Gribbin, supra note 91, at 382, 376–77. Moreover, in 1864, James Maxwell published his “tour de force paper, ‘A Dynamic Theory of the Electromagnetic Field,’ which summed up everything that it is possible to say about classical electricity and magnetism in a set of four equations, now known as Maxwell’s equations,” Id. at 431. These equations would be the basis for Albert Einstein’s work on the special theory of relativity. Id. at 435.
[123] Noonan, supra note 90, at 57–58, 65; see also Improvements in Rotary Steam-Engines, U.S. Patent No. 50,759 (issued Oct. 31, 1865) (see Figure 3). Railroad frogs are “used for keeping cars on the correct rails at intersections and track switches,” Noonan, supra note 90, at 57. Westinghouse’s invention was to use reversible cast steel frogs to prevent them from wearing out. Id.
Figure 3
Figure 3: U.S. Patent No. 50,759
 
[124] Carlisle, supra note 94, at 224; see also Improvement in Electrographic Vote-Recorder, U.S. Patent No. 90,646 (issued Jun. 1, 1869) (see Figure 4).
Figure 4
Figure 4: U.S. Patent No. 90,646
 
[125] Carlisle, supra note 94, at 224. Cf. Fisk, supra note 99, at 1139 (“In the late nineteenth century, research laboratories dedicated to invention and technological innovation—like Thomas Edison’s famed laboratory enclave at Menlo Park, New Jersey—were unusual”).
[126] 74 U.S. 583, 586 (1868). The invention in Agawam was a new mechanical combination for manufacturing wool and other fibrous materials using a carding machine that was fed a continuous roll. Id. at 584–86, 599–602.
[127] Id. at 587.
[128] Id.
[129] Id. at 588, 593. Interestingly, Goulding actually failed to apply for a renewal of his patent right due to “erroneous information given him by the Commissioner of Patents,” and the patent expired. Id. at 588, 594. Twenty-one years after the patent expired, Goulding persuaded Congress to pass a special act, authorizing the patents commissioner to entertain an application for extension, and the patent reissued. Id. at 588, 594; see also An Act for the Relief of John Goulding, ch. 88, 12 Stat. 904 (1862). In a previous case involving Goulding’s patent, the constitutionality of the grant was questioned, but the circuit court rejected the argument. Jordan v. Donson, 13 F. Cas. 1092, 1095–96 (C.C.E.D. Pa. 1870).
[130] Agawam, 74 U.S. at 589, 595.
[131] Agawam, 74 U.S. at 605–07. In so holding, the Court repeated over and over that while only the inventor is entitled to a patent, if he employed other persons to help and they made “valuable discoveries ancillary” to the employer’s plan, the employer who invented the original idea is regarded as the owner of them. Id. at 602–03.  Rather, the employee’s contribution would need to meet the derivation requirements to preempt the employer’s rights from vesting. Id. (citing various derivation cases, including Pitts v. Hall, 19 F. Cas. 754 (C.C.N.D.N.Y. 1851); Reed v. Cutter, 20 F. Cas. 435 (C.C.D. Mass. 1841); Alden v. Dewey, 1 F. Cas. 329 (C.C.D. Mass. 1840)).
[132] Agawam, 74 U.S. at 603 (“Persons employed . . . are entitled to their own independent inventions.”).
[133] See supra notes 75–79 and accompanying text.
[134] Carlisle, supra note 94, at 225.
[135] Id. at 226–27.
[136] See supra note 125 and accompanying text.
[137] Noonan, supra note 90, at 74. By 1872, “Edison had filed patents for 52 inventions, most of them on printing-telegraph improvements. In the next three years, Edison created inventions such as the electric pen and mimeograph for making copies of letters and the quadruplex telegraph. The quadruplex handled two messages simultaneously to and from each end of the telegraph line, making a total of four messages on a single wire,” Id.
[138] Fisk, supra note 99, at 1141 (citing B. Zorina Khan & Kenneth L. Sokoloff, “Schemes of Practical Utility”: Entrepreneurship and Innovation Among “Great Inventors” in the United States, 1790–1865, 53 J. Econ. Hist. 289, 295–301 (1993) and Naomi R. Lamoreaux & Kenneth L. Sokoloff, Inventors, Firms, and the Market for Technology in the Late Nineteenth and Early Twentieth Century United States, in Naomi R. Lamoreaux, et al., Learning by Doing in Firms, Markets, and Nations (University of Chicago Press 1999)).
[139] See United States v. Burns, 79 U.S. 246, 252 (1871) (hinting at the hired-to-invent doctrine in dicta).
[140] 90 U.S. 530, 536, 564 (1874). Andrew Evans had previously received a patent for an improvement in the creation of shirt collars, which turned out to be a flop. Id. The shirt collars were “made of parchment-paper and coated with varnish of bleached shellac,” which—according to the reporter—evidently did not look like linen, much less starched linen, became discolored over time, emitted an odor when wearers sweated, and was not amenable to the new fashion of turned down collars. Id. at 537. So Evans employed Crane & Co. to create “something which while it was paper and could be produced cheaply should yet have such a thickness, tenacity, pliability united with strength, and have moreover that polish of surface, and that exact bluish tint which is found in the best starched linen—as distinguished from yellowness and from dead white—which would deceive even critical observers who had no opportunity of judging otherwise than by eye,” Id. Crane & Co. after laborious effort were able to create just such a collar, and after it was created, Evans assigned his patent rights to the Union Paper Collar Company and they applied for and received a reissue of the patent. Id. at 537, 539. Collar Co. then brought suit against Van Dusen.
[141] Id. at 550, 564 (“[P]ersons employed, as much as employers, are entitled to their own independent inventions, and if the suggestions communicated constitute the whole substance of the improvement the rule is otherwise, and the patent, if granted to the employer, is invalid, because the real invention or discovery belongs to the person who made the suggestions.”). The court first noted that Evans likely knew nothing about the process of manufacturing paper and that the original patent did not describe the process added in the reissue. Id. at 561. It then found that Evans “did not communicate any information to [Crane & Co.] respecting the process by which such paper could be produced, nor did he give [them] any directions upon the subject,” because without knowledge of making paper he was unable to do so. Id. at 562.
[142] See Whiting v. Graves, 3 Ban. & A. 222 (C.C.D. Mass. 1878). In Whiting, an employer hired a machinist to maintain the machines in his new dry goods factors and to make any other machinery that would be necessary, but did not include a provision in his service contract related to patents. Id. at 223. Instead, the employer would have the machinist assign his inventions prior to application for patents on them, which worked fine for the employer’s dry goods machines, but then the machinist made an improvement on sewing-machines. Id. The court determined that while the employer would be afforded a shop right in the machinist’s inventions by virtue of his original contract, an additional assignment would be required for the employer to take legal title. Id. at 224. The court went on to hold, based on various testimony, that the employer was only entitled to a one-half interest in the patents, by virtue of an agreement to pay the patent expenses for one-half of the profits from the manufacture of the invention, and that the machinist was entitled to assign his interests to the defendants, which meant they could not infringe. Id. at 1061; see also Niagara Radiator Co. v. Meyers, 40 N.Y.S. 572, 573–76 (N.Y.S. Ct. 1896) (finding machine shop foreman not hired to invent); Connelly Mfg. Co v. Wattles, 23 A. 123, 124 (1891) (finding a shop foreman would not be hired to invent).
[143] Carlisle, supra note 94, at 226.
[144] Id.
[145] Noonan, supra note 90, at 60 (“By the 1880s, Westinghouse was involved in several enterprises. As well as companies and patents he originated on his own, he also started purchasing the companies and patents of others, too. He seemed to find opportunities everywhere.”).
[146] See Clark v. Fernoline Chem. Co., 5 N.Y.S. 190, 191 (N.Y. Sup. Ct. 1889) (In a case primarily concerned with a dispute over unpaid wages, the New York Superior Court held that although an employee had been hired as a chemical expert and chemist, his employer was only entitled to a shop right and not an assignment of any patents granted to the employee.); see also Gill v. United States, 160 U.S. 426 (1896) (holding that the patentee had acquiesced to the government’s use of his invention—a shop right); McAleer v. United States, 150 U.S. 424, 432 (1893) (finding an expressed contract permitting government use); Lane & Brodley Co. v. Locke, 150 U.S. 193, 196, 198–99 (1893) (finding implied license—or shop right—from inventor to partnership, which was incorporated while he was still working there); Solomons v. United States, 137 U.S. 342, 346–48 (1890) (finding an implied license permitting government use—a shop right). In Dalzell v. Dueber Watch Case Manufacturing Co., an individual, who had been employed at various times within a company as an electroplater, slider and toolmaker, invented various things and applied for and was granted patents on them. 149 U.S. 315, 315–19 (1893) (the inventions involved improvement in the making of watch case cores). The employee sued the employer fro infringement, and the employer responded by seeking specific performance of an oral agreement to assign the right to obtain patents on them. Id. at 317–20. The Court held that an oral agreement for the “assignment of the right to obtain a patent for an invention” was neither within the statute of frauds nor covered by the requirements that all patent assignments be in writing. Id. at 320. However, the Court determined that the testimony was in hopeless conflict and therefore refused to afford specific performance; thus, allowing the infringement action to proceed. Id. at 326. The Court did, however, leave open the possibility that a shop right had been granted. Id. (“It is proper to add that the question whether the [employer], by virtue of the relations and transactions between it and [the employee], had the right, as by an implied license, to use [the employee’s] patents in its establishment is not presented by either of these records, but may be raised in the further proceedings upon the bill against the [employer] for an infringement.”).
[147] See Eclipse Mfg. Co. v. Adkins, 44 F. 280, 281–82 (C.C.N.D. Ill. 1890) (determining that employer who came up with the idea of putting ornamenting on radiator pipes, but had his employees create the design and build the pipes, was entitled to his patent, as the patent was on the idea, not the specific ornamentation); Streat v. White, 35 F. 426, 426, 428 (C.C.S.D.N.Y. 1888) (involving a design patent for textiles intended to be employed to create imitation “seersucker” fabrics, the court determined that because the plaintiff only had the idea of imitation seersucker, which did not constitute a novel idea, and employed a cotton goods factory manager to produce it by whatever means were necessary, he could not patent the resulting design). In Streat, the factory manager and the designer “were furnished a sample of a seersucker, and with a photographic copy of the sample, and where told to imitate it, and that the way in which the imitation was to be effected was left with the designer, who was solely responsible for a successful result, and to whom the task of finding an idea or conception of the method of imitating the crinkle was solely committed,” Id. at 428; see also Streat v. Simpson, 53 F. 358, 359 (C.C.S.D.N.Y. 1893) (finding in a subsequent infringement suit that the plaintiff “had nothing to do about designing or engraving the tools for printing the imitation, but through [the factory manager]; and that the designer and engraver were controlled by anything but the sample and photograph given to them by [the factory manager] is not made to appear”).
[148] Compare Dalzell, 149 U.S. at 320 (“[A] manufacturing corporation which has employed a skilled workman, for a state compensation, to take charge of its works, and to devote his time and services to devising and making improvements in articles there manufactured, is not entitled to a conveyance of patents obtained for inventions made by him while so employed, in the absence of express agreement to that effect.”) with Solomons, 137 U.S. at 346 (1890) (stating in dicta that “[i]f one is employed to devise or perfect an instrument, or a means for accomplishing a prescribed result, he cannot, after successfully accomplishing the work for which he was employed, plead title thereto as against his employer. That which he has been employed and paid to accomplish becomes, when accomplished, the property of his employer. Whatever rights as an individual he may have had in and to his inventive powers, and that which they are able to accomplish, he has sold in advance to his employer.”); Annin v. Wren, 44 Hun. 352, 353 (N. Y. Sup. Ct. 2d Dept. 1887) (holding that equity required a skilled draughtsman—hired by an iron truck and wheelbarrow manufacturer to develop his business—to assign his patents on improvements to the manufacturer’s wheelbarrows and trucks to the manufacturer if they were based on their joint development, because it would be unreasonable to prevent an employer from using improvements made during the course of the draughtsman’s employment). The court’s decision in Annin may have been colored by the fact that the draughtsman assigned his patent application to his brother in order to deprive the manufacturer of it. Id. at 354.
[149] See Standard Parts Co. v. Peck, 264 U.S. 52, 58–60 (1924).
[150] See Lawrence M. Friedman, A History of American Law 446–59 (Simon & Schuster 1973) (discussing the right of the corporate form at the end of the nineteenth century); see also Christopher L. Tomlins, Law, Labor, and Ideology in the Early American Republic 259 (Cambridge Univ. Press 1993) (“By midcentury, on and off the farm, it seems likely that the proportion of productively engaged Americans employed by others—about one-third in 1820—had increased to about one-half.”).
[151] It cannot. See Hapgood v. Hewitt, 119 U.S. 226 (1886). In Hapgood, a mechanical expert had been hired by a corporation as vice president, and later superintendent, to create plows and other goods, and to supervise generally the manufacturing department. Id. at 229. While in the corporation’s employ, the expert was asked with creating an iron plow similar to the company’s existing wooden ones, and he did so within the corporation’s employ, during working hours, using employees and corporate materials. Id. at 230. While he was inventing these plows, he never claimed a property ownership, but when he left he made his own, patented the improvements, and sought money from the corporation for its use of the improvements. Id. at 232. The Court held that he was not required to assign his patent to the corporation, but the corporation had received a “naked license to make and sell the patented improvement,” Id. at 233. However, the corporation had, in the meantime, dissolved and the stockholders had created a new corporation, which was its successor in interest; the Court held that the new corporation did not benefit from the shop right. Id. at 232–34. Nevertheless, in Lane & Bodley Co. v. Locke, the Supreme Court determined that a shop right created during a partnership carried over after incorporation, because of the close affinity between the two corporate entities and because the inventor had ratified the transfer by acknowledging the incorporated entity was the successor to the partnership. 150 U.S. at 196.
[152] Carlisle, supra note 94, at 227.
[153] Dennis Crouch, The Changing Nature Inventing: Collaborative Inventing, Patent Law Blog (Patently-O) (Jul. 9, 2009, 9:28 AM), http://www.patentlyo.com/patent/2009/07/the-changing-nature-inventing-collaborative-inventing.html (“This data is derived from a sample of 750,000 patents issued from August 1971 through February 2009. All reported results are significant at the 99% confidence level.”); see also Dennis Crouch, Person(s) Skilled in the Art: Should the Now Established Model of Team-Based Inventing Impact the Obviousness Analysis?, Patent Law Blog (Patently-O) (May 17, 2011, 5:24 PM), http://www.patentlyo.com/patent/2011/05/persons-skilled-in-the-art-should-the-now-established-model-of-team-based-inventing-impact-the-obviousness-analysis.html [hereinafter Team-Based Inventing] (“In [1852], 82% of patents listed only one inventor and a mere 3% listed three or more inventors. By 2011, the statistics had inverted. Less than one-third (32%) of patents issued so far this year list just a single inventor and 43% identify three or more inventors. During this 60-year period, the average number of inventors per patent has more than doubled.”); Dennis Crouch, Cross-Border Inventors, Patent Law Blog (Patently-O) (Nov. 21, 2010, 8:06 PM), http://www.patentlyo.com/patent/2010/11/cross-border-inventors.html (“The number of inventors per patent has been steadily increasing over the past forty years. Patents issued during the past six months, have an average of 2.7 inventors per patent. In all, 68% of these patents list multiple inventors with 13% listing five or more inventors. Prior to 1990, most patents listed only one inventor.”).
[154] Team-Based Inventing, supra note 153.
[155] See Louis Galambes, The American Economy and the Reorganization of the Sources of Knowledge, in The Organization of Knowledge in Modern America, 1860–1920 269, 277 (Alexandra Oleson and John Voss eds., Johns Hopkins Univ. Press 1979).
[156] See Aalmuhammed v. Lee, 202 F. 3d 1227 (9th Cir. 2000), discussed supra note 84.
[157] Littlefield v. Perry, 88 U.S. 205 (1875) (finding agreement to assign future inventions enforceable); Hulse v. Bonsack Mach. Co., 65 F. 864, 867–68 (4th Cir. 1895) (finding pre-invention assignment contract enforceable); cf. Fisk, supra note 99, at 1191 (“At the end of the nineteenth century, the number of reported decisions enforcing employee patent assignments began steadily to increase.”). But cf. Aspinwall Mfg. Co. v. Gill, 32 F. 697, 700 (C.C.D.N.J. 1887) (“A naked assignment or agreement to assign, in gross, a man’s future labors as an author or inventor—in other words, a mortgage on a man’s brain, to bind all its future products—does not address itself favorably to our consideration.”); Eustis Mfg. Co. v. Eustis, 27 A. 439, 442–43 (N.J. Ch. 1893) (finding despite agreement that employee would give employing company “the benefit of any and all patents for cooking utensils made by or issued to him during the term of his office and employment,” company held only an exclusive license of the patents, not ownership).
[158] Noble, supra note 8.
[159] See id.; Rights of Employed Inventors: Hearing on H.R. 4732 and H.R. 6635 Before the H. Subcomm. on Courts, Civil Liberties, and the Administration of Justice, 97th Cong. 1 (1982) (remarks of Rep. Robert Kastenmeier quoting U.S. Patent Office report); see also William P. Hovell, Note, Patent Ownership: An Employer’s Rights to His Employee’s Inventions, 58 Notre Dame L. Rev. 863, 864 (1983); Robert L. Gullette, State Legislation Governing Ownership Rights in Inventions Under Employee Invention Agreements, 62 J. Pat. Off. Soc’y 732, 738–39 (1980).
[160] Crouch & Rantanen, supra note 9.
[161] Intellectual Property Owners Association, Top 300 Organizations Granted U.S. Patents in 2010, The IP Record, at 9–17 (2011).
[162] See supra Part I.B.
[163] See Stanford, 131 S. Ct. at 2192 (2011) (holding that Stanford did not hold the rights to a patent that was allegedly infringed by Roche’s HIV test kits); Abbott Point of Care Inc. v. Epocal Inc., 666 F.3d 1299, 1304 (Fed. Cir. 2012) (affirming the dismissal of Abbott complaint for lack of standing to bring suit based on blood sample testing patents); see also Banks v. Unisys Corp., 228 F.3d 1357, 1359 (Fed. Cir. 2000) (as employee did not sign employer’s “standard form for assignment of inventive rights,” fact issues precluded summary judgment determination of whether the hired-to-invent doctrine applied); Mallinckrodt Inc. v. Masimo Corp., 293 F. Supp. 2d 1102, 1106 (C.D. Cal. 2003) (employee’s release of ownership claims to product did not include release of ownership of ideas later patented by employee); cf. Teets v. Chromalloy Gas Turbine Corp., 83 F.3d 403, 409 (Fed. Cir. 1996) (reversing district court finding of employee ownership due to implied-in-fact contract to assign rights to employer). But see Preston v. Marathon Oil Co., 684 F.3d 1276, 1288–89 (Fed. Cir. 2012) (holding post-employment employee agreement that included automatic patent assignment provision with only continued employment as consideration was valid and enforceable).
[164] Stanford, 131 S. Ct. at 2192, 2199. The Supreme Court expressly did not address Roche’s “shop right” or license arguments. Id. at 2193 n.1.
[165] Abbott, 666 F.3d at 1300.
[166] Id.
[167] Id. at 1300, 1303–04.
[168] Id. at 1300.
[169] Id. at 1303.
[170] Cmty. For Creative Non-Violence v. Reid, 490 U.S. 730, 751–52 (1989).
[171] See supra Part I.A.

Conception and Misconception in Joint Inventorship

Conception and Misconception in Joint Inventorship
Aaron X. Fellmeth* A pdf version of this article may be downloaded here.  

Introduction

It is an axiom of patent law that utility patents protect “inventions” only; it is a paradox of patent law that the Patent Act does not define “invention” except tautologically: “The term ‘invention ’ means invention or discovery.”[1] Given that patents are designed to reward inventors for the disclosure of their inventions,[2] it would have been helpful had the Act settled just what constitutes an invention. Alas, it does not. After 220 years of progressively developed patent statutes and jurisprudence, the concept of invention, so central and fundamental to patent law, has become one of the most misunderstood. The Patent Act’s obscurity on this key point of law does not prevent the Act from employing the term “invention” ubiquitously. A person is entitled to a patent unless “the invention was known or used by others in this country” or the inventor “has abandoned the invention.”[3] Patentability may not be negated “by the manner in which the invention was made.”[4] A person infringes the patent by making or selling the “patented invention.”[5] Among the mysteries emanating from uncertainty about what kinds of technologies qualify as a patentable invention, identifying the proper inventors has developed into one of the most intractable. Because the U.S. Constitution allows only “inventors” to obtain patents on their inventions,[6] we must know what an invention is before we can decide who qualifies as its inventor. The inventor himself has a self-evident interest in certainty about his or her entitlement to a patent as well. In 2011, Congress supplemented the definitions section of the Patent Act through the Leahy-Smith America Invents Act (“AIA”), to take effect in March 2013, which defines the term “inventor” for the first time. However, that definition merely adds another point of confusion; it provides, “The term ‘inventor’ means the indiv­idual or, if a joint invention, the individuals collectively who invented or dis­cover­ed the subject matter of the invention.”[7] A definition stating that an inventor is the person who invents the invention would offer us nothing more than another tautology. However, the new amendment does not exactly do that; instead, it defines as the “inventor” the person who invents the “subject matter” of the invention.[8] It is not immediately clear how the subject matter of an invention differs meaningfully from the invention itself. Instead of resolving a difficult and core problem of law, the new definition merely adds a point of obscurity. The intuitive way to begin resolving this paradox is to define as an “invention” whatever satisfies the legal requirements for the grant of a utility patent. The Patent Act requires that, to qualify for a utility patent, an invention must be “novel,” “non-obvious” at the time of conception (after the AIA, the “effective filing date”), and “useful.”[9] But the Act and its implementing regulations also require the fulfillment of several more conditions for a patent to issue, including a written disclosure of the invention, an oath or declaration by the applicant, and the payment of a fee.[10] Surely, not all of these requirements speak to the definition of an invention. We can hardly say that a person has failed to invent something until he has signed an oath and paid a fee to the U.S. Patent and Trademark Office.[11] At the application stage, the doctrinal analysis for invention and inventorship is deceptively straightforward. If the application satisfies the requirements of the Patent Act and regulations, the invention will be considered whatever is described in the application, and the question of inventorship will be resolved by presumption. The PTO will not examine inventorship, but will usually accept without question the claims of the applicants that they originated the invention.[12] The trouble typically arises when the patent is enforced or challenged in litigation.[13] In the disputing context, the definition of the invention does not normally assume a pivotal role because the focus of litigation tends toward the “claims.” Claims are statements in the patent “particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.”[14] To infringe a utility patent, the accused infringer must generally have made, used, sold, or imported a product described in one or more of the patent’s claims. If the defendant challenges the validity of the patent, the strategy is to prove the invalidity of all allegedly infringed claims. Courts have similarly gravitated toward claims analysis to determine whether a member of a research team qualifies as an inventor. As consistently invoked by both the Supreme Court and the U.S. Court of Appeals for the Federal Circuit, the accepted patent law doctrine treats claims as if they were equivalent to the invention itself.[15] Yet, in inventorship disputes, the case for fixating on the claims is based more on pedigree than logic or policy. Claims are nearly always drafted by lawyers or patent agents after (sometimes years after) the invention was completed.[16] The claims may or may not express what the inventor conceived; they rarely express the full scope of the invention, and not infrequently capture more than what the inventor invented.[17] Despite the convenience of equating the claims with invention in infringement analysis, judicial opinions equating the two have propelled patent law down a treacherous path. The consequences of this wayward doctrine reach into the very processes of innovation. Most modern technology research is the product of collaborations between multiple scientists and engineers, usually in large research institutions.[18] Collaborations frequently involve teams from industry and universities or government agencies joining forces to advance technology despite divergent economic interests. Joint inventorship has accordingly come to assume a critical role in technological research and development. Approximately half of all patents now list more than one inventor.[19] As Congress has long recognized,[20] the patent system best achieves its goal of encouraging technological development by facilitating rather than hindering joint invention. To achieve this goal, inventorship doctrine needs to define its criteria with sufficient specificity to allow research teams to determine who qualifies as an inventor with some confidence. To achieve maximal efficiency and fairness, those criteria should be neither overinclusive nor underinclusive. All collaborators who contribute significantly to the conception of the invention, and only those collaborators, should qualify as inventors. The consequences for a bona fide inventor who is excluded from a patent application covering an invention to which he contributed significantly can be devastating for the omitted inventor. The financial rewards of being named an inventor may be substantial, indeed monumental, because the inventors are the presumptive owners of the patent to the invention as tenants in common.[21] In the absence of an agreement to the contrary, a joint inventor is entitled to exploit the patented invention commercially without the consent of, or owing a duty of contribution or accounting to, the other inventors.[22] In other words, exclusion from joint inventorship means that the putative inventor may lose all of the financial rewards of the invention, no matter how great that inventor’s investment of time, resources, or creativity in the invention.[23] Recognition as a joint inventor means not only a potential share of the economic rewards, but it may have career-altering effects. A scientist or engineer may invest months or years in a research project that ultimately results in a patent. He may expend institutional funds, exhaust grants, and stake his or her reputation on the success of the project. Frequently, hiring, promotion, or tenure depends on successful completion of the project. Exclusion from the inventorship on the resulting patent denies the researcher not only his or her just financial rewards, but career recognition that may dictate his future in academia or industry. Fixating on claims will in many cases have no effect on inventorship determinations. Frequently, a joint inventor’s contribution, if it appears anywhere in the issued patent, will be claimed expressly. But relying exclusively on claims to determine inventorship is like solving three adjacent sides of a Rubik’s Cube. From some viewpoints, the puzzle appears fully solved. But the apparent order masks a fundamental incoherence evident from other viewpoints. The full puzzle cannot be solved without twisting the ostensibly finished sides back into disorder. Inventorship doctrine suffers from the same fundamental incoherence. Fixating on the claims makes inventorship doctrine seem sound when the inventor’s contribution is recited expressis verbis in one or more claims. Other inventive contributions fall within the confused jumble of fragmentary judicial statements on the doctrine of invention. For reasons to be explained, there is no necessary correlation between the relative importance of an inventive contribution, or the effort or creativity involved in that contribution, and its expression in the claims. To deny such collaborators recognition as inventors is unfair to them, confuses basic patent doctrine, and undermines the patent law’s incentive structure. The purpose of this article is to disaggregate the patent claims from the invention and to restore the integrity of the concepts of invention and inventorship. This restoration will require abandoning the time-honored judicial fixation on claims. Giles Rich, frequently cited by reverent courts and patent bar alike,[24] famously said, “[T]he name of the game is the claim.”[25] That phrase reflects Rich’s zealous opposition to the concept of invention[26] as well as the key role of claims in limiting the patent monopoly. But the influence of his maxim has far surpassed its meager truth. As Oliver Wendell Holmes once observed, “It is one of the misfortunes of the law that ideas become encysted in phrases and thereafter for a long time cease to provoke further analysis.”[27] Claims are operationally important to patent law, but inventors do not invent claims, they invent inventions. And, to coin a rather different maxim: The prevention of any mention of invention causes misapprehension. This article will explore the causes of the confusion; its consequences for inventors; and its resulting effect on U.S. technological development. Part II of this article begins by summarizing the law governing invention and inventorship, and its development and interpretation by the Supreme Court and Federal Circuit. Joint inventorship receives special attention in this part because the case law in this area reveals especially well how fixating on the patent claims distorts patent law analysis more generally. Part II then relates how an idiosyncrasy in the standard of proof in patent litigation has reinforced distorting biases in inventorship analysis. Part III maps out how the broken link between inventorship doctrine and the public policy it is intended to promote can be repaired. Specifically, Part III.A analyzes the legislative history of the Patent Act provisions on joint inventorship to determine congressional intent and compares the current joint inventorship doctrine’s misdirected fixation on claims to the policy goals sought by Congress and discusses the policy consequences. Part III.B then evaluates how the question of inventorship affects the role of the patent system in technological development generally and proposes an interpretation of the Patent Act that better fulfills the goals of U.S. technology policy. Part IV concludes by summarizing the argument.

I. The Law of Invention and Inventorship

A. The Patent Law Concept of Invention

Although the 1952 Patent Act uses the term “invention” throughout, the closest the Act comes to defining the term nontautologically is in Section 101: “Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.”[28] This is not technically a definition, but it gives clues to one by specifying that an invention is a process or physical product that is new and useful.[29] The term “new” is almost universally interpreted as a term of art in patent law (equivalent to “novel”)[30], further defined in Section 102.[31] The term “useful” is not defined in the Act but has been construed in the case law to mean that the invention offers some immediate and practical benefit to the public.[32] While Sections 100 and 101 provides a starting point for defining invention, they are incomplete. As suggested in Part I, there are other requirements for patentability. Procedural requirements such as the inventor’s oath or the application fee obviously do not figure in the definition of the invention itself. But there are other substantive requirements for an invention to qualify for patent protection, the most important of which is nonobviousness. Despite its omission from Section 101 of the Patent Act, nonobviousness is in some ways the keystone of the definition of invention. The Patent Act, as amended by the America Invents Act, defines an invention as obvious:
if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. . . .[33]
The nonobviousness requirement lies so close to the heart of the definition of an invention because its purpose is to clarify the historically confused, and sometimes contradictory, judicial interpretation of the constitutional requirement of “invention.”[34] A new technology that would have been obvious to an ordinary scientist or engineer at the critical date may qualify as an innovation in the colloquial sense, but it is not an invention in the patent law sense.[35] A statutory analysis of the concept of invention might be expected to end here. Other requirements for obtaining a patent are usually considered procedural rather than substantive. The Patent Act itself classifies the novelty, utility, and nonobviousness requirements in a chapter of the Act (“Patentability of Inventions”) separate from the chapter specifying the procedural requirements (Chapter 11—“Application for Patent”). Yet, the definition of invention remains incomplete because nothing in Sections 101 to 103 obligates the inventor to understand and appreciate the significance of his own discovery.[36] An unrecognized product or process is not yet an invention at all in the patent law sense.[37] The relevant section of the Patent Act, Section 112, is found among the procedural provisions of Chapter 11. Section 112 codifies the disclosure requirement.[38] Specifically, it requires that every patent application include a “specification” section that sets forth a full written description of the invention, “and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use” that invention.[39] Section 112 also includes the requirement that the application terminate with “claims.”[40] Between the written description and the claims, the specification is supposed to alert the public to the outward bounds of the invention that will be protected by patent, and to teach the public how to make and use the claimed part of the invention. For this reason, Section 112 is commonly thought to embody the fundamental bargain of the patent system: the inventor discloses the invention to the public in exchange for a temporary legal monopoly over making, using, selling, offering to sell, or importing the invention.[41] To that extent, the enablement provisions of Section 112 are indeed procedural and not substantive. They establish a disclosure requirement that the applicant must fulfill in order to obtain a patent, but this requirement assumes that the inventor sufficiently appreciates what he has invented and how to reproduce it. In other words, an applicant who can fulfill the requirement to describe clearly how to make and use the invention proves that he understands the invention sufficiently, and that he “actually invented the invention claimed.”[42]
1. The Role of Conception
Section 112 thus implies an uncodified but essential aspect of the definition of an invention. Invention has not occurred until the inventor appreciates that he or she has invented something useful, functioning, worthwhile. This subjective component of invention is inherent in the idea of “conception,” which is integral to establishing the moment when the invention became entitled to a patent. Conception is considered the “touchstone” of invention,[43] and has been defined as “the complete performance of the mental part of the inventive act” and “the formation in the mind of the inventor of a definite and permanent idea of the complete and operative invention, as it is hereafter to be applied in practice.”[44] Conception is incomplete unless the idea of the invention “is so clearly defined in the inventor’s mind that only ordinary skill would be necessary to reduce the invention to practice, without extensive research or experimentation.”[45] In other words, even if the inventor has a complete mental picture of the invention’s physical configuration, the invention is not “conceived” in patent law until the inventor has also determined how to make and use the invention in practice.[46] Even when conception no longer defines the date at which the inventor is entitled to a patent, conception will still be necessary, because without it the applicant will lack the information necessary to complete the Section 112 disclosure. Therefore, while Section 112 is indeed a procedural provision, one of its other functions is to provide evidence of whether the application’s subject matter satisfies the substantive requirements of invention. As the Federal Circuit put it: “The conception analysis necessarily turns on the inventor’s ability to describe his invention with particularity. Until he can do so, he cannot prove possession of the complete mental picture of the invention.”[47] The relationship between the inventor’s understanding and the physical configuration of the invention can be illustrated through an example:
{1a}  Ira has processed several chemicals to create a novel and nonobvious compound through a complex series of steps. Ira believes the compound could prove to cure cancer, because its molecular structure is loosely related to the structures of other chemicals that have anticancer effects.
In example {1a}, Ira may be capable of fully describing how to make and use his new molecule, and so he can satisfy the procedural requisites of Section 112. However, if Ira’s belief in the molecule’s anticancer properties lacks any firm basis in science, the molecule does not satisfy the Section 101 utility requirement. Although under current law an inventor need not know with certainty or even probability that the invention will work for conception to be considered complete,[48] a mere wish or unsupported expectation does not suffice. In Hitzeman v. Rutter, the Federal Circuit held that a genetically-altered yeast intended to produce antigens having certain key size and sedimentation rates had not yet been conceived by the inventor, because the inventor had only a “hope,” not a reasonable expectation, that the yeast would perform in that manner.[49] In Scenario {1a}, then, unless Ira has some scientifically sound basis, however partial, for attributing anti-cancer properties to his molecule, there is not (yet) a patentable invention. Consider a slight alteration of example {1a}:
{1b}  On January 30, Ilsa processes several chemicals to create a novel and nonobvious compound through a complex series of steps. At the time, she has no idea whether the chemical has any effective use. On June 1, she discovers that the new chemical is extremely effective at treating cancer. Ilsa files a patent application on the chemical.
Ilsa has clearly created the precursor to a new cancer-fighting drug, but when was it invented? On January 30 or June 1? Example {1b} illustrates the subjective requirement of invention. A new product or process may meet the objective statutory patentability requirements (novelty, nonobviousness, and utility), but it is not a patentable “invention” until its creator appreciates the utility of the invention and understands how to reproduce it.[50] As a correlate, there have been cases in which the inventor created a new chemical substance unwittingly; the substance was not deemed to have been conceived until the inventor perceived its presence,[51] because conception has not occurred until the inventor gains the ability to describe the invention.[52] In Example {1b}, conception occurred on June 1. Another variation illustrates a concomitant principle:
{1c} On January 30, Ian conceives of a chemical compound that he has sound scientific reason to believe will be extremely effective at treating cancer. However, he has no expertise in synthesizing this class of chemicals. On June 1, Rowena assists Ian to synthesize the compound using new insights and techniques developed specifically for the purpose.
Example {1c} is in some ways the reverse of {1b}. In {1b}, the first inventor could visualize and make the compound but did not know how to use it. In {1c}, the first inventor could visualize and knew how to use the compound, but did not know how to make it. The outcomes are equivalent; in both cases, the compound is not “invented” until June 1; in scenario {1c}, Ian and Rowena are joint inventors.[53] The foregoing discussion, like any doctrinal restatement, gives an illusion of certainty about the law relating to invention. In practice, gray areas in the subjective prong of conception, coupled with frequent misunderstanding of the inventive process, plague the law and its application with inconsistencies. The explanation of example {1b} relies on the PTO’s or court’s ability to distinguish between sufficient and insufficient bases for attributing utility to the invention. From a scientific or engineering perspective, the question of whether an untested product or process will function as intended is not a “yes” or “no” one, but a sliding scale of probability. The Federal Circuit has adopted a murky test to determine where conception occurs on the scale between mere surmise and certainty of adequate functioning. For example, on one hand, in order to shift the burden of proof to the applicant, a PTO examiner challenging utility must produce evidence that a person having ordinary skill in the relevant art would reasonably doubt the alleged utility.[54] On the other hand, an applicant or patentee asserting the utility of a patented invention must produce at least some evidence, however inconclusive, that the invention works as intended.[55] This leaves the PTO, which bears the burden of proving an absence of utility,[56] without much guidance. To be safe, the PTO has effectively adopted the practice of virtually never rejecting pharmaceutical inventions on utility grounds unless the allegation of utility is incredible.[57] This is partly because the PTO does not test the utility of an invention disclosed in a patent application and so must rely primarily on the applicant’s representations and the (usually very general) background knowledge of the examiner. Such an “incredibility” test is inappropriate in judicial pro­ceedings, where the party challenging the patent may introduce evidence of inutility that was unavailable to a PTO examiner. Yet, the Federal Circuit and its predecessor have endorsed an exceedingly low standard of evidence to uphold an allegation of utility, especially in the active field of pharmaceutical research.[58] Congealing the uncertainty is the jurisprudence, discussed in Part II.C, creating a strong legal presumption of validity of the patent despite a complete or nearly complete vacuum of information about whether the invention was in fact useful at the time the patent was examined and issued. Another difficult doctrinal issue is raised in example {1b}. Suppose Ilsa did not discover the utility of the chemical; a new researcher, Ron, did. Does this not make Ron the inventor rather than Ilsa? Or does Ilsa remain the inventor? Or might they be considered joint inventors? From a logical and strictly doctrinal standpoint, the fact that conception was incomplete until Ron discovered the utility of the chemical implies that Ron is the sole inventor if he did not collaborate with Ilsa in any way, and a joint inventor with Ilsa if they did collaborate and Ron’s discovery was more than a straightforward application of well-known principles of chemistry. However, as discussed in Part II.C, Federal Circuit and Supreme Court precedents have created a basis for doubting whether courts would actually endorse this outcome.
2. The Role of the Claims
The problematic patent law conception of invention arises in part from the role of the claims in defining the invention. The claims, as noted above, are the part of the patent document that delimits the outer boundaries of the invention for which the applicant seeks protection. They are typically one sentence long and terse, if not cryptic, in their phrasing. Partly because the claims are drafted by patent lawyers or agents rather than the inventor, and partly due to the organic idiosyncrasies of language, the claims may or may not encompass all of the invention. In fact, they may encompass quite a bit more than the invention.[59] A claim may encompass subject matter that the inventor never conceived and never could conceive. There is no a priori reason why the claims will completely over­lap with the invention. An inventor who claims less than the invention may lose some of the benefit of the invention. There are several reasons why this might happen. One is poor drafting, although the consequences of poor drafting to the hapless patentee are mitigated by the doctrine of equivalents and the possibility of broadening reissue proceedings.[60] Another possible reason is that the inventor wishes to commercially exploit only a subset of an invention’s embodiments and to dedicate the remainder to the public.[61] In that case, patent coverage narrower than the invention as a whole poses no policy problems, because it is voluntary. In contrast, an inventor who claims more than the invention risks the overbroad claims being invalidated.[62] When the claims exceed what was enabled in the written description, they fail to honor the basic patent bargain of dis­closure of the invention in exchange for a legal monopoly over the claimed subject matter.[63] Patents with overbroad claims also fail to ensure that the inventor possessed the claimed subject matter, because there is no way to verify that the applicant did indeed invent what he cannot or will not describe. Indeed, such patents may suffer from a constitutional deficiency on the theory that Congress has no power to grant patents on subject matter the inventor did not invent.[64] Probably the most famous example of an inventor who tried to claim far more than he invented relates to the Morse telegraph.[65] Morse, recognized as the inventor of the electro-magnetic telegraph, obtained a reissue patent with eight claims. The first seven claimed various uncontroversial components and embodiments of the electric telegraph. The eighth claim, however, covered much broader subject matter:
I do not propose to limit myself to the specific machinery or parts of machinery described in the foregoing specifications and claims; the essence of my invention being the use of the motive power of the electric or galvanic current, which I call electro-magnetism, however developed for marking or printing intelligible characters, signs, or letters, at any distances, being a new application of that power of which I claim to be the first inventor or discoverer.[66]
The Supreme Court struck the claim as overbroad, because it covered subject matter that Morse had not described in his patent “and indeed had not invented.”[67]
Professor Morse has not discovered, that the electric or galvanic current will always print at a distance, no matter what may be the form of the machinery or mechanical contrivances through which it passes. You may use electro-magnetism as a motive power, and yet not produce the described effect, that is, print at a distance intelligible marks or signs. To produce that effect, it must be combined with, and passed through, and operate upon, certain complicated and delicate machinery, adjusted and arranged upon philosophical principles, and prepared by the highest mechanical skill. And it is the high praise of Professor Morse, that he has been able, by a new combination of known powers, of which electro-magnetism is one, to discover a method by which intelligible marks or signs may be printed at a distance. And for the method or process thus discovered, he is entitled to a patent. But he has not discovered that the electro-magnetic current, used as motive power, in any other method, and with any other combination, will do as well.[68]
If the claim had been upheld, Morse could prevent another inventor who had developed an entirely different and potentially better means of using electricity to transmit information for printing at a distance (for instance, by wireless radio frequency transmitter and receiver) from practicing his patent. The Court accordingly rejected the claim. The Morse case illustrates the interdependence of the Section 112 disclosure requirement and the concept of invention. The Court’s reasoning implied that the disclosure requirement does not only alert the public to the parameters of the protected intellectual property and teach the public how to make and use the invention, but tailors the scope of patent protection to what the inventor actually contributed to the technological field.[69] When the applicant claims more than he invented, the benefits of the patent to the public are attenuated and the possibility of a dysfunctional patent precedent arises. Patent prosecutors are typically more concerned about underclaiming than overclaiming. The patent examiner frequently whittles down an overly aggressive claim, and those exaggerated claims that survive prosecution will benefit from a strong presumption of validity, which gives the patentee useful leverage for negotiating with licensees and competitors, even if the claims are not strictly enforceable under a proper interpretation of the claim. At worst, the patentee can obtain a nar­row­ing reissue a few years down the road to ensure the claim’s enforce­ability.[70] An overly narrow claim, in contrast, risks dedicating the un­claimed subject matter to the public for all time. It is possible to seek a broadening reissue, but only within two years after the patent is granted.[71] As a result, many patent pro­secutors tend toward aggressively claiming more than the inventor actually con­ceived, so long as the overclaiming will not court rejection by the PTO examiner. The claims are important, but as the Federal Circuit has emphasized en banc, not all-important: “The claims . . . do not stand alone. Rather, they are part of ‘a fully integrated written instrument’” (i.e., the patent).[72] Yet, when not engaged in the narrow task of claim construction, courts have sometimes fetishized the claims as if they were the only legally material part of the patent. The importance of the claims in determining whether anticipation by prior art or patent infringe­ment has occurred has led many courts, including the Supreme Court, to use language that conflates the claims with the invention itself: the claims, they assert, “define the invention.”[73] The Federal Circuit has stated that a researcher must “conceive” of a “claim” (rather than an invention) to be an inventor,[74] and suggested that Section 116 of the Patent Act requires a coinventor to “make a contribution to the conception of the subject matter of a claim.”[75] The Patent Act says no such thing; it formerly referred to inventions and “subject matter” rather than claims, and now refers to these and to the “claimed invention” as well. Claims may or may not coincide with the full scope of the invention or even the “claimed invention.” Oskar Liivak has defined the invention as “the set of all of the embodiments that the inventor has invented. That is, the invention is the set of embodiments conceived by the inventor in enough detail so that each of those embodiments is capable of being reduced to practice.”[76] This definition, while incomplete, has the virtue of indicating how the invention is (or at least should be) more expansive than the claims, which are “written attempts to circumscribe varying subsets of the embodiments disclosed in the specification.”[77] Liivak’s point, that the claims (and indeed the “claimed in­ven­tion” of Section 100(j)) are a subset of the invention rather than the invention itself, is well taken. Courts have on a few occasions distinguished explicitly between the aspects of the invention protected by patent and the invention as a whole. For example, in a moment of clarity, a panel of the Federal Circuit observed that “the specification teaches an invention, whereas the claims define the right to exclude.”[78] The Supreme Court has also distinguished between the two, stating that claims are “construed . . . with a view to ascertaining the in­ven­tion.”[79] The necessity of distinguishing between the claims and the invention to which the claims pertain is also evident from the Patent Cooperation Treaty (PCT), to which the United States is a party.[80] The PCT allows inventors to file a single patent application that may result in patents in multiple countries. Under the “unity of invention” requirement, the patent application must relate to just one “invention” or a “group of inventions so linked as to form a single general inventive concept.”[81] Yet, in each state granting a patent resulting from the PCT application, the claims may differ to comply with local filing requirements.[82] A single invention could not receive the same patent protection in different countries using different claims unless the invention is something larger than the claims. The frequent judicial and scholarly statements that claims “define” the invention, then, are misleading.[83] The invention is more than the claims; the claims limit the invention but they do not alone define it. “[I]t is fundamental that claims are to be construed in the light of the specifications and both are to be read with a view to ascertaining the invention.”[84] The written description portion of the specification “‘is always highly relevant to the claim construction analysis. Usually, it is dispositive; it is the single best guide to the meaning of a disputed term.’”[85] If the claims must always be read in light of the specification, the claims cannot “define” the invention; they merely limit the extent to which the patent protects the invention. The Leahy-Smith America Invents Act, the relevant provisions of which take effect on March 16, 2013, suggests that Congress shares this perception of the distinction between the claims and invention. The Section 102 amendments substitute the term “invention” in the novelty provisions of the Patent Act with the term “claimed invention,”[86] which is in turn defined as “the subject matter defined by a claim in a patent or an application for a patent.”[87] From this, it appears that Congress considers the “invention” to be different from the “claimed invention.”[88] Because the patent protects only the claimed portion of an invention from infringement, it is possible to infringe on a “claim,” but not an “invention.” By extension,[89] prior art may anticipate specific claims in a patent application without rendering the invention unpatentable. Section 102 now more clearly suggests that novelty analysis consists in determining whether specific claims are anticipated by prior art, as opposed to analyzing whether the “invention” as a whole is anticipated by prior art, because each claim stands on its own. It follows that a person who infringes any one claim infringes the patent, even if the accused product or process does not infringe any other claim. For both anticipation and infringement purposes, then, claims do not “define” the invention in the conventional sense; they define the rough outlines of what part of the invention can overcome anticipatory prior art and accordingly what part of the invention the patent protects from infringement. This key distinction between the invention and the claims has unappreciated importance for determinations of inventorship, and especially of joint inventorship.

B. Inventorship and Joint Inventorship

1. The Evolved Doctrine
When a sole researcher conceives the entirety of an invention, she must satisfy all of the patentability requirements by herself. When the inventor fulfills these requirements, abstruse doctrinal or policy problems arise only in exceptional cases. But the image of the lone scientist confined to a laboratory for months, and ultimately emerging with a triumphant new invention has more romantic appeal than correspondence with reality. Although much government rhetoric exalts the independent inventor,[90] in modern industry, discoveries in applied science are usually a product of teamwork. Joint invention is now more common than sole invention.[91] In general, a joint invention “is the product of a collaboration between two or more persons working together to solve the problem addressed.”[92] Yet, the process of identifying whether a collaborator qualifies as a joint inventor was until very recently crude and uncertain, and even today the law remains under­developed. Before 1984, the Patent Act’s provisions on joint inventorship were entirely procedural; they offered no guidance to the judiciary on how joint inventorship should be understood. The Act merely stated: “When an invention is made by two or more persons jointly, they shall apply for patent jointly and each sign the application and make the required oath, except as otherwise provided in this title.”[93] It then went on to explain what procedural measures could be taken if a joint inventor refused to join in the application, as well as the consequences of a non-inventor being wrongly listed as an inventor (commonly known as “misjoinder”) or an inventor being erroneously omitted (“nonjoinder”).[94] For many years, courts struggled with the question of how to determine who was a joint inventor. Regarding the threshold of contribution required to qualify as a joint inventor, their formulated tests were frequently vague and sometimes contradictory, leading one district court judge to lament:
The exact parameters of what constitutes joint inventorship are quite difficult to define. It is one of the muddiest concepts in the muddy metaphysics of the patent law. . . . Perhaps one need not be able to point to a specific component as one’s sole idea, but one must be able to say that without his contribution to the final conception, it would have been less—less efficient, less simple, less economical, less something of benefit. . . . . . . . . . . . This situation does make it difficult to say . . . with real certainty, whether or not a given person “is” a joint inventor in a given case.[95]
Such guidance was indeed lean. “Less something of benefit” is rather overinclusive as a standard; it would reward routine, and indeed flatly obvious, contributions with coinventorship. Courts were plainly struggling with the doctrine. Another judge wrote:
The conception of the entire device may be due to one [inventor], but if the other makes suggestions of practical value, which assisted in working out the main idea and making it operative, or contributes an independent part of the entire invention, which is united with the parts produced by the other and creates the whole, he is a joint inventor, even though his contribution be of comparatively minor importance and merely the application of an old idea.[96]
These pronouncements were problematic not only because of their lack of guidance. The mere contribution of “an old idea” of “minor importance” is probably too low a standard, and has been at least called into doubt by Federal Circuit precedent.[97] If a joint inventor must contribute something nonobvious, then the nature of the contribution determines joint inventorship, not merely the effect of the contribution on the invention’s functioning. It would seem that a collaborator who contributed nothing more than information or designs well known in the art would not qualify as an inventor unless the application of that prior art to the problem at hand would have been nonobvious.[98] A decided improvement in the functioning of the invention is relevant to assessing the quality of the contribution but far from dispositive. While the historical standard of joint inventorship might have seemed low, in some ways it compensated for an aberration of patent law that discouraged team research. In case after case, federal courts denied joint inventors the right to file a single patent with multiple claims on the product of their collaboration, if all inventors did not contribute to each and every claim.[99] In other words, this “all claims rule” excluded from inventorship any collaborator who did not contribute to every claim of the patent. As a leading treatise of the time put it: “Only where the same single, unitary idea of means is the product of two or more minds, working pari passu, and in communication with each other, is the conception truly joint and the result a joint invention.”[100] One nineteenth century court even held that when two researchers invent distinct parts of the same new machine, they should obtain separate patents as sole inventors on each part of the machine rather than a single patent as coinventors on the machine as a whole.[101] In 1967, a federal trial court cleared up much of the doctrinal obscurity and tried to reverse the bias against collaborative research. Monsanto v. Kamp Co. arose from an interference proceeding before the U.S. Patent Office, predecessor to the PTO.[102] The invention in question was a specially coated plastic container for pharmaceuticals.[103] The plaintiffs challenged the designation of two defendants as joint inventors, which launched the court into an original discussion of the principles of joint invention:
A joint invention is the product of collaboration of the inventive endeavors of two or more persons working toward the same end and producing an invention by their aggregate efforts. To constitute a joint invention, it is necessary that each of the inventors work on the same subject matter and make some contribution to the inventive thought and to the final result. Each needs to perform but a part of the task if an invention emerges from all of the steps taken together. It is not necessary that the entire inventive concept should occur to each of the joint inventors, or that the two should physically work on the project together. One may take a step at one time, the other an approach at different times. One may do more of the experimental work while the other makes suggestions from time to time. The fact that each of the inventors plays a different role and that the contribution of one may not be as great as that of another, does not detract from the fact that the invention is joint, if each makes some original contribution, though partial, to the final solution of the problem.[104]
The district court decision in Kamp was (and is) cited with some frequency by other courts, including the Federal Circuit.[105] Although Kamp stood as a beacon of thoughtful interpretation, a district court decision could not unify patent doctrine nationwide. Eventually, Congress had to clarify the standards of joint invention. This took the form of the 1984 Patent Law Amendments Act (PLAA), which added the following language to what is now Section 116(a):
Inventors may apply for a patent jointly even though (1) they did not physically work together or at the same time, (2) each did not make the same type or amount of contribution, or (3) each did not make a contribution to the subject matter of every claim of the patent.[106]
This language has not been altered since and remains the main Patent Act provision defining the conditions for joint inventorship. The 1984 PLAA reversed the all claims rule, allowing a research team to apply for a single patent on a unitary inventive concept despite the fact that some team members may have contributed to only a single claim in a multiple claim patent. It also codified the trend advanced by the Kamp decision toward relaxing the standard of collaborative invention. It is now clear that direct interaction between or among collaborators is unnecessary. For example, consider the following scenario:
{2} Three scientists, Alpha, Beta, and Gamma, are involved in research calculated to lead to the discovery of a room-temperature superconducting material. Alpha is involved throughout the entire project, which takes a total of ten years before an invention has been conceived and is ready for patenting, but Alpha’s contribution, although of an inventive quality, is relatively minor. Quantitatively, Alpha’s contribution of effort, expertise, and creative thinking amounts to about 10% of all the effort, expertise, and creative thinking that went into the invention. Beta was involved from the beginning as well, but after nine years and six months, quit the project. Beta’s contribution was, however, over­whelming, accounting for 80% of the research effort, etc. During the last six months, Gamma became involved in the project and collaborated with Alpha to finalize the conception of the invention. Due to Gamma’s late arrival, Gamma contributed only 10% to the project. Beta and Gamma never directly communicated with one another; Gamma merely took up Beta’s almost-completed work.
In scenario {2}, a requirement of direct collaboration at the time the invention was completed would make Alpha and Gamma joint inventors, but not Beta, despite the fact that Beta individually contributed much more than the other two inventors combined. This result is not only unfair to Beta, it creates a powerful disincentive to collaborate in applied research, because, except in the rare situation that the research is certain to result in the successful development of a valuable invention in short order, a researcher’s contribution could be wasted if the researcher needs to leave the collaboration before a patentable result is produced. Many such reasons would be beyond the control of the collaborator, such as illness, financial distress during extended research, termination of employment with an employer sponsoring the research, and so forth. The PLAA establishes beyond peradventure that Beta is as much an inventor as his collaborators. Similarly, the PLAA suggests that in assessing inventorship in scenario {1b} above, if Ron rather than Ilsa discovered the utility of Ilsa’s compound, Ron would be a coinventor. Before the PLAA, courts could deny the importance of Ron’s inventive contribution. For example, in Eli Lilly & Co. v. Premo Pharma­ceutical Laboratories, Inc., two scientists had synthesized a compound (cephalexin) in “the hope that the resulting compound would be [an] effective oral antibiotic” for certain biochemical reasons.[107] A third scientist tested the compound and found that it was indeed an effective oral antibiotic for entirely different biochemical reasons than the first two scientists had guessed. The patent on the compound, however, listed only the first two scientists as inventors. When an accused infringer challenged the validity of the patent for nonjoinder, the Third Circuit held that the third scientist’s discovery of the invention’s “nonobvious trait” did not qualify him as a joint inventor of the compound. In attempting to justify this result, the court wrote:
it is without question that the named inventors, Drs. Morin and Jackson, were the only persons who performed the synthesis that created the patented product. In the words of § 116, cephalexin was ‘made by’ the two named inventors, not by the biochemist who first noted that the organic chemists’ predictions had been realized.[108]
After the PLAA, this result even less plausible than before. A compound whose utility or nonobvious traits are unknown has not yet been “conceived” as an invention under the patent law.[109] A collaborator who makes a significant, original contribution to conception, such as by discovering utility or the nonobvious trait, is a coinventor even if his contribution was different in type (testing of the compound versus synthesis of the compound) from that of his collaborators.[110] With regard to the level and kind of contribution required of a joint inventor, the PLAA added relatively little.[111] Instead, the newly organized Federal Circuit developed joint inventorship analysis in its jurisprudence—a jurisprudence that continues to dominate discussions of inventorship today.[112] Beginning in the early 1990s, the court decided a series of cases that delineated the basic substantive contours of joint inventorship.[113] For the most part, these did not radically depart from the main body of cases decided before the PLAA. Before the amendment, the concept of joint inventorship was rarely interpreted to require “give-and-take” between or among joint inventors. In a 1980 case, for example, a Florida district court found that when one researcher developed a prototype of the invention and passed the prototype to a second researcher, who perfected it, they were joint inventors despite the lack of any regular communication between them. “The ideas” of the first inventor “were presented daily” to the second “by the prototype, and the final result was a creation that exceeded the results of either inventor,” the court observed.[114] The court understood that, although collaboration always requires communication between the collaborators, communication can take nonverbal as well as verbal forms. There is no policy reason to prioritize verbal over nonverbal collaboration if both forms can accelerate technological development. The Federal Circuit continued this interpretation of joint inventorship law; so long as the inventors were consciously cooperating, one collaborator may simply read and build upon a report written by another, or profit from the suggestion of another at a meeting, to qualify them as joint inventors.[115] More importantly, the Federal Circuit unified and fleshed out the kind of contribution necessary to qualify a collaborator as a joint inventor. From Pannu v. Iolab,[116] three general conditions may be deduced. First, the alleged coinventor must contribute to the conception of the invention.[117] Second, his contribution to the invention must not be insignificant in quality, when measured against the dimension of the full invention.[118] Third, the contributor must do more than merely explain to the real inventors well-known concepts or the current state of the art.[119] The second and third conditions follow in a sense from the first. The alleged joint inventor’s contribution could not merely relate the state of the art to the collaborators, because providing well-known information “is not a contribution to conception.”[120] One does not invent merely by repeating public information or using ordinary skills; the contribution must go beyond well-known prior art. Similarly, non-technical contributions to the invention, such as suggest­ing a desirable goal of research (unless identifying that goal is technically difficult and not obvious) and providing management or financing, are not contributions to conception.[121] The Federal Circuit has repeatedly observed that Section 116 of the Patent Act “sets no explicit lower limit on the quantum or quality of inventive contribution required for a person to qualify as a joint inventor.”[122] However, the court has interpreted Section 116 to impose a second condition on the alleged joint inventor—that his contribution must be “not insignificant in quality, when that contribution is measured against the dimension of the full invention.”[123] Where to draw the line of significance on the scale between a critical contribution and a trivial one is a tricky and highly subjective exercise. Unfortunately, the case law interpreting significance remains notably thin. In one unpublished case, the U.S. Court of Appeals for the Fourth Circuit interpreted the significance factor to require that all joint inventors contribute to the specific aspect of the invention that made it patentable. In Levin v. Septodont, the developer of a new numbing cream to treat mouth ulcers, Dr. Kilday, was approached by a retired dentist, Dr. Levin, who had the idea of developing a numbing mouth rinse to reduce pain caused by routine dental procedures.[124] Working together, Kilday and Levin determined that the active ingredients in the rinse should be the antiseptic phenol and the anesthetic benzocaine.[125] Kilday determined the proportion of ingredients in the product.[126] However, Kilday also recognized that these ingredients could not be dissolved in an alcohol solution due to the burning sensation caused by alcohol. Kilday therefore approached Eastman Chemical to help develop a suitable suspension solution. After significant research, Eastman developed a new solution composed of a combination of polyethylene glycol and propylene glycol that solved the burning problem.[127] Eastman sought a patent on the resulting dental mouth rinse, but named only its employees, and not Kilday.[128] Septodont sought to correct the patent by joining Kilday as a coinventor. Levin resisted on the ground that the PTO examiner had opposed the patent application as obvious until a limitation was added specifically claiming the Eastman suspension solution. The court sided with Levin, interpreting joint inventorship doctrine to require contribution to the novel and nonobvious limitations of the claim—indeed, regardless of whether a specific collaborator’s contribution appears in the claims or indeed is the defining feature of the invention.[129] In other words, the Fourth Circuit would have interpreted an inventive contribution’s “significance” in light of its novelty and nonobviousness. Levin v. Septodont is not a precedential opinion,[130] but it illustrates a troubling misunderstanding of the concept of joint invention. The court effectively held that the collaborators who conceived the invention as a whole, determined its utility, and chose key ingredients and their proportions did not contribute significantly to conception. The problem arises from the Fourth Circuit’s dissection of the invention. The invention as a whole was novel and nonobvious—an anesthetic mouth rinse for routine dentistry—and it was conceived by Levin, Kilday, and Eastman scientists in collaboration. A collaborator’s contribution need not define the nonobvious aspect of the invention itself, so long as it satisfies the three Pannu conditions. No doubt the Eastman collaborators were inventors as well, given their contribution to an important and nonobvious aspect of the invention. Possibly their suspension solution was independently patentable, and in that case the Eastman scientists would stand alone as inventors of the solution. But the patent claimed not the solution itself but rather an “anesthetic and antiseptic mouth rinse” using a non-alcohol solution.[131] Simply put, the invention before the court would have been impossible without Levin and Kilday’s original contributions. In denying them status as coinventors, the Fourth Circuit not only deprived a valuable collaborator of the patent reward, but conferred that reward on Eastman for novel and nonobvious ideas that Eastman did not develop and probably could not have developed itself. It is probably fair to say that any contribution adding the novel, nonobvious, or useful element to the invention is ipso facto significant. To conclude from this, however, that any other contribution is insignificant is a non sequitur. The question of what kind of contribution does qualify as significant—a question the Federal Circuit has not had occasion to answer in great detail—remains open. Presumably, such factors as whether the collaborator:
• identified or solved a problem unrecognized by the others; • solved a problem that other collaborators could not solve; • added a nontrivial advantage to the invention that the other collaborators did not contemplate; or • contributed the novel, nonobvious, or useful aspect of the invention or a nontrivial part of it
would weigh heavily in favor of finding an inventive contribution “not insignificant” if directed to any functional aspect of the invention.[132] The contributor’s skill set and abilities relative to other collaborators may have some bearing on whether these criteria are satisfied. The kind and level of expertise of the collaborator, his effort expended relative to other collaborators, and the importance and originality of his insight to the advantages of the invention, would all be relevant as well.[133] The consequences of designating too many or too few collaborators as inventors on a patent application may be trivial or (much more rarely) fatal, depending on the provable intent of the applicants. Misjoinder or nonjoinder of joint inventors is supposed to invalidate a patent if done intentionally, but an accused infringer cannot realistically be expected to produce clear and convincing evidence of the applicant’s state of mind in most cases. As a result, successfully proving misjoinder or nonjoinder usually results in a correction of inventors in order to save the patent from invalidity.[134]
2. The Distorting Effect of Claims Fixation
The joint inventorship doctrine described in Pannu v. Iolab is sound as far as it goes. It meshes with both the constitutional limitations on the congressional patent power and the language and policy purposes of the Patent Act. The problem is not in the general standard of joint inventorship, but in its implementation. The Federal Circuit has sometimes diverted the inquiry away from a fact-intensive examination of the realities of the inventive process into a rigidly formulaic test, with the regrettable result of facilitating mistaken or fraudulent nonjoinder. The judicial decisions carrying the greatest potential to warp inventorship standards probably are not substantive, but procedural. This relates back to the manner in which courts have confused the claims with the invention. The Constitution and Patent Act refer to inventors and their inventions, but as noted, courts have frequently written the invention out of patent law and substituted the claims, so that, for example, an inventor in the Federal Circuit’s jurisprudence does not conceive an invention, but rather a “claim,”[135] and a coinventor must contribute not to the conception of the invention, but to the conception of “the subject matter of a claim.”[136] The standard procedure developed by the Federal Circuit for evaluating joint inventorship was explained in Trovan, Ltd. v. Sokymat S.A. Irori.[137] There, the court equated inventorship analysis to infringement analysis.[138] Infringement analysis begins with a binding interpretation of the claims through a Markman hearing.[139] With the meaning of the claims established, the court then compares the accused product or process to the claims on a claim-by-claim basis; if the accused product or process literally or equivalently satisfies all the limitations of a claim, then infringement has occurred.[140] Under Trovan, the procedure to be followed in inventorship analysis also begins with an interpretation of the claims “to determine the subject matter encompassed thereby.” The next step is “to compare the alleged contributions of each asserted coinventor with the subject matter of the properly construed claim to then determine whether the correct inventors were named.”[141] As one patent lawyer has observed, inventorship is now determined “on a claim-by-claim basis”[142] as well, despite the fact that an inventor on any one claim is considered an inventor of the entire invention (and, therefore, all of the claims).[143] Trovan analysis, which has become the accepted procedure for evaluating disputes about inventorship, creates an aberration in the Federal Circuit’s otherwise mostly sound jurisprudence on joint inventorship. In undertaking inventorship analysis, courts now tend to fixate on whether an alleged coinventor can point to specific claims language directly attributable to that alleged coinventor’s contributions.[144] The result of fixating on claims is frequently an overly simplistic understanding of the inventive process. In Eli Lilly & Co. v. Aradigm Corp., the Federal Circuit justified its practice of ruling out some collaborators as joint inventors based on the “quality” of inventive contribution:
The line between actual contributions to conception and the remaining, more prosaic contributions to the inventive process that do not render the contributor a coinventor is sometimes a difficult one to draw. Contributions to realizing an invention may not amount to a contribution to conception . . . if they are too far removed from the real-world realization of an invention, or if they are focused solely on such realization.[145]
The first point, relating to highly abstract or indefinite contributions, is intended to weed out collaborators who merely point out the goal without contributing concrete ideas for achieving the goal. A collaborator who does nothing more than to propose that the research team develop a vaccine for HIV or a more efficient solar receptor is not an inventor even if the team succeeds in inventing the vaccine or receptor. Yet, it is quite possible to interpret the court’s language to deprive legitimate inventors of recognition for their contributions. Sometimes the most difficult and important part of the inventive process is understanding the problem to be solved. The contribution of a collaborator who merely identifies the problem may indeed be far removed from resolving it, but his work may be the foundation or even keystone of the invention. For example, suppose a research team working for a chemical developer is trying to synthesize an effective insecticide for termites. The goal is to discover a compound to whose toxicity termites find it difficult to adapt by rapid evolution. After years of fruitless and expensive labor on this project, one team member, Carlos, observes that exposure to certain nontoxic chemicals causes the termites to avoid tree wood and preferentially consume other forms of organic matter. After running some experiments, Carlos determines that it is possible that some such chemicals are likely to resist evolutionary adaptation, although it is not clear which chemical compounds would do the trick. After communicating his findings to the team, Carlosis transferred to a different research team. However, based on Carlos’s observations, the original team reorients its entire research program and, after a few more months, invents a nontoxic compound that effectively makes lumber termite-proof. Carlos’ contribution was undoubtedly “far removed from the real-world realiza­tion” of the resulting compound; he does not even know what the compound looks like or how it was produced. Yet, to say that Carlos was not a joint inventor of the compound disserves the goals of patent law. In the scenario, Carlos’ contribution was not a mere recitation of the prior art and was integral to and necessary for the invention of the compound. If he is not considered an inventor, he has no incentive to reveal his key insight to the other members of the research team, and the policy goal of Section 116 is subverted. Moreover, he is unfairly denied a reward for a breakthrough discovery that enabled the patented invention. The Eli Lilly court’s last point, regarding contributions solely to the “realization” of the invention, is supposed to weed out collaborators who merely performed technical steps in the reduction to practice of an invention already fully conceived.[146] The superficial appeal of this limitation is deceptive as well. The court could merely have been reciting the tautology that, if the invention was already complete, then no further contribution could be significant. If the condition is not tautological (or redundant of the other requirements of joint inventorship), then the collaborator in question must have made both a technically significant and nonobvious contribution to making or using a product invention (or performing a process invention). If the resulting idea could not be made or used (or performed) without a significant and nonobvious contribution from a third party, it has not been fully conceived yet. If the collaborator’s contribution converts an incipient concept into a mature, invented product or process, then he has satisfied the conditions of joint inventorship. Because knowledge of making and using an invention is part of the invention itself, a significant and nonobvious contribution “solely” to the “realization” (or reduction to practice) of an invention may indeed qualify the contributor as a joint inventor.[147] The necessity of the contribution for patentability strongly suggests that the invention had not yet been fully conceived. The court’s statement on this point relates once again back to its fixation on the claims; claims do not necessarily teach how to make and use the invention, so if the claims “define” the invention, then a collaborator who pains­takingly discovers how to make or use the invention properly is no inventor if the means of making or using the invention is not recited explicitly in a claim. While the ambiguities in the Federal Circuit’s jurisprudence on joint inventorship may seem technical, their effects are magnified by the all-or-none consequences of inventorship determinations. Section 116 specifies that collaborators are not precluded from designation as inventors by the fact that they did not contribute to every claim in the patent. In Ethicon Inc. v. U.S. Surgical Corp., the Federal Circuit interpreted this provision to give every inventor who contributed to any claim in the patent equal ownership in all claims, including claims to which that inventor never contributed.[148] Despite frequent criticism of the court’s opinion,[149] the result was all but dictated by the language of Section 116 and the concept of unity of invention that underlies patent law.[150] But this policy has the unfortunate consequence of giving the most important contributor to an invention no greater rights than the least important.[151] The probable result is to dampen the incentive to engage in team research, because adding new members to a research team—especially those not obligated to assign their rights to the same employer—could result in a loss of meaningful patent rights.[152] The current jurisprudence therefore gives patent applicants a strong incentive, and no realistic disincentive, to fraudulently omit joint inventors from the patent application in order to avoid sharing the patent bounty.

C. The Strong Presumption of Validity in Inventorship Analysis

The grant of a patent by the PTO creates a presumption that the patent is correct and valid. The 1952 Patent Act provides expressly that anyone challenging the validity of a patent bears the burden of persuasion that the patent is invalid.[153] This burden originates from early patent law cases in which the U.S. Supreme Court established a general rule of deference to informed decisions of the Patent Office, most notably in Morgan v. Daniels:
Upon principle and authority, . . . it must be laid down as a rule that where the question decided in the Patent Office is one between contesting parties as to priority of invention, the decision there made must be accepted as controlling upon that question of fact in any subsequent suit between the same parties, unless the contrary is established by testimony which in character and amount carries thorough conviction.[154]
The patent applicant actually benefits from a presumption in his favor beginning at the moment the application is filed. The Federal Circuit noted that the PTO bears the burden of challenging the patent applicant’s “presumptively correct” allegations regarding the utility of the claimed invention.[155] A private party challenging an issued patent before a court bears an even heavier burden. Morgan v. Daniels has long been interpreted by the Federal Circuit and its predecessor, the Court of Customs and Patent Appeals, to mean that anyone challenging the validity of a patent must satisfy an elevated burden of proof, by supplying clear and convincing evidence of invalidity.[156] The “clear and convincing” standard of proof is nowhere to be found in the Patent Act. Its origin is instead a long line of judicial decisions holding that a government agency such as the PTO is “presumed to do its job” correctly.[157] As with other aspects of patent validity, a private party challenging a patent on grounds of misjoinder or nonjoinder bears the clear and convincing burden of persuasion.[158] In enacting the 1952 Patent Act, Congress chose not to specify the standard of proof for parties alleging patent invalidity for reasons unknown. In the recent case Microsoft v. i4i, the Supreme Court inferred from Congress’s silence that the legislature intended to incorporate (part of) the judicially invented standard of proof into the statute.[159] Indeed, the Court asserted that it “must” presume—apparently as a canon of statutory construction—that Congress intended to continue the status quo unless the statute otherwise specifies.[160] It is doubtful that, whenever Congress fails to comment on a matter within the scope of a statute, Congress thereby automatically endorses all case law not expressly disclaimed by the statute.[161] However, the fact that Congress has overlooked numerous opportunities to change the highly visible standard of proof suggests that the clear and convincing standard generally aligns with congressional intent in this instance. Doug Lichtman and Mark Lemley have argued against the strong presumption of validity, except with respect to a narrow category of patent challenges, on policy grounds.[162] Although the courts have never articulated a fully developed policy basis for the strong presumption, Lichtman and Lemley infer the two most likely justifications as being deference to the PTO’s expertise and a desire to strengthen patent rights in order to encourage commercialization.[163] But Lichtman and Lemley point out that the PTO does not, probably with current funding cannot, and in any event lacks sufficient motivation to, invest the resources necessary to render an authoritative determination on all questions of patentability.[164] The absence of third-party information in most cases further limits the ability of PTO examiners to assess thoroughly the validity of the patent. As for the disincentive to invest in a patent of uncertain validity, Lichtman and Lemley observe that such uncertainty exists in any case and is only one (probably minor) factor affecting commercialization.[165] To the extent that the patent’s validity has been insufficiently verified by the PTO, strengthening the presumption of validity distorts patent markets arbitrarily. Regardless of whether this argument persuades, the justification for the strong presumption of validity fails utterly with respect to matters never considered by the PTO. The PTO relies heavily or entirely on the thoroughness, honesty, and candor of the patent applicant with regard to some patent matters. Even for those confident in the PTO’s expertise and rigor, a general presumption that the PTO does its job correctly does not justify a presumption of omniscience. Why should the mere submission of the patent application by self-interested applicants create a prima facie presumption of accuracy and truthfulness?[166] On this logic, parties have sometimes challenged the strong presumption of validity with respect to questions not examined by or raised before the PTO. In Eli Lilly & Co. v. Aradigm Corp., the plaintiff made the exceedingly narrow argument that a preponderance of evidence standard should be used “‘when there are two co-pending patent applications claiming the same subject matter’ in front of the [PTO], one of which issues as a patent allegedly omitting the inventor, and the other of which was filed by the allegedly omitted inventor.”[167] The plaintiff, in short, was not challenging the heightened burden of proof in general or even with respect to inventorship questions not considered by the PTO. Its argument was limited to the case in which an interference should have been declared but was not. The Federal Circuit rejected even that limited argument, pointing to the alleged coinventor’s “strong temptation” to misrepresent the extent of its involvement in the collaboration and the absence of any risk to its rights in its own patent application, which it could lose in an interference.[168] Neither of the Eli Lilly court’s contentions justifies the clear and convincing evidence standard. The patent applicant seeking to exclude joint inventors has as strong a temptation to misrepresent its role in the inventive process as the alleged coinventors. The same advantages are at stake for each of them—the ability to commercialize (or suppress) the claimed invention. As for the absence of risk, this argument is simply irrelevant. If the issued patent survives challenge by the alleged coinventor, then the patentee may enjoy its patent rights without sharing them with a coinventor. The alleged coinventor will have no right to obtain a junior patent on the same subject matter, so it risks its rights either way. Moreover, the costs of a full-scale patent litigation dwarf the costs of a review or reexamination, and the alleged coinventor faces as much litigation expense as the patentee. Notwithstanding the Eli Lilly decision, a panel of the Federal Circuit faced another, more aggressive challenge to the clear and convincing evidence standard in inventorship determinations in Vanderbilt University v. ICOS Corp.[169] Pointing to the PTO’s inability to consider the status of alleged coinventors not listed in the patent application, Vanderbilt proposed applying the normal burden of proof.[170] The court again rejected the argument, but this time noted that the plaintiff was “of course free to seek en banc reconsideration of our settled law on this issue.”[171] Although the Federal Circuit has always rejected these challenges to the standard of proof since its creation in 1982, courts frequently expressed doubt with regard to matters not before the PTO both before and after 1952. Morgan itself required deference only to questions “decided in the Patent Office,” and not to questions never considered by the agency.[172] Since then, numerous courts have observed that the rationale for the strong presumption of validity was weakened in such circumstances,[173] and in several cases have actually given reduced deference to the PTO.[174] In KSR International Co. v. Teleflex, Inc., the Supreme Court itself joined the chorus, holding that when a matter had not been raised before the PTO, “the rationale underlying the presumption—that the PTO, in its expertise, has approved the claim—seems much diminished.”[175] Nonetheless, the Supreme Court has recently enshrined a uniform clear and convincing evidence standard for all aspects of the issued patent. The Court in Microsoft Corp. v. i4i Ltd. declined to interpret prior cases to cast doubt on the clear and convincing evidence standard, despite seeming to recognize in the same pre-1952 cases many doubts about the logic of the clear and convincing evidence standard applied to matters unexamined by the PTO.[176] By the majority opinion’s logic, Congress should also be “presumed” to have codified these cases. Instead, the i4i Court adopted an expedient, holding that when a matter was not considered by the PTO, the PTO’s judgment “may lose significant force” and “the challenger’s burden to persuade the jury of its invalidity defense by clear and convincing evidence may be easier to sustain.”[177] Just how the challenged part of the patent “los[ing] significant force” differs meaningfully from lowering the burden of proof is hard to comprehend. Operationally, the Court has intimated that the trial judge can, or perhaps must, notify the jury that the facts at issue were never considered by the PTO while at the same time insisting that the jury apply the clear and convincing standard of proof.[178] But the jury never bases its judgment of patent validity on an assumption of the PTO’s infallibility. The facts are presented to the jury, and the court explains its interpretation of the law, including the burden and standard of proof. The jury may be expected always to decide the question of patent validity based on the facts presented to it, regardless of whether the PTO considered the same question earlier based on different (or no) facts. The jury is not called upon to judge the PTO’s thoroughness or accuracy. If the standard of proof does not vary, neither will the jury’s judgment. How could it? By holding that the 1952 Patent Act codified the clear and convincing evidence standard without equally codifying the case law doubting the uniform applicability of that standard, the Court has seriously undermined a key element of the patent system’s logic. The Court has at best created arbitrary doctrine favoring patent owners and at worst created an incentive for dishonest prosecution practices. The applicant controls the patent application process and decides which information to reveal to the PTO and which to withhold. The PTO, badly overburdened with applications[179] and staffed with examiners who are rarely expert in the specific technology at issue, does not always thoroughly examine such questions as novelty and nonobviousness, and never questions a superficially credible allegation of utility or inventorship. The clear and convincing evidence standard gives applicants a perverse incentive to obtain a patent fraudulently and enjoy special legal protection for their misdeeds. Recently, the Federal Circuit managed to compound the errors of the i4i jurisprudence under the bizarre circumstances of Sciele Pharma v. Lupin Ltd.[180] The patent applicant in that case had received nonobviousness rejections on some of its claims by the PTO examiner. In response, the applicant canceled the claims, effectively creating a public record that the applicant and examiner agreed that the claims were invalid. Nevertheless, by a PTO error, the patent as ultimately issued reinstated the canceled claims.[181] The patentee, instead of seeking a certificate of correction, reexamination, or reissue to correct the error, ignored it. When the patentee attempted to assert these claims in an infringement litigation, the accused infringer attacked the claims as invalid. The claims, the defendant argued, should not benefit from the strong presumption of validity, because they were clearly invalid by public acknowledgement of both the patent applicant and PTO. Nonetheless, the Federal Circuit insisted that admittedly invalid claims should be presumed valid with the elevated standard of proof under i4i.[182] One wonders whether the next doctrinal step will be for courts to uphold as valid claims that the patentee openly admits are invalid during the litigation itself. The consequences for joint inventors alleging unlawful nonjoinder to the patent will be especially unfor­tunate. The flaws in using an enhanced burden of proof for unexamined patent matters are multiplied in the inventorship context by two key differences between the PTO’s inventorship “determination” and any other possible challenge to the patent. First, a charge of nonjoinder or misjoinder does not usually invalidate the patent. The normal remedy is merely correction by judicial amendment of the patent.[183] The consequences of being forced to share patent rights with a coinventor are rarely as dire as the complete invalidation of the patent or its operative claims, and so justify less extraordinary protections for the patentee. Second, as noted, the PTO virtually never questions the applicant’s assertions of inventorship when examining the patent. It relies on the oath of the inventors, which must identify each inventor of the invention for which a patent is sought.[184] Patent applicants seeking to exclude a coinventor for self-serving purposes may have little compunction about misrepresenting inventorship. Because the PTO is well known not to examine inventorship, the chances of discovery are slim unless the patent is challenged in court under the enhanced burden of proof. Even if the patent is challenged, as noted, the result will not be any kind of sanction but mere correction of inventorship. There is no minimally convincing justification for applying the clear and convincing evidence standard to questions of inventorship. What is perhaps most ironic about the strong presumption of validity of the PTO’s fictitious inventorship “determination” is that the Federal Circuit has consistently treated inventorship as a question of law rather than a question of fact. Specifically, it has called the “overall inventorship determination” a question of law, which is “premised on underlying questions of fact.”[185] The court has also observed that the “determination of whether a person is a joint inventor is fact specific, and no bright-line standard will suffice in every case.”[186] These pronouncements suggest some confusion as to the distinction between questions of law and fact. Questions of law necessarily reference the facts that must be present to satisfy the test of legality. Whether these facts are present in any given case is, by definition, not a question of law. The question of fact in the case of joint inventorship is the alleged coinventor’s role in conceiving the invention. The manner of and degree to which a contribution enables or improves the conception of the invention, or the process of conception, are also questions of fact. Whether the facts satisfy the legal standard of significance requires an assessment of how those facts match up to the legal standard—traditionally, a matter within the province of the factfinder, not the court.[187] The Federal Circuit is mistaken, then, in concluding that “whether a person is a joint inventor” is a question of law. The standard of law is indeed a bright line: either the alleged coinventor made a significant contribution to conception of the invention, or she did not. There is nothing in between. The pertinent question of law is typically what kind of contribution to conception is sufficiently “significant” to qualify the contributor as a joint inventor. What the court means in saying that there is no bright line, rather, is that it feels unable to articulate with great precision what is the standard of “significance” as a rule for all cases. The question of law is what the standard of significance should be, not whether any specific alleged coinventor’s contribution is significant. In any case, given that the Federal Circuit has adopted the position that inventorship is a question of law, it follows that the appellate court should give no deference to the PTO’s decision on inventorship and should thus review challenges to inventorship under the normal preponderance of evidence standard. Appellate courts defer to lower tribunals on questions of fact only, while reviewing determinations of law de novo.[188]

II. Relinking Inventorship Analysis to Technology Policy

The previous part revealed how the dysfunction in joint inventorship law can be traced to two key doctrines. The first is the treatment of claims as if they defined the invention rather than merely limited the scope of its patent protection. The second is the strong presumption of validity for all aspects of the patent, including inventorship, in the absence of any cogent policy justification for such a presumption. Combined, these two positions threaten to warp the logic of inventorship and to undermine the incentives for collaborative research. In Part III.A, the alignment of modern inventorship doctrine with the legislative intent underlying the adoption of the 1984 amendments to Section 116 of the Patent Act will be evaluated. Part III.B follows with an analysis of the policy consequences of the current inventorship doctrine and a proposal for how to correct it to conform to legislative intent and the public policy favoring team research.

A. Legislative History of the 1984 Patent Law Amendments Act

Congress faces relatively few constitutional limitations on its ability to grant and enforce patent rights. The journals of the 1787 constitutional convention record no debate on the intellectual property clause, which appeared in the earliest drafts of the Constitution and was unanimously approved and passed. The clause empowers Congress “To promote the Progress of . . . useful Arts, by securing for limited Times to . . . Inventors the exclusive Right to their . . . Discoveries.”[189] It has long been accepted that the clause means approximately what it says; it authorizes Congress to give to inventors an “exclusive” right to practice their inventions[190] for the purpose of encouraging the development of new applied technology (the “useful Arts”). The Constitution thus seems to mandate that patent legislation be interpreted in the manner most likely to advance technology. In pursuing this goal, patents must be granted to the true inventors of the invention patented; the Constitution nowhere authorizes the grant of exclusive rights to persons other than inventors.[191] Both the first Congress and the Supreme Court accordingly interpreted the clause to require the patent reward to be directed to the true inventor of the invention for which a patent is sought.[192] The legislative history of the 1984 PLAA tells much about its policy purposes, and these align well with the Constitution’s mandate. The PLAA began as a pair of bills introduced in the Senate and House of Representatives in 1983 “to make certain clarifications with respect to joint inventors.” The bills in each case merely provided that each inventor “need not have made a contribution to each claim contained in the application.”[193] These early drafts were intended to reverse the judicial “all claims rule,” requiring every collaborator to have contributed to every claim in a patent in order to qualify as an inventor on that patent.[194] The all claims rule forced research teams to struggle to parse out each collaborator’s contribution to each claim and to file multiple patents on closely related inventions, with a resulting risk that some patents would become prior art and preempt others by the same collaborators. The 1983 bills were popular with the patent bar and inventing industries, but these also viewed the bills as an opportunity to codify certain other key pre-Federal Circuit judicial decisions, such as Monsanto v. Kamp.[195] The patent bar lobbied Congress through the American Intellectual Property Law Association (AIPLA) and other bar and industry organizations. On March 15, 1984, AIPLA submitted a proposed amendment to the bills specifically intended to open the door to joint inventorship in the kinds of collaboration not involving direct contact between joint inventors or an equal contribution by each. The amendment’s goals were described in testimony by AIPLA’s president:
Researchers in an organization sometimes work on one aspect of an invention, while others may work on a different aspect. Personnel are continually added to the research team, while others may leave the team. Concepts and development are often generated through brainstorming and cannot accurately be attributed to a particular inventor or inventors. The criteria for joint inventorship, as the amendments to Section 116 would state such criteria, have been judicially recognized.[196]
The main judicial recognition to which the AIPLA president referred was Monsanto v. Kamp. His testimony made clear that paragraphs (1) and (2) of the new Section 116 were adopting the criteria cited by the Kamp district court.[197] The AIPLA report to the House Subcommittee on Courts, Civil Liberties, and the Administration of Justice repeated these concerns:
A research project in today’s environment may include many inventions and some inventions may have contributions which are made by some individuals who were not involved in other aspects of the invention. It is appropriate to recognize the contribution of each individual even though the individual may not have been involved in, or may not have contributed to, all aspects of the invention.[198]
High-ranking PTO officials supported the AIPLA amendments with testimony that paralleled or duplicated the AIPLA rationales for amending Section 116.[199] In his written statement to Congress, the Commissioner of Patents echoed AIPLA’s point about the difficulty of pinpointing the source of inventive contributions during collaborative research:
Scientists or researchers in an organization often work on a particular aspect or embodiment of the invention, or on only a portion of the invention, while others work on different aspects, embodiments or portions. Scientists are continually added to a research team, while other scientists leave the team. Concepts and development plans generated through brainstorming cannot always be accurately attributed. . . . H.R. 4527 would eliminate the need for making these sometimes chancy, complex and time-consuming determinations by specifying that joint inventors need not have contributed jointly to each claim in an application. As we understand the provision, inventors would also be regarded as joint inventors whether or not they physically worked together at the same place or at the same time in developing the invention. Further, joint inventorship would not require that each inventor make the same type or amount of contribution to the invention or that each make a contribution to the subject matter of each claim of the patent.[200]
The Commissioner was mistaken in asserting that the amendment would “eliminate” the need to make a determination of inventive contribution, because, after the amendment, the applicants would still have to determine whether each collaborator listed as an inventor had made the threshold contribution to conception. The amendment would, however, create a more uniform standard and lower threshold for qualifying as a joint inventor, which would reduce the risk of nonjoinder and misjoinder. Identical statements were made by the Commissioner of Patents before the Senate Judiciary Committee.[201] Industry representatives also testified overwhelmingly in favor of the amendments.[202] Representatives of industries in which collaborative research is extremely common if not critical to commercial success, such as industrial chemicals and pharmaceuticals, supported the bill on these grounds. The American Chemical Society’s letter of support asserted:
The proposed modification of Section 116 of 35 U.S.C. is appropriate and just, for it recognizes that much research that results in an invention is conducted on a team basis. Team members may each contribute to a significant stage of the research, but seldom does each team member contribute to each stage. The ACS supports this modification for it removes the inequity of depriving an individual of the status of joint inventor when that person was a significant contributor to an invention.[203]
In response to these suggestions, the sponsor of House Bill 4527 (Rep. Kastenmeier) let that version die in committee and introduced a new bill, H.R. 6286, in November 1984 containing AIPLA’s language verbatim.[204] In doing so, Kastenmeier also mentioned the new bill’s expected benefits to “universities and corporations which rely on team research”[205]—a purpose also reflected in the official analysis of the bill.[206] The Senate bill was similarly amended, with a parallel explanation of the purpose of the bill described as recognizing “the realities of modern team research.”[207] The report of the post-conference bill made identical references to these “realities” and made explicit that paragraphs (i) and (ii) adopted “the rationale of decisions such as Monsanto v. Kamp.”[208] The President’s statement on signing the bill into law, too, emphasized its utility for removing technical obstacles to team research.[209] With strong industry support and minimal opposition to the bills, Congress enacted the amendment to Section 116 in the precise words proposed by AIPLA.[210] The legislative history indicates that the primary purpose of the PLAA was to codify a flexible standard of joint invention in order to encourage team research and, ultimately, efficient technological development. Congress evidently perceived no coherent policy purpose in forcing large research companies and university science and engineering laboratories to conform to arbitrary rules about which researcher may participate on a team without precluding a valid patent or at least raising the costs of patenting. Ideally, the sole determinant of research collaboration should be the most efficient use of research personnel and resources. Interfering with the research efficiency hardly advances the patent law’s purpose of fostering technological development. With its exclusive concern the promotion of collaborative research, the PLAA amendment to Section 116 left the substantive definition of conception untouched. The amendment does confirm that different collaborators may make different types and amounts of contributions without losing their status as coinventors, and it does not limit such contributions to any specific subclass of inventive input. Beyond that, the PLAA had nothing to say about the standard of inventorship. Yet, the Federal Circuit soon began interpreting inventorship in a manner that privileges certain kinds of inventive contributions while dismissing other, equally valuable contributions. In Ethicon v. U.S. Surgical Corp.,[211] the court’s analysis focused on whether the putative inventor’s contribution appears in the claims. The court qualified this by stating elsewhere in the case that coinventorship depends not on a contribution to the claim itself or specific language in the claim, but rather on whether the collaborator “contributed to the invention” defined and limited by a claim.[212] Nonetheless, the court in Trovan reinforced the doctrinal fixation on claims by laying out a formal procedure for determining inventorship that seemed to rigidly ignore every inventive contribution not reflected in specific claim language.[213] As discussed, Ethicon and Trovan set forth a procedure for determining inventorship. However, the Federal Circuit has never explicitly held that an inventor’s contribution must appear expressis verbis in the claims. In both Ethicon and Trovan, the inventor had allegedly contributed specific, identifiable limitations or elements to a combination invention. In Trovan, the alleged contribution was the addition of a wire support feature to an electronic transponder. In Ethicon, the alleged contribution was the addition of a safety feature (a “detaining means”) to a surgical trocar. In both cases, the contribution was a specific physical feature integrated into the invention. Such features are almost invariably set forth explicitly in the claims, and the Federal Circuit accordingly focused on whether the alleged coinventor’s contribution was encompassed by the claims. The contribution to the claims in such cases served as a satisfactory proxy for an analysis of whether the inventor contributed to the invention itself. In neither case did the Federal Circuit repudiate the importance of contributing to the invention when that contribution was not recited explicitly in the claims. Indeed, the court has recognized that a coinventor must contribute to the conception of the “invention” rather than the claims.[214] The picture that emerges from these cases is one of a jurisprudence mired in confusion about the critical distinction between the claims and the invention. The most problematic aspect of Trovan analysis, which has become the normal procedure for determining inventorship, is its elision of the distinction between claims and invention through the inapposite analogy of inventorship to infringement analysis. Infringement analysis is not a good parallel to inventorship analysis for several reasons. Most importantly, the two analyses serve entirely different policy purposes. Infringement analysis focuses tightly on the claims because the Patent Act’s disclosure requirements limit the scope of the patent. In the infringement context, focusing on the claims makes good sense. The claims are meant to limit the inventor’s enforcement powers. By definition, infringement means that the accused product or process has trespassed on one or more of the specific claims. It follows that infringement analysis should revolve around the claims. However, even in the infringement context, the claims are not the only or even last word on the scope of patent protection. Courts have generally recognized that, as important as claims are to determining whether patent infringement has occurred, they do not exist in a vacuum, but rather should be interpreted in the context of the invention as a whole, as described in the specification and prosecution history.[215] Moreover, using the doctrine of equivalents, courts have expanded patent protection to encompass more of the invention than the inventor technically claimed.[216] Why should the patent owner have the right to enforce the patent against a product or process not explicitly claimed in that patent? The answer can only be that the doctrine is designed to protect the invention when the claims do not fully capture that invention. Such a theory is possible only if the invention is something greater than the claims. The theory underlying the doctrine of equivalents is precisely that the invention and the claims are different but overlapping concepts. The need for claims relates especially to their public notice function. Claims serve the purpose of putting the public on notice of what practices the patent forbids, while the patent document as a whole also teaches the public how to practice the invention. An infringement determination is the precondition to enforcing the patent, and so focusing on the claims protects accused infringers against being subjected to damages and penalties that were unforeseeable based on the claim language. Limiting infringement actions to the claims requires the inventor, as the person with the most control over the phrasing of the patent application, to define as clearly as possible the scope of the invention so that competitors will be given fair notice of the subject matter protected by the patent.[217] Accordingly, infringement (at least, literal infringement) means intrusion on the claims by definition. If the accused infringer’s activities did not fall within the scope of the claims as reasonably interpreted, and did not equivalently infringe, the accused infringer need fear no sanctions. In contrast, an inventorship determination is designed neither to protect the public’s expectations about which technologies they are free to use nor to maintain the patent’s value against threats by competitors. It is instead intended to encourage innovation and, in the context of Section 116 of the Patent Act, to preserve the incentives for collaboration between persons having complementary skill sets or whose research work otherwise synergizes well. Accordingly, Section 116 of the Patent Act nowhere says or implies that the inventive contribution must appear in the claims portion of the specification. Getting inventorship right encourages inventive activity by rewarding those persons who make a significant contribution to the invention and serves equity by ensuring that no collaborator free-rides off another collaborator’s important contribution. For these purposes, the appearance of a collaborator’s contribution in the claims may indicate that the collaborator is a bona fide coinventor if the contribution is significant, goes to conception, and does more than recite the state of the art. Nothing in the pre-1984 case law or the PLAA suggests the adoption of a change in the substantive standard of inventorship to exclude any collaborator whose contributions do not appear in the claims. The policy rationale for not limiting inventorship analysis to the claims is further explored in the next section. Unfortunately, the clarity of the jurisprudence on joint inventorship has been further degraded by imprecise language in American Board of Education v. American Bioscience, Inc.,[218] where the Federal Circuit suggested that a collaborator could not be a joint inventor of a class of chemical compounds unless it “conceived” one “of the claimed compounds.”[219] These dicta do not actually reflect the trend of Federal Circuit jurisprudence on the whole. The court stated in Ethicon that each joint inventor “needs to perform only a part of the task which produces the invention,”[220] as provided in Section 116 of the Patent Act. In Fina Oil, the court was even clearer: “One need not alone conceive of the entire invention, for this would obviate the concept of joint inventorship.”[221] Imagine the probability of all joint inventors simultaneously forming identical pictures in their individual minds of the entire operative invention.

B. Reconstructing Inventorship

It will be helpful here to summarize the main points of the argument so far. Claims limit the protectable aspect of the invention; they do not “define” the invention, as courts frequently assert. Inventorship analysis cannot be analogized to infringement analysis, as the Federal Circuit has erroneously suggested in Ethicon, Trovan, and a long line of subsequent cases. It is unnecessary for a collaborator on a research team to have contributed specific language to a claim in order to qualify as a joint inventor, so long as he or she satisfies the three Pannu conditions. These observations lead to the conclusion that a collaborator who makes a significant contribution to the conception of an invention may qualify as a coinventor without the need for Trovan analysis. More specifically, a collaborator who makes a sufficient contribution to the utility or nonobvious aspect of the invention, or to the method of making or using it, may qualify as an inventor even if his contribution does not appear in the claims. Without such contributions, there might have been no invention at all, and therefore no patent with claims to enforce. Claims limit the patentee’s enforcement powers; they do not, however, suffice to qualify the invention for patent protection. That protection is contingent on the invention’s novelty, nonobviousness, utility, and adequate disclosure. Unless the coinventors collectively satisfy all of these requirements, no patent may issue in the first place. A uniform requirement to conduct Trovan inventorship analysis, in con­fusing the claims with the invention, undermines this dynamic. If the invention were nothing more than the claims, the analogy to infringement analysis would make sense. But the invention is different, sometimes radically different, from the claims. The resulting merging of the claims with the invention disenfranchises any coinventor whose contribution the patent applicant did not expressly describe in the claims as a limitation, no matter how important that contribution might have been. This section will discuss how a correction of the judicial understanding of “invention” can point the way toward a correction of determinations of joint inventorship. With such a correction, the incentives toward collaborative research and disincentives toward strategic nonjoinder of inventors can be restored.
1. Equity and Strategic Nonjoinder
Dan Burk and Mark Lemley have explained some practical consequences of overemphasizing the claims and ignoring the invention as a whole. As they point out, patent applicants can draft claims narrowly in order to get around objections by the examiner and to obtain a patent, and then later interpret the claims broadly in light of the specification or through the doctrine of equivalents:
The shift in focus from the invention to the claim language allows both sides to game the process. It permits—and indeed even encourages—over claiming by patentees, particularly patentees drafting or interpreting claims years after the invention itself. If the focus is on the language of my claims, not the product that I actually built or described, I can interpret the language creatively to claim, in retrospect, to own inventions that I didn’t have in mind when I wrote the patent claims.[222]
The result is to undermine the disclosure and public notice function of Section 112 of the Patent Act. The patent bar is well aware of these consequences and capitalizes on them as a matter of course. Although some patent lawyers avoid overclaiming to strengthen the patent against potential challenges, most seek to draft claims broadly by default.[223] With regard to inventorship determinations, the focus on the claims would allow legally sophisticated researchers to exclude naïve collaborators from inventorship by drafting claims in a manner that underplays or ignores the latter’s crucial inventive contributions, without which the patent might have been unobtainable. Consider the following scenario:
{3} Three scientists, Delta, Epsilon, and Zeta, jointly begin a research project for the discovery of a flexible, bulletproof fabric. Through extensive testing, Delta and Epsilon rule out hundreds of fiber candidates and suggest several molecular and material characteristics that a bulletproof fabric would have, with clues to how one could fabricate such fibers. With this guidance in hand, Zeta successfully creates the material.
In this case, all three collaborators qualify as joint inventors on the resulting patent. The contributions of each are ex hypothesi more than mere recitations of well-known principles and are plainly significant to the invention’s novelty, nonobviousness, and utility. The Federal Circuit has accurately observed that “a person is not precluded from being a joint inventor simply because his or her contribution to a collaborative effort is experimental,”[224] so the fact that Delta and Epsilon were not involved in the final steps in inventing and crafting the patented material does not deny them joint inventor status.[225] Yet, suppose Zeta, who is the only inventor in possession of a complete mental picture of the fabric, were to apply for a patent without crediting Delta and Epsilon as coinventors. Zeta could rely on Trovan and its progeny to support his self-identification as the sole inventor of the material. Focusing on the claims, a court would compare the inventive contributions of Delta and Epsilon to the claim language. Nothing in the claims would necessarily point to the contributions of Delta and Epsilon, especially if Zeta drafted the claims strategically. The claims would certainly not mention anything about the fiber candidates that Delta and Epsilon had so painstakingly ruled out; claims describe what the applicant regards as the invention, not what the applicant regards as irrelevant or nonfunctional. Yet, much of the inventive process is typically “negative discoveries” that winnow the paths toward the inventive goal to the smallest number possible.[226] Nor would the claims necessarily refer to the material’s bulletproof characteristics; claims (especially in the chemical, metallurgical, and pharmaceutical arts) commonly describe an invention without reference to its utility or purpose. This information would most likely appear in the written description portion of the patent. The reason for its omission from the claims makes sense in the context of patent law, where a patent grants exclusionary rights to the patented invention used for any purpose, regardless of whether it was a purpose contemplated by the inventor. Patent applicants accordingly omit reference to the utility and other details of the invention in the claims to avoid narrowing the scope of the claims. The motivation of seeking a patent that covers unforeseen uses of, or methods of making, a patented product is logically unconnected to, and has no effect on, the determination of inventorship of that product. The utility of an invention, and the knowledge of how to make and properly use it, are essential aspects of the invention, even if they do not necessarily limit its protections. As the only collaborator with knowledge of the material’s molecular structure and fibers, and perhaps the only one with knowledge of how to fabricate the material, then, Zeta may rely on a great deal of Federal Circuit precedent to arrive at precisely the wrong conclusion—that he is the sole inventor.[227] The effort and creativity that Zeta put into the invention may be dwarfed by those of his collaborators, but the logic of Trovan analysis, strictly applied, awards him the entire patent right anyway. This problem does not arise as readily in patents covering mechanical devices or processes, because a coinventor will usually have contributed an identifiable physical component to a device or a step to a process. For example, in Ethicon v. U.S. Surgical Corp., the alleged coinventor of surgical trocar could point to a specific safety means limitation, incorporated into the relevant claim, that he contributed to the trocar.[228] Moreover, mechanical device patents frequently specify their utility in the claims themselves,[229] so a collaborator’s contribution to utility would survive Trovan analysis. But no great imagination is required to envision hypothetical cases posing the same problem in the mechanical engineering field. For example, a novel device invented by Iago may be obvious over the prior art until Charles discovers a nonobvious use for the device.[230] Or a novel manufacturing process invented by Ilsa may be nonfunctioning when applied to certain materials, but if Calvin discovers that the process functions well when applied to a nonobvious choice of materials, Calvin is a coinventor. The inventive contributions of Charles and Calvin rendered their respective inventions patentable, but those contributions need not appear in the claims. The arbitrary exclusion of some inventors caused by overreliance on the claims creates special problems in materials, chemical, and pharmaceutical development. As noted, patents in these fields typically claim the compound itself with no mention of its utility. Discussion of utility is relegated to the written description portion of the patent for such inventions.[231] Claims that omit any mention of the invention’s utility, or the method of making and using it, highlight the merely partial role played by the claims as a definition of the invention. They also prevent a collaborator who discovers a compound’s utility from qualifying as an inventor under a strict Trovan analysis. Pharmaceutical inventions illustrate this point with special aptness. In developing a new pharmaceutical invention, the scientific problem is not to create a novel compound in a vacuum. A pharmaceutical compound, once created, does not “do” anything; it has no utility outside of its target biological organism. In order for a patent covering such a compound to issue, the applicant must credibly allege that the compound has some biochemical effect on a target organism. In other words, a pharmaceutical compound is not a functioning invention until introduced into its target organism. The claims, to the extent that they merely recite some molecular structure without mention of a therapeutic utility, limit the patentable scope of the invention but cannot “define” it in any meaningful sense. The incomplete role of the claims in “defining” pharmaceutical inventions reveals its significance clearly in any description of the inventive process. The process of conceiving and synthesizing the claimed compound is often a relatively small part of drug design. In order to complete conception of the compound through trial-and-error testing, the inventors may need to perform all of the following steps: (1) understand the biological mechanism or agent causing the disorder; (2) select or develop a target chemical molecule (“candidate”) potentially capable of treating the disorder; (3) develop an efficient method for producing the candidate on a mass scale; (4) develop or select a series of systems for screening the candidate against the biological mechanism, or an environment that simulates the mechanism; and (5) modify the candidate as needed to increase its potency; reduce toxicity, side effects, or contraindications; reduce the molecular half-life; and determine proper dosage and delivery method.[232] As a result, pharmaceutical inventions usually result from team research involving biologists, biochemists, medicinal chemists, and pharmacologists performing different tasks. Inventors may have contributed to any of these steps. In some cases, the biological mechanism underlying the disorder to be treated may be well understood in the art, but the kind of molecule that will remedy the biological disorder will be the subject of uncertainty. In other cases, the biological mechanism will be mysterious or misunderstood, and the conception of the invention may require unique insights into the biological source of the disorder. The latter class of inventions is by far the most challenging to develop and perfect; all pharmaceutical inventions require the selection, testing, and possible modification of a bioactive compound, but not all require extensive original research into human or other animal biochemical functioning. Yet, because the claims will typically recite only the chemical structure of the molecule, an inflexible Trovan-type analysis would arbitrarily privilege the second and fifth steps over the others, despite the fact that in any given case the former may require the least effort, expertise, creativity, or other inventive contribution.[233]
2. Restoring Incentives for Collaborative Research
Claim fixation results in an inventorship analysis obviously inequitable to the excluded inventors, but it also poses a threat to collaborative research. Often, private companies seek to collaborate with university scientists and engineers to develop innovative products. For example, drug developers frequently collaborate with university biologists and chemists to discover diagnostics and treatments for persistent human diseases and disorders. University researchers are typically involved in exploring human or other animal biology or biochemistry, while the drug companies have expertise in medicinal chemistry and pharmacology. In such collaboration, Trovan analysis seems to suggest that the university researchers could never qualify as inventors if they did not contribute some specific molecular arrangement to the claimed compound. The patent incentive to collaboration with drug companies disappears. An interesting illustration of the flaws in Trovan analysis occurred in a previously mentioned case, Board of Education ex rel. Florida State University v. American Bioscience.[234] There, FSU scientists and their post-doctoral fellow, Dr. Tao, had been experimenting with variants of paclitaxel, a naturally-occurring anticancer compound. At the FSU lab, Tao learned a “semi-synthetic” process for producing paclitaxel analogs. During this time, FSU scientists undertook a research project to develop nitro-taxols as chemo­thera­peutic radio­sensitizing taxanes. Within a relatively short time, the FSU scientists discovered that a specific compound they had synthesized, PNIP, showed promise as a radiosensitizing agent. One of the FSU scientists (Dr. Nadizadeh) had developed a “secret” method for synthesizing PNIP, of which Tao was informed. At some point during this research, one of the FSU scientists spoke at a conference regarding the synthesis of paclitaxel. Scientists at a pharmaceutical developer, VivoRx, had attended and decided to start researching radiosensitizing agents using a variant of paclitaxel, known generically as docetaxel, as the parent structure. Soon after having begun research, VivoRx hired Tao to assist in the project, and with his help in synthesizing compounds (allegedly using Nadizadeh’s secret method of making nitro-taxols), VivoRx obtained patents on several docetaxel analogs. In evaluating whether any FSU scientists qualified as joint inventors of the patented compounds, the Federal Circuit discounted FSU’s experimental work: “general knowledge regarding the anticipated biological properties of groups of complex chemical compounds is insufficient to confer inventorship status with respect to specifically claimed compounds.”[235] The court then observed that “invention does require conception, and there is no evidence that FSU’s inventors conceived any of the claimed compounds.”[236] With regard to the use of FSU’s methods to make the compounds, the court concluded that this was irrelevant, because the patent did not claim the method of making the compounds.[237] Specifically, FSU alleged that Tao had used Nadizadeh’s secret method to synthesize the patented compounds. To this, the court replied:
despite the fact that Nadizadeh may have developed a method of making PNIP and other taxol derivatives, the record in the present case indicates that he did not conceive the claimed compounds; only ABI’s inventors were in possession of both the structure of the claimed compounds and an operative method of making those compounds.[238]
Although the court acknowledged that conception of an invention requires disclosure of the method of making it, it held that the inventor of the method could not be a coinventor of the compounds made by his method unless he “conceive[d] of the claimed compounds” himself.[239] The American Bioscience court’s focus on the “claimed compounds” and discounting of the (unclaimed) method of making them is symptomatic of claims fixation. Whether FSU scientists conceived of an entire claimed compound is not the pertinent question; the patented compounds were conceived by at least some of the inventors. The FSU scientists need only have contributed to the conception of the invention in some way; conception of the entire compound by any specific inventor is entirely irrelevant, as long as the research team collectively conceived the claimed compounds. A proper analysis of inventorship would have acknowledged that the contribution of necessary background biological knowledge, useful suggestions for functional groups of the claimed compounds, and techniques for making the compounds, could all qualify the contributors as coinventors if the contributors had collaborated with the VivoRx team and satisfied the three Pannu conditions of coinventorship. Under this analysis, it is possible that some of the FSU scientists would not have qualified as joint inventors. It is also possible that some would have qualified. The court’s repeated distortion of the law of inventorship in the case did considerable mischief on the facts before it, where FSU scientists who may have contributed substantially to the VivoRx inventions were excluded from consideration as joint inventors. In effect, the Federal Circuit sanctioned what may have been the misappropriation of a university’s valuable research, gathered over the course of many years, by a private company that promptly commercialized it for its own benefit and the preemption of any benefit by the university scientists. More troubling still, the decision has had a predictably perverse effect on the development of joint inventorship doctrine. For example, it misled the district court in Vanderbilt University v. ICOS Corp. into concluding that failure of any alleged collaborator to independently conceive of the entire claimed invention negated any possibility of joint inventorship.[240] Very recently, the Federal Circuit has retreated from the American Bioscience reasoning and partially rectified the wayward doctrine. In Falana v. Kent State University, the plaintiff, a university researcher, developed an original method of making certain compounds for use in liquid crystal displays.[241] After the plaintiff (Dr. Falana) left the university’s employment, his supervisor used Falana’s method to synthesize a compound that he patented without naming Falana as a coinventor. The university read American Bioscience to dictate that, because it had not sought to patent Falana’s method, and Falana had never conceived the patented compounds, he was not a coinventor of any compound resulting from the use of his method. The Federal Circuit disagreed with this understanding of American Bioscience.[242] Actually, the university’s reading of the case was quite reasonable. The American Bioscience opinion was plainly dismissive of a researcher who had contributed a method of making the patented compound without conceiving of the compound itself. To get past American Bioscience without reversing itself, the court had to somehow reconcile its own holding with the fact that Dr. Falana, like Dr. Nadizadeh in American Bioscience, had never made or seen the university’s patented compounds. To do this, the court resorted to the time-honored tactic of distinguishing the cases on the facts. In American Bioscience, Nadizadeh never actually worked with the patentee’s inventive team directly; he worked with Tao, who in turn worked with the patentee to develop the compound. In Falana, the eponymous plaintiff worked directly with the university research team, although he left before the compound was discovered. This distinction is, of course, irrelevant; the Patent Act does not require that all joint inventors work together, merely that they are exposed to each other’s work in the course of some cooperative enterprise. The Falana court also suggested that, in American Bioscience, Nadizadeh had simply “taught skills or general methods” that merely facilitated the later invention “without more.”[243] But this too is irrelevant; the operative question is whether the skills and methods taught by the collaborator (a) contributed more than well-known principles, and (b) made a significant contribution (c) to conception. As the court itself recognized in Falana, “Where the method requires more than the exercise of ordinary skill . . . the discovery of that method is as much a contribution to the compound as the discovery of the compound itself.”[244] Problematic as the distinction between the cases may be, it is salutary that the Federal Circuit is beginning to perceive a more accurate concept of invention. The misfortune is that, in not reversing the mistakes of American Bioscience unambiguously, the court leaves standing contradictory approaches to the role of claims in determining inventorship. Also problematical is the potential of claims-fixation to lead to an oxymoronic “inventorless invention.” A slight extension of scenario {3} demonstrates how this paradox might arise:
{4} Eta, Theta, and Iota are collaborating to discover a cure for lung cancer. Through diligent and unusually insightful research, Eta discovers a hitherto unknown biochemical mechanism that makes cancer cells susceptible to certain kinds of polypeptides. Theta invests several months into creating a series of nonobvious screening systems for testing compounds against the discovered vulnerability in cancer cells. Together, Eta and Theta propose to Iota, a medicinal chemist, the characteristics that such a compound should have. Following these instructions, Iota pulls a well-known drug compound off his shelf, modifies it into a series of novel derivative compounds using well known techniques, and runs them through Theta’s screening systems. One of the derivative compounds turns out to be a wonder drug that cures lung cancer with a very high probability in humans. Together, they seek a patent on the compound only.
Under the facts of scenario {4}, neither Eta nor Theta ever pictured in their minds the finished compound. Nor did they contribute any specific physical structure to the claimed molecule. They lacked the requisite knowledge of medicinal chemistry to do so. Neither Eta nor Theta would be coinventors under a strict Trovan analysis or the American Bioscience approach, because none of their contributions appear in the claims. Iota synthesized the patented molecule, and was the first and only collaborator to conceive it in his mind, but nothing in his contribution exceeded the state of the art of medicinal chemistry. His work was the least inventive of the three, and arguably did not rise to a contribution to conception. If the claims merely recite the molecular structure of the claimed compound, the result will be that an invention of undoubted novelty, nonobviousness, and utility, might have no inventor at all.[245] Under an expansive interpretation of Trovan, the cure for cancer described in scenario {4} could well be doctrinally unpatentable.[246] The result is diametrically opposite to any patent policy of encouraging collaborative research and the equitable sharing of its benefits.

Conclusion

The argument of this Article—that the concept of invention and inventorship has been erroneously construed as “defined” by the claims—has relied on an argument sufficiently complex to proceed in stages. First, I showed that the invention cannot be defined by the claims, because the claims do not necessarily express all the prerequisites for a patentable invention. These omitted elements include most typically utility, a method for making the invention, and a method for using the invention. Without these, no novel and nonobvious product or process has yet attained the status of an “invention” ready for patenting. Therefore, a research collaborator who contributes one or more of these aspects to an invention qualifies as a co-inventor even if his contribution does not appear in the claims. Moreover, treating the claims as defining the invention contradicts both the unity of invention rule and the rationale underlying the doctrine of equivalents. The Trovan sequence of analysis relied upon by the Federal Circuit consequently cannot be the only valid method for determining inventorship. Second, I explained how the consequences of fixating on the claims undermined Congress’ clearly expressed intentions in amending Section 116 of the Patent Act to encourage collaborative technological research. Third, I discussed the aggravating effects of two erroneous and contradictory doctrines adopted by courts: (a) the strong presumption of validity accorded to all aspects of an issued patent, regardless of whether the PTO examined the aspect at issue, and (b) the treatment of inventorship as a question of law. Finally, I suggested how inventorship doctrine could be rectified, most generally by treating the claims as defining the outward limits of patent protection for the invention rather defining than the invention itself. Procedurally, courts should hew to Falana analysis and resort to Trovan analysis only when the alleged co-inventor’s sole contribution appears expressis verbis in the claims. Until the Supreme Court or Federal Circuit decisively clarifies the distinction between the invention and the claims, and the role of joint inventors in the former, the law of inventorship will continue to impose unnecessary costs on collaborative research and to skew inventorship determinations in favor of legally sophisticated but unprincipled researchers to the disadvantage of their less urbane collaborators. The problems resulting from claim fixation are further aggravated by adherence to an unjustifiably strong presumption that the persons named as inventors in the patent are the true and only inventors. There is no statutory basis for a strong presumption of validity on matters unexamined by the PTO, and no cogent rationale for judicially imposing a “clear and convincing evidence” standard. Because all coinventors have equal and nonexclusive ownership rights in the patent by default,[247] the stakes for nonjoinder are potentially very high. Absent an assignment agreement, all inventors own rights to all claims in their patent, regardless of whether they contributed to those claims in any way. The Federal Circuit’s decision in Ethicon giving inventors ownership rights to claims to which they did not contribute was perhaps inevitable given the unity of invention principle and the phrasing of Sections 116 and 262 of the Patent Act, but it had the unfortunate consequence of increasing the risks of collaborative research outside of an employment contract. Collaborations between private companies, or between research universities and private industry, stand to suffer unnecessarily. Any adjustment of the ownership rights in a patent with multiple inventors would best be accomplished legislatively, but many of the current doctrine’s ill effects can be remedied by rectifying the wayward judicial conception of the invention. The overwrought fixation on claims and strong presumption of validity in inventorship determinations appear nowhere in the Patent Act. They are progeny of the judiciary and correctible by it. The judicial attitude toward overemphasizing claims should mirror that of Bill Cosby’s father to the young Bill: “I brought you into this world, and I can take you out.”[248]
* Professor, Arizona State University College of Law; J.D., Yale Law School; M.A., Yale University; B.A., University of California, Berkeley. The author thanks Dennis Karjala and Mark Lemley for helpful comments on an earlier draft of this article.
[1] 35 U.S.C. § 100(a) (2006). Half of this definition is semicircular, the other half compounds its own obscurity by failing to define a “discovery,” resulting in much confusion and misunder­standing as to the scope of patentable subject matter. See Linda J. Demaine & Aaron X. Fellmeth, Reinventing the Double Helix: A Novel and Nonobvious Conception of the Biotechnology Patent, 55 Stan. L. Rev. 303, 368-77 (2002-2003).
[2] See Grant v. Raymond, 312 U.S. 218, 241-42 (1832); Graham v. John Deere Co., 383 U.S. 1, 5-8 (1966).
[3] 35 U.S.C. § 102(a), (c) (2006). The America Invents Act substitutes the term “claimed invention” for “invention,” and defines that term in the new Section 101(j). See Leahy-Smith America Invents Act, § 3, Pub. L. 112-29, 125 Stat. 284 (codified in scattered sections of 35 U.S.C.). The implications of this change will be analyzed below.
[4] 35 U.S.C. § 103(a) (2006).
[5] 35 U.S.C. § 271(a) (2006).
[6] U.S. Const. art. I, §. 8, cl. 8; see also 35 U.S.C. § 111(a)(1) (2006) (“An application for patent shall be made, or authorized to be made, by the inventor, except as otherwise provided in this title . . . .”).
[7] America Invents Act, supra note 3, § 3, to be codified as 35 U.S.C. § 101(f).
[8] Id.
[9] 35 U.S.C. §§ 101-03 (2006). To supplement the Patent Act, it might be thought useful to look to treaties mandating standards of patent protection among state parties. The most relevant such treaty, the WTO Agreement on Trade-Related Aspects of Intellectual Property Rights, does not define “invention,” but it does specify that to qualify for a patent, an invention must be “new, involve an inventive step, and [be] capable of industrial application.” TRIPS: Agreement on Trade-Related Aspects of Intellectual Property Rights art. 27(1), Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization [hereinafter WTO Agreement], Annex 1C, The Legal Texts: The Results of the Uruguay Round of Multilateral Trade Negotiations 320 (1999), 1869 U.N.T.S. 299 [hereinafter TRIPS Agreement]. Unfortunately, this adds nothing to our understanding of what constitutes an invention.
[10] 35 U.S.C. §§ 112, 115 (2006); 37 C.F.R. § 1.17. The TRIPS Agreement also requires a written disclosure, although it does not require all of the information provided for in the 1952 Patent Act, such as a best mode disclosure. Compare TRIPS Agreement, supra note 9, art. 29(1) with 35 U.S.C. § 112 (2006) (amended 2011).
[11] Intuitively, we might be tempted to draw a distinction between substantive and procedural requirements of patentability, with only the former qualifying as components of the definition of invention. As will be discussed, this approach would miss some core features of invention that do not clearly fall into the category of substantive criteria.
[12] The PTO’s Manual of Patent Examining Procedure does not even contemplate examination of inventorship. See Manual of Patent Examining Procedure ch. 700 (8th ed., July 2010 rev.), available at http://www.uspto.gov/web/offices/pac/mpep/index.htm [hereinafter MPEP].
[13] It may also arise in a stage of administrative review, such as in derivation proceedings conducted under 35 U.S.C. § 135, or possibly post-grant review conducted under 35 U.S.C. ch. 32.
[14] America Invents Act, supra note 3, § 118, to be codified at 35 U.S.C. § 112(b).
[15] See Phillips v. AWH Corp., 415 F.3d 1303, 1312 (Fed. Cir. 2005) (citing many Supreme Court and earlier Federal Circuit cases for the proposition that the claims measure the invention). However, one writer has rightly pointed out that excessive fixation on the claims tends to encourage patent applicants to claim more than they have enabled. Oskar Liivak, Rescuing the Invention from the Cult of the Claim, 42 Seton Hall L. Rev. 1 (2012); see also Christopher A. Cotropia, What Is the “Invention”?, 53 Wm. & Mary L. Rev. 1855 (2012) (discussing the functional advantages of defining an invention more broadly than as the claims).
[16] An example is seen in this case study of the typical process from invention to patent. Greg Myers, From Discovery to Invention: The Writing and Rewriting of Two Patents, 25 Social Studies of Science 1, 57, 69 (1995).
[17] See Robert P. Merges & Richard R. Nelson, On the Complex Economies of Patent Scope, 90 Colum. L. Rev. 839, 848 (1990) (noting that current practice permits the stretching of claims beyond what the inventor has disclosed).
[18] See George Franck, Scientific Communication — A Vanity Fair?, 286 Sci. 53, 53 (1999); Edward T. Lentz, Inventorship in Laboratory Research, in Understanding Biotechnology Law 187, 188 (Gale R. Peterson ed., 1993).
[19] See John R. Allison & Mark A. Lemley, Who’s Patenting What? An Empirical Exploration of Patent Prosecution, 53 Vand. L. Rev. 2099, 2117 (2000).
[20] See Part I.B.1, infra.
[21] 35 U.S.C. § 262 (2006); see Lawrence M. Sung, Collegiality and Collaboration in the Age of Exclusivity, 3 DePaul J. Health Care L. 411, 425 (2000).
[22] Drake v. Hall, 220 F. 905, 906 (7th Cir. 1914). The right of joint patent owners contrasts with common law rights of joint owners of other kinds of property. Tenants in common of real property, for example, have a duty of accounting for and sharing profits with other tenants. See 20 Am. Jur. 2d Cotenancy and Joint Ownership § 42 (2012). Even joint copyright owners share an undivided interest in the work. Childress v. Taylor, 945 F.2d 500, 505 (2d Cir. 1991); Weinstein v. University of Ill., 811 F.2d 1091, 1095 (7th Cir. 1987).
[23] “May” lose, because the inventor may be entitled to some compensation by prior contractual arrangement.
[24] See, e.g., Arlington Indus., Inc. v. Bridgeport Fittings, Inc., 632 F.3d 1246, 1256 n.2 (Fed. Cir. 2011); In re Hiniker Co., 150 F.3d 1362, 1369 (Fed. Cir. 1998); Hilton Davis Chem. Co. v. Warner-Jenkinson Co., 62 F.3d 1512, 1539 (Fed. Cir. 1995) (Plager, J., dissenting); Two-Way Media LLC v. America Online, Inc., 508 F. Supp. 2d 526, 528 (S.D. Tex. 2007).
[25] Giles S. Rich, The Extent of the Protection and Interpretation of Claims-American Perspectives, 21 Int’l Rev. Indus. Prop. & Copyright L., 497, 499 (1990).
[26] See, e.g., Giles S. Rich, Why and How Section 103 Came to Be, in Nonobviousness: The Ultimate Condition of Patentability 1, 201 (John F. Witherspoon ed., 1980).
[27] Hyde v. United States, 225 U.S. 347, 391 (1912) (Holmes, J., dissenting).
[28] 35 U.S.C. § 101 (2006).
[29] See In re Nuijten, 500 F.3d 1346, 1354 (Fed. Cir. 2007) (“The four categories [process, machine, manufacture, and composition of matter] together describe the exclusive reach of patentable subject matter. If a claim covers material not found in any of the four statutory categories, that claim falls outside the plainly expressed scope of § 101 even if the subject matter is otherwise new and useful.”).
[30] See S. Rep. No. 82-1979 (1952), reprinted in 1952 U.S.C.C.A.N. 2394, 2410 (“Section 102, in general, may be said to describe the statutory novelty required for patentability, and includes, in effect, an amplification and definition of ‘new’ in section 101.”). But see Demaine & Fellmeth, supra note 1, at 385-88 (arguing that the term “new” in Section 101 should be read more broadly to exclude products of nature and natural processes from patentability).
[31] 35 U.S.C. § 102 (2006).
[32] See Nelson v. Bowler, 626 F.2d 853, 856 (C.C.P.A. 1980).
[33] 35 U.S.C. § 103 (2006). Prior to the 2011 amendments, the Patent Act referred to “subject matter” rather than “claimed invention”; there is no indication that this amendment was intended to change the substantive standard of patentability.
[34] See Rich, supra note 25, at 497-99; Demaine & Fellmeth, supra note 1, at 365-67.
[35] The Patent Act applies to “inventions” and “claimed inventions” — it nowhere creates or deals with a concept of “unpatentable inventions.” However, it may be inferred that some “inventions” are unpatentable if the applicant is either not the true inventor or fails to satisfy the procedural requirements in the patent application.
[36] See Invitrogen Corp. v. Clontech Labs., Inc., 429 F.3d 1052, 1063 (Fed. Cir. 2005) (“[C]onception requires that the inventor be able to define the compound so as to distinguish it from other materials, and to describe how to obtain it . . . . In other words, conception requires that the inventor appreciate that which he has invented.”); see also Dow Chem. Co. v. Astro-Valcour, Inc., 267 F.3d 1334, 1341 (Fed. Cir. 2001) (“[T]he date of conception of a prior inventor’s invention is the date the inventor first appreciated the fact of what he made.”).
[37] A distinction should be made between an invention, the novelty of which the inventor has not yet recognized and an invention, the novelty and operative characteristics of which the inventor recognizes, but whose mode of operation the inventor does not fully understand. The latter can qualify as a patentable invention, because understanding precisely how an invention achieves its purpose is not required; this is why patent applicants may frame their claims in “means-plus-function” or “step-plus-function” format. See 35 U.S.C. § 112(f) (2013).
[38] 35 U.S.C. § 112 (2013).
[39] Id. § 112(a).
[40] Id. § 112(b)
[41] See, e.g., Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 150-51 (1989); Kewanee Oil Co. v. Bicron, 416 U.S. 470, 480-81 (1974). The value of disclosure as a basis for the patent grant has been recurrently questioned, however. See, e.g., Brenner v. Manson, 383 U.S. 519, 534 (1966) (Harlan, J., dissenting); Alan Devlin, The Misunderstood Function of Disclosure in Patent Law, 23 Harv. J.L. & Tech. 401, 403 (2010); Mark A. Lemley & Ragesh K. Tangri, Ending Patent Law’s Willfulness Game, 18 Berk. Tech. L.J. 1085, 1100-02 (2003); Rebecca S. Eisenberg, Patents and the Progress of Science: Exclusive Rights and Experimental Use, 56 U. Chi. L. Rev. 1017, 1028–29 (1989). Mark Lemley has argued that the real benefit of the patent to most inventors may be to avoid the risk of being excluded from practicing the technology at issue. Mark Lemley, The Myth of the Sole Inventor, 110 Mich. L. Rev. 709, 755 (2012).
[42] Ariad Pharms., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1351 (Fed. Cir. 2010) (en banc); see In re Wertheim, 541 F.2d 257, 263 (C.C.P.A. 1976).
[43] Burroughs Wellcome v. Barr Labs., Inc., 40 F.3d 1223, 1227 (Fed. Cir. 1994).
[44] Townsend v. Smith, 36 F.2d 292, 295 (Ct. Cust. & Pat. App. 1930); Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1376 (Fed. Cir. 1986).
[45] Burroughs Wellcome, 40 F.3d at 1228; Hiatt v. Ziegler, 179 U.S.P.Q. 757, 763 (Bd. Pat. App. 1973). Reduction to practice means that the inventor actually creates the physical invention (if it is a product) or performs the steps of the invention (if it is a process). Alternatively, an inventor can constructively reduce an invention to practice by filing a valid patent application clearly describing how to make and use the invention. MPEP, supra note 12, § 2138.05.
[46] The Federal Circuit has held that conception requires both understanding the invention’s structure and an operative method of making it. See, e.g., Amgen, Inc. v. Chugai Pharm. Co., 927 F.2d 1200, 1206 (Fed. Cir. 1991). It would be more accurate to add that conception also requires understanding how to use the invention. The Patent Act requires that the invention be useful, 35 U.S.C. § 101. Although, as noted, the Federal Circuit interprets the Act not to require the inventor to know with confidence that the invention is useful, the Act clearly requires that the inventor disclose how to use the invention, assuming it does work. See 35 U.S.C. § 112 (2006). Therefore, conception of a product properly requires knowledge of: (1) the physical structure of the invention, (2) how to make the invention, and (3) how to use the invention. Cf. Space Systems/Loral, Inc. v. Lockheed Martin Corp., 271 F.3d 1076, 1080 (Fed. Cir. 2001) (“To be ‘ready for patenting’ the inventor must be able to prepare a patent application, that is, to provide an enabling disclosure as required by 35 U.S.C. § 112. . . . [If] the inventor himself [is] uncertain whether it could be made to work, a bare conception that has not been enabled is not a completed invention ready for patenting.”).
[47] Burroughs Wellcome, 40 F.3d at 1228.
[48] See, e.g., id. (“[A]n inventor need not know that his invention will work for conception to be complete.”); Cross v. Iizuka, 753 F.2d 1040, 1051 (Fed. Cir. 1985) (holding that in vitro testing of a potential therapeutic compound, though far from establishing actual benefit to any person, suffices to show utility).
[49] Hitzeman v. Rutter, 243 F.3d 1345 (Fed. Cir. 2001).
[50] See Silvestri v. Grant, 496 F.2d 593, 596 (Ct. Cust. & Pat. App. 1974); Invitrogen, Corp. v. Clontech Labs., Inc., 429 F.3d 1052, 1064 (Fed. Cir. 2005) (holding that establishing conception requires evidence that the inventor understood the invention to have the features that comprise the inventive subject matter at issue). Chisum has also interpreted the Supreme Court jurisprudence to hold that “discovery of a practical utility is part of the act of inventing.” Donald Chisum, Chisum on Patents § 2.02[5] (Rel. 110-9, 2007); see also Lentz, supra note 18, at 190 (“Thus, a person who conceives of or actually something, e.g., a new chemical entity, but who does not know of a use for the thing probably has not made a patentable invention and is not, therefore, an inventor unless and until a utility is discovered.”). The European Patent Office seems to interpret the European Patent Convention’s requirement of “susceptibility to industrial application”—the parallel of the utility requirement—as integral to the concept of invention as well. See Michigan State U./Euthanasia Compositions, Eur. Pat. Off. Bd. App., Case No. T 0866/01, para. 4.6 (unrep. May 11, 2005), available at http://www.epo.org/law-practice/case-law-appeals/pdf/ t010866eu1.pdf.
[51] See, e.g., Langer v. Kaufman, 465 F.2d 915, 918 (C.C.P.A. 1972).
[52] Burroughs Wellcome, 40 F.3d at 1228.
[53] See Falana v. Kent State Univ., 669 F.3d 1349, 1358 (Fed. Cir. 2012) (holding that the putative inventor who envisioned the genus of compounds and contributed the method of making it contributes to the conception of that genus); cf. Bd. of Educ. ex rel. Bd. Of Trustees of Fla. State Univ. v. Am. Bioscience, Inc., 333 F.3d 1330, 1342 (Fed. Cir. 2003) (observing that, if some inventors conceived a chemical compound but were unable to make it without the help of another scientist, the assisting scientist might qualify as a coinventor); Fina Oil & Chem. Co. v. Ewen, 123 F.3d 1466, 1473 (Fed. Cir. 1997) (“Conception of a chemical substance requires knowledge of both the specific chemical structure of the compound and an operative method of making it.”).
[54] In re Brana, 51 F.3d 1560. 1566 (Fed. Cir. 1995).
[55] Id.; see also In re ‘318 Patent Infringement Litig., 583 F.3d 1317, 1327 (Fed. Cir. 2009) (“Thus, at the end of the day, the specification, even read in the light of the knowledge of those skilled in the art, does no more than state a hypothesis and propose testing to determine the accuracy of that hypothesis. That is not sufficient.”); Envirotech Corp. v. Al George, Inc., 730 F.2d 753, 762 (Fed. Cir. 1984) (holding that lack of utility is shown when there is a complete absence of data supporting the assertion of utility). But see Univ. of Pittsburgh v. Hedrick, 573 F.3d 1290, 1298 (Fed. Cir. 2009) (“An inventor need not know that his invention will work for conception to be complete . . . the discovery that the invention actually works is part of its reduction to practice.”).
[56] In re Bundy, 642 F.2d 430, 433 (C.C.P.A. 1981).
[57] See MPEP, supra note 12, § 2107.01.
[58] See, e.g., Brana, 51 F.3d at 1566-68 (“Usefulness in patent law, and in particular in the context of pharmaceutical inventions, necessarily includes the expectation of further research and development. The stage at which an invention in this field becomes useful is well before it is ready to be administered to humans.”); Nelson v. Bowler, 626 F.2d 853, 856 (C.C.P.A 1980) (holding that inconclusive and statistically insignificant tests of the utility of a chemical compound were sufficient to establish utility); In re Gazave, 379 F.2d 973, 978 (C.C.P.A. 1967) (“[I]n the usual case where the mode of operation alleged can be readily understood and conforms to the known laws of physics and chemistry, operativeness is not questioned, and no further evidence is required.”).
[59] Merges & Nelson, supra note 17.
[60] On reissue proceedings, see 35 U.S.C. § 251 (2006); see generally 37 C.F.R. § 1.176.
[61] Any part of the invention disclosed in the specification but not claimed is deemed dedicated to the public. See Maxwell v. J. Baker, Inc., 86 F.3d 1098, 1106 (Fed. Cir. 1996). For an interesting exploration of this issue, see generally Robert A. Migliorini, The Dedication to the Public Doctrine and Lessons for Patent Practitioners, 87 J. Pat. & Trademark Off. Soc’y 825 (2005).
[62] See O’Reilly v. Morse, 56 U.S. (15 How.) 62 (1853).
[63] See sources cited supra note 10.
[64] See U.S. Const. art. I, § 8, cl. 8.
[65] Morse, 56 U.S. (15 How.) at 62.
[66] Id. at 112.
[67] Id. at 113.
[68] Id. at 117.
[69] See id.
[70] See 35 U.S.C. § 251 (2006).
[71] Id.
[72] Phillips v. AWH Corp., 415 F.3d 1303, 1315 (Fed. Cir. 2005) (en banc) (quoting Markman v. Westview Instruments, Inc., 52 F.3d 967, 978 (Fed. Cir. 1995)); see also United States v. Adams, 383 U.S. 39, 49 (1966).
[73] See, e.g., Altoona Publix Theatres, Inc. v. American Tri-Ergon Corp., 294 U.S. 477, 487 (1935); Purdue Pharma L.P. v. Endo Pharm. Inc., 438 F.3d 1123, 1136 (Fed. Cir. 2006) (“[I]t is the claims ultimately that define the invention.”); Netscape Commc’ns Corp. v. Konrad, 295 F.3d 1315, 1323 (Fed. Cir. 2002) (“However, it is the claims that define a patented invention.”).
[74] Univ. of Pittsburgh v. Hedrick, 573 F.3d 1290, 1297-98 (Fed. Cir. 2009); see also Ethicon, Inc. v. U.S. Surgical Corp., 135 F.3d 1456, 1460 (Fed. Cir. 1998) (“[T]he critical question for joint conception is who conceived, as that term is used in the patent law, the subject matter of the claims at issue.”) (emphasis added).
[75] Eli Lilly & Co. v. Aradigm Corp., 376 F.3d 1352, 1361-62 (Fed. Cir. 2004).
[76] See Liivak, supra note 15.
[77] Id.
[78] Univ. of Rochester v. G.D. Searle & Co., 375 F.3d 1303, 1306 (Fed. Cir. 2004).
[79] United States v. Adams, 383 U.S. 39, 49 (1966).
[80] Patent Cooperation Treaty, done on June 19, 1970, 28 U.S.T. 7645, 1160 U.N.T.S. 231, available at http://www.wipo.int/pct/en/texts/ articles/atoc.htm [hereinafter PCT].
[81] See id. arts. 2(i), 3(4)(iii), 17(3). See also Paris Convention for the Protection of Industrial Property art. 4(G), done on July 14, 1967, 21 U.S.T. 1583, 828 U.N.T.S. 305, available at http://www.wipo.int/treaties/en/ip/paris/trtdocs_wo020.html (establishing unity of invention requirement for priority patent applications).
[82] See PCT, supra note 80, art. 28.
[83] See, e.g., Sjoland v. Musland, 847 F.2d 1573, 1582 (Fed. Cir. 1988) (noting that claims define the invention); Mark J. Stewart, The Written Description Requirement of 35 U.S.C. § 112(1): The Standard After Regents of The University of California v. Eli Lilly & Co., 32 Ind. L. Rev. 537, 541 (1998-1999).
[84] Adams, 383 U.S. at 49.
[85] Phillips v. AWH Corp., 415 F.3d 1303, 1315 (Fed. Cir. 2005) (en banc) (quoting Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996)).
[86] America Invents Act, supra note 3, § 3, to be codified as 35 U.S.C. § 102.
[87] Id., to be codified as 35 U.S.C. § 101(i).
[88] Technically, to avoid an internal contradiction in the Patent Act, Congress should have amended Section 112 to say: “the subject matter for which the applicant seeks patent protection.” As the AIA amends the Patent Act, what the inventor “regards as his invention” is irrelevant; what matters are the aspects of the invention for which the applicant seeks and can obtain patent protection.
[89] See Peters v. Active Mfg. Co., 129 U.S. 530, 537 (1889) (“That which infringes if later anticipates if earlier.”).
[90] See Mark Janis, Patent Abolitionism, 17 Berkeley Tech. L.J. 899, 910-20 (2002) (describing governmental statements and policies paying homage to the “independent inventor”).
[91] See Stefan Wuchty, Benjamin F. Jones & Brian Uzzi, The Increasing Dominance of Teams in Production of Knowledge, 316 Science 1036, 1036-37 (2007); Dennis D. Crouch & Jason Rantenen, The Changing Nature [sic] Inventing: Collaborative Inventing, Patently-O Blog (July 9, 2009, 9:28 AM), at http://www.patentlyo.com/patent/2009/07/the-changing-nature-inventing-collaborative-inventing.html.
[92] Burroughs Wellcome Co. v. Barr Labs., Inc., 40 F.3d 1223, 1227 (Fed. Cir. 1994).
[93] Act July 19, 1952, ch. 950, § 116, 66 Stat. 792 (codified at 35 U.S.C. § 116 (1952)).
[94] Id.
[95] Mueller Brass Co. v. Reading Indus., 352 F. Supp. 1357, 1372-73 (E.D. Pa. 1972), aff’d, 487 F.2d 1395 (3d Cir. 1973). The reason for the quotation marks around “is” is unclear. See also Jamesbury Corp. v. United States, 518 F.2d 1384, 1396 (U.S. Ct. Cl. 1975) (commenting on “[t]he difficulty . . . in attempting to define a standard to be used to determine who is a joint inventor and the type of contribution that is necessary to qualify as a coinventor . . . .”).
[96] De Laski & Thropp Circular Woven Tire Co. v. Williamm R. Thropp & Sons Co., 218 F. 458, 464 (D.N.J. 1914), aff’d, 226 F. 941 (3d Cir. 1915).
[97] Cf. Hess v. Advanced Cardiovascular Sys., Inc., 106 F.3d 976, 981 (Fed. Cir. 1997).
[98] The timing of the supply of information or ideas by the collaborator determines whether it qualifies as a contribution to conception for purposes of establishing joint inventorship. Specifically, a not insignificant contribution to conception establishes joint inventorship if the contribution was not in the prior art at the time contributed. If the contribution entered the public domain later, after its contribution to the research team but before the team files a patent application, it qualifies the contributor as an inventor. As the Federal Circuit has explained: “Contributions to realizing an invention may not amount to contribution to conception if they merely explain what was ‘then state of the art.’” Eli Lilly & Co. v. Aradigm Corp., 376 F.3d 1352, 1359 (Fed. Cir. 2004) (quoting Hess, 106 F.3d, at 981)(emphasis added). The Patent Act, as amended by the AIA, provides explicitly that an invention does not necessarily sacrifice patentability merely by public exposure. See 35 U.S.C. § 102(b)(1) (2013) (requiring that disclosures by the inventor (or derived from the inventor) do not constitute a bar to patentability unless made one year prior to the effective filing date of the patent application). The same is true a fortiori for a contribution to conception of a patentable invention, because the contribution assisted in the final conception regardless of whether it later became public knowledge.
[99] For a review of the case law interpreting joint inventorship before 1984, see generally John O. Tresansky, Joint Invention, 7 AIPLA Q.J. 96 (1979).
[100] 1 William C. Robinson, The Law of Patents for Useful Inventions § 396 (1890).
[101] Worden v. Fisher, 11 F. 505, 508-09 (E.D. Mich. 1882).
[102] Monsanto Co. v. Kamp, 269 F. Supp. 818 (D.D.C. 1967).
[103] Id.
[104] Id. at 824.
[105] Cited 158 times as of this printing. See, e.g., Vanderbilt University v. Icos Corp., 601 F.3d 1297, 1302 (Fed. Cir. 2010); Hyatt v. Doll, 576 F.3d 1246, 1271 (Fed. Cir. 2009). Cited in the MPEP as well, supra note 12, § 605.07.
[106] Patent Law Amendments Act of 1984, Pub. L. 98-622, § 104, 98 Stat. 3385 (current version at 35 U.S.C. § 116 (2006)).
[107] Eli Lilly & Co. v. Premo Pharma. Labs., Inc., 630 F.2d 120, 135 (3d Cir. 1980).
[108] Id.
[109] See supra text accompanying note 45.
[110] This is a point well explained by Edward Lentz:
[A] person who conceives of or actually makes something, e.g., a new chemical entity, but who does not know of a use for the thing probably has not made a patentable invention and is not, therefore, an inventor unless and until a utility is discovered. At that time, a patentable invention will have been made and the person who conceived of the utility is probably a joint inventor with the person who conceived the thing.
In some cases, the novelty of an invention is not appreciated by the person who conceives of the invention but rather is discovered subsequently. In these cases, conception of the invention is not complete until the novelty is appreciated. Lentz, supra note 18, at 190.
[111] Cf. Burroughs Wellcome Co. v. Barr Labs., Inc., 40 F.3d 1223, 1227 (Fed. Cir. 1994) (“The statute does not set forth the minimum quality or quantity of contribution required for joint inventorship.”).
[112] In the 2011 America Invents Act, Congress adopted the first formal definition of a “joint inventor” or “coinventor,” meaning “any 1 of the individuals who invented or discovered the subject matter of a joint invention.” 35 U.S.C. § 100(g) (2012). As with the definition of “inventor,” this amendment does little beyond present a tautology and the mysterious additional concept of “subject matter” of an invention.
[113] Kimberly-Clark Corp. v. Procter & Gamble Distrib. Co., 973 F.2d 911 (Fed. Cir. 1992).
[114] Clairol Inc. v. Save-Way Indus., 210 U.S.P.Q. 459, 465 (S.D. Fla. 1980).
[115] Kimberly-Clark Corp., 973 F.2d at 915-17; see also Arbitron, Inc. v. Kiefl, No. 09-CV-04013, 2010 U.S. Dist. LEXIS 83597, at *16-17 (S.D.N.Y., Aug. 13, 2010) (holding that one scientist who reviewed and built on a report about another’s discovery collaborated sufficiently to qualify as joint inventors).
[116] Pannu v. Iolab Corp., 155 F.3d 1344 (Fed. Cir. 1998)..
[117] Id. at 1351.
[118] Id.
[119] Id.
[120] Eli Lilly & Co. v. Aradigm Corp., 376 F.3d 1352, 1362 (Fed. Cir. 2004); Pannu, 155 F.3d at 1351.
[121] See generally Robert W. Harris, Conceptual Specificity as a Factor in Determination of Inventorship, 67 J. Pat. & Trademark Off. Soc’y 315 (1985) (discussing the kinds of contributions to invention that do not rise to the level of contributions to conception due to lack of sufficient specificity).
[122] Fina Oil & Chem. Co. v. Ewen, 123 F.3d 1466, 1473 (Fed. Cir. 1997); Aradigm Corp., 376 F.3d at 1358.
[123] Fina Oil & Chem., 123 F.3d at 1473; Pannu, 155 F.3d at 1351.
[124] Levin v. Septodont, Inc., 34 Fed. Appx. 65 (4th Cir. 2002).
[125] Id. at 67.
[126] Id. at 70.
[127] Id. at 67.
[128] Id.
[129] Id. at 71-73.
[130] Id. at 67. Aside from the fact that the opinion was not selected for publication, the Fourth Circuit cannot bind the Federal Circuit. Normally, this issue would have been decided by the Federal Circuit, but the case was not primarily an infringement action, but rather a breach of contract action. 28 U.S.C. § 1295(a)(1) (2006) (noting that the Federal Circuit shall have exclusive jurisdiction from the appeal of a final decision of a district court in any civil action arising under, or in any civil action in which a party has asserted a compulsory counterclaim arising under, any Act of Congress relating to patents or plant variety protection.)
[131] U.S. Patent No. 5,547,657 (filed Oct. 11, 1994).
[132] The patent laws of most countries, like that of the United States, are underdeveloped in defining who qualifies as a joint inventor. For example, Canadian, German, and Japanese patent laws are all vague on the concept of joint inventorship. See, e.g., Motorkettensäge Case, Fed. Ct. Just. (Bundesgerichtshof) Decision of June 20, 1979 (Fed. Rep. of Ger.) (contributions that have not influenced the overall success of the invention or solved any significant problem do not support coinventorship). However, some offer more guidance than others. The first three of the factors listed in the text are used by Australian courts in determining the “material effect” of the collaborator’s contribution. See Row Weeder Pty., Ltd. v Nielsen, (1997) 39 IPR 400 (Austl.).
[133] Cf., e.g., Rhone-Poulenc Agro, S.A. v. Monsanto Co., 445 F. Supp. 2d 531, 548 (M.D.N.C. 2006) (expertise); Ethicon Inc. v. U.S. Surgical Corp., 135 F.3d 1456, 1460 (Fed. Cir. 1998) (skills possessed by one inventor and not the other).
[134] 35 U.S.C. §§ 116, 256 (2006); see Pannu v. Iolab Corp., 155 F.3d 1344, 1350 (Fed. Cir. 1998); MCV, Inc. v. King-Seeley Thermos Co., 870 F.2d 1568, 1570 (Fed. Cir. 1989).
[135] Univ. of Pittsburgh v. Hedrick, 573 F.3d 1290, 1298 (Fed. Cir. 2009).
[136] Eli Lilly & Co. v. Aradigm Corp., 376 F.3d 1352, 1361-62 (Fed. Cir. 2004).
[137] Trovan, Ltd. v. Sokymat S.A., Irori, 299 F.3d 1292, 1301-02 (Fed. Cir. 2002).
[138] Id.
[139] Markman v. Westview Instruments, Inc., 517 U.S. 370, 372 (1996).
[140] Trovan, 299 F.3d at 1302.
[141] Id.
[142] George M. Sirilla, How the Federal Circuit Clarified the “Muddy Concept” of Joint Inventorship, 91 J. Pat. & Trademark Off. Soc’y 509, 509 (2009).
[143] Ethicon, Inc. v. U.S. Surgical Corp., 135 F.3d 1456, 1460 (Fed. Cir. 1998).
[144] See, e.g., Bd. of Educ. ex rel. Bd. of Trustees of Fla. State Univ. v. Am. Bioscience, Inc., 333 F.3d 1330, 1339 (Fed. Cir. 2003); Ethicon, 135 F.3d at 1463 (“[T]his court must determine what Choi’s contribution was and then construe the claim language to determine if Choi’s contribution found its way into the defined invention.”) (emphasis added).
[145] Eli Lilly & Co. v. Aradigm Corp., 376 F.3d 1352, 1359 (Fed. Cir. 2004) (citations omitted).
[146] See Ethicon, 135 F.3d at 1460 (“[O]ne does not qualify as a joint inventor by merely assisting the actual inventor after conception of the claimed invention.”).
[147] The Federal Circuit very recently affirmed this point with respect to “making” the invention in Falana v. Kent State University, 669 F.3d 1349, 1358 (Fed. Cir. 2012)
[148] Ethicon, 135 F.3d 1456.
[149] See, e.g., id. at 1469 (Newman, J., dissenting); Gregory N. Mandel, Left-Brain versus Right-Brain: Competing Conceptions of Creativity in Intellectual Property Law, 44 U.C. Davis L. Rev. 283, 294-95 (2010).
[150] 37 C.F.R. §§ 1.475(a), 1.499 (2011). U.S. law on this subject reflects the requirements of important international patent treaties. See PCT, supra note 80, arts. 2(i), 3(4)(iii), 17(3); Patent Cooperation Treaty Regs. Rule 13.1 (“The international application shall relate to one invention only or to a group of inventions so linked as to form a single general inventive concept (“requirement of unity of invention”)); Paris Convention for the Protection of Industrial Property, supra note 81, art. 4(G).
[151] Gregory Mandel has proposed giving proportional rights to joint inventors who made contributions of varying levels to the invention. Mandel, supra note 149, at 353. In principle, such a rule could mitigate the problem to a degree, but suggesting the adoption of equitable apportionment methods is much easier in the abstract than operationalizing the concept. Judgments about the relative creativity of different collaborators’ inventive contributions would be a highly subjective exercise even with perfect information. In practice, it would be impossible due to the frequent inability to differentiate the source of inventive contributions in collaborative research and the temptation of any given collaborator to “remember” that another’s contribution was actually his own. After all, the main purpose of the modern Section 116 of the Patent Act is to eschew nice parsing of inventive contributions among coinventors. Moreover, not all inventive contributions are creative; some may take the form of perceptive observations of a helpful phenomenon by a prepared mind, or simply by systematic experimentation. Such contributions may qualify as inventive regardless of the absence of creativity. See 35 U.S.C. § 103(a). How is a court to weigh, on minimal objective evidence, the incomparable values of one collaborator’s creative insight against another’s tireless experimentation? Apportionment is, in short, sound in theory but not a practical solution to the problem of skewed inventorship rewards.
[152] See Mandel, supra note 149, at 347.
[153] 35 U.S.C. § 282(1) (2006).
[154] Morgan v. Daniels, 153 U.S. 120, 121, 125 (1894).
[155] In re Brana, 51 F.3d 1560, 1566 (Fed. Cir. 1995).
[156] See Nartron Corp. v. Schukra U.S.A., Inc., 558 F.3d 1352, 1356 (Fed. Cir. 2009).
[157] Am. Hoist & Derrick Co. v. Sowa & Sons, Inc., 725 F.2d 1350, 1359 (Fed. Cir. 1984); see also Radio Corp. of Am. v. Radio Eng’g Labs., Inc., 293 U.S. 1, 2 (1934).
[158] Eli Lilly & Co. v. Aradigm Corp., 376 F.3d 1352, 1358 (Fed. Cir. 2004); Fina Oil & Chem. Co. v. Ewen, 123 F.3d 1466, 1472 (Fed. Cir. 1997).
[159] Microsoft Corp. v. i4i Ltd. P’ship, 131 S.Ct. 2238, 98 U.S.P.Q.2d 1857, 1862 (2011).
[160] Id., 98 U.S.P.Q.2d at 1863.
[161] Only Justice Thomas declined to read Section 282 as implicitly incorporating the judicially developed standard of proof. Id. at 1868 (Thomas, J., concurring).
[162] Doug Lichtman & Mark A. Lemley, Rethinking Patent Law’s Presumption of Validity, 60 Stan. L. Rev. 45 (2007).
[163] Id. at 52-53.
[164] Id. at 53-55. Lemley and Sampat’s empirical research indicates that the PTO effectively weeds out a small but significant percentage of applications. See generally Mark A. Lemley & Bhaven Sampat, Is the Patent Office a Rubber Stamp?, 58 Emory L.J. 181 (2008) (evaluating the PTO’s patent rejection rate and finding it at 15%-20%).
[165] Lichtman & Lemley, supra note 162, at 52-53.
[166] See Brown v. Edeler, 110 F.2d 858, 861 (C.C.P.A. 1940) (finding that a duly executed joint application is prima facie evidence of joint inventorship); Van Otteren v. Hafner, 278 F.2d 738, 741 (C.C.P.A. 1960) (finding that a duly executed joint application is prima facie evidence of joint inventorship).
[167] Eli Lilly & Co. v. Aradigm Corp., 376 F.3d 1352, 1364 (Fed. Cir. 2004).
[168] Id. at 1366-67.
[169] Vanderbilt Univ. v. ICOS Corp., 601 F.3d 1297 (Fed. Cir. 2010).
[170] Id. at 1305.
[171] Id. n.3.
[172] Morgan v. Daniels, 153 U.S. 120, 125 (1894).
[173] See Microsoft, 98 U.S.P.Q.2d at 1866 (collecting cases).
[174] See, e.g., Mfg. Research Corp. v. Graybar Elec. Co., 679 F.2d 1355, 1360-61 (11th Cir. 1982); NDM Corp. v. Hayes Prods., Inc., 641 F.2d 1274, 1277 (9th Cir. 1981); Lee Blacksmith, Inc. v. Lindsay Bros., Inc., 605 F.2d 341, 342-43 (7th Cir. 1979).
[175] KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 426 (2007).
[176] Microsoft, 98 U.S.P.Q.2d at 1865-66.
[177] Id. at 1866.
[178] Id.
[179] See Mark A. Lemley, Can the Patent Office Be Fixed?, in Rules for Growth 367 (Robert Litan ed., 2011); Allison & Lemley, supra note 19 at 2118–19.
[180] 684 F.3d 1253 (Fed. Cir. 2012).
[181] Id. at 1256-57.
[182] Id. at 1260-61.
[183] See 35 U.S.C. §§ 116, 256.
[184] 35 U.S.C. § 115; 37 C.F.R. § 1.63(a)(2).
[185] See Eli Lilly & Co. v. Aradigm Corp., 376 F.3d 1352, 1362 (Fed. Cir. 2004).
[186] Fina Oil & Chem. Co. v. Ewen, 123 F.3d 1466, 1473 (Fed. Cir. 1997). Specifically, as one court observed, “[t]he findings relating to joint inventorship issues such as conception and inventive contributions are dependent upon an understanding of the scientific problem or problems the parties were trying to resolve.” Rhone-Poulenc Agro, S.A. v. Monsanto Co., 445 F. Supp. 2d 531, 535 (M.D.N.C. 2006).
[187] Judge Rader offered a similar objection to the Federal Circuit’s treatment of experimental use as a question of law. See Lough v. Brunswick Corp., 103 F.3d 1517, 1532-33 (Fed. Cir. 1997) (Rader, J., dissenting from the denial of rehearing en banc) (“With factual considerations dictating the outcome of the underlying analysis, logic relegates the standard of review to the factual realm as well. . . . Frankly the proposition that experimental use—a judge-made doctrine without any express basis in the Patent Act—is a question of law is absurd on its face. It is hard to imagine how case law decisions made by judges to resolve unique factual cases on a fact-driven issue could create questions of law.”). His reasoning applies a fortiori to the issue of inventorship. On the disputed standard of review of mixed questions of law and fact, see generally Evan Tsen Lee, Principled Decision Making and the Proper Role of Federal Appellate Curts: The Mixed Questions Conflict, 64 S. Cal. L. Rev. 235 (1991).
[188] See Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1376 (Fed. Cir. 1986).
[189] U.S. Const. art. I, § 8, cl. 8.
[190] On the seemingly odd choice of the word “Discoveries,” see Demaine & Fellmeth, supra note 1, at 367-74.
[191] This is not to say that the Constitution forbids Congress to allow inventors to assign their rights to others, of course.
[192] See An Act to Promote the Progress of Useful Arts (Patent Act of 1790), ch. 7, sec. 1, 1 Stat. 109 (Apr. 10, 1790); Grant v. Raymond, 31 U.S. (6 Pet.) 218, 242 (1832) (“[The patent] is the reward stipulated for the advantages derived by the public for the exertions of the individual [inventor], and is intended as a stimulus to those exertions.”).
[193] H.R. 4527, 98th Cong. (1983); S.B. 1535, 98th Cong., 1983 Cong. Rec. S9006.
[194] See, e.g., In re Sarett, 327 F.2d 1005, 1010, n.7 (Ct. Cust. & Pat. App. 1964).
[195] Congress established the U.S. Court of Appeals for the Federal Circuit in 1982 in order to centralize appeals for patent cases. Before 1982, the U.S. Court of Customs and Patent Appeals handled all appeals from administrative actions by the PTO, but appeals arising from infringement litigation were decided by the various circuit courts. The Federal Circuit was designed to remedy the lack of uniformity in patent doctrine arising from this fracturing of authority. See generally The U.S. Judicial Conf. Comm. On The Bicentennial Of The Constitution Of The U.S., The United States Court Of Appeals For The Federal Circuit: A History 1982-1990 (1991) (describing advent of the Federal Circuit); Daniel J. Meador, Origin of the Federal Circuit: A Personal Account, 41 Am. U.L. Rev. 581 (1992) (same).
[196] Statement of Bernarr R. Pravel, President, Am. Intel. Prop. L. Ass’n, before the House Subcommittee on Courts, Civil Liberties, and the Administration of Justice (June 13, 1984), in Innovation and Patent Law Reform (Mar. 28, Apr. 26, & June 6, 27, 1984), 98th Cong., 2d Sess., Pt. 3, Serial No. 105, at 2690, 2719 [hereinafter “Innovation and Patent Law Reform Hearings”].
[197] Id. at 2735-36.
[198] Id. at 2737.
[199] Testimony Of Gerald J. Mossinghoff, Assist. Sec’y & Comm’r of Pats. & Trademarks, Accompanied by Donald Quigg, Dep’y Asst. Sec’y & Dep’y Comm’r of Pats. & Trademarks, & Rene D. Tegtmeyer, Asst. Comm’r for Pats. (Mar. 28, 1984), in Innovation and Patent Law Reform Hearings, supra note 196, at 2, 7.
[200] Statement of Gerald J. Mossinghoff, in id. at 28-29.
[201] See Patent Law Improvements Act, Hearing before the Subcomm. on Patents, Copyrights and Trademarks of the Comm. on the Judiciary, U.S. Senate, 98th Cong., 2d Sess., on S.1535 and S. 1841 (Apr. 3, 1984), Serial No. J-98-107, at 18, 32-34 (Mossinghoff), 70 (AIPLA) [hereinafter “Senate Judiciary Comm. PLAA Hearings”].
[202] E.g., 130 Cong. Rec. 28,075 (1984) (statement of Rep. Moorhead); see W. Fritz Fasse, The Muddy Metaphysics of Joint Inventorship: Cleaning Up After the 1984 Amendments to 35 U.S.C. § 116, 5 Harv. J.L. & Tech. 153, 175 (1992).
[203] Letter from Warren D. Niederhauser, Pres.-Elect, Am. Chem. Soc’y, to Robert W. Kastenmeier, Chair, House Subcomm. on Cts., Civ. Libs., & the Admin. of Just. (Apr. 20, 1984), in Senate Judiciary Comm. PLAA Hearings, supra note 202, at 2617.
[204] H.R. 6286 (1984).
[205] 1984 Cong. Rec. 28073 (Oct. 1, 1984) (statement of Rep. Kastenmeier).
[206] Section-by-Section Analysis of H.R. 6286, Patent Law Amendments Act of 1984, 130 Cong. Rec. 10525-29 (1984), reprinted in 1984 U.S.C.C.A.N. 5827, 5833.
[207] S. Rep. No. 998-663, 98th Cong., 2d Sess. (Oct. 5, 1984), at 8.
[208] 1984 U.S.C.C.A.N. 5827, 5834-35.
[209] President’s Statement on Signing H.R. 6268 into Law, 20 Weekly Comp. Pres. Doc. 1818 (Nov. 9, 1984).
[210] See Innovation and Patent Law Reform Hearings, supra note 196, at 2720 (introducing the amendment to Section 116 of the Patent Act as adopted by Congress).
[211] In Ethicon v. U.S. Surgical Corp., the court wrote that it must construe the claim language to determine whether the putative coinventor’s contribution “found its way into the defined invention.” Ethicon Co. v. U.S. Surgical Corp., 135 F.3d 1456, 1463 (Fed. Cir. 1998).
[212] Id. at 1461-63.
[213] Trovan, Ltd. V. Sokymat S.A. Irori, 299 F.3d 1292 (Fed. Cir. 2002). See, e.g., Eli Lilly & Co. v. Aradigm Corp., 376 F.3d 1352, 1361-62 (Fed. Cir. 2004) (“[T]he law requires only that a coinventor make a contribution to the conception of the subject matter of a claim.”) (emphasis added).
[214] Bd. of Educ. ex rel. Bd. of Trustees of Fla. State Univ. v. Am. Bioscience, Inc., 333 F.3d 1330, 1337 (Fed. Cir. 2003) (“[E]ach joint inventor must generally contribute to the conception of the invention.”) (emphasis added).
[215] See, e.g., United States v. Adams, 383 U.S. 39, 49 (1966); Phillips v. AWH Corp., 415 F.3d 1303, 1315 (Fed. Cir. 2005) (en banc); Retractable Techns., Inc. v. Becton, Dickinson & Co., 653 F.3d 1296, 1303 (Fed. Cir. 2011), reh’g en banc denied, 659 F.3d 1369 (Fed. Cir. 2011). But cf. id. at 1312-13 (Rader, J., dissenting in part) (cautioning against importing limitations on claims from unclear language in the written description).
[216] See Graver Tank & Mfg. Co. v. Linde Air Prod. Co., 339 U.S. 605 (1950).
[217] See Gen. Elec. Co. v. Wabash Appliance Corp., 304 U.S. 364, 369 (1938); Hoganas AB v. Dresser Indus., 9 F.3d 948, 951 (Fed. Cir. 1993).
[218] Bd. of Educ., 333 F.3d 1330 (Fed. Cir. 2003).
[219] Id. at 1340-41.
[220] Ethicon, Inc. v. United States Surgical Corp., 135 F.3d 1456, 1460 (Fed. Cir. 1998).
[221] Fina Oil & Chem. Co. v. Ewen, 123 F.3d 1466, 1472 (Fed. Cir. 1997); see also Rhone-Poulenc Agro, S.A. v. Monsanto Co., 445 F. Supp. 2d 531, 549 (M.D.N.C. 2006) (“[N]either the statute nor the cases relating to joint inventorship requires the simultaneous presence or awareness of all who have contributed significantly toward conception when the last piece of the conception puzzle slips into place.”); Regents of the Univ. of Mich. v. Bristol-Myers Squibb Co., 301 F. Supp. 2d 633, 642 (E.D. Mich. 2003) (“[A] joint inventor need not have had the definite and permanent conception of the full invention – otherwise, he would be the sole inventor.”); Tavory v. NTP, Inc., 297 F. Appx. 976, 979, 2008 WL 4710761, **4 (Fed. Cir., Oct. 27, 2008) (unpub.) (“[N]o individual coinventor need have a ‘definite and permanent idea of the complete and operative invention’ so long as all of the coinventors collectively satisfy that requirement.”).
[222] Dan L. Burk & Mark A. Lemley, Fence Post or Sign Posts? Rethinking Patent Claim Construction, 157 U. Pa. L. Rev. 1743, 1762 (2009).
[223] See, e.g., Thalia V. Warnement & Troy E. Grabow, Drafting the Patent Specification, Am. Intell. Prop. L. Ass’n, Practical Patent Prosecution Training For New Lawyers, Aug. 2008, at http://www.aipla.org/learningcenter/library/papers/bootcamps/08patentbootcamp/ Documents/Grabow-paper.pdf; Rajiv Sarathy, Broad Patents Can Be Both Lucrative and Expensive, Perkins Coie Patent Law Insights, Nov. 21, 2009, at http://www.patentlawinsights.com/tags/broad-claims/.
[224] Fina Oil & Chem. Co. v. Ewen, 123 F.3d 1466, 1473 (Fed. Cir. 1997); see Burroughs Wellcome Co. v. Barr Labs., Inc., 40 F.3d 1223, 1229 (Fed. Cir. 1994).
[225] A few district courts have misread the Patent Act to allow courts to exclude as joint inventors collaborators who left the research team after the experimental stage. See, e.g., Huang v. California Inst. of Tech., 72 U.S.P.Q.2d 1161, 1177 (C.D. Cal. 2004) (“To judge the significance of an alleged joint inventor’s contribution to the complete invention, a court may consider whether the alleged joint inventor was able at the relevant time period to understand and articulate the inventive team’s final operative embodiments.”). This misses the entire point of the 1984 amendment to Section 116, which was based the desire to grant equal reward to early stage team members, as the Kamp decision upon which the amendment was based indicated. See Monsanto Co. v. Kamp, 269 F. Supp. 818 (D.D.C. 1967).
[226] See Sean B. Seymore, The Null Patent, 53 Wm. & Mary L. Rev. 2041 (2012).
[227] This scenario is not merely fanciful. A federal district court recently misinterpreted the law of joint inventorship in just such a manner in Vanderbilt Univ. v. ICOS Corp., 594 F. Supp. 2d 482 (D. Del. 2009), aff’d, 601 F.3d 1297 (Fed. Cir. 2010). Relying on Burroughs Wellcome and American Bioscience, the district court held that university researchers who had collaborated with a drug company to identify and test a series of pharmaceuticals could not be joint inventors because none of the university scientists had “conceived the ‘specific chemical structure of the compound’ claimed.” Id. at 505. Although the court recognized that the university researchers must have made a valuable contribution to the discovery of the patented compounds and found the drug company’s denial of benefit from the university “troubling,” it could find no basis in law for treating the university scientists as joint inventors even if they contributed the very scaffolding of the drug. Id. at 505-07. Judge Clevenger, writing for a panel of the Federal Circuit, corrected this error by pointing out that every coinventor need not individually conceive the final and entire invention; someone in the “inventorship team” need merely conceive the complete and final invention. Vanderbilt Univ. v. ICOS Corp., 601 F.3d 1297, 1306-07 (Fed. Cir. 2010). Yet, relying on the facts on record which were tainted by the trial court’s misunderstanding of the law, the Federal Circuit affirmed the district court opinion despite the absence of a sufficiently developed record based on a proper interpretation of the law, id. at 1307-08, an unfortunate choice, as Judge Dyk pointed out in dissent, id. at 1310 (Dyk, J., dissenting in part). This outcome results mainly from a jurisprudential problem already discussed—the Federal Circuit treating inventorship as a question of law rather than a question of fact.
[228] Ethicon, Inc. v. U.S. Surgical Corp., 135 F.3d 1456, 1463 (Fed. Cir. 1998).
[229] E.g., U.S. Pat. No. 7,975,315 (July 5, 2011) (“The present invention relates to an atomic force microscope . . . .”); U.S. Pat. No. 7,947,218 (May 24, 2011) (“The present invention relates to a portable cooling device . . . that provides air flow, and a shroud . . . at an article . . . .”).
[230] Many device patents do omit any mention of utility from the claims. See, e.g., U.S. Pat. No. 7,947,349 (May 24, 2011) (“The invention relates to a moulded piece forming a negative mould . . . .”).
[231] See, e.g., U.S. Pat. No. 7,973,166 (July 5, 2011) (claim 1); U.S. Pat. No. 6,740,654 (May 25, 2004) (claims 1-20); U.S. Pat. No. 5,972,658 (Oct. 26, 1999) (claims 1-8); U.S. Pat. No. 4,024,272 (May 17, 1977) (claims 1-9, 12-20). European practice is consonant. See Mobil Oil/Friction Reducing Additive, Eur. Pat. Off. Bd. App. Case No. G 0002/88, para. 5 (Dec. 11, 1989), [1990] E.P.O.R. 73 (“The discovered use of [a] compound or composition will normally be described in the patent, but may not be expressly claimed.”).
[232] See generally Shayne Cox Gad, Introduction: Drug Discovery in the 21st Century, in Drug Discovery Handbook 1, 1-7 (Shayne C. God ed., 2005). In rational drug design, the use of computer modeling may eliminate the need for and reduce reliance on some of these steps. See generally Ulf Madsen, Povl Krogsgaard-Larsen & Tommy Liljefors, Textbook of Drug Design and Discovery (2002).
[233] For example, the fifth step typically requires repeated recourse to the system developed in step 4, in order to verify that the modifications of the molecule perform the function desired in an acceptable manner. In any given case, the inventors may be able to use a prior art, off-the-shelf assay system to test and modify a nonobvious compound, or they may have to develop a nonobvious system for testing and modifying a prior art compound. Either step may constitute a valid contribution to conception.
[234] Bd. of Educ. ex rel. Bd. Of Trustees of Fla. State Univ. v. Am. Bioscience, Inc., 333 F.3d 1330 (Fed. Cir. 2003).
[235] Id. at 1340.
[236] Id. (emphasis added). Throughout this portion of the opinion, the court repeated and elaborated on the mistaken conclusion that the FSU scientists had to conceive the entire compounds to be joint inventors. Id. at 1340-42.
[237] Id. at 1341.
[238] Id. at 1342.
[239] Id.
[240] Vanderbilt Univ. v. ICOS Corp., 594 F. Supp. 2d 482, 504-06 (D. Del. 2009), aff’d, 601 F.3d 1297 (Fed. Cir. 2010).
[241] Falana v. Kent State Univ., 669 F.3d 1349, 1358 (Fed. Cir. 2012)
[242] Id. at 1357.
[243] Id. at 1358.
[244] Id..
[245] But see William R. Thropp & Sons Co. v. De Laski & Thropp Circular Woven Tire Co., 226 F. 941, 949 (3d Cir. 1915) (“One may conceive a general or imperfect outline of an entirely novel thing, which, without the conception of another developing it and giving it body, might never amount to invention; but if the conceptions of one supplement and complement the conceptions of the other, the result might be invention and therefore joint invention.”).
[246] In practice, a court might stretch the law to cover this case through the judicial fiction that Eta and Theta “conceived” the compound by producing all the information required to lead a person having ordinary skill in the art to the compound. However, Trovan analysis in no way sanctions such an interpretation of inventorship, which illustrates again how Trovan crams inventorship analysis into a Procrustean bed. Trovan, Ltd. v. Sokymat S.A., Irori, 299 F.3d 1292, 1302 (Fed. Cir. 2002).
[247] See supra text accompanying notes 144-146.
[248] Bill Cosby: Himself (20th Century Fox 1983).

Caught in the Middle: Reducing the Uncertainty Created by the FDA and the Patent System for Genetic Diagnostic Test Makers

Caught in the Middle: Reducing the Uncertainty Created by the FDA and the Patent System for Genetic Diagnostic Test Makers
By Sharonmoyee Goswami* A pdf version of this article may be downloaded here.  

INTRODUCTION

Genetic diagnostic tests promise a wealth of benefits. They can figure out if a particular drug will work in your body, tell you what disease you have, or one you might get. There are two main types of genetic diagnostic tests, commercially developed “kits”—complete test systems with all the components and instructions needed to conduct the test that are sold to multiple labs—and “laboratory-developed tests” (LDTs)—preassembled test systems intended for use at a single laboratory.[1] Unlike kits, LDTs are sold to individual health care providers and directly to patients.[2] Until recently, the Food and Drug Administration (FDA)  regulated commercially developed kits, but did not regulate LDTs, even when the two performed the same function.[3] In 2010, the FDA decided to start regulating all laboratory-developed tests. However, regulation does not occur in a vacuum, and increased regulation necessitates increased outlay for companies seeking to market genetic diagnostic tests. For drug companies, the risk that a product will require extensive ex ante investment in clinical trials but may not be approved for market release is mitigated by the presence of drug patents, and to a lesser extent, by statutory data exclusivity.[4] Both ensure that if a company succeeds in getting a drug past clinical trials and onto the market, then a competitor will not be able to copy their drug and unfairly profit without an initial outlay. Because the patentability of diagnostic tests is uncertain,[5] diagnostic test companies do not enjoy the same level of protection as drug companies. If the FDA expands the scope of regulation without the backstop of patent protection, it seems likely that the current booming market for genetic diagnostic tests will wane, to the detriment of consumers. To combat this result, I address three principle options:
  1. Require no regulation, but create no change in the status of patentable subject matter. Effectively, this option maintains the status quo and responds to critics’ suggestions that the FDA is expanding its reach too far by regulating genetic diagnostic tests.
  2. Pursue regulation, but create a per se patentable subject matter rule for diagnostic tests to balance the additional expense (notwithstanding additional barriers to patentability), creating a baseline similar to that for pharmaceutical drugs.
  3. Create a new regulatory regime that balances the additional expense of regulation through data exclusivity and limiting additional regulation to only the most necessary areas.
In exploring these options, this paper proceeds in five parts. Part I outlines the current regulatory framework and the FDA’s recent shift in regulation. Part II covers the main problems with the current unregulated market for genetic diagnostic tests and concludes by eliminating the first principal option. Part III addresses the current framework for patentable subject matter and discusses whether a per se rule would fit within this regime. In Part III, I determine that a twenty-year patent term may provide too much protection for genetic diagnostic tests, particularly considering that enforcement of such patents serves to limit consumer choice and may stifle innovation. Part III further finds an appropriate alternative to be FDA-mediated data exclusivity, as seen with the Hatch-Waxman Act. Part IV addresses what regulatory changes are necessary to accommodate the unusual needs of genetic diagnostic tests. The solutions suggested would reduce the ex ante uncertainty facing diagnostic test companies by clarifying the classification regime in the FDA, creating mandatory maximum times for approval within the FDA, streamlining complaint procedures to ensure that any issues with tests are quickly corrected, particularly for software and internet-based tests, and most importantly, creating a data exclusivity backstop for diagnostic tests that fall outside the scope of patentable subject matter. Corresponding solutions would benefit consumers by strengthening labeling requirements and genetic counseling for particular test varieties and increasing methods of keeping up with new technology in this field. Finally, Part V applies this new regulatory system to a purely computational genetic diagnostic test, which I believe to be emblematic of genetic diagnostic tests in the future.

I. Current Regulatory Framework for Genetic Diagnostic Tests and the Dangers of Increased Regulation

In this section, I discuss the current regulatory framework for genetic diagnostic tests, beginning with a short background of how institutional relationships led to the FDA’s current framework. Subsequently, I address the main problems with the regulatory framework that increased regulation would exacerbate and the negative impact that expansion could have on the genetic diagnostic test industry, namely, (1) the delay in time-to-market for the tests; (2) the uncertainty as to the classification of a particular test, and (3) the absence of any exclusivity backstop to prevent copying by competitors after approval.

A.  Institutional Frameworks at the FDA

According to the traditional “transmission belt” theory,[6] Congress creates rules that govern agency action, the agency adopts procedures that adhere to those Congressional directives, and these processes are kept in check through judicial review. However, unlike other agencies that adhere to the traditional “transmission belt” theory, the FDA is a particularly powerful agency that takes a greater role in governance than mere transmission.[7] The increasingly difficult subject matter that the FDA regulates creates a situation where Congress and the courts simply do not have the necessary expertise to create specific policies governing drugs and devices. Indeed, well-known practitioner Thomas Austern has criticized the FDA’s substantial power, alleging that it was “delegation running riot”.[8] Historically, Congress and the courts have stepped in to curtail FDA power in one of two situations: (1) when a high-risk product creates public outcry and (2) when FDA regulation injures industry substantially. The FDA originated out of concern about high-risk products—Congress created the agency in response to Upton Sinclair’s 1906 book The Jungle and its exposé of the meatpacking industry.[9] Subsequent expansion continued in response to other high-risk disasters. The 1938 Act[10] responded to medicines for children that had been mixed with antifreeze.[11] The 1962 Kefauver-Harris Amendments responded to the narrowly averted Thalidomide disaster.[12] Further, during the 1960s, all but one of the Congressional oversight hearings were conducted to criticize the FDA for failing to take adequate regulatory action against products that the committee concluded were safe or ineffective.[13] Congress also traditionally steps in when FDA involvement threatens to substantially injure industry. This second situation  may be a result of agency capture and Congressional interests in the industry. Established companies who get many products approved by the FDA have developed long-standing relationships with the agency, particularly given a “revolving door” between the agency and corresponding companies. Since these established companies are familiar with existing FDA procedures, they are unlikely to lobby Congress to change that framework. Even when FDA procedures are onerous, established companies are likewise better equipped to deal with them, often to the detriment of smaller entities. As those small companies are unlikely to have sway in Congress, the framework remains unchanged until problems become large enough to attract the attention of the general public or established companies. The high-risk disasters discussed supra are examples of FDA change in response to problems that affect the general public. More recently, changes at the FDA have come from established companies that helped enact legislation to make the FDA more efficient, most notably the Prescription Drug User Fee Act,[14] or to reduce adverse effects on industry from increased regulation, as with the Hatch-Waxman Act,[15] discussed infra. Because of the significant discretion given to the FDA by Congress and the courts, the agency is adept at responding to technological changes. Instead of relying on rulemaking or case-by-case adjudication like most agencies, the FDA promulgates new policies through guidelines, which often have a de facto binding effect because of the close relationship those companies have with the agency. If the company is able to follow the guidelines, it is in their best interest to do so because going against the FDA’s policies could put their entire line of products at risk. Of course, the agency would not retaliate by refusing to approve a particular product, but may delay the approval or requests for additional clinical trials, which increase expenses to the company with little corresponding benefit to the consumer. Finally, the FDA and its counterpart state agencies have fared extremely well in the courts.[16] When faced with complex scientific and technical issues, judges have been reluctant to overrule decisions made by the FDA, and the Supreme Court has repeatedly upheld the power of the FDA to protect the public from “dangerous products.”[17] Furthermore, because judges generally do not choose which cases they will adjudicate (although the Supreme Court does to a certain extent), the courts cannot provide a reliable mechanism for effecting policy, except in a reactive fashion. Despite these limitations, the courts have acted to influence policy in this area in two ways: (1) by curbing FDA procedural irregularities and (2) by marshalling public opinion of high-risk products through lawsuits. The main target of courts is the non-binding guidelines discussed supra, which have been struck down by courts when they become too much like binding regulations without any of the requisite procedural mechanisms.[18] The result has been an FDA policy in which the guidelines are never actively enforced, but upheld through extra-legal mechanisms. As for the marshalling of public opinion of high-risk products through lawsuits, many tort suits and class-actions have created changes in the way that the FDA approaches regulation. Although most of these suits target the individual companies rather than the FDA, because companies seek to avoid similar situations in the future, large entities may actually seek increased regulation from the FDA to avoid tort liability. This backdrop has led to the current state of genetic diagnostic test regulation. Until recently, the FDA had not exercised its enforcement discretion to regulate most genetic diagnostic tests. Because the adverse effects from the unregulated market for genetic diagnostics are not likely to surface for many years due to their predictive nature, there is not likely to be a “Thalidomide moment” in Congress or the courts for such an industry. Furthermore, most industry players have benefited financially from the FDA’s neglect, and consumers are much less able than either large or small companies to lobby Congress for increased regulation.

B.  FDA Shift in Regulating Genetic Diagnostic Tests

The majority of genetic diagnostic tests fall within the category of “lab-developed tests,” which until recently, were completely unregulated by the FDA. Instead, laboratory procedures used in such tests were subject to minimal regulation under the Clinical Laboratory Improvement Amendments of 1988 (CLIA) through the Centers of Medicare and Medicaid Services (CMS).[19] CLIA applies to all clinical laboratories that operate or provide testing services in the United States[20] and mandates that laboratories not accept materials from the human body for testing without adequate certification. In 1992, CMS issued regulations for implementing CLIA, creating “specialty areas” for laboratories that perform high-complexity tests, but did not include genetic diagnostic tests.[21] CMS had not instituted specific requirements for molecular or biochemical genetic testing laboratories by 2010.[22] Likewise, CLIA can only regulate a laboratory’s analytical validity (whether a test properly measures the characteristic it was intended to measure[23]), leaving clinical validity considerations (whether or not the test actually diagnoses the condition) up to the laboratory director.[24] The FDA currently regulates clinical validity for experimental diagnostics to some extent.[25] For example, suppose that a laboratory hypothesizes that a given gene sequence is associated with a particular disease. If the laboratory wished to offer this test to volunteers to bolster their confidence in this correlation, the FDA would require that the Internal Review Board of the institution oversee all such studies and inform the volunteers of the test’s experimental status.[26] More generally, the FDA regulates devices using biological materials outside the body (in vitro) differently depending on whether they are commercially available kits or lab-developed tests. The FDA regulates kits as medical devices,[27] which are organized into three classes. In contrast, lab-developed tests presently fall within the discretion given to the laboratory director under CLIA,[28] leading to two separate options for genetic diagnostic tests: one for commercially available kits and another for lab-developed tests administered by CLIA-regulated labs.[29] Many have complained that this creates an uneven playing field between these two categories.[30] In 2007, the FDA changed course and decided to regulate lab-developed tests more stringently to address the uneven field. There were a number of reasons for this change. Early lab-developed tests were limited to small entities, with close relationships between the physician or technician performing the test and the patient.[31] More recently, it became clear that the laboratory developing the test was actually a large corporation interacting with the patient only through the postal system.[32] The complexity of the tests had also increased in recent years.[33] Finally, the Secretary’s Advisory Committee on Genetic Testing had recommended that the FDA become more involved in pre-market review of these tests.[34] This 2007 guidance stated that the FDA would require pre-market review for a limited subset of lab-developed tests known as in vitro diagnostic multivariate index assays (hereinafter, “algorithm assays”). These algorithm assays analyze laboratory data using an algorithm to generate a result for diagnosing, treating, or preventing disease[35]. The agency was particularly concerned about this subset of lab-developed tests because they use proprietary methods to calculate patient-specific results that healthcare providers are unable to independently derive or confirm.[36] In 2010, the FDA decided not to issue final guidance on algorithm assays, and instead chose to pursue comprehensive regulation of all laboratory-developed tests.[37] In doing so, it recognized that the absence of oversight may make it easier for laboratories to develop and offer tests quickly, but believed that comprehensive regulation would create a level playing field.[38]

C.  The Three Classes of Devices for FDA Approval

This section examines the framework currently used for those genetic diagnostic tests (kits) that are regulated by the FDA. Because the FDA will likely proceed under a similar framework for future genetic diagnostics that fall within its enforcement discretion, I treat this classification system as a baseline for future suggestions for regulation in Part IV. Under the Medical Device Amendments of 1976,[39] once the FDA classifies a test as a medical device, it places the product into one of three classes of regulation. Class I, or low-risk, products require no approval before sale, although the FDA monitors adverse effect reports after sale.[40] Class I devices includes items like bandages where there are likely to be few ill effects even from device misuse.[41] Class II, or moderate-risk, products require clearance of a pre-market notification submission known as a 510(k).[42] The 510(k) submission requires a comparison of the submitted device with a legally marketed device and a showing that they are substantially equivalent.[43] There are also de novo classifications for Class II items that have no identifiable predicate device.[44] Class II devices may require additional clinical testing, but generally do not. Clinical testing in this context does not necessarily mean randomized, double-blind studies; submissions can rely on published studies or earlier data, although the FDA prefers studies where samples are prospectively collected.[45] The 510(k) requirements were envisioned as an efficient regulatory option: allowing companies to build upon established clinical and scientific evidence of safety.[46] As a result, the 510(k) option is more widely used than the Pre-Market Approval (PMA) option, discussed infra.[47] Finally, Class III, or high-risk, products require submission of an application for Pre-Market Approval.[48]  PMA is the FDA process of scientific and regulatory review to evaluate safety and effectiveness. These devices are those that support or sustain human life, are of substantial importance in preventing impairment of human health, or present an unreasonable risk of illness or injury, including patient misuse.[49] Examples include diagnostics for Hepatitis B and C and for HPV.[50] Regardless of whether the company follows the Class II or Class III pathway, the FDA may request that the company provide clinical data to support clearance or approval.[51] After the introduction of any product (of any class), adverse events must be reported to the FDA, even if such malfunctions do not cause any injury.[52] The FDA estimates that 50 percent of regulated devices are Class I, 42 percent are Class II, and only eight percent are Class III.[53] In the 2007 draft guidance, the FDA stated that most algorithm assays would be Class II or Class III devices, depending partly on the seriousness of the disease measured.[54] Although the requirements after classification are very clear, the FDA has been unclear about why certain devices fall into certain classes. Furthermore, given the deference accorded the FDA by the courts, it is difficult to change the classification of a device subsequent to FDA classification. The overlap between Class II and Class III indicates a need for increased clarity in device classification, including ex ante categories for genetic diagnostic tests. This is particularly necessary given the letters that the FDA has sent out to diagnostic test companies requesting that they submit information for approval and regulation pursuant to its enforcement discretion, discussed infra.

D.  Choice of Classification Delays Market Entry and Can Be Uncertain

The FDA’s decision to place a product into one of these three classes can make a substantial difference in both the expense associated with clinical trials and the amount of time that it takes before the product can enter the market. At the same time, empirical data shows that the European Union (EU) regulates many of the same products, but that these products are introduced many months or even years earlier in Europe than in the United States.[55] Because the FDA is planning to expand this regulation further, this indicates that previously unregulated products may be particularly affected by this disparity in time-to-market. The data suggests two solutions, (1) that the FDA create more stringent time limits on the approval of medical devices and (2) that the FDA provide greater transparency as to why they have chosen to classify a particular device into a given class, thereby enabling companies to better predict their expenses. A recent survey of 204 public or venture-backed medical technology companies compared the efficiency of the FDA’s regulation to equivalent EU practices and concluded that despite having similar safety outcomes, the FDA’s current practices are less efficient than the agency’s European counterparts.[56] Although some delays were caused by personnel changes during the approval process,[57] the average review time for a Class II product from first filing to clearance was ten months, suggesting that the FDA requirements are particularly onerous.[58] For those companies who communicated with the FDA prior to making a 510(k) submission, the total time from that first communication was nearly three years compared to seven months in Europe.[59] The average total cost for participants to bring a low-to-moderate risk 510(k) product from concept to clearance was approximately $31 million, with $24 million spent on FDA dependent activities.[60] For Class III products, the average review time increased to 54 months, almost four years longer than European agency review for the same products.[61] At least some of this delay stems from increased risk aversion to new products at the FDA.[62] For higher-risk products that require pre-market approval, the average total cost from concept to approval was approximately $94 million, with $75 million spent in stages linked to the FDA.[63] Because of the additional time spent in the regulation process, as well as the additional funds used, earlier expansions in FDA regulatory authority have been financially detrimental to the regulated industry. This extensive time frame is exacerbated by the uncertainty as to how the FDA will classify the newly regulated diagnostic tests. In 2010, the FDA sent several letters to genetic testing companies, including a letter to the genomic chip manufacturer Illumina.[64] The letter to Illumina stated that their chip (used to sequence genes) was classified as a device under the FDCA and had not been submitted for pre-market clearance or approval.[65] The letter further suggested that the agency could consider the chip a Class III device.[66] Similar letters were sent to deCODE Genetics, Navigenics and 23andMe, which use the Illumina chip for use in testing kits.[67] These letters stated that these products did not fall under the category of lab-developed tests because they were not “developed by and used in a single laboratory,”[68] and because the collection kits were distributed through a website or by a third party distributor.[69] All three letters intimated that such kits were considered Class III devices and required Pre-Market Approval submissions. A fifth letter was sent to Knome, which provides whole-genome sequencing and software to interpret this data.[70] Here, the FDA specifically stressed that as a software program that analyzes genetic test results generated by an external laboratory, the product is a diagnostic device that requires pre-market approval under the FDCA, and because it was not within a single laboratory, it was not considered a laboratory-developed test (which were then unregulated).[71] Fourteen additional letters were sent to other companies suggesting that their devices may require pre-market approval under Section 201(h) of the FDCA, although they did not suggest that they were Class II or Class III devices.[72] These letters suggest that these genetic diagnostic test manufacturers may need to go through the onerous Class III approval process, likely delaying the time-to-market for these devices even more.

E.  Historical Expansion of Regulation Has Been Alleviated by Commensurate Expansion in Patent and FDA-Mediated Data Exclusivity

This section shows that historical instances of regulatory expansion of previously unregulated products did have negative impact on those industries, and suggests that similar concerns about increased regulation of genetic diagnostic tests today are well-founded. Congress responded to negative results and complaints from industry by providing for greater patent exclusivity and for a new data-exclusivity backstop, which helped to reverse the trend in fewer drug approvals following increased drug testing regulation. These historical results suggest that there is no need to wait for a negative impact on diagnostic tests to aid the industry, particularly after many of these companies may choose to relocate to Europe, and provides support for the exclusivity solutions explored infra in Part III. As mentioned, the expansion of pre-market scrutiny by the FDA is not new. In 1962, the revolutionary Kefauver-Harris Amendments significantly expanded the FDA’s ability to regulate drugs before they went to market.[73] As a result, thirteen out of fourteen new products regulated by the FDA took longer to get to market in the United States than in other countries.[74] Without the protection of patents, it is unclear whether manufacturers would have invested the significant funds required for product approval.[75]Indeed, the delay in reaching market was already hurting the pioneer drug manufacturers by eating away at their patent term. In response, because generic manufacturers were subject to similarly stringent requirements, the pioneer drug manufacturers obtained court decisions preventing generic manufacturers from beginning clinical trials while a brand drug was still under patent.[76] The solution to these tactics was the Drug Price Competition and Patent Term Restoration Act of 1984 (hereinafter “Hatch-Waxman,” the names of its two sponsors), which tied the extension of patent terms to the period spent in the FDA approval stage.[77] The unusual thing about Hatch-Waxman is that in addition to providing a handy mechanism for resolving conflicts between pioneer drug manufacturers and generic drug manufacturers, it also provides for three-year data exclusivity for performing further clinical studies.[78] This means that for a generic company to gain FDA approval even after the expiry of the pioneer patent term, they must perform their own clinical trials rather than rely on the data from clinical trials performed by the pioneer drug manufacturer, or wait five years until this period of exclusivity expires. The Hatch-Waxman exclusivity period does not apply to medical devices, but the subsequent Safe Medical Devices Act of 1990 created a six-year data exclusivity provision for Class III devices[79] subject to PMA requirements.[80] As discussed earlier, there are very few Class III devices approved each year, but some Class II devices still require clinical trials. A baseline level of exclusivity for any diagnostic test that needs clinical trials for approval or reclassification by the FDA would help to alleviate the uncertainty created by the current patentability status discussed in Part III. To ensure that the expansion of regulation does not hurt companies, it is insufficient to simply apply the current medical device framework used for kits to all genetic diagnostic tests (including LDTs). Instead, we should pursue a framework that would reduce the significant times required in device approval, as well as clarify device classifications to ensure that diagnostic companies are prepared for the investment that the approval process would require. In addition, companies’ investments in clinical trials should be protected with either a patent or a data-exclusivity backstop.

II. Why We Need Better Regulation of Genetic Diagnostic Tests

As discussed in Part I, the FDA is currently overhauling the regulation of diagnostic tests to include additional regulation of genetic diagnostic tests. Additional regulation, particularly where FDA classifications of such devices is unclear, could be damaging for the industry. Critics find the expansion of FDA oversight into this sector as emblematic of FDA overreach stifling industry and suggest that the agency avoid expanding its enforcement discretion to regulate laboratory-developed tests.[81] However, this section discusses why additional regulation is necessary for these laboratory-developed genetic diagnostic tests and why this expansion is beneficial for consumers. Diagnostic tests detect the presence of a condition by measuring some quantity associated with the condition. Similarly, genetic diagnostic tests can predict an individual’s propensity for a condition based on their nucleotide sequence, as discussed infra. If such tests worked perfectly to predict disease and provided completely clear information to patients who use the tests, there would be no need for regulation. Unfortunately, this is not the case. First, I address common errors made in diagnostic genetic testing and the role that regulation may play in minimizing such errors. These common errors include: (1) errors caused by using molecular markers rather than sequencing the whole genome; (2) errors due to variations in gene expression between individuals; and (3) errors resulting from statistical studies relied upon in designing the test, which are compounded by the complexity of gene interactions. Oversight by a regulatory agency, such as the FDA, would reduce the occurrence of these errors and increase the utility of such tests for consumers. To understand these problems, this section discusses the principles that enable genetic diagnostic tests to function and explains where these principles break down, resulting in errors in the test results. All humans have strands of DNA comprising four fundamental nucleotides, Adenine (A), Thymine (T), Guanine (G) and Cytosine (C).[82] The sequence of these nucleotides form genes, which determine which proteins the body produces, and thus determines how the body functions.[83] Within every species, there are variants of genes known as alleles.[84] To use a simplified example, not all humans have the same eye color. However, we all have genes that affect eye color, such that some of us have a blue-eye color allele whereas others may have alleles for brown or green eyes.[85] Certain alleles are more likely to lead to diseased conditions than others. [86] For example, a mutant allele of a gene known to repair damaged DNA may be less effective at repairing such damages than the normal allele.[87] Because the DNA cannot be repaired, individuals with the mutant allele may have diseases caused by the damaged DNA, including colon cancer.[88] Therefore, a genetic diagnostic test may be able to predict the development of  cancer by detecting the presence of the mutant allele. One method for detecting the presence of a mutant allele is by sequencing all of the nucleotides. A possible issue with sequencing is that a genetic diagnostic test may return an incorrect sequence, reading “ATAC” when a subject actually has “ATGC”, demonstrating a failure of analytical validity.[89] A recent study compared the data of five individuals sequenced by two different genetic testing companies for thirteen diseases.[90] The results of the tests were 99.7% the same between the two companies, suggesting that the analytical validity for unregulated tests (the status quo) is quite high.[91] Even those companies that have been criticized for their inaccuracy do not have analytical validity thresholds below 99%.[92] Therefore, the issues in this section focus on issues aside from analytical validity.

A.  Molecular Markers That Do Not Correctly Predict the Presence of a Disease Allele

Although whole genome sequencing (where every nucleotide is sequenced) is the best way to detect the presence of certain genes, it currently costs approximately $20,000 per person.[93] To save costs, companies use molecular markers that are statistically associated with the presence of certain mutant alleles.[94] The use of molecular markers means that rather than sequencing A, T, G, and C, a company may only sequence the starting “A” and determine the remaining nucleotides based on their statistical analyses. A simplified example can help to illustrate how molecular markers are able to replicate full sequencing. Suppose that, in addition to having brown eyes and blonde hair, your mother also had an unusual allele that protected against the bubonic plague.[95] If your mother’s ancestors lived in areas where the plague was prevalent, those individuals who did not have this allele would be more likely to die. Soon, the population would have a high prevalence of the plague protection allele and the alleles surrounding that allele on the chromosome. If those alleles were for blonde hair and brown eyes, this population would likely share those characteristics as well. However, if the correlation between blonde hair and the plague allele were perfect, there would be no error from using the blonde hair as a marker for the actual allele. As discussed infra, such a perfect correlation, however, is seldom the case. Sometimes the statistical assumptions used may be incorrect, causing errors in the genetic results reported to consumers. Events such as “cross-over” between genes may cause such errors. Humans have two copies of each gene, one from each parent.[96] For example, while your mother might have blonde hair and brown eyes, your father might have black hair and blue eyes. Let us also assume that the genes corresponding to hair and eye color are next to each other on the same chromosome.[97] If you inherited one copy of each gene from each parent, you would have one copy with blue eyes and black hair and the other copy with brown eyes and blonde hair. Assuming that your father’s genes were dominant, no child would ever have blue eyes with blonde hair or brown eyes with black hair.[98] Therefore, to create beneficial variation in the population, chromosomes “cross-over” one another in sex cells, yielding new chromosome combinations like blue eyes with blonde hair and brown eyes with black hair.[99] Such natural variation makes the job of genetic diagnostic test companies more difficult. Returning to your mother’s hypothetical plague protection allele associated with her blonde hair and brown eyes, a genetic test company may always assume that any blonde-haired, brown-eyed individual possesses the unusual allele. However, if an individual has either had a cross-over event between the plague allele and their hair color allele, the genetic diagnostic test will yield incorrect results. An error like this can lead to a false negative or a false positive result. Such errors are more likely when the company is using a “weak-effect” marker—one where the correlation between the marker and the full nucleotide sequence is statistically tenuous. Additional regulation can alleviate this concern by either requiring full sequencing for certain regions of the genome that incur many cross-over events or by requiring that companies rely on multiple markers (or a single “strong-effect” marker) rather than a single “weak-effect” marker.[100]

B.  Mistakes in Genetic Test Interpretation Occur Because of Genetic and Environmental Variation

Another regulatory criterion is clinical validity, the accuracy of a test in actually diagnosing a particular condition.[101] Regulating clinical validity is more difficult because of the interpretation involved, and because of genetic concepts known as “penetrance” and “expressivity.” This is largely because consumers can find it difficult to understand that a heightened risk of breast cancer from a single gene can mean anything from never developing breast cancer to dying from the disease in the next ten years. Penetrance is the percentage of people with a particular genotype (nucleotide sequence) who show the diseased phenotype (physical manifestation of the allele).[102]Expressivity, on the other hand, is the intensity of the phenotype expressed by someone who possesses a particular genotype.[103] These variations among individuals present the most difficult challenges for genetic test regulation. Genetic diagnostic tests would be easier for consumers to understand if explained by a physician intermediary, who could show that the same mutation could lead to widely varying results in onset of the disease, as discussed in the earlier example. However, unlike similar non-genetic diagnostic tests, genetic diagnostics are sold and marketed directly to consumers, presenting a challenge for regulators. This problem is exacerbated because an interpretation error can lead to dire results. Suppose a company sequenced a subject’s genes for repairing damaged DNA and found that the woman carried the disease allele for BRCA1, a gene associated with the onset of breast cancer.[104] The woman may not develop breast cancer in her lifetime—a penetrance problem—or may develop a form that does not rapidly metastasize—an expressivity problem.  Indeed, very few diseases have 100% penetrance, where all of the individuals with a particular genotype have the corresponding disease.[105] As a result, interpretation of the test results remains difficult even with the knowledge of the full sequence.[106] Expressivity is a lesser concern for clinical validity because such variation arises once the individual has the disease. However, for serious diseases, individuals may take action based on these disease risk profiles when in reality their disease expression may be very low based on other risk factors, such as age and lifestyle.[107] As a result, much of the regulation in this area will depend on what information the company provides to consumers. If the genetic diagnostic testing company only provides the sequence with no additional information, very little regulation may be necessary. On the other hand, if the company provides extensive information, such as “breast cancer will appear at age thirty-four,” such claims require more regulation because consumers are more likely to rely on such information for treatment purposes.[108] The molecular markers problem compounded with expressivity and penetrance strengthens the case for regulation; even with a perfect correlation between “A” and “ATGC”, it is not clear that the disease associated with “ATGC” may present in the individual. It is even less likely that the disease associated with “ATGC” will present in an individual who actually has “AAGT”. Because of the research used to create these correlations, an individual with characteristics different from the majority population, particularly race, has a much higher likelihood of having test results that are not actually correlative.[109] These factors support the use of strong-effect markers, which move diagnostic companies closer to “perfect correlation”. As it stands, when companies sequence strong-effect markers, the agreement between multiple testing companies is higher than when companies use weak-effect markers (that have more moderate effects).[110] To use the earlier example, suppose your mother’s ancestors did not live in an environment where the plague was prevalent. Rather than facing evolutionary pressures when the plague allele was not present, individuals without the allele would survive at the same rate as the rest of the population. As a result, the population would show a very different distribution of alleles from the original example and the molecular marker of blonde hair would be less helpful. Regulation can mitigate such problems by requiring that consumers receive additional information about the studies that produced the genetic testing data and by creating greater incentives to include a wider population in initial genome-wide-association studies.[111]

C.  Mistakes in Genetic Test Interpretation Occur Because of Unforeseen Interactions between Genes

Not all genetic tests simply convey the likelihood of a single disease based on a single gene. Many tests predict propensity for a disease or for reoccurrence of a condition based on algorithms that approximate the disease pathway.[112] For example, scientists believe that many genes affect the development of schizophrenia.[113] By putting together multiple genotype markers from different locations on the genome, a company could predict an individual’s likelihood of developing schizophrenia. However, if there were errors in developing this algorithm, this clinical prediction would be wrong. Indeed, the chance of error is very high because the interactions between genes can be unpredictable and small-effect genes can be difficult to find. For instance, one study found that a predictive test using the cumulative effects of thousands of small-effect genes was more accurate at predicting the existence of schizophrenia than a test using only major-effect genes.[114] One way that this sort of error can occur is when a lab relies on a meta-analysis, a type of study that puts together data from multiple studies to deduce the interactions between different genes. The problem with meta-analyses is that the actual interaction between two particular genes may not have been tested directly, which could lead to incorrect conclusions. One possible cause for this error is when the two genes are on the same pathway that leads to the disease.[115] In most cases, if gene Z affects a function earlier in a molecular pathway than gene Y, it does not matter if the allele at gene Y carries a mutation, since the pathway would have already been altered due to a mutated allele at gene Z.[116] For a simple example, consider the genes that control certain properties of human hair: suppose that gene Z governs hair growth and gene Y governs hair texture. If the individual carries an allele at gene Z that prevents hair growth entirely, it does not matter if the allele at gene Y is for curly or straight hair, the individual will be unaffected. A meta-analysis of genes like Z and Y might infer increased effects from mutations at both genes when the actual effects remain the same when both genes are present. Even with perfect testing results, there are dangers in how companies present their data to patients.[117] Is it sufficient to say that the patient has an increased likelihood of developing disease sometime in their life? This may depend on the seriousness of the disease tested, and even the character of the testing service. The common errors with (1) molecular marker technology, (2) disease phenotype variation from expressivity and penetrance, and (3) the interaction between different genes show that regulation of genetic diagnostic tests is necessary. Therefore, continuing the status quo of minimal regulation is not a valid option, despite the possible adverse effects on industry.

III. Is a Per Se Patentable Subject Matter Rule for Genetic Diagnostic Tests the Solution?

Now that we have established that there must be some regulation for genetic diagnostic tests, this section examines e statutory or judge-made exclusivity for diagnostic tests to help underwrite the costs of such regulation and bolster the financial viability of genetic testing companies. One such option is a per se patentable subject matter rule. As we discussed earlier, with increased regulation of diagnostic tests, companies become more interested in protecting their investment that they have made in new technologies because of the additional financial outlays required to fund additional regulatory submissions and possible clinical trials. Patents can be one form of protection for these companies. Under the patent system, companies in the United States may file a specification and claims (which must fall within the scope of patentable subject matter and meet patentability criteria of novelty, non-obviousness, enablement and written description) in exchange for twenty years of exclusive rights to the claimed technology.[118] The patent system protects the patent-holder from others who copy their invention by allowing the patent-holder to sue for damages or an injunction if another company’s product infringes their patent.[119] Aside from patenting the genetic identity associated with the disease, companies may also be interested in patenting the algorithm that predicts the propensity for metastasis of tumors or the interaction of multiple genes. While the Hatch-Waxman framework helped counter he delay to market for drugs through patent-term extension, this solution is insufficient for genetic diagnostic tests, given the current uncertainty as to whether such tests fall within the scope of patentable subject matter. Therefore, to provide the same protection to diagnostic test companies as we do drug companies, we must create an equally effective exclusivity backstop. This may be accomplished by bringing all genetic diagnostic tests within the scope of patentable subject matter through the creation of a per se rule (making all genetic diagnostic tests “patent-eligible”), or through FDA-mediated data exclusivity, a feature also found within Hatch-Waxman. This section explores two issues with respect to the creation of a per se rule: (1) does it fit within the current patentable subject matter jurisprudence and (2) is patentability really necessary for the financial success of genetic diagnostic tests, or will the lesser protection of data exclusivity be sufficient? As this section will show, data exclusivity is a superior option because of the possibility of patent over-enforcement stifling innovation in the diagnostic test arena.

A.  A Per Se Rule Does Not Fit within the Patentable Subject Matter Framework

The current era of controversy over patentable subject matter began in 1980 with the seminal Supreme Court case, Diamond v. Chakrabarty, where the inventor, Chakrabarty, filed a patent application claiming “a bacterium from the genus Pseudomonas”—not merely the process claims for producing the genetically modified bacterium, but the bacteria themselves.[120] The Court determined that this bacterium was not an unpatentable natural phenomenon, but rather a non-naturally occurring manufacture or composition of matter, and therefore held it was patentable.[121] A related case from many years earlier is Parke-Davis & Co. v. H.K. Mulford Co., where Judge Learned Hand reasoned in dicta that purified adrenaline was patentable subject matter because it was materially different from the naturally-occurring adrenaline found in the human body.[122]Parke-Davis provides some of the support for the modern patenting of genes, reasoning that such genes are patentable because they are the purified forms—the inventor has separated the gene from the rest of the chromosome and from the other cellular matter in the naturally occurring state.[123] Four recent cases have brought into question the patentability of genetic diagnostic tests. This section discusses each case and determines that not only is the framework for patentable subject matter unpredictable and fact-specific for diagnostic tests, but also that a per se patentability rule would be unfeasible to implement without disrupting the patentability of unrelated industries. The first such case is Laboratory Corporation v. Metabolite Laboratories,[124] where the patent at issue claimed a process that detected deficiency of two vitamins, folate and cobalamin, by measuring the level of homocysteine in the body fluid. If the homocysteine levels fell within a particular range, the patent suggested that there was a vitamin deficiency.[125] In his dissent from the Court’s unusual decision to dismiss the case for certiorari improvidently granted, Justice Breyer wrote that the patent office should not have permitted this patent as it “claim[ed] a monopoly over a basic scientific relationship,” yielding an interpretation that would cause doctors’ medical diagnoses to infringe the patent.[126] Although Justice Breyer’s view was not ultimately adopted by the Court, his remark suggests that the area of patentable subject matter was far from settled and created an opening for subsequent cases.[127] The second major case relevant to our analysis of patentable subject matter is In re Bilski (later Bilski v. Kappos), which did not deal with diagnostic tests, but rather software patents.[128] The inventor, Bilski, claimed an algorithm for hedging.[129] The Federal Circuit disposed of Bilski’s patent by determining that an algorithm fell within the scope of patentable subject matter so long as it was implemented with a non-trivial machine or involved the transformation of matter, reaffirming the so-called “machine-or-transformation test”.[130] As Bilski’s method did not involve any such transformation of matter nor was it implemented in a non-trivial machine, the patent-in-suit was therefore held to be unpatentable.[131] Immediately, it was unclear how Bilski would apply to diagnostic tests and methods. The Federal Circuit applied the “machine-or-transformation test” in a third major case, Prometheus Laboratories v. Mayo Collaborative Services.[132] Prometheus is the exclusive licensee of two patents that calibrate the dosage of thiopurine drugs.[133] Although doctors had used such drugs to treat autoimmune diseases for many years, their efficacy was limited by non-responsiveness and drug toxicity in some patients.[134] To address this complication, the patents claimed the following “method of treatment”: (1) administer a particular dosage of the drug; (2) determine the levels of the drug metabolites found in the body fluid; (3) depending on the levels of those metabolites, either increase or decrease the dosage of the drug subsequently administered.[135] In the original Federal Circuit decision in 2009, Judge Lourie applied the “machine-or-transformation test” and found the requisite transformation in the method of treatment.[136] This overruled the District Court judgment that the patent was invalid because the administration was a trivial “data-gathering step” (rather than fundamental to the invention) and the claims relied on a naturally-occurring correlation.[137] Subsequently, when In re Bilski was appealed to the Supreme Court in Bilski v. Kappos, the Court found the “machine-or-transformation” test was too restrictive, and held that although the test could provide “a useful and important clue” as to the patentability of an invention, it was not the sole test for patentability.[138] In dicta, Justice Kennedy noted that the formalistic “machine-or-transformation test” may have the unfortunate side-effect of rendering “advanced diagnostic medical techniques” unpatentable.[139] Indeed, he was not alone in this fear: when a lower court invalidated Prometheus’ patent based on the Federal Circuit’s machine-or-transformation test, one response alleged that this decision would “threaten to invalidate the entire field of medical treatment and diagnostic patents on which the innovative and lifesaving biotech industry is built.”[140] When the Federal Circuit revisited Prometheus Labs after Bilski v. Kappos, they upheld the court’s initial finding of validity.[141] The court recognized that Congress envisioned a permissive approach to patent eligibility to ensure that “ingenuity should receive a liberal encouragement.”[142] The Federal Circuit found that the decision turned on whether the claims covered a natural phenomenon, whose patenting would entirely preempt the use of that correlation, or whether it was only a particular application of that phenomenon.[143] The court held that it was the latter, and therefore within the scope of patentable subject matter.[144] Whether the Supreme Court upholds the Federal Circuit’s decision is another question. The final major case I address is Association for Molecular Pathology v. United States Patent and Trademark Office, or “the Myriad Case.”[145] The patents enforced by Myriad Genetics claimed the particular nucleotide sequence associated with genes for breast cancer, in effect claiming isolated forms of the genes in the manner prefigured by Parke-Davis.[146] These patents enabled Myriad to have a monopoly over nearly all diagnostic tests for breast cancer caused by the two major breast cancer genes, BRCA1 and BRCA2,[147] causing outrage in the scientific community, as well as amongst patients seeking testing for susceptibility to breast cancer. Because this case deals explicitly with genetic diagnostic tests, it best illustrates the state of affairs that arises in the absence of regulation. Myriad offers a laboratory-developed test, and although the lab is accredited under the CLIA discussed supra, the FDA has never approved the actual test offered by Myriad.[148] Therefore, there is a greater chance that the test conducted by Myriad Genetics leads to incorrect results, and without a legally available second opinion, a patient is left without recourse.[149] In Judge Sweet’s opinion in the Myriad Case, he determined that genes were outside the scope of patentable subject matter and struck down the nucleotide sequence claims (and many other claims in the two patents).[150] The argument was that purified genes, despite whatever Judge Learned Hand may have said in Parke-Davis, are not “markedly different” from naturally-occurring genes, and thus the patent covered a product of nature rendering it invalid.[151] After this decision was appealed to the Federal Circuit, the United States Department of Justice (the “DOJ”) filed an amicus brief that effectively straddled the positions of the two parties, the DOJ supported the decision to disallow purified gene patents but felt that complementary DNA patents should remain valid.[152] The distinction that the DOJ made is that since complementary DNA does not contain introns,[153] it is sufficiently different from naturally-occurring genomic DNA.[154] In contrast, isolated DNA is no different from cotton fibers isolated from the cotton plant; it may be necessary to isolate a substance to make use of it, but by itself does not yield an invention.[155]. The DOJ distinguished the purified adrenaline in Parke-Davis from the purified gene in Myriad by raising a crucial point: with genes it is the similarity to the naturally-occurring substance, rather than the differences from purification, that yield the benefits of the patent for diagnostic and medical purposes.[156] Subsequently, the Federal Circuit reversed Judge Sweet’s District Court decision in part, holding that Myriad’s composition claims to “isolated” DNA molecules were patent-eligible, but finding that those claims comparing DNA sequences were patent-ineligible, as they only involved mental steps.[157] These inconsistent positions have created a great deal of uncertainty about which products would fall within the scope of patentable subject matter, which makes it difficult for companies to make ex ante investments in this technology. Indeed, the positions adopted by Judge Sweet in the Myriad Case and in the DOJ amicus brief might be seen as arbitrary. Both interpret the words of the Constitution, “to promote the progress of science and the useful arts” to exclude a type of invention that they do not feel would promote that progress without any clear scientific demarcation.[158] This analysis shows that there is no clear answer regarding whether genes should or should not be patentable, especially given that the patentability of genes and diagnostic tests is not clearly prohibited by the language of the patent statute.  If one conceptualizes the relationship between a segment of DNA and a disease as correlations, then they are no different from the correlations upheld by the Federal Circuit in Prometheus. If they are conceptualized as products of nature, the boundary shifts depending on how natural something must be to fall within that unpatentable class. Therefore, even where courts declare otherwise, the patentable subject matter question must be driven by policy: would it be beneficial for the progress of innovation to allow the patentability of these technologies? Given the flux of the law on this matter, it would be difficult to create a per se rule declaring that all genetic diagnostic tests were patentable subject matter. Not only would Congress or the courts need to deal with the many scientists and academics who oppose such a position, the rule itself would throw into question the entire jurisprudence dealing with patentable subject matter, with effects on other technologies, as discussed by Justice Kennedy in Bilski.[159] Therefore, it would be unwise for the genetic diagnostic test industry to rely on predicting the direction of the Supreme Court in deciding whether their particular test would fall within the scope of patentable subject matter. Most importantly, even if Congress supported such a rule, it is not clear that blanket patent-eligibility would “promote the Progress of Science and the useful Arts,” as the Intellectual Property Clause of the Constitution requires,[160] as shown by the innovation-stifling actions taken by some patentholders in this arena, as I discuss next.

B.  In the Absence of a Per Se Rule, Companies Still Need a Baseline Level of Exclusivity

In the absence of a per se rule, is the solution to have extensive regulation of genetic diagnostic tests, but not patentability? It is crucial to remember that the Myriad Case took place against a backdrop of no regulation, and even then, companies were concerned about their initial investment in the technology. This section discusses the possibility of data-exclusivity as a lower tier of protection that would enable diagnostic test companies to recoup investment in new tests even if courts find that such tests fall outside of the scope of patentable subject matter. Based on this discussion, I conclude that FDA-mediated data exclusivity for all tests that require clinical testing for approval is the best option. Many articles have discussed blocking patent issues in gene patents.[161] Some have argued that patents are not necessary to incentivize the development of genetic diagnostic tests.[162] Gene patents’ claims cover nearly twenty percent of the human genome, and those interested in doing research on those claimed genes must overcome the hurdle of whatever costs the patent-holder has placed on her patent.[163] University researchers have also asserted that “[the prospect of] patents do not affect research in this area” as most research is done in a university setting, funded by government grants.[164] At the same time, private companies conduct a large amount of research building on university research in developing diagnostic tests,[165] which definitely does depend on patents.[166] A recent study found that at least one patent in either Europe or the United States covered 19 of the 22 most prevalent hereditary diseases, and that many were covered by several patents.[167] In both the United States and Europe, universities are the top patent-holders,[168] and most genetic diagnostic patents originate in the United States,[169]which the study authors speculate stems from the liberal patent policy in this country. In Heller and Eisenberg’s seminal article, the anti-commons is characterized as many cross-cutting patents that require new entrants into the field to license from each of these actors.[170] However, if patents can be circumvented or “designed-around,” no such licenses are necessary. Therefore, the current state of genetic diagnostic patents is not necessarily an anti-commons as imagined by Heller and Eisenberg: as the Huys study recognizes, although 25% of patents claim particular genes, only 3% of these patents cannot be circumvented.[171] However, 38% of these gene patents claimed diagnostic methods, which are generally more difficult to circumvent.[172] The Huys study made no substantive conclusions about the existence of any patent thicket, but noted that the uncertainty associated with the patentability of such tests created more difficulties for inventors in the development of technologies associated with gene patents.[173] More recently, a series of extensive studies, led by Professor Cook-Deegan, focused on ten hereditary diseases and the effects of patents on their treatment and research.[174] Here, I focus firstly on breast cancer and associated cancers, and then on Alzheimer’s Disease tests, given their high rates of occurrence and significant impact on the population. Breast cancer is a particularly lucrative disease for patent-holders like Myriad. For breast cancer, Myriad is the exclusive licensee and sole provider of tests based on BRCA1 and BRCA2 in the United States. For colorectal cancer, tests are available from multiple laboratories (including Myriad), but none have the same monopoly position as Myriad does in breast cancer testing.[175] Therefore, the comparison between breast cancer and colorectal cancer testing is a useful metric of the effects of diagnostic patents. Even as a monopoly entity, Myriad often acts in the public interest. As of August 2008, Myriad has submitted over 18,000 entries for the 2,600 unique mutations to the Breast Cancer Information Core database (a publicly available central repository for information regarding mutations and polymorphisms in breast cancer susceptibility genes).[176] However, Myriad has also limited certain types of research by using a very broad definition of what constitutes infringing—when the Genetic Diagnostics Laboratory began testing patients using National Cancer Institute protocols, while additionally providing breast cancer results to patients, Myriad claimed that this constituted patent infringement.[177] Myriad does not enforce its patents against non-commercial research or against laboratories providing tests that it does not sell,[178] although its ambiguous policy may still create a chilling effect. In contrast, the studies found no similar chilling effect with the two tests offered for colorectal cancer.[179] In House Judiciary Committee hearings, some scientists argued that Myriad purposely did not adopt more cost-effective testing methods.[180] However, the sequencing methods used by Myriad for breast cancer tests are actually cheaper than those used by Myriad and other providers for similar colorectal cancer tests; if the technology has advanced, Myriad’s competitors have not adopted it either.[181] Still, even assuming that Myriad is using the most cost-effective testing methods, its tests remain prohibitively expensive compared to molecular marker tests, breast cancer gene sequencing as offered by Myriad costs $2400 per patient on average, compared with $99 for the 23andMe test that includes preliminary results for various breast cancer risk factors, including breast tissue density. Similarly, Alzheimer’s Disease, which cost the U.S. healthcare system $61 billion in 2002, is a disease where patents can be very lucrative.[182] The majority of those with the disease have late-onset Alzheimer’s, which has only one clearly established risk factor known as APOE.[183] A small percentage of cases arise from early-onset Alzheimer’s, for which there are three dominant mutations found in one of three genes.[184] In the United States, genetic testing for Alzheimer’s is provided almost exclusively by Athena, who has licenses to three major Alzheimer’s gene patents,[185] and offers the test for $475. However, Graceful Earth, a direct-to-consumer testing company, produces a non-FDA approved test for Alzheimer’s for $280,[186] using indirect markers rather than the strong-effect markers used by Athena.[187] In a well-regulated environment, consumers could expect near certainty in their risk results provided by Athena or Myriad, and a reasonable certainty from their marker tests from Graceful Earth or 23andMe. However, such regulation is costly for companies. In the absence of any data exclusivity or patent protection, it is likely that Athena or Myriad would need to provide extensive clinical data to the FDA in order to market their products, while Graceful Earth and 23andMe would have used that data free of charge and benefited from Athena or Myriad’s initial investment. Such copying by other actors might make inventors like Athena or Myriad reluctant to take on the initial burden under additional FDA regulation. The crucial question becomes whether in the presence of additional regulation we need the complete protection offered by patents—20 years of exclusivity in exchange for the information disclosed to the public—or whether a lower-tier of exclusive protection such as clinical data exclusivity would suffice.[188] The answer seems to be that a lower tier of protection would be sufficient, even beneficial. Patents are often enforced indiscriminately against copyists and innovators alike, stifling new growth in this field. A lower tier of exclusive protection would enable new genetic diagnostic tests to be developed in a short period, but allow new innovators to protect their investment in the technology and in producing clinical data for the FDA. Furthermore, the disclosure provided by the patent system in this field is minimal compared to the information found in scientific publications, suggesting that the bargain struck by the patent system may not be fulfilled in the area of genetic diagnostic tests.[189] Furthermore, concerns about lower test quality from lack of competition could be allayed by stringent FDA regulation of these tests.[190] Therefore, a per se patentable subject matter rule is not the solution for protecting the interests of consumers and genetic diagnostic test developers. Not only would it fit poorly within the current patentable subject matter framework, but it is also not the best promoter of innovation. At the same time, a baseline level of exclusivity for a shorter period is required to allay some of the uncertainty of diagnostic test developers, particularly given extensive regulation by the FDA. For these reasons, innovation policy would be better served by granting genetic diagnostic test developers data exclusivity rather than patent protection.

IV. Suggestions for a New Regulatory Regime for Genetic Diagnostic Tests

With the earlier discussions in mind, this section presents a comprehensive framework for the regulation of diagnostic tests. In response to the regulatory problems discussed in Part II, some legislators have suggested the creation of an entirely new department within the FDA to address the unique needs of genetic diagnostic tests.[191] However, such drastic measures are unnecessary. Notably, I do not recommend that the FDA change current procedures for testing analytical validity (allowing the CMS to continue such regulation), nor should the FDA change significantly the current three-class system for medical devices before applying it to genetic diagnostic tests. Instead, I propose the following five simple changes to the regulatory system for genetic diagnostic tests at the FDA.
  1. Increase Requirements for Labeling and Genetic Counseling That Keep up with Developing Technology through Monthly Meetings
With a genetic diagnostic test, adverse consequences from an incorrect result can be significant, leading to mismanagement of the disease or unnecessary surgery.[192] In light of the known errors in the genetic test methods outlined earlier, I would add more to the FDA’s analysis. Firstly, there should be different oversight for those tests that rely on molecular markers and for those that rely on sequencing. The former should have clear labeling outlining how molecular markers may not predict your risk at all, particularly for patients of certain backgrounds. Molecular marker tests that frequently lead to disparate results when using different markers should be required to use multiple-effect markers for the same gene, or consistently use strong-effect markers. Secondly, the FDA should mandate expansive labels outlining factors required in the interpretation of diagnostic tests. In a long-term study of effects from a representative genotyping test, researchers found no indications of test-related distress in 90.3% of participants.[193] However, the Bloss study’s authors acknowledged the unusual demographics of the study cohort: most of those who completed the study had some post-graduate education, and of the 44% of participants who failed to follow up after testing, most had completed a four-year college degree.[194] To address this artifact, the FDA should expand the labeling requirements in the test results for the presentation of certain disease alleles. Many companies that provide genotyping for multiple markers have associated websites and blogs that are continually updated with new information associated with the subject’s disease alleles. If the initial test results are provided on the website, this format is adequate. If they are communicated via mail, the FDA should mandate associated genetic risk information and context for the most severe disease alleles in this printed report. Thirdly, certain tests should require consultations with a genetic counselor. Informally, the FDA has indicated that the availability of consultation with certain tests will be taken into account in the approval process, but no official guidelines have mentioned such treatment.[195] I recommend these consultations despite the Bloss study’s finding that only 10% of cohort participants elected free genetic counseling,[196] partly because 26% of cohort participants elected instead to show their test results to their physician. Because most physicians in the United States feel ill-equipped to deal with such genomics data,[197] these results suggest that genetic counselors should attempt to augment the role of the physician, which includes taking into account family histories and environmental factors—elements that can affect the expression of the gene in question and help reduce errors from penetrance and expressivity—when interpreting test results. Occasionally, the FDA can choose to allow a company to only offer the particular genetic diagnostic test to members of a particular population or racial cohort or require additional scientific studies for support from the manufacturer, as many genetic diagnostic tests have fidelity within certain populations.[198]  The FDA has a reasonably well-defined set of non-binding regulations governing the collection of race and ethnicity in clinical trials, and in these regulations it requires application sponsors to analyze whether dosage must be altered for certain sub-groups.[199] However, although the FDA is cognizant of these issues, it may be difficult to give proper weight to each issue when a test offers data for multiple traits, each prefaced on different studies.[200] Finally, every month, a committee should convene in the FDA to determine future regulations and options for genetics-based diagnostic tests. This committee should have limited conflict-regulation options.[201] There should be options for regulations that deal with race and sex-based differences in the test data. Although the FDA has stated that it would be difficult to hold these meetings with regularity to address new developments in the field, it would be difficult to regulate this technology without them.[202] Until a decade ago, most genes were discovered using extensive family histories from cultures that kept good records and often had a history of inbreeding. With the advent of better sequencing technology, genome-wide association studies have become more widespread.[203] As a result, the information that is translated into diagnostic tests, particularly genetic diagnostic tests, is rapidly advancing. This means that a command-and-control approach will likely become obsolete every few years. This suggestion allows the FDA to move incrementally, detailing how the present medical device framework will apply to diagnostic tests, with frequent consultation from advisory groups. These groups can specifically examine the problems with genetic tests discussed earlier: expressivity, penetrance and, complex traits involving multiple genes. As genes become better-defined, these committees can ensure that tests do not rely on outdated data.
  1. Provide Data-Exclusivity for Class II Tests That Require Additional Clinical Studies.
The FDA has admitted that it does not take into account costs or the patent system in its analysis.[204] However, as discussed supra, biotechnology and pharmaceutical companies do take into account the protections that the patent system offers when deciding which products to pursue on a commercial scale.[205] Furthermore, the patentability of these diagnostic tests that the FDA seeks to regulate is not clear. While Judge Sweet and the subsequent amicus brief by the Department of Justice have made a reasonable argument to limit the patentability of the “purified form of a gene,” this position does not clarify the patentability status of other genetic products. Nor will such questions be resolved quickly. Indeed, the biggest lament of biotechnology companies has been that the status of these products is so uncertain that it makes it difficult to invest in these new technologies. At the same time, numerous studies on the development of genetic diagnostic tests for breast cancer, colorectal cancer, and Alzheimer’s Disease have shown that much of the research that has led to these discoveries is not privately funded, but rather funded by the government.[206] In fact, it is not until the commercialization stage that private actors generally enter the picture. In the previously open regulatory climate, the comparatively small investment required for transferring a laboratory invention to a commercial product meant that patents on such technologies were not always beneficial, and occasionally harmful when enforced stringently. However, the regulatory situation has changed, which means that those interested in providing diagnostic tests must invest more to bring their tests to market. Given the comparatively smaller investment required for developing diagnostic tests compared to pharmaceutical drugs,[207] as well as the smaller scope of clinical trials that must be completed, there is an appropriate role for the FDA to provide a baseline level of data exclusivity, expanding the current Safe Medical Devices Act regime and exclusivity regimes under the Orphan Drug Act or the Pediatric Exclusivity Act.[208]In the latter models, companies receive data exclusivity from the FDA when they voluntarily undertake clinical studies suggested by the FDA to allow the use of their current drugs on children or to treat rare diseases. The Pediatric Exclusivity Act provides six months exclusivity for all uses of the drug, not just pediatric uses, while the Orphan Drug Act provides an extraordinary seven years of exclusivity.[209] Here, a three-year exclusivity term for Class II diagnostics can incentivize companies and even university groups to go the extra step of preparing data for regulation without worrying about rival groups using their test data for their own approval, creation of a performance standard or the classification of their device.
  1. Create Mandatory Maximum Times for Class II and Class III Approval Processes or Increase the Scope of Class II Devices to Decrease Overall Approval Times.
Suggested guidelines for Class II and III approval processes include: (a) no 510(k) clearance for an in vitro diagnostic should take longer than 6 months from start to finish and (b) no PMA for an in vitro diagnostic should take longer than 12 months from start to finish. It is crucial that in providing an additional safety factor, these additional regulations do not also institute a new obstacle. The Center for Devices and Radiological Health has a two-fold mission: to protect public health while promoting public health through new medical technologies.[210] This second mission cannot be fulfilled through an inefficient approval process. In implementing these new time-frame limitations, the FDA must also implement a more transparent monitoring system to track how long the approval process really takes. The FDA has often argued that this additional time provides additional safety for consumers. Firstly, from studies discussed supra, it is evident that although some additional time can be useful, the extreme amounts of time that the FDA has taken have not translated into commensurate benefits for consumers.[211] Furthermore, it is not clear that the safety gains from such delays outweigh the adverse effects on patients from delays in approving life-saving drugs.[212] The timelines suggested here are based on the average review times for equivalent products in the European Union and the review times that the FDA itself espouses in the FDA white paper discussed earlier. However, it may be insufficient to require shorter review times. As an FDA white paper previously addressed after the passage of the Prescription Drug User Fee Act, the review times suggested have been contemplated by the agency.[213] Therefore, I also suggest that in order to classify genetic diagnostic tests into Class III, the burden be placed on the FDA to show why prior adverse actions by individuals have necessitated this stringent classification, creating Class II and the 510(k) clearance process as a default position for most genetic diagnostic tests.
  1. Solicit Complaints Directly from Consumers
The complaint procedures for faulty tests must be streamlined because under the proposed scheme consumers with direct access to genetic diagnostic tests will use the software rather than just healthcare providers. All companies with web-presences—even if they do not have a software or computational test—must clearly link patients to a complaint page provided by the FDA, currently available through the MedWatch system.[214] Indeed, experts estimate that the current voluntary reporting system by physiciansvastly reduces the number of complaints reported to the agency, or only about ten percent of all adverse events.[215] Consumers may submit such complaints for slow-turnaround in tests, a frequent complaint made about small-testing operations, in addition to incorrect test results and misleading language.  Turnaround time can be important: for example, PGxHealth, a commercial provider of genetic tests, returns results within two months, compared to one year for research-based providers of the same genetic tests.[216]Many companies have offered direct-to-consumer genetic testing earlier in the treatment cycle than other diagnostic tests, treating them as a “first-pass-through” before submitting a patient to extensive medical treatment or physician analysis.[217] If genetic diagnostic test results are not available quickly, this role as a preliminary diagnostic measure is lost.
  1. Require Minimal Regulation of Purely Software-Based Genetic Diagnostic Tests
Despite some jurisdictional issues, it is clear that software-based tests are “systems intended for use in the diagnosis of disease or other conditions” within the meaning of the Medical Device Amendments of 1976, even though they do not directly involve the withdrawal of samples from a patient. As whole-genome sequencing becomes an increasingly viable option for consumers,[218] such software-based tests will become more common. Armed with a copy of their full-sequence, users can run their data through frequently updated software that takes into account more recent meta-analyses and scientific studies. Indeed, such software is similar to the PrometheASE program already used by many in analyzing the raw data provided by 23andMe and Navigenics.[219] If the FDA could only regulate the companies who actually take samples from their clients to create the whole-genome sequences, they would not be regulating clinical validity at all, but the analytical validity of high-throughput sequencing machines, a mostly useless task. Furthermore, it would be unusual to require clinical trials from entities that may consist exclusively of programmers. If the programmers have properly designed the software from existing data in published papers, the FDA should easily complete the verification of such data. This situation may require the FDA to alter the types of individuals who work in their testing division and require a more ad hoc approach to such tests, such as including more individuals familiar with computer programming or statistical techniques and involving the creation of trials that depend on paper data, or data that is not from clinical trials of the particular genetic test, but previously published research on the particular genetic correlation. If the test cannot be verified based on published data, then the programmer cannot make them available to consumers. If the programmer wishes to take on the burden of clinical testing to have the option of offering it to consumers, this option should be left open. Additionally, with software-based tests and other tests that are updated frequently, the FDA should implement a simple online submission system for previously verified tests. This will function differently from pre-market or post-market options, and act as more of a constant monitoring system: the entity seeking approval should submit the change to the test along with the study that supports the change, allowing minimal FDA oversight to verify test changes.

V. Application of Proposed Regulatory Regime to a Challenge for the FDA

This section analyzes the application of this regulation to a test that flouts the conventions for both patents and FDA regulation, what Nature describes as “The Renegade Gene Test.” This test—also known as the Salzberg test is a computer program that can check any genome for 68 gene mutations that increase the risk of breast cancer and other cancers.[220] Although there are over 1000 gene mutations allegedly linked with the occurrence of breast cancer, the programmers believe that their program can be easily expanded to include the effects of those mutations when the time comes. Their software is freely available under an open source license that allows others to use, modify, and redistribute the program.

A.  Regulation of Analytical Validity and Classification of the Tests

Under the suggested regime, neither the Myriad test nor the Salzberg test would require any testing for analytical validity. Since the Salzberg test is an open source framework, this might create problems, but would likely be resolved through the current user generated content model that has been successful with Linux and Wikipedia. Since breast cancer is a serious disease, both the Myriad test and the Salzberg test could fall under either Class II or Class III. Because women may undergo preemptive mastectomies and other serious treatment after taking the Myriad test, but might not pursue the same options under the Salzberg test, Salzberg might argue for a lower level of regulation. As a result, Myriad could be regulated as a Class III device, while Salzberg could be regulated as a Class I or II device. Regardless, the Myriad test would likely be approved with little difficulty within the time frames suggested supra, although the FDA might require warnings addressing the utility of a negative result on the test, as many breast cancer patients do not necessarily have mutations in either BRCA1 or BRCA2, two genes associated with the onset of breast cancer. Salzberg would first argue that computational tests fall outside of FDA’s regulatory jurisdiction. This argument would likely be unsuccessful because computational tests clearly fall within the language of the Medical Device Amendments of 1976.[221] The approval of the Salzberg test would also be somewhat difficult. Because it only looks at 68 gene mutations when 1000 are possible, the test does not “represent a comprehensive list of BRCA mutations.”[222] Since the clinical studies published on breast cancer do not rely on such a small subset of gene mutations, the FDA could not rely on the computational test providing the same accuracy as if it had 1000 mutations. At this juncture, this Note has recommended that the FDA refrain from requiring full clinical trials from computational test providers. Instead, the FDA may require that Salzberg either cite an earlier study that demonstrates that 68 mutations can predict the disease with the same accuracy as 1000 mutations, or at least show the level of accuracy provided using these 68 mutations (or molecular markers, as the case may be). It is likely that Salzberg would need to refine the test to include all 1000 mutations because of the considerable expense involved in conducting their own clinical trials. Even if Salzberg and Pertea provided a test that did use all 1000 mutations, the FDA would likely require them to place certain warnings on their website detailing the possible errors from inputting in the raw genome data on one’s own, as well as a link to the FDA complaint page suggested supra.

B.  Possible Patent Protection and Data Exclusivity

At minimum, the Salzberg test would infringe upon a number of already existing gene patents. However, because individuals would have to input their own genetic data to perform the actual test covered by the patents, Myriad could only sue the programmers Salzberg and Pertea for inducing infringement, rather than direct infringement.[223] Let us suppose that the Supreme Court strikes down the Federal Circuit’s Myriad decision and that the only option is for data-exclusivity under the plan outlined in Part IV. Given the clear correlations available between the BRCA1/BRCA2 genes and breast cancer, it is unlikely that Myriad would have needed additional clinical trials for approval at the FDA. Since it does not need to submit new data and can rely on published data, there is no option for data exclusivity at the FDA and Myriad would receive no protection from those who wish to replicate its test. This is likely a good thing, given that the earlier studies (discussed supra in Part III) show that more options for breast cancer tests would be available at cheaper options if the Myriad patents were not enforced as stringently. However, it could also cause future companies like Myriad to avoid publishing its results when a new gene is discovered, particularly as the patent system is no longer an option and thus there is no mechanism forcing disclosure. However, this threat is largely illusory, as the identity of the gene can be derived from the publicly available test. Therefore, the proposed regulatory scheme in this paper the proper incentives for both innovators and consumers.

Conclusion

This paper could not possibly cover all of the issues arising from the regulation of diagnostic tests or the questions of their patentability. The proposed regulation aims to minimize the effects of three common errors in genetic diagnostic tests affecting consumers by increasing labeling and genetic counseling requirements and adopting new testing protocols, while reducing ex ante uncertainty for genetic diagnostic test companies by clarifying classification regimes and approval times and providing for a data exclusivity back-stop of three years if the test is found to be outside the scope of patentable subject matter. The regulation also aims to improve FDA administration by enhancing adverse effect reporting and maintaining monthly meetings to address the rapid improvements in the field.
 
* Copyright © 2012 by Sharonmoyee Goswami. J.D. Candidate, 2012, New York University School of Law; B.S.E., 2009 (Chemical Engineering), Princeton University 2009. Many thanks to the Honorable Robert A. Katzmann for his invaluable feedback, and to Professor Richard Epstein for his helpful comments. I would also like to thank the staff of the Journal of Intellectual Property & Entertainment Law, the Notes Editors of the Journal of Legislation & Public Policy, and Ambrose Carr for their thoughtful edits.
[1] Juliana Han, Note, The Optimal Scope of FDA Regulation of Genetic Tests: Meeting Challenges and Keeping Promises, 20 HARV. J. L. & TECH. 423, 427 (2006-2007).
[2] Id.
[3] Meeting Notice, 75 Fed. Reg. 34463 (June 17, 2010).
[4] Data exclusivity means that competitors are not permitted to rely on the pioneer drug company’s trials for their own approval process. Rather, they must perform their own trials for FDA approval.  See Small Business Assistance: Frequently Asked Questions for New Drug Product Exclusivity, FDA.GOV, http://www.fda.gov/Drugs/DevelopmentApprovalProcess/Small BusinessAssistance/ucm069962.htm (last visited Nov. 29, 2011).
[5] See Prometheus Labs., Inc. v. Mayo Collaborative Servs., 628 F.3d 1347, 1349–50 (Fed. Cir. 2010), cert. granted, 131 S.Ct. 3027 (June 20, 2011) (No. 10-1150).
[6] Richard B. Stewart, The Reformation of American Administrative Law, 88 HARV. L. REV. 1667, 1671–76 (1975).
[7] See H. Thomas Austern, Philosophy of Regulation: A Reply to Mr. Hutt, 28 FOOD & DRUG COSMETIC L.J. 189 (1973).
[8] Id. at 191 (quoting Justice Cardozo).
[9] FDA History Part I: The 1906 Food and Drugs Act and Its Enforcement, FDA.GOV (June 18, 2009), http://www.fda.gov/AboutFDA/WhatWeDo/History/Origin/ucm054819.htm.
[10] Federal Food, Drug, and Cosmetic Act, Pub. L. No. 75-717, 52 Stat. 1040 (1938).
[11] FDA History Part II: The 1938 Food, Drug and Cosmetic Act, FDA.GOV (June 18, 2009), http://www.fda.gov/AboutFDA/WhatWeDo/History/Origin/ucm054826.htm (the culprit was a miracle drug known as “Elixir Sulfanilamide” mixed with ethylene glycol, or antifreeze, that caused the deaths of 100 people, many of them children).
[12] MORRIS UDALL, 87TH CONG., THALIDOMIDE: A CRIPPLING DRUG PROMISES GREATER PROTECTION FOR CONSUMERS (Aug. 17, 1962), available at http://www.library.arizona.edu/exhib its/udall/congrept/87th/620817.html.
[13] Peter Barton Hutt, The Transformation of United States Food and Drug Law, J. ASS’N FOOD & DRUG OFFICIALS, Sept. 1996, at 1. Indeed, this has not changed even today.  In April 2011, the FDA was criticized at a Senate Hearing for allowing high-risk artificial hips to be sold without sufficient stress-testing. Barry Meier, Group Faults FDA on Oversight of Devices, N.Y.TIMES, Apr. 12, 2011, at B10; see also A Delicate Balance: FDA and the Reform of the Medical Device Approval Process Before the S. Special Comm. on Aging, 112th Cong. (2011) (statement of William Maisel, Deputy Center Director for Science and Chief Scientist of the Center for Devices and Radiological Health, FDA).
[14] PDUFA created a fee system for the FDA to allow the agency to increase its small budget and approve drugs faster. Prescription Drug User Fee Act § 502, 21 U.S.C. § 379g (1992). A similar framework was created some years later for medical devices.  Medical Device User Fee and Modernization Act of 2002, Pub. L. No. 107-250, 116 Stat. 1588 (codified as amended in scattered sections of 21 U.S.C).
[15] Known as the Drug Price Competition and Patent Term Restoration Act, Pub. L. 98-417, 98 Stat. 1585 (1984).
[16] Hutt, supra note 14, at 1.
[17] Id.
[18] A good example is FDA regulation of Good Manufacturing Practices, where the agency created a minimum Safety Assurance Level (SAL) through a guideline, which they sought to enforce. United States v. Bioclinical Sys., Inc., 666 F.Supp. 82, 83 (D.Md. 1987) (finding that the FDA may not enforce the guideline).
[19] Stuart Hogarth et al., The Current Landscape for Direct-to-Consumer Genetic Testing: Legal, Ethical & Policy Issues, 9 ANNU. REV. GENOMICS HUM. GENETICS 161, 170 (2008).
[20] Id.
[21] Id.
[22] Misha Angrist et al., Impact of Gene Patents and Licensing Practices on Access to Genetic Testing for Long QT Syndrome. 12 GENETICS IN MEDICINE S111, S124 (Supp. 2010).
[23] Han, supra note 2, at 429.
[24] Id.; Hogarth et al., supra note 20, at 168.
[25] Courtney Harper, Remarks at the Public Meeting on Oversight of Laboratory Developed Tests 35 (July 19, 2010) [hereinafter Meeting Day 1] (transcript available at http://www.fda.gov/downloads/medicaldevices/…/ucm226203.pdf).
[26] Id.
[27] 21 C.F.R. § 809 (2006); Hogarth et al., supra note 20, at 172.
[28] Gail H. Javitt & Kathy Hudson, Federal Neglect: Regulation of Genetic Testing, ISSUES IN SCIENCE AND TECHNOLOGY (2006), http://www.issues.org/22.3/javitt.html.
[29] Meeting Day 1supra note 26, at 14–15.
[30] Id. at 35.
[31] Id. at 17.
[32] See Id. at 26–27.
[33] Id. at 17–18.
[34] Id. at 23; SEC’Y’S ADVISORY COMM. ON GENETICS, HEALTH, AND SOCIETY, REVISED DRAFT REPORT ON GENE PATENTS AND LICENSING PRACTICES AND THEIR IMPACT ON PATIENT ACCESS TO GENETIC TESTS 10 (2010) [hereinafter SACGHS REPORT].
[35] Medical Devices; Classification/Reclassification; Restricted Devices; Analyte Specific Reagents, 62 Fed. Reg. 62,243, 62,250 (1997).
[36] Hogarth et al., supra note 20, at 173; Draft Guidance: In Vitro Diagnostic Multivariate Index Assays, 72 Fed. Reg. 41081 (proposed July 26, 2007), 4 [hereinafter Algorithm Assay Draft Guidance], available at http://www.fda.gov/downloads/MedicalDevices/DeviceRegulation andGuidance/GuidanceDocuments/ucm071455.pdfMeeting Day 1supra note 26, at 25–26.
[37] Meeting Notice, 75 Fed. Reg. 34463 (June 17, 2010).
[38] Id.
[39] 21 U.S.C. § 360(k).
[40] See Nutraceutical Corp. v. Von Eschenbach, 459 F.3d 1033, 1036 (10th Cir. 2006) (discussing the numerous Adverse Event Reports that the FDA received from customers after taking Ephedra); Adverse Event Reporting System (AERS), FDA.GOV, http://www.fda.gov/Drugs/ GuidanceComplianceRegulatoryInformation/Surveillance/AdverseDrugEffects/default.htm (last visited Nov. 29, 2011).
[41] LARS NOAH, LAW, MEDICINE, AND MEDICAL TECHNOLOGY 269 (2d ed. 2007).
[42] 21 U.S.C. § 360(k) (2010); Algorithm Assay Draft Guidance, supra note 37, at 7–8.
[43] 21 C.F.R. §§ 807.81-807.100; Algorithm Assay Draft Guidance, supra note 37 at 11.
[44] Katherine Serrano, Office of In Vitro Diagnostic Device Evaluation and Safety, Meeting Day 1supra note 26, at 50 (prior to the de novo process, all new devices were by default Class III devices even if they were not high-risk). The FDA is currently overhauling some of its 510(k) approval pathway, including streamlining the de novo application, creating a panel of external experts and a Center Science Council, and establishing a public database of medical device information. Plan of Action for Implementation of 510(k) and Science Recommendations (January 2011), available at http://massdevi.server265.com/podcast/510(K)_Implementation_ Chart.pdf.
[45] Katherine Serrano, Meeting Day 1supra note 26, at 54–55.
[46] JOSH MAKOWER ET AL., FDA IMPACT ON U.S. MEDICAL TECHNOLOGY INNOVATION: A SURVEY OF OVER 200 MEDICAL TECHNOLOGY COMPANIES 13 (2010).
[47] Id.
[48] 21 U.S.C. § 360e (2010).
[49] Algorithm Assay Draft Guidance, supra note 37, at 13.
[50] Katherine Serrano, Meeting Day 1supra note 26, at 49.
[51] MAKOWER ET AL., supra note 47, at 12.
[52] Katherine Serrano, Meeting Day 1supra note 26, at 57.
[53] Id. at 45.
[54] Algorithm Assay Draft Guidance, supra note 37, at 8.
[55] MAKOWER ET AL., supra note 47, at 7.
[56] Id. at 20.
[57] Id.
[58] Id. at 22. This delay is much longer than the three month timeline claimed by the FDA’s own data. See id. at 21.
[59] Id. at 22.
[60] Id. at 7.
[61] Id. at 22.
[62] Id. at 25. This risk-averse approach is partly due to the influence from other institutions, including Congress and the courts, which have often overruled the FDA for allowing a risky product onto the market, but rarely for failing to allow a safe product to enter the market. Richard A. Epstein, The Case for Field Preemption of State Laws in Drug Cases, 103 NW. U. L. REV. 463, 469-470 (2009).
[63] MAKOWER ET AL., supra note 47, at 7.
[64] Chips comprise thousands of probes for cDNAs which bind to sections on a DNA sample that are identical to the cDNA. This enables companies to determine easily whether the sample-holder’s genome contains the probe regions. cDNA, or complementary DNA, is DNA synthesized from an mRNA template (the blueprint for the protein product made from the DNA strand) in a reaction catalyzed by the enzyme reverse transcriptase and the enzyme DNA polymerase. Chromatin Immunoprecipitation Sequencing, ILLUMINA.COM, http://www.illumina .com/technology/chip_seq_assay.ilmn (last visited Nov. 7, 2011).
[65] § 210(h) of the Food, Drug and Cosmetic Act defines what falls under the “device” definition. 21 U.S.C. § 321(h). Since Illumina’s chip ostensibly fell under this definition, it was eligible for regulation by the FDA.
[66] Letter from Dr. Alberto Gutierrez , Office of In Vitro Diagnostic Device Evaluation and Safety, Center for Device and Radiological Health to Mr. Jay T. Flatley, President & CEO, Illumina, Inc. (June 10, 2010), available at http://www.fda.gov/downloads/MedicalDevices/Reso urcesforYou/Industry/UCM215242.pdf. Letters were also apparently sent to other chip manufacturers, according to the FDA. Mary Carmichael, DNA Dilemma: The Full Interview with the FDA on DTC Genetic Tests, NEWSWEEK, August 5, 2010, http://www.newsweek.com/blogs/ the-human-condition/2010/08/05/dna-dilemma-the-full-interview-with-the-fda-on-dtc-genetic-tests.print.html.
[67] FDA Letters to Industry (June 10, 2010), available at http://www.fda.gov/Medical Devices/ResourcesforYou/Industry/ucm111104.htm.
[68] See Letter from Dr. Alberto Gutierrez, Office of In Vitro Diagnostic Device Evaluation and Safety, Center for Device and Radiological Health to Ms. Anne Wojcicki, President and Co-Founder, 23andMe, Inc. (June 10, 2010), available at http://www.fda.gov/downloads/Medical Devices/ResourcesforYou/Industry/UCM215240.pdf; Letter from Dr. Alberto Gutierrez, Office of In Vitro Diagnostic Device Evaluation and Safety, Center for Device and Radiological Health to Mr. Earl M. Collier, Jr., deCODE Genetics (June 10, 2010), available at http://www.fda.gov/ downloads/MedicalDevices/ResourcesforYou/Industry/UCM215241.pdf.
[69] Id.
[70] Letter from Dr. Alberto Gutierrez, Office of In Vitro Diagnostic Device Evaluation and Safety, Center for Device and Radiological Health to Mr. Jorge Conde, Co-Founder & CEO, Knome, Inc. (June 10, 2010) (on file with author), available at http://www.fda.gov/downloads/MedicalDevices/ResourcesforYou/Industry/UCM215239.pdf.
[71] Id.
[72] Letters to Manufacturers Concerning Genetic Tests (July 19, 2010) (on file with author), available at http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/InVitroDiagno stics/ucm219582.htm.
[73] Robert Temple, Policy developments in regulatory approval, 21 STATISTICS IN MEDICINE 2939, 2939–40 (2002).
[74] DAVID FRUM, HOW WE GOT HERE: THE 70S 180 (2000).
[75] Rebecca Eisenberg, Patents, Product Exclusivity, and Information Dissemination: How Law Directs Biopharmaceutical Research and Development, 72 FORDHAM L. REV. 477, 479–80 (2003).
[76] The Federal Circuit in Roche v. Bolar effectively eliminated the judge-made experimental use exemption as a protection for generic pharmaceutical manufacturers. 733 F.2d 858, 863 (Fed. Cir. 1984) (finding that the experimental use exception only covers use for amusement, idle curiosity, or strictly philosophical inquiry, but not for business purposes). The ruling in Roche was crucial because the previous year, the Supreme Court had held that generic drugs needed to provide data to the FDA for approval and could not rely on the data previously provided by the pioneer drug manufacturer. United States v. Generix Drug Corp., 460 U.S. 453, 460 (1983).
[77] Glaxo, Inc. v. Novopharm, Ltd., 110 F.3d 1562, 1568 (Fed. Cir. 1997).
[78]  Small Business Assistance: Frequently Asked Questions for New Drug Product Exclusivity, FDA.GOV, http://www.fda.gov/Drugs/DevelopmentApprovalProcess/SmallBusiness Assistance/ucm069962.htm (noting that a 505(b)(2) or 21 U.S.C. § 355(b)(2) application for new indications provides the basis for three additional years of data exclusivity).
[79] See 21 U.S.C. § 360j(h)(4).
[80] NOAH, supra note 35, at 871.
[81] See Victoria Pratt, Remarks at the Public Meeting on Oversight of Laboratory Developed Tests 17 (July 20, 2010) [hereinafter Meeting Day 2] (transcript available at http://www.fda.gov/ downloads/MedicalDevices/NewsEvents/WorkshopsConferences/UCM226204.pdf).
[82] ANTHONY J.F. GRIFFITHS ET AL., INTRODUCTION TO GENETIC ANALYSIS 3 (9th ed.2007).
[83] Id. at 4–5, 8.
[84] Id. at 8.
[85] Id. Note that eye-color is actually governed by many genes (it is a polygenic trait); this example is not indicative of the actual complex genetics of eye-color. Although polymorphisms in the OCA2 gene account for 74% of variation in eye color, other genes such as HERC2 also play a role in eye color variations. See Duffy et al., A Three Single-Nucleotide Polymorphism Haplotype in Intron 1 of OCA2 Explains Most Human Eye-Color Variation, 80 AM. J. HUMAN GENETICS 241, 248 (2007).
[86] Jeff M. Hall et al., Linkage of Early Onset Familial Breast Cancer to Chromosome 17q21, 250 SCIENCE 1684, 1684 (1990).
[87] See generally Annegret Müller & Richard Fishel, Mismatch Repair and the Hereditary Non-polyposis Colorectal Cancer Syndrome (HNPCC), 20 CANCER INVESTIGATIONS 102 (2002).
[88] Id.
[89] SACGHS REPORT, supra note 35, at 56. As mentioned supra, analytical validity is the ability of a test to measure the characteristic that it was designed to measure, without drawing any particular conclusions from that characteristic.
[90] Pauline C. Ng et al., An agenda for personalized medicine, 461 NATURE 724, 724 (2009).
[91] Id.
[92] Angrist et al., supra note 23, at S112. Myriad Genetics, one of the defendants in the A.M.P. v. U.S.P.T.O. case alleging the invalidity of patents claiming human genes, has greater than 99% analytical validity, despite its current unregulated status. See Robert Cook-Deegan et al., Impact of gene patents and licensing practices on access to genetic testing for inherited susceptibility to cancer: Comparing breast and ovarian cancers with colon cancers. 12 GENETICS IN MEDICINE S15, S22 (2010); see also Telephone Interview with Alberto Gutierrez, Director of the Office of In Vitro Diagnostics, Food and Drug Administration and Katherine Serrano, Office of In Vitro Diagnostics, Food and Drug Administration (Jan. 13, 2011) [hereinafter Gutierrez Telephone Interview].
[93] Denise Gellene, Illumina Slashes Cost of Individual Genome Sequencing Service. XCONOMY (June 11, 2010), http://www.xconomy.com/san-diego/2010/06/09/illumina-slashes-cost-of-individual-genome-sequencing-service/ (last visited Nov. 29, 2011). However, the price of whole-genome sequencing is expected to rapidly decrease with improving technology, perhaps even to $1,000.  See Andrew Pollack, Taking DNA Sequencing to the Masses, N.Y. TIMES, Jan. 5, 2011, at B1, available athttp://www.nytimes.com/2011/01/05/health/05gene.html [hereinafter Pollack I].
[94] Personal Genome Service, 23ANDME, https://www.23andme/you/faqwin/sequencing/ (last visited Nov. 4, 2011).
[95] S.K. Cohn & L.T. Weaver, The Black Death and AIDS: CCR5-?32 in genetics and history, 99 Q.J. Med. 497, 499 (2006). Note that although Northern Europeans are likely to carry CCR5, the gene is not actually associated with blonde hair, as with this example. Id. at 498–99 (suggesting that Northern Europeans are likely to carry CCR5); see also S. Ito & K. Wakamatsu, Diversity of human hair pigmentation as studied by chemical analysis of eumelanin and pheomelanin, 25 J. European Acad. Dermatology & Venereology 1369, 1369.
[96] Humans are diploid; they have two copies of each gene except in the sex chromosomes. GRIFFITHS ET AL., supra note 83, at 3, 35; see also Gregor Mendel, Verhandlungen des naturforschenden Vereines in Brünn [Experiments in Plant Hybridization], ABHANDLUNGEN, 3, 12-13 (1865).
[97] They are not. The primary hair color gene, MC1R, is found on chromosome 16. Ira Gantz et al., Mapping of the Gene Encoding the Melanocortin-1 (a-Melanocyte Stimulating Hormone) Receptor (MC1R) to Human Chromosome 16q24.3 by Fluorescence in Situ Hybridization, 19 GENOMICS 394, 394 (1994). The primary eye color gene, OCA2 is found on chromosome 15. Carolina Rooryck, et al., High resolution mapping of OCA2 intragenic rearrangements and identification of a founder effect associated with deletion in Polish albino patients, 129 HUMAN GENETICS 199, 199 (2011).
[98] GRIFFITHS ET AL., supra note 83, at 32 (explaining the concept of dominance using Mendel’s experiments with peas).
[99] Id. at 119.
[100] Ng. et al., supra note 91, at 725.
[101] Han, supra note 2, at 429; Hogarth et al., supra note 20, at 167.
[102] DANIEL L. HARTL & ELIZABETH W. JONES, GENETICS: ANALYSIS OF GENES AND GENOMES 106 (6th ed. 2005).
[103] GRIFFITHS ET AL., supra note 83, at 247.
[104] Hall et al., supra note 87, at 1684.
[105] GRIFFITHS ET AL., supra note 83, at 247.
[106] Cook-Deegan et al., supra note 93, at S26.
[107] Amy L. McGuire et al., Regulating Direct-to-Consumer Personal Genome Testing, 330 SCIENCE 181, 181 (2010).
[108] Gutierrez Telephone Interview, supra note 93.
[109] Esteban Gonzáles Burchard et al., The Importance of Race and Ethnic Background in Biomedical Research and Clinical Practice, 348 NEW ENG. J. MED. 1170, 1171 (2003).
[110] Ng et al., supra note 91, at 725.
[111] Burchard et al., supra note 110, at 1174–75.
[112] Letter from The International Schizophrenia Consortium, Common Polygenic Variation Contributes to Risk of Schizophrenia and Bipolar Disorder, 460 NATURE 748, 748 (2009).
[113] Id.
[114] Id.
[115] GRIFFITHS ET AL., supra note 83, at 242.
[116] Id.
[117] Compare Brendan Maher, The Pitfalls of Tracing Your Ancestry, NATURE NEWS (Nov. 13, 2008), http://www.nature.com/news/2008/081113/full/news.2008.1227.html (suggesting that individuals may be afraid of genetic tests telling them that they are actually of a different genetic background then they expected), with Cinnamon S. Bloss et al., Effect of Direct-to-Consumer Genomewide Profiling to Assess Disease Risk, 364 NEW ENG. J. MED., 524, 524 (2011) (finding that 90.3% of individuals were not upset by the results of their genetic test).
[118] See generally  35 U.S.C. §§ 1-376. Patentable Subject matter is addressed in § 101, novelty in § 102, non-obviousness in § 103, and the enablement and written description requirements in § 112.
[119] Id. at § 271.
[120] Diamond v. Chakrabarty, 447 U.S. 303, 305 (1980). As a result, anyone else using these bacteria—a living thing—would infringe Chakrabarty’s patent.
[121] Id. at 309-10.
[122] Parke-Davis & Co. v. H.K. Mulford Co., 189 F. 95, 98 (S.D.N.Y. 1911).
[123] Ass’n for Molecular Pathology v. USPTO, 702 F.Supp. 2d 181, 185 (S.D.N.Y. 2010).
[124] Lab. Corp. v. Metabolite Labs, 548 U.S. 124, 125 (2006) (Breyer, J., dissenting from dismissal of certiorari as improvidently granted).
[125] Id.
[126] Id. at 132.
[127] Gene Quinn, Bilski Tea Leaves: Remembering the Lab Corp Non-Decision. IP WATCHDOG (Jan. 18, 2010), available at http://ipwatchdog.com/2010/01/18/bilski-tea-leaves-remembering-the-lab-corp-non-decision/id=8479/.
[128] In re Bilski, 545 F.3d 943, 961 (Fed. Cir. 2008) (en banc), aff’d in part sub nomBilski v. Kappos, 130 S.Ct. 3218 (2010).
[129] Bilski, 545 F.3d at 949–950.
[130] Id. at 961, 963 (discussing the constitutional basis for the machine-or-transformation test and applying the test to the facts of the Bilski patent).
[131] Id.
[132] Prometheus Labs., Inc. v. Mayo Collaborative Servs., 581 F.3d 1336, 1350 (Fed. Cir. 2009), vacated, 130 S.Ct. 3543 (2010).
[133] Prometheus Labs., Inc. v. Mayo Collaborative Servs., 628 F.3d 1347, 1349–50 (Fed. Cir. 2010), cert. granted, 131 S.Ct. 3027 (June 20, 2011) (No. 10-1150).
[134] Id. at 1350.
[135] Id.
[136] Prometheus Labs., Inc. v. Mayo Collaborative Servs., 581 F.3d at 1350, cert. granted, 131 S.Ct. 3027 (June 20, 2011) (No. 10-1150).
[137] Id. at 1341.
[138] Bilski v. Kappos, 130 S.Ct. 3218, 3221 (2010).
[139] Id. at 3227.
[140] William Simmons, Bilski Blundering Biotech, 27 NATURE BIOTECH. 245, 247 (2009).
[141] Prometheus Labs., Inc. v. Mayo Collaborative Servs., 628 F.3d 1347, 1349 (Fed. Cir. 2010).
[142] Id. at 1353 (quoting Bilski v. Kappos, 130 S.Ct. at 3225 (quoting Diamond v. Chakrabarty, 447 U.S. 303, 308 (1980) (quoting 5 Writings of Thomas Jefferson 75–76 (H. Washington ed. 1871)))).
[143] Prometheus Labs., 628 F.3d at 1354.
[144] Id. A parallel case not discussed here is Classen Immunotherapies, Inc. v. Biogen IDEC, Nos. 2006-1634, 2006-1649, 2011 WL 3835409, (Fed. Cir. Aug. 31, 2011) (holding that claims of patents directed towards a method of lowering the risk of chronic immune-mediated disorder, including the physical step of immunization in accordance with a lower risk schedule was patent-eligible, while the abstract principle of varying immunization schedules having consequences on certain diseases was not). Nonetheless, I believe that it fits within my general framework of courts creating subject-specific patent eligibility decisions that make it difficult for genetic diagnostic test manufacturers to predict patent eligibility ex ante.
[145] Ass’n for Molecular Pathology v. USPTO, 702 F.Supp. 2d 181 (S.D.N.Y. 2010), rev’d in part, aff’d in part, 653 F.3d 1329 (Fed. Cir. 2011).
[146] Id. at 185.
[147] BRCA1 and BRCA2 account for the majority of inherited breast cancers in the United States. H.T. Lynch et al., Hereditary breast cancer: Part I. Diagnosing Hereditary Breast Cancer Syndromes. 14BREAST J. 3 (2008).
[148] Gutierrez Telephone Interview, supra note 93.
[149] Ass’n for Molecular Pathology v. USPTO, 702 F.Supp. 2d at 207.
[150] Id. at 238.
[151] Id. at 222.
[152] Brief for the United States as Amicus Curiae in Support of Neither Party, Ass’n for Molecular Pathology v. USPTO, No. 2010-1406, slip op. (Fed. Cir. Oct. 29, 2010).
[153] Introns are portions of the gene that are not transcribed to form mRNA, which eventually creates proteins—effectively excess DNA, although scientists predict that these portions have some unknown utility. See GRIFFITHS ET AL., supra note 83, at 6–7, 465. They are also subject to variance and modification by related genes.
[154] Brief for the United States, supra note 142, at 14–17.
[155] Id. at 23–24.
[156] Id. at 31.
[157] Ass’n for Molecular Pathology v. USPTO, 653 F.3d 1329, 1334 (Fed. Cir. 2011). The plaintiff’s attorneys have stated that they will petition for certiorari, although earlier petitions for rehearing at the Federal Circuit were denied. Turna Ray, ACLU, PUBPAT Seek to Challenge Patentability of Isolated Genes Before Supreme Court, PHARMACOGENOMICS REPORTER, October 12, 2011.
[158] U.S. CONST. art. I, § 8, cl. 8.
[159] Bilski v. Kappos, 130 S.Ct. 3218, 3227 (2010).
[160] U.S. CONST. art. I, § 8.
[161] Most notably, see the seminal Science magazine article, which has been cited 1,449 times as of this writing. Michael A. Heller & Rebecca S. Eisenberg, Can Patents Deter Innovation: The Anticommons in Biomedical Research, 280 SCIENCE 698 (1998).
[162] Mildred K. Cho et al., Effects of Patents and Licenses on the Provision of Clinical Genetic Testing Services, 5 J. MOLECULAR DIAGNOSTICS 3, 3 (2003).
[163] Kyle Jensen & Fiona Murray, Intellectual Property Landscape of the Human Genome, 310 SCIENCE 239, 239 (2005).
[164] SACGHS REPORT, supra note 35, at 20. An important aside to this story is that university research is not protected by the very narrow research exception to patent infringement. Madey v. Duke University, 307 F.3d 1351, 1362 (Fed. Cir. 2002) (allowing the research exemption only when the infringement was done for “amusement, to satisfy idle curiosity, or for strictly philosophical inquiry” and finding that the university’s business interest in research precluded such protection).
[165] Indeed, the BIO survey cited by Myriad states that the majority of companies generally in-licensed (licensed from universities) projects that are in the pre-clinical stage of development and thus require substantial investment and commercialization risk. 77% of the respondents indicated that they intended to spend 5–15 years and over $100 million in developing a commercial product, which may dwarf the spending by the federal government through research grants. Ass’n for Molecular Pathology v. USPTO, 702 F.Supp. 2d 181, 210-11 (S.D.N.Y. 2010).
[166] SACGHS REPORT, supra note 35, at 22-23.
[167] Isabelle Huys et al., Legal Uncertainty in the Area of Genetic Diagnostic Testing, 27 NATURE BIOTECH. 903, 903–05 (2009).
[168] Id. at 906.
[169] Id. at 907.
[170] Heller & Eisenberg, supra note 161.
[171] Huys et al., supra note 167, at 905.
[172] Id. at 908.
[173] Id. at 909.
[174] SACGHS Report, supra note 35, at 7.
[175] Cook-Deegan et al., supra note 93, at S15.
[176] Id. at S27–S28.
[177] Id. at S28.
[178] Id.
[179] Id.
[180] Id.
[181] Id. at S29.
[182] Katie Skeehan et al., Impact of gene patents and licensing practices on access to genetic testing for Alzheimer disease, 12 GENETICS IN MEDICINE S71, S71 (2010).
[183] Id. This risk factor also has a reasonably small effect on Alzheimer’s onset.
[184] Id.
[185] Id. at S74.
[186] Id.
[187] Skeehan et al., supra note 182, at S77. Graceful Earth received one of fourteen letters from the FDA concerning regulation of its tests on July 19, 2010. Letter from Dr. Alberto Gutierrez, Office of In VitroDiagnostic Device Evaluation and Safety, Center for Device and Radiological Health to Mr. Kenneth W. Friedenberg, Owner, Graceful Earth, Inc. (July 19, 2010), available at http://www.fda.gov/downloads/MedicalDevices/ProductsandMedicalProcedu res/InVitroDiagnostics/UCM219585.pdf.
[188] William E. Ridgway, Note, Realizing Two-Tiered Innovation through Drug Regulation, 58 STAN. L. REV. 1221, 1224 (2005–06). Note however, that Ridgway’s system is actually the opposite of the one suggested here—in this model, the FDA exclusivity will provide the baseline tier of protection, with the patent system providing an enhanced level, rather than the reverse.
[189] SACGHS REPORT, supra note 35, at 20. There is some question as to whether this lower level of exclusivity also has potential ill-effects on innovation, but this needs to be further explored. See Heidi Williams,Intellectual Property Rights and Innovation: Evidence from the Human Genome (Nat’l Bureau of Econ. Research Working Paper No. 16213, 2010).
[190] Id. at 43.
[191] Turna Ray, Draft Bill Proposes FDA Create New Division to Review “Advanced Personalized Diagnostics”, PHARMACOGENOMICS REPORTER (June 23, 2010), available at http://www.genomeweb.com/dxpgx/draft-bill-proposes-fda-create-new-division-review-advanced-personalized-diagnos?page=show (discussing a draft bill by Representative Orrin Hatch (R-UT) entitled “Better Evaluation and Treatment Through Essential Regulatory Reform for Patient Care Act,” which would create a new division to review diagnostics. No date for the introduction of the bill has been set yet).
[192] Meeting Day 1supra note 21, at 71 (statement of Elizabeth Mansfield, Director of Personalized Medicine, Office of In Vitro Diagnostic Device Evaluation and Safety).
[193] Bloss et al., supra note 118, at 7.
[194] Id. at 6–7, 10.
[195] Gutierrez Telephone Interview, supra note 93.
[196] Bloss et al., supra note 118, at 10.
[197] Id. at 8–9.
[198] For example, the data obtained by the International HapMap project, which seeks to develop a haplotype map (HapMap) of the human genome that will describe the common patterns of human genetic variation, necessarily depends on the study participants. See Groups Participating in the International HapMap Project, http://hapmap.ncbi.nlm.nih.gov/groups.html (last visited Oct. 18, 2011).
[199] 21 C.F.R. § 314.50 (d)(5)(v) (2009).
[200] See Maher, supra note 118 (discussing the limitations of genetic ancestry testing and recommendations on the delivery of test results).
[201] See Richard A. Epstein, How Conflict-of-Interest Rules Endanger Medical Progress and Cures, Manhattan Institute, Project FDA Report No. 3 (2010) (alleging that conflict-of-interest provisions from the FDA hurt medical progress by preventing the agency from having the best scientists. Because both the FDA and the pharmaceutical companies look for the best talent, stringent conflict-of-interest protocols may actually weed out the best actors).
[202] Gutierrez Telephone Interview, supra note 93.
[203] See, e.g., Mark Schena et al., Quantitative Monitoring of Gene Expression Patterns with a Complementary DNA Microarray, 270 SCIENCE 467, 467 (1995) (detailing the use of cDNA in microarrays to measure the quantitative expression of numerous genes at one time).
[204] Carmichael, supra note 67; Gutierrez Telephone Interview, supra note 93.
[205] Eisenberg, supra note 76, at 480.
[206] See generally Cook-Deegan et al., supra note 93.
[207] Gutierrez Telephone Interview, supra note 93.
[208] Alexander T. Tabarrok, The Blessed Monopolies, 24 REGULATION 10, 10-12 (2001), available at http://www.cato.org/pubs/regulation/regv24n4/v24n4-1.pdf.
[209] Id.
[210] MAKOWER ET AL., supra note 47, at 11.
[211] Id.
[212] See Abigail Alliance v. Von Eschenbach, 445 F.3d 470 (D.C. Cir. 2006) (finding that a terminally-ill patient has a constitutional right to access potentially life-saving, post-Phase I drugs).
[213] FDA WHITE PAPER, PRESCRIPTION DRUG USER FEE ACT (PDUFA): ADDING RESOURCES AND IMPROVING PERFORMANCE IN FDA REVIEW OF NEW DRUG APPLICATIONS 4 (2005).
[214] MedWatch Voluntary Reporting Form 3500https://www.accessdata.fda.gov/scripts/med watch/medwatch-online.htm (last visited Oct. 18, 2011).
[215] This itself is a marked improvement over the 1% reporting statistic found by the GAO in a 1986 study. NOAH, supra note 42, at 281.
[216] Angrist et al., supra note 23, at S111.
[217] Meeting Day 1supra note 21, at 29 (statement of Courtney Harper, Division Director for FDA Division of Chemistry and Toxicology Devices).
[218] Pollack I, supra note 94. Although, newspapers are often eager to report that this technology is ready before its time. See, e.g., Andrew Pollack, When Gene Sequencing Becomes a Fact of Life, N.Y. TIMES, Jan. 17, 2001, at C1, available at http://www.nytimes.com/2001/01/ 17/business/17AIDS.html (discussing the advance in H.I.V. resistance tests through the use of genome sequencing).
[219] Promethease is a tool that builds a report using SNPedia (which uses data taken from published studies) and a file of genotypes of the kind calculated by 23andMe and Navigenics. Promethease, http://www.snpedia.com/index.php/Promethease (last visited Oct. 18, 2011).
[220] Alla Katnelson, The renegade gene test, NATURE NEWS (Oct. 14, 2010), http://www.nature.com/news/2010/101014/full/news.2010.540.html.
[221] 21 U.S.C. § 360(k).
[222] Steven L. Salzberg & Mihaela Pertea, Do-it-yourself genetic testing, 11 GENOME BIOLOGY 1, 2 (2010).

[223] 35 U.S.C. § 271(b)–(c) (2006) (outlining the requirements for actionable induced infringement, including the requirement that the underlying direct infringement be successful).

“Term”inal Illness: Curing the Patent Term Using Empirical Analysis of Patent Globalization

By Wesley D. Markham* A pdf version of this article may be downloaded here. I.  Introduction This paper takes an empirical approach to a policy-based question: how long should patents last in the United States, especially given changes in the international patent law regime? The overarching, even constitutionalized, policy behind the U.S.’s patent system is to promote the progress of science and the useful arts.[FN1] This is a laudable goal, but the devil is in the details.  Utilizing an intellectual property regime to maximize innovation requires a delicate balancing act.  On the one hand, inventors and the firms for which they work need an incentive to innovate.  In the United States, one such incentive is a limited monopoly in the form of patents for new and useful inventions.  On the other hand, every patent takes something away from the public domain, thereby making it more difficult for others to build on prior discoveries.  In other words, patents both encourage and discourage innovation.  The key to a successful patent system is determining the correct balancing point, i.e., the smallest reward necessary to spur inventors to invent. The term of patent protection embodies a primary variable in this balancing act.  A relatively long patent term greatly rewards an inventor, but it also imposes a high burden on others who wish to use or improve upon patented technology.  Under both U.S. and international law, patent terms generally run for approximately twenty years.[FN2] But while U.S. patent terms have remained fairly – though not entirely – stable for the last fifteen years, international patents have become increasingly lucrative.  To remain relevant, patent law in the U.S. must adapt to this patent globalization, in part by recalibrating the duration of the patent monopoly.  I tackle this issue by asking whether we should reduce the patent term in the U.S. to compensate for the enhanced potential for patent exclusivity outside the U.S.  Specifically, I developed a new metric, termed “global patent term” (GPT), to assess whether the average “reward per invention” has significantly increased over the last fifteen years given an expanding global patent regime.  If it has become relatively easy for firms to secure a high level of protection for each invention, e.g., by receiving multiple patents on the same invention in many different countries, the “reward per invention” is high.  In a well-functioning patent system, a high “reward per invention” should result in increased innovation.  If the “reward per invention” has increased but the level of innovation has remained stagnant or decreased, then we might conclude that the current level of patent protection is too high and should respond by reducing the patent term in the United States.  A shorter patent term would allow patented inventions to fall into the public domain more quickly, enabling others to take full advantage of the patented technology. Furthermore, I conducted this analysis on three firms in three very different industries – Pfizer, the pharmaceutical giant; International Paper Company, a worldwide leader in paper products; and UNISYS, a large technology services provider.  These three case studies indicate that patent globalization discriminates based on technology, i.e., rewards firms in some industries but not others.  Given my results, I conclude that we should seriously consider implementing technology-specific patent terms in the U.S.  For industries in which patent globalization has made a real difference, such as the pharmaceutical industry, the increased “reward per invention” outside the U.S. should offset any innovation incentives lost by shortening the duration of U.S. patents.  For industries that depend on patent exclusivity in the U.S. but not elsewhere, a longer U.S. patent term may provide the optimum innovation incentives. II.  A Short History of the Patent Term in the United States & Abroad A. The U.S. Story The following history of the patent term in the U.S. serves to illuminate two important points.  First, the chosen length of patent protection in this country is, and has always been, rather arbitrary, a product of compromise and historical accident instead of sound economic analysis.  Second, as a practical matter, any proposal that might be interpreted as weakening the patent system in the U.S., e.g., an attempt to shorten the patent term, is likely to meet serious political resistance.[FN3] Like so much of American law, the original duration of patent protection in the United States was borrowed from England.[FN4] The Patent Act of 1790 provided that the exclusive rights associated with a patent shall last “for any term not exceeding fourteen years.”[FN5] In 1836, a revised Patent Act made available a seven year patent term extension, above and beyond the original fourteen year term, in certain circumstances.[FN6] This lasted until 1861, at which time Congress declared that “[a]ll patents hereafter granted shall remain in force for the term of seventeen years from the date of issue; and all extension of such patents is hereby prohibited.”[FN7] One commentator proposes that the change from a fourteen year patent term to a seventeen year patent term was made, in part, because extensions were so common under the old “fourteen plus seven” system.[FN8] Yet another opines that Congress selected the seventeen year term because seventeen is “the number midway between 14 and 21.”[FN9] Between 1790 and 1875, Congress also passed approximately seventy-five private bills extending the patent terms of particular inventors who successfully lobbied Congress that the generally applicable statutory term had not afforded time to generate sufficient income from their inventions for one reason or another.[FN10] Although many of these private bills extended the term of patents that had already lapsed into the public domain, a practice held categorically unconstitutional in Graham v. John Deere Co. of Kansas City,[FN11] Congress’s historical practice of enacting discriminatory, inventor-specific patent terms adds a degree of constitutional legitimacy to the more principled, industry-specific terms I ultimately propose. The patent term of “seventeen years from issue date” remained in effect until 1994, when Congress brought U.S. patent law into compliance with the TRIPS Agreement.[FN12] Implementation of TRIPS modified the term of patent protection in the U.S. from “seventeen years from issue date” to “a term beginning on the date on which the patent issues and ending 20 years from the date on which the application for the patent was filed in the United States…”[FN13] The change in patent term from “seventeen years from issue date” to “twenty years from filing date” sparked controversy.  Opponents of the new “twenty years from filing date” patent term stressed that pioneering technology typically takes a long time to evaluate and approve, which would result in shorter terms under the new system.[FN14] Indeed, the biotechnology sector would be particularly hard hit “because the most commercially attractive patents can take over fourteen years to issue.”[FN15] Weaker patents, the opponents continued, would in turn reduce the royalties that are the lifeblood of private research and development (R&D) funds.[FN16] To address these concerns, Congress passed the American Inventors Protection Act of 1999 (AIPA). [FN17] Specifically, the AIPA adds a new provision to compensate applicants fully for PTO-caused administrative delays, and, for good measure, includes a new provision guaranteeing diligent applicants at least a 17-year term by extending the term of any patent not granted within three years of filing.  Thus, no patent applicant diligently seeking to obtain a patent will receive a term of less than the 17 years as provided under the pre-GATT3 standard; in fact, most will receive considerably more.  Only those who purposely manipulate the system to delay the issuance of their patents will be penalized…[FN18] Today, the duration of patent protection in the United States remains “twenty years from filing date,” subject to the aforementioned patent term adjustment provisions.[FN19] B. Patent Globalization Under TRIPS and the PCT Patenting has gone global.  The Patent Cooperation Treaty[FN20] (PCT) makes it possible for inventors and firms to seek patent protection simultaneously in many different countries by filing “international” patent applications.[FN21] As of 1980, only 30 countries adhered to the PCT.[FN22] By 1995, the number of PCT contracting states had grown to 82.[FN23] Today, 142 countries have signed on to the PCT.[FN24] According to the World Intellectual Property Organization (WIPO), the body that administers the PCT, the PCT “postpones the major costs associated with international patent protection” and “brings the world within reach” (emphasis added).[FN25] In effect, the proliferation of PCT contracting states makes it relatively easy for a patent applicant to pursue de facto worldwide patent protection on her invention. Not only has it become easier to patent around the globe, but the minimum level of patent protection afforded by most countries has never been higher.  The 1994 Agreement on Trade- Related Aspects of Intellectual Property Rights, or TRIPS, sets a floor below which intellectual property standards shall not fall.[FN26] For example, TRIPS requires that “patents shall be available for any inventions, whether products or processes, in all fields of technology,” with only limited exceptions.[FN27] Additionally, TRIPS mandates a minimum patent term of twenty years, measured from the patent application filing date.[FN28] III:  Methodology: GPT as a Proxy for “Reward Per Invention” R&D Expenditures as a Proxy for “Amount of Innovation” In this article, I investigate whether the average “reward per invention” has significantly increased over the last fifteen years, due in large part to the 1994 TRIPS Agreement and the proliferation of PCT contracting states.  More specifically, I evaluate whether patent applicants are actually seeking and obtaining more widespread patent protection today than they were at the time of the TRIPS Agreement fifteen years ago.  As stated above, if it has become relatively easy for firms to secure a high level of protection for each invention, e.g., by receiving multiple patents on the same invention in many different countries, the “reward per invention” is high.  In a well-functioning patent system, a high “reward per invention” should result in increased innovation.  If the “reward per invention” has increased but the level of innovation has not, then the prevailing level of patent protection may be too high, and we should respond by reducing the patent term in the United States.  A shorter patent term would bring patented inventions into the public domain more quickly, enabling others to take early advantage of the patented technology. To measure the “reward per invention,” I developed a metric called the “global patent term,” or GPT.  The GPT is the sum of the terms of protection of all the patents, worldwide, on a single invention.  Of course, not all patents are created equal.  For example, a twenty year monopoly in the United States is likely more valuable than a twenty year monopoly in Belgium.  To account for this difference in value, each patent term in the GPT calculation is scaled according to the gross domestic product (GDP) of the corresponding country, using the United States GDP as a baseline. A short hypothetical will illustrate the calculation.  Suppose that in 1995, Company XYZ typically applied for patent protection for its inventions only in the United States.  In this case, the GPT equals the U.S. patent term, or twenty years.  Now, in 2009, suppose that Company XYZ typically applies for patent protection in the United States, China, and Canada.  The GPT equals [(U.S. patent term)(U.S. GDP / U.S. GDP) + (China patent term)(China GDP / U.S. GDP) + (Canada patent term)(Canada GDP / U.S. GDP)], or [(20 years)($14,204 billion / $14,204 billion) + (20 years)($4,326 billion / $14,204 billion) + (20 years)($1,400 billion / $14,204 billion)], or [20 years + 6.1 years + 2.0 years], or approximately twenty-eight years. In this hypothetical, the “global patent term” for Company XYZ’s invention has risen from twenty years in 1995 to twenty-eight years in 2009, an increase of 40%.  If the patent system is working properly, we would expect to see a higher level of innovation from Company XYZ as a result of this 40% increase in “reward per invention.”  To test whether this is actually the case, we can examine the R&D spending of Company XYZ during the relevant timeframe. The TRIPS Agreement and the steady growth in number of PCT contracting states have arguably facilitated patenting around the globe.  If the reward for inventing, as measured by the “global patent term” metric, is greater than it has ever been before, then the level of innovation, as measured by R&D spending, should be similarly high.  However, if innovation lags in the face of ever-increasing patent protection, the extra rewards are not fostering innovation but rather unjustly enriching some patent holders.  If this is the case, we should consider shortening the patent term in the United States. IV:  Three Case Studies: Pfizer, International Paper, and UNISYS To shed light on these difficult issues, I analyzed the “global patent terms” and R&D spending of three different firms in three very different industries: Pfizer, a pharmaceutical giant; International Paper Company, a global paper product producer; and UNISYS, a worldwide information technology provider.  In particular, I compared each firm’s average GPT and R&D spending in 1995, just after TRIPS was enacted, and 2009, after the proliferation of PCT contracting states and the entrenchment of the TRIPS regime.  The results vary widely among the three firms. Of the three firms, Pfizer has most clearly taken advantage of the increased availability of patents around the globe.  The average (mean/median) GPT of Pfizer’s 1995 U.S. patents was 95.7/85.4 years.  The average GPT rose significantly to 131.4/108.7 years in 2009.  Even excluding U.S., EP, and WO filings, which heavily influence the overall GPT calculations due to their high GDPs, Pfizer’s average GPT increased from 23.7/22.6 years in 1995 to 29.9/25.7 years in 2009.  Not surprisingly, the average number of countries in which Pfizer filed for patent protection also increased, from 13 in 1995 to 16 in 2009. Additionally, between 1995 and 2009, Pfizer drastically altered its approach regarding where to file for patent protection, turning away from filing in many Western European countries such as Austria, Germany, and Denmark, and towards filing in Central and South American countries such as Mexico, Argentina, and Brazil.  For example, 66% of Pfizer’s 1995 U.S. patents had counterparts in Germany.  That number fell to just 13% in 2009.  On the other hand, only 14% of Pfizer’s 1995 U.S. patents had counterparts in Brazil, but by 2009, 72% of Pfizer’s U.S. patents had related Brazilian applications. The aforementioned data strongly suggests that Pfizer reaps the rewards of the strong global patent system buttressed by the Patent Cooperation Treaty and the TRIPS Agreement.  In turn, we should see enhanced innovation in the form of increased R&D spending by Pfizer if the patent system is functioning properly.  In one respect, we do.  Pfizer greatly increased R&D spending from approximately $1.4 billion in 1995 to approximately $7.8 billion in 2009.  However, measured as a percentage of total revenues, the increase in Pfizer’s R&D spending from 1995 to 2009 was minimal – 14.4% to 15.7%. If Pfizer were the only company playing the patent game, then it might make sense to reduce the term of patent protection in the United States.  After all, the data shows that Pfizer gets significantly more worldwide patent protection for each invention today than it did before the TRIPS Agreement took effect.  In return, Pfizer has only minimally increased its R&D spending as a fraction of its total revenues.  This looks a lot more like unjust enrichment than enhanced innovation.  And for a company like Pfizer, which relies heavily on patenting both inside and outside the United States, the increased availability and strength of patents around the globe will likely offset any harm to innovation that may come from reducing the patent term in the U.S.  But Pfizer is not the only company in town.  Other firms, such as International Paper Company and UNISYS, that rely more heavily on the U.S. patent system and not on patenting in foreign countries, do not benefit nearly as much as Pfizer does from the worldwide patent regime ushered in by the Patent Cooperation Treaty and the TRIPS Agreement. While International Paper’s average GPT (mean/median) rose from 42.9/20.0 years in 1995 to 66.1/40.0 years in 2009, most of that increase can be attributed to multiple related U.S. patents on the same technology, not higher global patenting activity on the part of the company.  Excluding U.S., EP, and WO filings, International Paper’s average GPT remained relatively flat, rising slightly from 4.0/0.0 years in 1995 to 5.6/0.0 years in 2009.  The average number of countries in which International Paper filed for patent protection was essentially flat as well – 3.2/1.0 in 1995 and 3.4/1.0 in 2009. In stark contrast to Pfizer, International Paper does not appear to have benefited from heightened levels of global patent protection.  The company simply does not pursue patents in enough countries outside the United States for the TRIPS Agreement and the proliferation of PCT contracting states to make an appreciable difference in terms of innovation incentives.  Therefore, reducing the patent term in the United States may have a deleterious effect on innovation at a company such as International Paper, which has already seen R&D expenditures plummet from $111 million (0.56% of net sales) in 1995 to $13 million (0.06% of net sales) in 2009. UNISYS finds itself in a similar position to International Paper when it comes to patenting on a global scale – it didn’t do it in 1995, and it doesn’t do it now.  UNISYS’ average GPT (mean/median) actually fell from 54.1/20.0 years in 1995 to 42.0/20.0 years in 2009.  Excluding U.S., EP, and WO patent application filings, UNISYS’ average GPT was flat and basically non-existent – 1.8/0.0 years in 1995 and 2.8/0.0 years in 2009.  In general, UNISYS only seeks patents for its inventions in the United States, so the enhanced global patent protection made possible by TRIPS and the PCT has little effect on the innovation decisions and R&D expenditure incentives at UNISYS.  This becomes painfully obvious when one considers that R&D expenditures at UNISYS have fallen from $405 million (6.4% of revenue) in 1995 to $102 million (2.2% of revenue) in 2009. V:  A Modest Proposal To summarize, not all companies benefit from the availability of strong global patent protection facilitated by the Patent Cooperation Treaty and mandated by TRIPS.  In fact, not even all companies that pursue patents benefit from the aforementioned regime.  Easy access to strong patents around the globe favors firms like Pfizer that seek protection for their inventions in many different countries.  For companies like International Paper and UNISYS that file for patents primarily in the United States, the theoretical presence of strong patent regimes in other countries has little if any practical effect on the companies’ incentives to innovate. Therefore, even if it were possible under TRIPS to decrease the U.S. patent term to less than twenty years, an across the board patent term reduction would not be advisable.  Such a move would likely have little influence on innovation at firms such as Pfizer because the enhanced innovation incentives provided by increased patent protection in the rest of the world would counterbalance the effect of a shorter U.S. patent term.  However, for firms that believe it to be in their best interest to seek patents only or primarily in the United States, reducing the U.S. patent term might stifle their drive to innovate. To solve this problem, we should consider setting different patent terms for different technology sectors.  A relatively short U.S. patent term might provide the ideal level of innovation for industries such as pharmaceuticals in which the typical inventions (drugs) have worldwide profit-making power.  In other industries, such as information technology and software, patenting in many less-developed countries just does not make economic sense because the market is so limited.  A longer term of patent protection in the U.S. might be necessary to achieve the maximum amount of innovation in technology sectors such as these.  By adopting technology-specific patent terms, it may be possible to increase innovation and to bring patented subject matter into the public domain more quickly.  
* I would like to thank Rochelle Dreyfuss, Barton Beebe, Oren Bar-Gill, and all those who participated in the Lederman/Milbank Fellowship workshop and the Colloquium on Innovation Policy for their insightful comments and critiques.  All errors are, of course, my own. [FN1] U.S. Const. art. I, § 8, cl. 8. [FN2] See 35 U.S.C. § 154 (2006) (subject to certain exceptions, patents run for 20 years from application filing date); Agreement on Trade-Related Aspects of Intellectual Property Rights, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1C, Legal Instruments – Results of the Uruguay Round, 33 I.L.M. 1197 (1994) [hereinafter TRIPS Agreement or TRIPS], art. 33 (prescribing a minimum term of patent protection of 20 years from application filing date). [FN3] The intense debate over recently-proposed patent reform legislation exemplifies this point.  See, e.g., Amy Schatz, Patent Overhaul Gets Close, Draws Opposition, Wall St. J., Mar. 28, 2011, at B3 (noting that Corporate America is divided over Congress’s current patent law reform efforts, despite “calling for an overhaul of U.S. patent laws for years”). [FN4] See David S. Abrams, Did TRIPS Spur Innovation An Analysis of Patent Duration and Incentives to Innovate, 157 U. Pa. L. Rev. 1613, 1618 (2009). [FN5] Patent Act of 1790, ch. 7, § 1, 1 Stat. 109, 110. [FN6] See Patent Act of 1836, ch. 357, § 17, 5 Stat. 117, 124-25. [FN7] See Act of Mar. 2, 1861, ch. 88, § 16, 12 Stat. 246, 249. [FN8] See Dana Rohrabacher and Paul Crilly, Congressional Commentary: The Case for a Strong Patent System, 8 Harv. J. Law & Tech. 263, 264 (1995). [FN9] Eric E. Johnson, Calibrating Patent Lifetimes, 22 Santa Clara Computer & High Tech. L.J. 269, 283 (2006). [FN10] See, e.g., An Act for the Relief of Oliver Evans, 6 Stat. 70 (1808).  See also Eldred v. Ashcroft, 537 U.S. 186, 234-35 (2003) (Stevens, J., dissenting) (providing historical analysis of early U.S. patent and copyright law). [FN11] 383 U.S. 1, 6 (1966) (“Congress may not authorize the issuance of patents whose effects are to remove existent knowledge from the public domain”). [FN12]See S. Rep. No. 103-412 (1994) (discussing Section 532 of the Uruguay Round Agreements Act). [FN13] See Uruguay Round Agreements Act § 532, Pub. L. No. 103-465, 108 Stat. 4809 (1994). [FN14] See H.R. Rep. No. 104-887 (1997) (discussing Hearing Volume No. 104-58, particularly the testimony of Ms. Gardner). [FN15] Patricia Montalvo, How Will the New Twenty-Year Patent Term Affect You?  A Look at the TRIPS Agreement and the Adoption of a Twenty-Year Patent Term, 12 Santa Clara Computer & High Tech. L.J. 139, 160 (1996). [FN16] See Dana Rohrabacher and Paul Crilly, Congressional Commentary: The Case for a Strong Patent System, 8 Harv. J. Law & Tech. 263, 264 (1995). [FN17] See generally American Inventors Protection Act of 1999, Pub. L. No. 106-113, 113 Stat. 1501. [FN18] Id. at 49-50. [FN19] See 35 U.S.C. § 154 (2006). [FN20] Patent Cooperation Treaty, June 19, 1970, 28 U.S.T. 7645, 1160 U.N.T.S. 231. [FN21] About the Patent Cooperation Treaty, http://www.wipo.int/pct/en/treaty/about.htm (last visited May 5, 2010). [FN22] See PCT Contracting States (2010), http://www.wipo.int/pct/guide/en/gdvol1/annexes/annexa/ax_a.pdf. [FN23] See id. [FN24] See id. [FN25] World Intellectual Prop. Org., Protecting Your Inventions Abroad: Frequently Asked Questions About the Patent Cooperation Treaty (PCT) 16 (2006), available at http://www.wipo.int/export/sites/www/pct/en/basic_facts/faqs_about_the_pct.pdf. [FN26] See generally TRIPS Agreement, available at http://www.wto.org/english/docs_e/legal_e/27-trips.pdf. [FN27] TRIPS art. 27. [FN28] TRIPS art. 33.

Patent Eligibility of Molecules: “Product of Nature” Doctrine After Myriad

By Can Cui* A pdf version of this article may be downloaded here. I.  Introduction The 1952 Patent Act sets forth four categories of subject matter that are patent-eligible: process, machine, manufacture, and composition of matter.[FN1] The Supreme Court has interpreted the subject-matter eligibility to be very broad[FN2]; instead of delineating what is patent-eligible, the Supreme Court has crafted a number of categories of “patent-ineligible” subject matter.[FN3] These judicially crafted categories, however, are not well defined.  The boundaries of these categories are also constantly challenged by the advancement of science and technology, particularly biotechnology.  While the Patent Act has remained largely unchanged since its codification in 1952, the discovery of the structure of deoxyribonucleic acid (“DNA”) in 1953 has revolutionized our lives in a great number of ways.  The complete mapping of the human genome has also changed our way of thinking about life and dealing with disease.  Discoverers of new genes, especially those associated with disease susceptibilities, rushed to the patent office to protect their discoveries.  Whether DNA molecules should be patented had been the topic of active policy debate for more than twenty years until the early 2000’s, when the question about their subject-matter eligibility was first seriously raised in the academy.[FN4] On March 29, 2010, in a 156-page-long decision, the U.S. District Court in the Southern District of New York held that claims directed to isolated DNA molecules were invalid under 35 U.S.C. § 101 for failing to claim patent-eligible subject matter, because the claimed isolated DNA molecules were not “markedly different” from native DNA molecules as they exist in nature, according to Senior Judge Robert W. Sweet.[FN5] This Note discusses the patent-eligibility of the subject matter set forth in these DNA-molecule claims. In determining that these claims-in-suit were invalid under 35 U.S.C. § 101, Judge Sweet first analyzed the scope of § 101, focusing heavily on the so-called “product of nature” doctrine.[FN6] He then reviewed Supreme Court cases and some lower court opinions that are purportedly related to the “product of nature” doctrine, and provided a test for determining whether a product purified from nature is patent-eligible: “the purified product must possess ‘markedly different characteristics’ in order to satisfy the requirements of § 101.”[FN7] Finally, Judge Sweet applied this test to the subject matter in the claims-in-suit.  Focusing almost exclusively on “DNA’s unique qualities as a physical embodiment of information,” Judge Sweet concluded that “none of the structural and functional differences cited by [defendant] Myriad between native BRCA1/2 DNA and the isolated BRCA1/2 DNA claimed in the patents-in-suit render[ed] the claimed DNA ‘markedly different’” from its naturally-occurring counterpart, thus failing the test.[FN8] Needless to say, the outcome of this case, if upheld upon appeal, would have a profound impact on the biotechnology industry.  This Note, however, does not try to delve into any policy arguments, but will focus on the more fundamental question about the “product of nature” doctrine, how it has been applied to different types of claimed subject matter prior to Judge Sweet’s opinion, and what an appropriate test should be for applying the doctrine to claims directed to molecules.  This Note will propose such a test and apply the proposed test to the DNA-molecule claims-in-suit, leading to a legal conclusion contrary to that of Judge Sweet.  This Note will also make an effort to address certain issues in Judge Sweet’s opinion, and to provide some practical tips to patent practitioners in drafting DNA molecule claims. II.  Analysis A.  The “Product of Nature” Doctrine: Does It Really Have a Legal History? Is there a “product of nature” doctrine?  At the very least, it is fair to say that the doctrine has not been articulated very well by courts.  Judge Sweet cited four cases in which “product of nature” or “products of nature” was recited, none of which has a claim-in-suit directed to a molecule.[FN9] Because the subject matter of the claims-in-suit in Myriad is DNA molecules, this Note will include judicial treatment of molecule claims in the history review of relevant case law in the context of the “product of nature” doctrine.  Instead of a chronological order, this Note will apply a brand new analytical framework to the history review, categorizing the relevant cases based on whether the claimed subject matter is an element, a molecule, or a microorganism.[FN10] 1. Naturally-occurring chemical elements are generally patent-ineligible subject matter. One of the earliest cases ever mentioning a “product of nature” is Gen. Elec. Co., where a claim-in-suit is directed to “[s]ubstantially pure tungsten having ductility and high tensile strength.”[FN11] Even though pure tungsten as claimed had not been found in nature, the Third Circuit’s view was that since tungsten is an element, all of its properties are natural by definition.[FN12] Judge Sweet cited this case to prove the “well established” “product of nature” doctrine.[FN13] One sentence from this case is, however, particularly relevant and worthy of analysis: “a patent cannot be awarded for a discovery or for a product of nature, or for a chemical element.”[FN14] First, the second “or” suggests that the patent-ineligibility of a product of nature is dicta, because the opinion is about the patent-ineligibility of a chemical element, tungsten.  Second, the patent-ineligibility of a discovery expressed in this sentence makes the entire sentence susceptible to constitutional attack.[FN15] The patent-ineligibility of naturally-occurring chemical elements was followed by the Court of Customs and Patent Appeals, which seemed to have more readily relied on the “product of nature” doctrine in its decisions.[F16] 2. Claims to molecules, including those to DNA molecules, had never been invalidated by courts under the “product of nature” doctrine before Judge Sweet’s opinion. Somewhat surprisingly, case law on the patent-eligibility or –ineligibility of molecules does not prove to be abundant.  The only case from the Supreme Court that touched on the issue was decided in 1884, where a claim to a molecule was found invalid based on lack of novelty.[FN17] Similarly, molecule claims were often not attacked before lower courts under the “product of nature” doctrine, but simply for alleged lack of novelty[FN18] or for obviousness.[FN19] One case, Merck & Co. v. Olin Mathieson Chemical Corp., 253 F.2d 156 (4th Cir. 1958), specifically mentioned the “product of nature” doctrine.  There, the claimed “composition comprising recovered elaboration products” (emphasis added) was found to be patent-eligible because it “did not exist in nature in the form in which the patentees produced it.”[FN20] Of particular relevance to claims to DNA molecules are two Federal Circuit cases decided in the last decade of the 20th century.  In Amgen, Inc. v. Chugai Pharmaceutical Co., 927 F. 2d 1200 (Fed. Cir. 1991), the Federal Circuit held that a claim directed to “[a] purified and isolated DNA sequence” was novel and thus valid.[FN21] Subject matter patent-eligibility was not an issue in this case.  Nine years later, the Federal Circuit accepted a district court’s interpretation of “substantially pure DNA sequences” in a patent claim as including “both naturally-occurring and non-naturally occurring sequences.”[FN22] This was an affirmance of a summary judgment of non-infringement, however, and so validity was not an issue in this case at all. 3. Microorganisms: Few but highly relevant cases. The Supreme Court has thus far taken up only two cases where patent claims reciting microorganisms came under its scrutiny.  In Funk Bros. Seed Co. v. Kalo Inoculant Co., 333 U.S. 127 (1948), the Supreme Court held that a product claim directed to “[a]n inoculant . . . comprising a plurality of selected . . . strains of different species of bacteria” did not “disclose an invention or discovery within the meaning of the patent statutes.”[FN23] The opinion did not recite “product of nature,” however, and Justice Douglas’ analysis deserves a closer look.  First, the “product” claim-in-suit was not directed to a single bacterial species, but to a composition comprising a plurality of species, which do not co-exist in nature in the form in which they are found in the claimed composition.[FN24] Second, what rendered the product patent-ineligible was not the bacterial species per se, and Justice Douglas did not exclude them from being patent-eligible.[FN25] Thirty-two years later, the Supreme Court recognized the patent-eligibility of a genetically-engineered bacterium in Chakrabarty.  The claimed bacterium was a new strain of Pseudomonas into which two stable energy-generating plasmids[FN26] had been introduced to give the bacterium enhanced “hydrocarbon degradative” properties.[FN27] Chakrabarty confirmed the patent-eligibility of what Justice Douglas did not exclude as being patent-ineligible in Funk Bros., i.e., a new bacterium, a changed bacterium, or a bacterium with improved functioning beyond its natural functioning.[FN28] Chakrabarty does not recognize Funk Bros. as good law because of the “product of nature” doctrine, however, because Funk Bros. does not articulate the doctrine and the claimed subject matter in Funk Bros. is not a product of nature.[FN29] B.  A Critique of Judge Sweet’s Test and the Proposal of a New Test. Judge Sweet readily took one sentence out of Chakrabarty and formulated the “markedly different characteristics” [FN30] test.  What led Judge Sweet to reach the conclusion of patent-ineligibility of the claimed DNA molecules is not the lack of markedly different characteristics between them and their naturally-occurring counterparts, however, but some overwhelmingly identical function between them – the function of DNA as genetic information carrier.[FN31] This categorical exclusion of all differences in light of one common function is not the correct approach, however, because where subject matter patent-eligibility is an issue, there is almost always more similarity than difference between the claimed subject matter and its naturally-occurring counterpart.  Judge Sweet’s test is also inherently difficult to apply because there is no bright line between a difference that is marked and a difference that is not. Taking for granted that the “product of nature” doctrine exists, this Note proposes a two-part test in determining subject matter patent-eligibility of molecules under the doctrine.  This test puts structure before function, because the latter is at least in part determined by the former.  Under this test, the first step is to determine whether there is any structural difference between the claimed molecule(s) and its naturally-occurring counterpart(s) in the absence of any isolation or purification.  If a structural distinction is found, the entire analysis is complete.  Because of the structural difference, the claimed molecule, by definition, is not a “product of nature,” and is thus patent-eligible.  If a claimed molecule has the same structure as its naturally-occurring counterpart in the absence of any isolation or purification, then the next step is to determine whether there is any functional difference between the two.  Only a functional difference which would not have existed “but for” the isolation or purification puts the claimed molecule outside the scope of “product of nature.”  If the claim passes both parts of the test, i.e., the claimed molecule is structurally identical to its naturally-occurring counterpart and there is no functional difference between the two which would not have existed but for the isolation or purification, then the claimed molecule is a product of nature and patent-ineligible. C.  Application of the Proposed Test Would Have Yielded Results Consistent with the Vast Majority of Relevant Case Law on Molecule Claims. Retroactive application of the proposed test to molecule-claim case law would have yielded results consistent with the vast majority of relevant cases.  For example, the claim in Cochrane would have been found invalid if the “product of nature” doctrine were forced upon it and the proposed test applied, because it would likely have passed both parts of the test, i.e., the claimed molecule is structurally no different from and is used for exactly the same purpose as its naturally-occurring counterpart in the absence of any isolation or purification – as a red dye.  In both Parke-Davis and Kuehmsted, the product claims-in-suit could not have been invalidated under the “product of nature” doctrine because the products claimed were structurally different from anything found in nature.  In In re Williams, the validity of the product claim-in-suit would have been upheld, too, if the proposed test were applied.  Structurally, the claimed molecule, the laevo rotary form of lactone, is different from its naturally-occurring counterpart, which is in racemic form.  Even if one could argue that the racemic form necessarily contains the laevo form, the laevo form has a drastically and noticeably different function from the racemic form, which would not have existed but for the purification of it from the racemic form.[FN32] The only case where the claims-in-suit are true molecule claims and application of the proposed two-part test may have yielded a result inconsistent with the decision of the court is In re Bergstrom.  In this case, the claimed PGE2 and PGE3 molecules are structurally no different from their respective naturally-occurring counterparts, and there is no evidence of any functional difference between the claimed molecules and the naturally-occurring molecules either.[FN33] The reason for the inconsistent results does not lie in the proposed test, however, but in the court’s belief that compounds that “do not exist in nature in pure form” are not “naturally occurring.”[FN34] As some of the most fervent proponents of the “product of nature” doctrine clearly pointed out, In re Bergstrom was “not a section 101 patentable subject matter case.”[FN35] The proposed test is applicable specifically to molecule claims and should not be used on claims directed to elements, because an element, when in pure form (assuming that it does not exist in nature in such pure form), can have very different functions from its status as part of a molecule, and so would have been patent-eligible under the test.[FN36] The inapplicability of the proposed test to element claims is of no significance to patent applicants today, however, because all naturally-occurring elements have been identified by now. It is also hard to apply the proposed test to claims directed to microorganisms, because microorganisms are so complex that it is meaningless to define if two are “structurally” the same.  For now, Chakrabarty has made some bright-line rules, i.e., a difference as large as a few exogenous plasmids would make a new bacterium.  It remains to be seen if any smaller difference, e.g., only one exogenous plasmid, or addition or deletion of genomic DNA sequences, or even one or more point mutations, would distinguish a bacterium from its naturally-occurring counterpart under the “product of nature” doctrine. D.  Application of the Proposed Test to the Composition Claims-in-Suit Yields a Legal Conclusion Contrary to That of Judge Sweet’s. The proposed test can be easily applied to the DNA molecule claims-in-suit in Myriad.  The claimed DNA molecules are patent-eligible because they are all structurally different from their naturally-occurring counterparts, i.e., the chromosomes on which BRCA1 and BRCA2 genes are located.[FN37].  Such structural differences are at least two-fold.  First, the sequences of the claimed DNA molecules are different from the sequences of the respective chromosomes.  For example, the specification defines “isolated DNA,” which is recited in all composition claims-in-suit, as follows: “An ‘isolated’ . . . DNA . . . is one which is substantially separated from other cellular components which naturally accompany a native human sequence or protein, e.g., ribosomes, polymerases, many other human genome sequences and proteins.”[FN38] In the case of DNA, only other sequences on the same chromosome as a certain native sequence are considered naturally accompanying that native sequence.  When a DNA molecule comprises the native sequence substantially separated from the naturally accompanying sequences, such DNA molecule cannot have the same sequence as the chromosome on which the native sequence is located. Second, the claimed DNA molecules do not contain intron sequences that are parts of the chromosomes.  Judge Sweet argues that because native, intron-containing DNAs by definition encode the BRCA1 and BRCA2 proteins, such claims as Claim 1 of the ’282 Patent and Claim 1 of the ’492 Patent read on the native DNAs and so the claimed DNAs are patent-ineligible.[FN39] This argument indeed has some merit with regard to Claim 1 of the ’282 Patent.  In contrast to that claim, the distinction between molecule and sequence in Claim 1 of the ’492 Patent helps the patentee obviate such an attack.  Native, intron-containing chromosomal DNA falls outside the scope of this claim because even though it encodes the recited polypeptide, the polypeptide is not encoded by a single nucleic acid sequence as required by the claim but collectively by a number of distinct exon sequences interspersed by the introns.  This claim represents an excellent example of how careful claim drafting can obviate a “product of nature” rejection. The claimed DNA molecules are also patent-eligible because there exist functional differences between them and their naturally-occurring counterparts, the chromosomes, which would not have existed but for the isolation.  Defendant Myriad focused on the gain of function as a result of the isolation.[FN40] The isolation also causes loss of function, e.g., the isolated DNAs no longer encode, let alone express, the hundreds of genes other than BRCA1 and BRCA2 expressed by their respective chromosomes.  Neither the gain of function nor the loss of function would have existed but for the isolation.  Therefore, the claimed DNA molecules are not products of nature and are patent-eligible. III.  Conclusion The legal history of the “product of nature” doctrine is messy, if not extremely thin.  This Note at the outset reviewed judicial opinions on the patent-eligibility and –ineligibility of various claimed subject matter related to the “product of nature” doctrine.  Of particular relevance to the DNA molecule claims in the patents-in-suit in Myriad are those cases in which the claims-in-suit were directed to molecules.  Significantly, not even one molecule claim was invalidated by courts under the “product of nature” doctrine before Judge Sweet’s opinion. Assuming the existence of the “product of nature” doctrine, this Note proposed a two-part test applicable to claims directed to molecules that would have unified the holdings of relevant case law and yielded results consistent with them, with one justified exception.[FN41] Applying the proposed test to the composition claims-in-suit in Myriad leads to only one legal conclusion – the claimed DNA molecules are not products of nature and are patent-eligible.  The key to the successful understanding and application of the proposed test is to know what the naturally-occurring counterpart is.  The Myriad court compared the sequences of the claimed molecules with certain sequences on the chromosomes.  Such comparison is fundamentally flawed.  When the claimed subject matter is a molecule, its naturally-occurring counterpart must also be a molecule, not part of a molecule, e.g., a partial sequence of a DNA molecule. Judge Sweet’s opinion has also been objected to by a wide array of educational institutions and companies in the biotech and the pharmaceutical industries, as evidenced by the vast majority of the amicus briefs filed before the Federal Circuit that are in support of reversal.  This Note did not focus on analyzing those briefs.  Rather, this Note proposed a test that can be applied to molecule claims under the “product of nature” doctrine and unify the vast majority of relevant case law.  The proposed test can, hopefully, draw some bright line in judicial determination of the patent eligibility of molecules.  
…* B.Sc., Peking University; Ph.D., Harvard University; J.D. candidate, New York University. [FN1] 35 U.S.C. § 101. [FN2] J.E.M. Ag Supply, Inc. v. Pioneer Hi-Bred Int’l, Inc., 534 U.S. 124, 131 (2001). [FN3] See, e.g., Gottschalk v. Benson, 409 U.S. 63, 67 (1972) (“Phenomena of nature, . . . mental processes, and abstract intellectual concepts are not patentable . . . .”); Diamond v. Chakrabarty, 447 U.S. 303, 309 (1980) (“The laws of nature, physical phenomena, and abstract ideas have been held not patentable.”). [FN4] See John M. Conley & Roberte Makowski, Back to the Future: Rethinking the Product of Nature Doctrine as a Barrier to Biotechnology Patents (Part I), 85 J. Pat. & Trademark Off. Soc’y, 301 (2003) [hereinafter Conley & Makowski, Back to the Future (I)]; John M. Conley & Roberte Makowski, Back to the Future: Rethinking the Product of Nature Doctrine as a Barrier to Biotechnology Patents (Part II), 85 J. Pat. & Trademark Off. Soc’y, 371 (2003) [hereinafter Conley & Makowski, Back to the Future (II)].  Those articles discuss the patent-eligibility of “genes” and “DNA sequences.”  See, e.g., Conley & Makowski, Back to the Future (I)supra, at 309.  “Gene patenting” is simply a misnomer.  No patent claim has ever been directed to a gene, partly because the word “gene” itself has always been and will perhaps always be an indefinite word even to those skilled in the art.  See, e.g., Helen Pearson, What Is A Gene?, 441 Nature 399 (2006).  While the word “sequence” is used by many patent practitioners and appears in the preamble of many patent claims, the author of this Note takes the position that “sequence” refers only to the order in which nucleotides are arranged in a DNA molecule, and so it is the DNA molecule that should be the claimed subject matter, not any sequence.  This basic concept must be understood first before analyzing the patent-eligibility of these claims. [FN5] Ass’n for Molecular Pathology v. USPTO (Myriad), 702 F. Supp. 2d 181, 232 (S.D.N.Y. 2010). [FN6] It appears that Judge Sweet thinks that “products of nature” encompass all patent-ineligible subject matter.  See id. at 220 (“[T]he proper analysis requires determining . . . whether the claimed invention constitutes statutory subject matter . . . or whether the claimed invention instead falls within the judicially created ‘products of nature’ exception to patentable subject matter, i.e., ‘laws of nature, natural phenomenon, and abstract ideas,’ Chakrabarty, 447 U.S. at 309.”).  The author of this Note takes the position that “product of nature” is but one judicially created category of patent-ineligible subject matter, if at all.  For example, it is obvious that “product of nature” does not encompass “abstract ideas.” [FN7] Id. at 227 (quoting Chakrabarty, 447 U.S. at 310). [FN8] Id. at 229.  “BRCA1” and “BRCA2” refer to the human breast cancer susceptibility genes 1 and 2, respectively. [FN9] Id. at 218 n.40.  In particular, the claim-in-suit in Chakrabarty is directed to a microorganism.  The claims-in-suit in Gen. Elec. Co. v. De Forest Radio Co., 28 F.2d 641 (3d Cir. 1928), In re Marden, 47 F.2d 957 (C.C.P.A. 1931), and In re Marden, 47 F.2d 958 (C.C.P.A. 1931) are all directed to elements. [FN10] These categories are by no means an exhaustive list of the subject matter that may appear in patent claims where the “product of nature” doctrine may be applied.  The choice of these three categories for analysis is based on the available case law where subject matter patent-eligibility was or could have been discussed by the relevant authorities. [FN11] U.S. Patent No. 1,082,933 p.9 l.120-21 (filed Jun. 19, 1912). [FN12] See Gen. Elec. Co., 28 F.2d at 643. [FN13] Myriad, 702 F. Supp. 2d at 219 n.40. [FN14] Gen. Elec. Co., 28 F.2d at 642. [FN15] “Discoveries” is explicitly written in the “IP Clause” of the Constitution.  U.S. Const. art I, § 8, cl. 8. [FN16] See, e.g.In re Marden, 47 F.2d 957, 1046 (C.C.P.A. 1931) (“Uranium is a product of nature, and the appellant is not entitled to a patent on the same, or upon any of the inherent natural qualities of that metal.”); In re Marden, 47 F.2d 958, 959 (C.C.P.A. 1931) (“[P]ure vanadium is not new in the inventive sense, and, it being a product of nature, no one is entitled to a monopoly of the same.”). [FN17] Cochrane v. Badische Anilin & Soda Fabrik, 111 U.S. 293, 311 (1884) (“Artificial alizarine,” though “made artificially for the first time,” is “an old article” that could be “eliminated from the madder root,” and is not patentable as “a well-known substance.”).  Judged by today’s standard, the claim-in-suit is a product-by-process claim and would have been novel over the same product produced by known methods.  See Abbott Labs. v. Sandoz, Inc., 566 F.3d 1282 (Fed. Cir. 2009).  Another often cited case is American Wood-Paper Co. v. Fibre Disintegrating Co., 90 U.S. 566 (1874).  The product claim-in-suit in this case, however, is not directed to a molecule, but “a pulp suitable for the manufacture of paper.”  Id. at 594.  The claim was invalidated because the pulp was not novel.  Were it novel, it “might have been patented.”  Id. at 596.  These two cases, as some of the most fervent proponents of the “product of nature” doctrine clearly stated, “were not really product of nature cases.”  Conley & Makowski, Back to the Future (I)supra note 4, at 328 n.193. [FN18] See, e.g.In re Williams, 171 F.2d 319 (C.C.P.A. 1948) (holding that laevo rotary form of lactone was patentable over the known racemic form); In re Bergstrom, 427 F.2d 1394 (C.C.P.A. 1970) (holding that purified prostaglandins PGE2 and PGE3 were not anticipated by those previously extracted from human or animal prostate glands).  Another often cited case is Parke-Davis & Co. v. H.K. Mulford Co., 189 F. 95 (S.D.N.Y. 1911), holding that purified adrenaline in base form was not anticipated by the existence of adrenaline in suprarenal glands, because adrenaline in suprarenal glands exists as a natural salt, not in base form.  Id. at 103.  Yet another often cited case that is further removed from the “product of nature” doctrine is Farbenfabriken of Elberfeld Co v. Kuehmsted, 171 F. 887 (7th Cir. 1909) (holding that the claim directed to acetyl salicylic acid (“aspirin”), the product of a new process which for the first time produced it in a pure state and rendered it valuable for medicinal use, is valid and infringed), because aspirin is not a product of nature, but man-made. Id. at 890. [FN19] See, e.g.In re Kratz, 592 F.2d 1169, 1172 (C.C.P.A. 1979).  The “synthetically produced” limitation in the composition claim-in-suit would have brought the scope of the claim outside the ambit of “product of nature,” so that the doctrine would not have applied in this case in the first place.  Judge Sweet agreed that Kratz was not attacked or discussed as a subject matter patent eligibility case, but did not make it clear whether it could have been attacked under the “product of nature” doctrine.  Ass’n for Molecular Pathology v. USPTO (Myriad), 702 F. Supp. 2d 181, 227-28 (S.D.N.Y. 2010). [FN20] Merck & Co. v. Olin Mathieson Chemical Corp., 253 F.2d 156, 157, 164 (4th Cir. 1958); U.S. Patent No. 2,703,302 col.21, l.70 through col.22 l.6 (filed Dec. 8, 1952).  The claims-in-suit are not really molecule claims, as evidenced by the highlighted plural form of “products.”  The court also agreed that the claimed composition was not pure.  Merck & Co., 253 F.2d at 160. [FN21] Amgen, Inc. v. Chugai Pharmaceutical Co., 927 F. 2d 1200, 1202, 1204, 1206 (Fed. Cir. 1991).  To ensure consistency and to minimize confusion, the claim should have been drafted to be directed to a purified and isolated DNA molecule.  See supra note 4. [FN22] Schering Corp. v. Amgen Inc., 222 F.3d 1347 (Fed. Cir. 2000); Schering Corp. v. Amgen Inc., 18 F. Supp. 2d 372, 400 (D. Del. 1998). [FN23] Funk Bros. Seed Co. v. Kalo Inoculant Co., 333 U.S. 127, 128 n.1, 132 (1948). [FN24] See id. at 128 n.1. [FN25] See id. at 131 (“The combination of species produces no new bacteria, no change in the six species of bacteria, and no enlargement of the range of their utility. . . . Their use in combination does not improve in any way their natural functioning.”).  This language suggests that once the conditions expressed in the negative are met, the subject matter may be patent-eligible. [FN26] A plasmid is a kind of DNA molecule. [FN27] Diamond v. Chakrabarty, 447 U.S. 303, 305 (1980). [FN28] See supra note 25 and accompanying text. [FN29] See supra notes 23-25 and accompanying text.  Contra Conley & Makowski, Back to the Future (II)supra note 4, at 376. [FN30] Chakrabarty, 447 U.S. at 310. [FN31] Myriad, 702 F. Supp. 2d at 228-29 (“[DNAs] are physical carriers of information . . . . This informational quality is unique among the chemical compounds found in our bodies, and it would be erroneous to view DNA as ‘no different[]’ than other chemicals previously the subject of patents.  Myriad’s argument that all chemical compounds . . . necessarily conveys [sic] some information ignores the biological realities of DNA in comparison to other chemical compounds in the body.  The information encoded in DNA is not information about its own molecular structure incidental to its biological function . . . .  Rather, the information encoded by DNA reflects its primary biological function: directing the synthesis of other molecules in the body – namely, proteins . . . .”) (second alteration in original) (emphasis in original) (footnote omitted) (citations omitted) (internal quotation marks omitted).  While there is no doubt that DNA’s function as a carrier of genetic information is of great importance, this function is determined by its molecular structure.  After all, the course that teaches DNA and the central dogma is called Molecular Biology, not Informational Biology.  No molecular structure is incidental to the molecule’s biological function; the molecular structure at least in part determines the biological function.  The biological function may be directing the synthesis of proteins, as in the case of DNA molecules, or binding some targets in the body, as in the case of many other biologically active chemical compounds.  The ability of a certain chemical compound to bind some targets and not others is also information about other molecules in the body.  Such information is the subject of study in the fields of chemoinformatics and pharmacoinformatics.  Even though this information is not “genetic” information, it nevertheless is of great importance to our understanding of the human body. [FN32] The laevo rotary form of a compound has the property of deflecting polarized light to the left, while the racemic form produces no deflection of polarized light.  In re Williams, 171 F.2d 319 (C.C.P.A. 1948). [FN33] In re Bergstrom, 427 F.2d 1394, 1398, 1402 (C.C.P.A. 1970). [FN34] Id. at 1401. [FN35] Conley & Makowski, Back to the Future (II)supra note 4, at 387-88. [FN36] For example, pure tungsten has ductility and high tensile strength not observed in a tungsten compound.  See supra note 11 and accompanying text. [FN37] A chromosome is a naturally-occurring DNA molecule. [FN38] U.S. Patent No. 5,747,282 (filed Jun. 7, 1995) (the “’282 Patent”) col.19 l.8-18; U.S. Patent No. 5,837,492 (filed Apr. 29, 1996) (the “’492 Patent”) col.17 l.62 through col.18 l.5 (emphasis added). [FN39] Ass’n for Molecular Pathology v. USPTO (Myriad), 702 F. Supp. 2d 181, 230 (S.D.N.Y. 2010); ’282 Patent col.153 l.56-58 (“An isolated DNA coding for a BRCA1 polypeptide, said polypeptide having the amino acid sequence set forth set forth in SEQ ID NO:2.”); ’492 Patent col.167 l.16-19 (“An isolated DNA molecule coding for a BRCA2 polypeptide, said DNA molecule comprising a nucleic acid sequence encoding the amino acid sequence set forth in SEQ ID NO:2.”) (emphasis added). [FN40] See Myriad, 702 F. Supp. 2d at 230-31 (quoting Myriad Defendants’ Memorandum in Reply to Plaintiffs’ Opposition to Myriad Defendants’ Motion for Summary Judgment at 9, Myriad, 702 F. Supp. 2d 181 (No. 109CV04515)). [FN41] See supra notes 33-35 and accompanying text.

How to Protect Against a Licensing Partner’s Bankruptcy: Patent Licenses and the Bankruptcy Code

By Jordan Markham* A pdf version of this article may be downloaded here. I.  Introduction In uncertain times, or when dealing with uncertain partners, planning for the possibility of bankruptcy ex ante can provide a very real benefit to a party contemplating a patent licensing agreement.  Since the financial crisis of 2008, many contractual partners who formerly looked rock solid have experienced major cash-flow problems.  In addition, it has always been the case that in some fields of technology, such as biotech, a significant number of businesses are expected to fail.  Thus, it is important to think through at the outset how a license might be treated by a bankruptcy court, and where possible, to structure the agreement accordingly.  How to best do this will depend primarily on whether a party is the patentee or the licensee, and on the extent to which rights are transferred (i.e. whether the transaction results in a sale or merely a license agreement).  As we shall see, in the context of a bankruptcy proceeding, the patentee is generally better served by a greater, and a licensee by a lesser, transfer of rights. This article will explore these dynamics in detail.   It will first consider the effects of a bankruptcy court’s categorization of a patent rights transfer agreement as either an executory contract or a completed sale, and then it will consider how courts make this categorization.  In particular, it will begin by describing the provisions in the Bankruptcy Code that deal with executory contracts and how the parties’ rights depend on whether the debtor is the licensor or the licensee.  It will then consider, from the view of patent law, the difference between assignments and licenses.  It will move on to explore how courts in bankruptcy categorize assignments and different types of license agreements as either executory contracts or completed sales.  Finally, it will conclude with practical suggestions that flow from the preceding analysis. II.  The Treatment of Executory Contracts Under the Bankruptcy Code A.  Section 365(a) and the Power to Accept or Reject an Executory Contract Under the bankruptcy code, a transfer of intellectual property rights is either a completed sale or an executory contract.  But agreements respecting patent rights do not naturally present themselves as one or the other.  Transactional intellectual property lawyers usually speak of transfers of patent rights in terms of assignments and licenses.  For present purposes, the following simplifications are useful.  An assignment is a transfer of all significant rights under a patent, including the right to sue.  For bankruptcy purposes, an assignment is usually treated as a completed sale.  A license is an agreement whereby a patentee agrees to refrain from enforcing her right to exclude the licensee from exploiting her invention in some way.  The vast majority of effective license agreements are considered executory contracts for bankruptcy purposes. Assignments result in fewer complications than licenses under bankruptcy law.  If the bankrupt party is the buyer/assignee and the assignment does indeed amount to a completed sale, the rights under the assignment become property of the bankruptcy estate and may be disposed of accordingly.[FN1]Meanwhile, the seller/patentee need have no dealings with the bankruptcy proceeding except insofar as she is owed compensation by the bankrupt party (in which case her claims will be treated as any other prepetition claim of like kind).[FN2] Where the nondebtor is the buyer/assignee, unless the assignment amounts to a fraudulent transfer,[FN3] it is similarly free from entanglements with the bankruptcy proceeding since the bankrupt seller has already assigned over its rights.[FN4] In contrast to the simplicity of completed sales, executory contracts are governed by the complex provisions of section 365 of the Bankruptcy Code[FN5] and are the subject of considerable controversy in bankruptcy courts.  Section 365(a) permits “the trustee [or debtor in possession[FN6]], subject to the court’s approval, [to] assume or reject any executory contract or unexpired lease of the debtor.”[FN7] Note that section 365(a) pertains to executory contracts, as do the other subsections of section 365.  These provisions, therefore, govern the vast majority of license agreements.[FN8] The effect of 365(a) is to allow the debtor[FN9] to assume or reject a contract depending on whether, from the debtor’s perspective, the net gain outweighs the liabilities under the contract. The power placed in the hands of the debtor by section 365(a) to accept or reject a contract without regard to the preferences of the opposite party is considerably limited by protections built into the remainder of the section or read into it by courts that serve that party’s interests.  Courts have held that the debtor must have some minimal business justification for its choice[FN10] and may not, for example, reject a contract for the sole purpose of harming a competitor.[FN11] Further, if the debtor chooses to assume a contract under 365(a), it must assume the entire contract with all of its burdens.[FN12] To comply with section 365(b), the debtor must cure any prior defaults and provide adequate assurance of future performance.[FN13] In the event that the debtor rejects an executory contract, under subsection 365(g), the nondebtor is entitled to a prepetition unsecured damages claim for breach of that contract.[FN14] Finally, as will be discussed below, the power granted to the debtor under 365(a) is further limited by the carve-outs described in 365(c) and 365(n), which apply when the debtor is the licensee and licensor, respectively. B.  When the Nondebtor is Licensee: Section 365(n) Section 365(n)[FN15] provides an important protection to a licensee from the potentially severe consequences of having rights for which it contracted and on which it may depend suddenly yanked away by an insolvent licensor.  The protection of section 365(n) was added relatively recently as part of the Intellectual Property Bankruptcy Protection Act of 1988 in response to a particularly alarming decision by the Fourth Circuit Court of Appeals.[FN16] In Lubrizol, the debtor licensor sought to reject its non-exclusive license agreement with Lubrizol for no other reason than that its “continued obligation to Lubrizol under the agreement would hinder [its] capability to sell or license the technology on more advantageous terms to other potential licensees.”[FN17] Meanwhile, Lubrizol argued that it would be seriously damaged by having to renegotiate with the debtor at this juncture.  The Fourth Circuit was not unsympathetic, stating that “allowing rejection of such contracts … imposes serious burdens upon contracting parties such as Lubrizol.”[FN18] It was further concerned that such a result would “have a general chilling effect upon the willingness of such parties to contract at all with businesses in possible financial difficulty.”[FN19] However, when a statute enacted by Congress is clear, courts will usually apply it literally, regardless of its results or questionable policy.  This is what the Fourth Circuit did with section 365(a), allowing the debtor to reject the contract and leaving Lubrizol in a difficult position. Congress could do nothing for Lubrizol, but it did mitigate the problem for future licensees in Lubrizol’s position.  Section 365(n) allows a licensee of “intellectual property”[FN20] to retain its rights under the license, at least to the extent that they do not require specific performance of affirmative obligations of the debtor.[FN21] Exclusive licenses remain exclusive, confidentiality agreements remain in effect, and licensees even retain any option to extend the license built into the agreement.[FN22] In exchange, of course, licensees must continue to pay any royalties due under the agreement.[FN23] Or, alternatively, if the licensee prefers to accept the debtor’s rejection, it has the usual option of treating the rejection as a breach giving rise to prepetition unsecured damages claim under section 365(g).[FN24] C.  When the Nondebtor is the Licensor: Section 365(c) Section 365(c) is an extraordinary provision which, when read literally, in most cases gives a licensor of patented technology the ability to refuse to honor a contract with a licensee in bankruptcy.  Section 365(c) provides:
The trustee may not assume or assign any executory contract or unexpired lease of the debtor, whether or not such contract or lease prohibits or restricts assignment of rights or delegation of duties, if— (1)(A)  applicable law excuses a party, other than the debtor, to such contract or lease from accepting performance from or rendering performance to an entity other than the debtor or the debtor in possession, whether or not such contract or lease prohibits or restricts assignment of rights or delegation of duties; and (B) such party does not consent to such assumption or assignment…[FN25]
The provision would more readily make sense if it read, “the trustee may not assume and assign any executory contract….”  In that case, the licensor would simply be extended the same right under bankruptcy law – to refuse performance from a party with whom it did not contract – that it had prior to the bankruptcy (assuming it had that right).  But the statute reads otherwise.  It says that if the nondebtor licensor objects to the assumption of the contract by the debtor, and if non-bankruptcy law excuses that licensor from accepting performance from some other (hypothetical) third-party, the debtor itself may not assume the license. This is a severe limitation on the rights granted to the debtor under section 365(a).  The pertinent non-bankruptcy “applicable law” for patent licenses is federal common law.[FN26] It is a well-established principle under this law that unless the terms of the agreement provide otherwise, a non-exclusive patent license is personal in nature and therefore not freely assignable.[FN27] This rule has been extended to exclusive licenses by at least one lower court,[FN28] and that court was almost certainly correct in its decision.[FN29] Although an exclusive license represents a more complete transfer of rights, it is still personal in nature.  It would be a serious and unjustifiable burden to ask a patentee to accept performance on an exclusive license from a competitor who sought to suppress rather than exploit the invention.  Of course, the code is correct to honor this right under federal common law.  But it does not follow that patentees should be free to reject performance from the debtor itself.  The terms of section 365(c)(1) allows a licensor, at its whim, to force a renegotiation of terms of access to patented technology to which the debtor had already secured a right.  This flies in the face of the core bankruptcy policy in favor of maximizing the value of the estate.  Nevertheless, many courts have felt constrained to follow this “plain meaning” of the statute. This literal interpretation of section 365(c) is usually called the “hypothetical test.”  The test asks whether the licensor would be excused from doing one thing (acquiescing to assignment to a hypothetical third party) in order to determine whether it is excused from doing another (acquiescing to assumption).  The Third, Fourth, Ninth, and Eleventh Circuits take the “hypothetical test” approach.[FN30] This approach has the merits of leaving lawmaking to a representative elected body and of holding Congress to its words.[FN31] Plain meaning dogmatists would also claim that it has the advantage of predictability since anyone can read the statute and know what it means without delving deep into the case law or engaging in some kind of mystical divination of Congressional purpose.[FN32] Such a divination was attempted by the First Circuit Court of Appeals in Summit Inv. & Dev. Corp. v. Leroux.[FN33] The court in this case looked to two pieces of statutory history[FN34] to justify the conclusion that subsection 365(c) – and, more to the point in Leroux, 365(e) as well[FN35]– contemplates a “case-by-case inquiry” into whether the nondebtor party was actually being forced to accept performance under its executory contract from someone other than the debtor party with whom it originally contracted.[FN36] This approach is referred to as the “actual test.” Before looking to statutory history, the court in Leroux concluded that 365(c) is ambiguous and capable of alternative readings.[FN37] It is not true that 365(c) alone is ambiguous.  It speaks clearly.  However, as has been noted by advocates and courts, section 365(f) conflicts with section 365(c).[FN38]Section 365(f)(1) provides: Except as provided in subsections (b) and (c) of this section, notwithstanding a provision in an executory contract or unexpired lease of the debtor, or in applicable law, that prohibits, restricts, or conditions the assignment of such contract or lease, the trustee may assign such contract or lease under paragraph (2) of this subsection.[FN39] That 365(f) allows the assignment of contracts notwithstanding applicable law indicates a drafting error.  Section 365(f)(1) is conditioned on 365(c), and 365(c) expressly prohibits both assumptions and assignments when they are excused by applicable law.  Thus, 365(f)(1), by way of 365(c)(1), renders itself superfluous.[FN40] One or the other provision is in error.  But the way the statute conflicts does not lead logically to the conclusion that the “actual test” is correct.  Policy intuition, on the other hand, does lead to this conclusion.  The actual test does not unduly burden the estate by permitting a hold-up style renegotiation, and it also does not force a patentee to accept performance from a third party where federal common law would excuse it.  It is in keeping with sound bankruptcy policy and, not surprisingly, it is the test favored by respected commentators and academics.[FN41] Of course, from a licensor’s point of view, the hypothetical test is preferable to the actual test.  It has the virtue of insuring the nondebtor against being locked into an agreement with a partner who is only marginally viable, or whose management or ownership has changed in a way that is disagreeable to the licensor.[FN42] Particularly for a patentee who has granted an exclusive license, and thereby promised not to license the technology to another party, the burden of having to accept performance from a defunct partner is substantial.[FN43] That these burdens probably do not justify the terms of 365(c) as Congress drafted it,[FN44] and that not all courts apply the hypothetical test, should give pause to a potential licensor.  The area of law is unstable and probably not to be depended upon. In fact, at least one court in a “hypothetical test” jurisdiction has allowed a work-around solution to the debtor’s 365(c) problem.  In In re Hernandez,[FN45] the court permitted the debtor to intentionally neglect to accept or reject an exclusive license agreement under 365(a) so that the license would simply “ride through” the bankruptcy unaffected.[FN46] The so called “ride through” doctrine has been approved by courts of appeals in other contexts[FN47] and could provide an effective shield for debtor licensees against 365(c).  The major drawback from the debtor’s perspective is that since the contracts are not accepted under 365(a), the protection against ipso facto clauses in 365(e) does not apply.[FN48] III.  Conclusion What lessons can we learn from the material just surveyed?  As a nondebtor licensee, it would be helpful to structure an agreement as an assignment when possible, since an assignee is beyond the reach of the bankruptcy court.  However, when this is not a realistic option, the licensee’s rights are reasonably well protected by section 365(n).  The only further precaution that might be helpful is to provide for the option to extend the length of the agreement, since these options will be honored by a bankruptcy court. As a licensor, it is advantageous to have the agreement in the form of an executory contract.  This is because section 365(c) allows for at least some continuing control over the technology.  In a jurisdiction where the “actual test” is followed, the licensee will be prevented from assigning the license without the licensor’s consent.  In a “hypothetical test” jurisdiction, the licensor will have the right to refuse performance even from the licensee.  Executory contracts include practically all licenses agreements in most jurisdictions, but it would be helpful to have at least some of the licensee’s payments in the form of royalties, rather than a single lump sum.  Specifying any other significant ongoing obligations imposed on the licensor would also be helpful.  Finally, in order to avoid a “ride-through” situation, a licensor should include an ipso facto clause in the contract.  
* Jordan Markham earned his J.D. from NYU Law School in May of 2010. He will be joining Milbank Tweed LLP as an associate in January of 2011, and in the interim has been working at the White Plains IP boutique, Leason Ellis LLP. [FN1] See 11 U.S.C. § 541(a)(1) (2000) (defining the property of a bankruptcy estate as including “all legal or equitable interest of the debtor in property…”). [FN2] See 11 U.S.C. § 507 (2000) (establishing claim priorities). [FN3] See id. § 548. [FN4] Of course, if the buyer has not completed its obligations under conditions of the sale (for example, the buyer owes cash payments), these will remain owed to the estate. [FN5] 11 U.S.C. § 365 (2000). [FN6] See id. § 1107 (2000). [FN7] Id. § 365. [FN8] For the sake of readability this paper shall speak of licenses rather than patent rights transfers in the context of executory contracts.  Similarly, it shall speak of licensees and licensors rather than transferees and transferors. This simplification is justified by the extreme rarity of patent rights transfers that are not licenses but are treated as executory contracts. [FN9] A debtor in possession, as a general rule, has the same rights and duties as a trustee under the bankruptcy code.  See 11 U.S.C. § 1107.  For the purpose of readability, this paper shall use the term debtor rather than trustee or debtor in possession. [FN10] See Johnson v. Fairco Corp. (In re Johnson), 61 B.R. 317, 320 (Bankr. N.D. Ill. 1986) (describing the business judgment standard as a “lax standard” that disturbs a debtor’s actions “[o]nly where [such actions] are in bad faith or in gross abuse of its managerial discretion.”). [FN11] See In re Petur U.S.A. Instrument Co., 35 B.R. 561, 563 (Bankr. W.D. Wash. 1983); Lubrizol Enterprises, Inc. v. Richmond Metal Finishers, Inc., 756 F.2d 1043, 1047 (4th Cir. 1985) (the question is  “whether the decision of the debtor that rejection will be advantageous is so manifestly unreasonable that it could not be based on sound business judgment, but only on bad faith, or whim or caprice.”). [FN12] 3 Collier on Bankruptcy ¶ 365.03[1] (15th ed. rev. 2006). [FN13] 11 U.S.C. § 365(b). [FN14] However, even if specific performance would ordinarily be available (which is unlikely in an intellectual property licensing context), it is not available with respect to a party in bankruptcy.  Lubrizol, 756 F.2d at 1048 (citing H.R. Rep. No. 95-595, at 349 (1978), reprinted in 1978 U.S.C.C.A.N. 5963, 6305) (“Under 11 U.S.C. § 365(g), [the licensee] would be entitled to treat rejection as a breach and seek a money damages remedy; however, it could not seek to retain its contract rights in the technology by specific performance even if that remedy would ordinarily be available upon breach of this type of contract.”). [FN15] 11 U.S.C. § 365(n). [FN16] 3 Collier on Bankruptcy P 365.14 (“In enacting section 365(n), Congress effectively overruled the decision of the Court of Appeals for the Fourth Circuit in Lubrizol Enterprises, Inc. v. Richmond Metal Finishers, Inc. (In re Richmond Metal Finishers, Inc.), which had permitted a debtor in possession to reject a technology licensing agreement and terminate the licensee’s right to use the license.”). [FN17] 756 F.2d at 1047. [FN18] Id. at 1048. [FN19] Id. [FN20] Intellectual property under the code includes patent, copyright and trade secret rights but, notably, excludes trademark rights.  11 U.S.C. § 101(35A) provides: The term “intellectual property” means— (A) trade secret; (B) invention, process, design, or plant protected under title 35; (C) patent application; (D) plant variety; (E) work of authorship protected under title 17; or (F) mask work protected under chapter 9 of title 17; to the extent protected by applicable nonbankruptcy law. [FN21] 11 U.S.C. § 365(n)(1)(B). [FN22] Id. § 365(n)(1)(B)(ii). [FN23] Id. § 365(n)(2)(B). [FN24] Id. § 365(n)(1)(A). [FN25] § 365(c) (emphasis added). [FN26] Everex Sys., Inc. v. Cadtrak Corp. (In re CFLC, Inc.), 89 F.3d 673, 679-80 (9th Cir. 1996) (holding, after lengthy analysis, that the statutory “applicable law” is federal law, and that federal law does not permit free assignment of patent licenses); Institut Pasteur v. Cambridge Biotech Corp., 14 F.3d 489, 492 (1st Cir. 1997) (same conclusion). [FN27] Troy Iron & Nail Factory v. Corning, 55 U.S. (14 How.) 193, 216 (1852) (“A mere license to a party, without having his assigns or equivalent words to them, showing that it was meant to be assignable, is only the grant of a personal power to the licensees, and it is not transferable by him to another.”); Lane & Boley Co. v. Locke, 150 U.S. 193, 195-96 (1883) (holding that patent licenses, as a form of personal property, are not assignable unless expressly made so); In re CFLC, Inc., 89 F.3d at 679-80; Institut Pasteur, 14 F.3d at 492. [FN28] In re Hernandez, 285 B.R. 435 (Bankr. D. Ariz. 2002). [FN29] See Peter S. Menell, Bankruptcy Treatment of Intellectual Property Assets: An Economic Analysis, 22 Berkeley Tech. L.J. 733, 798 (2007) (“Unless the patentee has transferred so much of the patent bundle to an exclusive licensee as to constitute an assignment, it seems appropriate for a bankruptcy court to read the non-assignability of patent license rule to apply whether or not the license agreement granted exclusive or nonexclusive rights.”). [FN30] Third Circuit:  In re W. Elec., Inc., 852 F.2d 79 (3d Cir. 1988); see also In re Access Beyond Techs., Inc., 237 B.R. 32 (Bankr. D. Del. 1999). Fourth Circuit: RCI Tech. Corp. v. Sunterra Corp. (In re Sunterra Corp.), 361 F.3d 257, 262 n.9 (4th Cir. 2004); see also In re Travelot Co., 286 B.R. 447 (Bankr. S.D. Ga. 2002). Ninth Circuit:  Perlman v. Catapult Entm’t (In re Catapult Entm’t), 165 F.3d 747, 747 (9th Cir. 1999). Eleventh Circuit:  City of Jamestown v. James Cable Partners, L.P. (In re James Cable Partners, L.P.), 27 F.3d 534, 537, reh’g denied, 38 F.3d 575 (11th Cir. 1994). [FN31] See In re Catapult Entm’t, 165 F.3d at 754. [FN32] It is the view of this author that the rigid application of statutory nonsense is more likely to result in outrage and confusion than comfortable predictability. [FN33] 69 F.3d 608 (1st Cir. 1995). [FN34] S. Rep. No. 95-989, at 59 (1978), reprinted in 1980 U.S.C.C.A.N. 5787, 5845 (“This section will require the courts to be sensitive to the rights of the nondebtor party to executory contracts and unexpired leases.  If the trustee is to assume a contract or a lease, the court will have to ensure that the trustee’s performance under the contract or lease gives the other contracting party the full benefit of his bargain.”); H.R. Rep. No. 95-1195, at § 27(b) (1980) (“This amendment makes it clear that the prohibition against the trustee’s power to assume an executory contract does not apply where it is the debtor that is in possession and the performance to be given or received under a personal service contract will be the same as if no petition had been filed because of the personal nature of the contract.”). [FN35] Section 365(e) mirrors 365(c): (e) … (2)      Paragraph (1) of this subsection does not apply to an executory contract or unexpired lease of the debtor, whether or not such contract or lease prohibits or restricts assignment of rights or delegation of duties, if— (A) (i)      applicable law excuses a party, other than the debtor, to such contract or lease from accepting performance from or rendering performance to the trustee or to an assignee of such contract or lease, whether or not such contract or lease prohibits or restricts assignment of rights or delegation of duties; and (ii)      such party does not consent to such assumption or assignment; … [FN36] 69 F.3d at 612-13; see also Institut Pasteur v. Cambridge Biotech Corp., 104 F.3d 489 (1st Cir. 1997), cert. denied, 521 U.S. 1120 (1997) (following and applying the Leroux interpretation of 365(c) with respect to an executory nonexclusive patent license). [FN37] 69 F.3d at 612-13. [FN38] Perlman v. Catapult Entm’t (In re Catapult Entm’t), 165 F.3d 747, 751 (9th Cir. 1999); In re Catron, 158 B.R. 629, 636 (E.D. Va. 1993), aff’d, 25 F.3d 1038 (4th Cir. 1994); In re Cardinal Indus., Inc., 116 B.R. 964, 976-77 (Bankr. S.D. Ohio 1990). [FN39] 11 U.S.C. § 365(f)(1) (emphasis added). [FN40] Attempts to reconcile the provisions are thoroughly unconvincing.  See In re Catapult Entm’t, 165 F.3d at 751, and cases cited therein. [FN41] See David G. Epstein, Steve H. Nickles & James J. White, Bankruptcy § 5-15, at 258-59 (1993) (advocating that “Congress should amend section 365(c)(1) to make clear that ‘applicable law’ prohibitions on assignment do not preclude assumption in bankruptcy” and urging courts to permit a trustee or debtor to assume a contract notwithstanding a prohibition on assignment in the “applicable law” so as to ensure a harmony between subsections (c) and (f) of section 365); 3 Collier on Bankruptcy ¶ 365.06[1][d][iii]; Daniel J. Bussel & Edward A. Friedler, The Limits on Assuming and Assigning Executory Contracts, 74 Am. Bankr. L.J. 321, 322 n.6 (2000) (arguing that the “’actual test’ yields correct results from the point of view of bankruptcy policy and allows results in bankruptcy to effectively mirror the results outside of bankruptcy”). [FN42] For example, a provision in the non-exclusive license in Institute Pasteur allowed a competitor of the patentee to gain access to the patentee’s coveted technology by acquiring the debtor. 104 F.3d at 494-95. [FN43] A patentee has limited time to commercially exploit her patent.  Section 154(c)(2) of the Patent Act provides that the patent term begins when the patent is issued and ends 20 years from the date that the patent application was filed.  Since it take years to procure a patent from an overburdened patent office, patent terms are typically considerably fewer than 20 years. [FN44] Congress, in overriding ipso facto clauses in section 365(e), has expressed the view that, in general, the burden to a nondebtor party of having to accept performance from or render performance to a party in bankruptcy is outweighed by the need to maximize the value of the estate and to rehabilitate the debtor. [FN45] 287 B.R. 795 (Bankr. D. Ariz. 2002). [FN46] Id. [FN47] See, e.g., Stumpf v. McGee (In re O’Connor), 258 F.3d 392 (5th Cir. 2001), and cases cited therein. [FN48] See 11 U.S.C. § 365(e)(1).

Software Developers, On Guard!: Offering Software for Sale Can Trigger a Bar to Patentability Even If the Software Is Untested and Incomplete

By Paul A. Ragusa and Jack Chen* A pdf version of this article may be downloaded here. In Pfaff v. Wells Elecs., Inc.[FN1] the Supreme Court formulated the now well known test for determining when an invention cannot be patented due to a sale or offer for sale more than one year prior to the filing of a patent application. Specifically, the Court held that an invention need not be “reduced to practice” at the time of the sale or offer to create a statutory bar against patent protection. [FN2] Instead, a sale or offer of an invention “ready for patenting,” is sufficient to raise a statutory bar. [FN3] The purpose of the on-sale bar is to encourage early disclosure of inventions to the public as well as to prevent a de facto patent term extension by those who would commercially exploit an invention for an extended period of time prior to filing a patent. During the past twelve years, the courts have applied the Pfaff test to various technologies, some with more clarity than others. One thorny area involves the application of the Pfaff test in the context of software related inventions. Although the Pfaff Court rejected precedent that an invention needed to be “substantially complete” to provide a statutory bar, [FN4] it did not address how a software related invention can be ready for patenting where the code is incomplete, and untested. As a practical matter, how is a court to determine whether an unfinished software-related invention is ready for patenting and therefore can operate to trigger an on-sale bar? A recent district court decision addresses this issue head-on and is discussed below. I. Background An applicant may be barred from patenting a software method after offering it for sale even if the computer code implementing the method is incomplete at the time of the offer. The patent statute states, “[a] person shall be entitled to a patent unless…the invention was…on sale in this country, more than one year prior to the date of the application for patent in the United States.” [FN5] The purpose of the on-sale bar is to encourage early disclosure and to prevent inventors from removing existing knowledge from public use. [FN6] The on-sale bar applies when two conditions are met: (i) the invention is the subject of a commercial offer for sale and (ii) the invention is ready for patenting prior to the statutory one-year period. [FN7] An invention is “ready for patenting” where either (i) the invention has been reduced to practice or (ii) the inventor had prepared drawings or other descriptions of the invention that were sufficiently specific to enable a person skilled in the art to practice the invention. [FN8] II. Robotic Vision Systems The Court of Appeals for the Federal Circuit addressed the Pfaff test in the context of an invention involving software or computer programming in Robotic Vision Sys., Inc. v. View Eng’g, Inc. [FN9] There, the court addressed a software method for scanning leads on integrated circuit devices. The application for Robotic’s patent was filed on June 24, 1992, establishing a one-year date of June 24, 1991. [FN10] In March of 1991, Robotic sold one of its scanning devices to Intel Corporation and agreed to deliver the patented device to Intel by June 3, 1991, thereby establishing a commercial offer for sale in March 1991, prior to the critical date of June 24, 1991. [FN11] The Federal Circuit determined whether the claimed invention was ready for patenting prior to the critical date. Some time between March and April of 1991, a co-inventor of the patented scanning method, William Yonescu, disclosed the claimed method to Daniel Briceno of Robotic and asked him to write the software to implement the method. [FN12] It was undisputed that Briceno ultimately completed the software program according to Yonescu’s description in March-April 1991, thereby establishing that Yonescu’s description was sufficiently specific to allow Briceno to practice the invention and that the claimed invention was ready for patenting prior to the critical date of June 24, 1991. [FN13] According to the court, the second Pfaff requirement may be satisfied even though there is no “actual completion of such software…, provided that there is a disclosure that is sufficiently specific to enable a person skilled in the art to write the necessary source code to implement the claimed method.” [FN14]Under the Pfaff test, Robotic’s invention was therefore invalid due to an on-sale bar. [FN15] III. Netscape Commc’ns Corp. Netscape Commc’ns Corp. v. ValueClick, Inc. [FN16] involved the cookies feature of the popular Netscape browser in which a piece of data, called a “cookie” and stored on the user’s local computer, could be sent to the remote web server to enable the remote server to remember previous interactions with the user. For example, a remote server could identify a particular user by his or her cookie and present the user with his or her stored shopping cart of merchandise. ValueClick argued that the cookie feature was the subject of an offer for sale as early as July or August of 1994, more than one year prior to the filing of the corresponding patent application on October 6, 1995. [FN17] Netscape countered that no software product could have included the cookie feature prior to the one-year date (October 6, 1994) because the public release of the Netscape browser did not occur until after that date (October 13, 1994). Thus, Netscape argued that the cookie feature was not reduced to practice prior to the on-sale bar date. [FN18] According to the District Court, “this statement misunderstands the law governing the ‘ready for patenting’ Pfaff prong because it assumes that an invention is only reduced to practice, and thus the on-sale bar can only be applied, after the source code has been perfected.” [FN19] Testimony by Netscape’s expert revealed that a draft version of source code pertaining to the cookie feature was entered into Netscape’s software repository on October 4, 1994. [FN20] Netscape also testified that part of the method was completed by October 6, 1994, and that an early version of the code was entered into the software repository on October 3, 1994. [FN21] Thus, the District Court concluded that the existence of the draft source code prior to October 6, 1994, although perhaps incomplete, demonstrated that the method was ready for patenting prior to the critical date. The court went further. “Moreover, with respect to inventions involving computer code, Pfaff simply requires complete conception of the invention, not the source code’s actual completion, provided that there is an enabling disclosure that would allow one skilled in the art to complete the invention.”[FN22] In a declaration to the Patent and Trademark Office, Netscape’s Chief Technology Officer, John Giannandrea, stated that Netscape’s software developer, Lou Montulli, disclosed the cookie invention, which corresponded to claim 1 of the patent-in-suit, during design review meetings in July and August of 1994. [FN23] According to Giannandrea, the meetings involved Giannandrea and Montulli drawing the software architecture for the cookie invention on a white board. [FN24] The District Court found that Giannandrea, a software developer with more years of experience than Montulli, was a person of ordinary skill in the art for purposes of the Robotic test. [FN25] The July/August disclosure to Giannandrea and Montulli’s completion of the source code in October of 1994, coupled with Giannandrea’s years of software programming experience, constituted an enabling disclosure that would have enabled Giannandrea to write the source code. Accordingly, the District Court held that the evidence showed that the cookies invention was “ready for patenting” under the second prong of the Pfaff test. IV. Conclusion The Netscape case highlights the risk of offering for sale software products that are under development. A simple investigation concerning whether software code was complete is insufficient to establish the critical date for the purposes of § 102(b). Instead, an investigation should determine when the invention was disclosed in sufficient detail to enable one of ordinary skill in the art to write software, regardless of the state of software development. *** *Paul A. Ragusa is a partner and Jack Chen an associate in the New York office of Baker Botts LLP, where they practice intellectual property law with an emphasis on patent litigation, patent portfolio management, counseling, and licensing. [FN1] 525 U.S. 55 (1998). [FN2] Id. at 67. [FN3] Id. at 68. [FN4] Id. at 66. [FN5] 35 U.S.C. § 102(b). [FN6] Pfaff, 525 U.S. at 64. [FN7] Id. at 67-68. [FN8] Id. at 68. [FN9] 249 F.3d 1307, 1312 n.2 (Fed. Cir. 2001). [FN10] Id. at 1309. [FN11] Id. at 1311. [FN12] Id. [FN13] Id. at 1311-12. [FN14] Id. at 1312 n.2. [FN15] Id. at 1312-13. [FN16] No. 1:09cv225, 2010 U.S. Dist. Lexis 8733 (E.D. Va 2010). [FN17] Id. at 8779-80. [FN18] Id. at 8779. [FN19] Id. at 8780. [FN20] Id. at 8779. [FN21] Id. [FN22] Id. at 8780 (citing Robotic Vision Sys., 249 F.3d at 1311-13). [FN23] Id. at 8745-8746 n.8, 8782-83. [FN24] Id. at 8782-83. [FN25] Id. at 8782.