We are working through Bayh-Dole’s definition of “subject invention” at 35 USC 201(e). For an invention to be a subject invention, it must meet every element of the definition. Folks can’t just say something is a subject invention and by doing so make it a subject invention–though this is what many university administrators try to do in practice. Similarly, just because folks say an invention is not a subject invention, it does not make it so. The definition of subject invention is placed in federal patent law. (Bayh-Dole amends federal patent law–that in itself is really, really strange.) Thus, “subject invention” defines a new category of invention in patent law–invention and subject invention.
Here’s 35 USC 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.
Here’s 35 USC 261:
Subject to the provisions of this title, patents shall have the attributes of personal property.
Bayh-Dole introduces new conditions and requirements on inventions that meet the definition of subject invention. The conditions upon which patents may be obtained, and the attributes that those patents have, are then based in whether Bayh-Dole’s conditions and requirements apply. If an invention is a subject invention, then the attributes of any patent obtained are subject not only to patent law generally but specifically to Bayh-Dole’s statement of policy, other conditions, and requirements. Bayh-Dole makes these conditions imposed on subject inventions a matter of statute, not federal contract.
We now turn to two restrictions on subject inventions having to do with when an invention is determined to have been “made”:
conceived or first actually reduced to practice
The terms “conceived” and “first actually reduced to practice” come from now obsolete patent law having to do with establishing who has been first to make a given invention in cases where multiple patent applications have been filed on the same subject matter. Thus, these are not terms of scope but rather terms of chronological priority that are being repurposed in Bayh-Dole. The reasons are long and involved, so like most everything else in Bayh-Dole, we are in TL;DR territory. The short of it is that Bayh-Dole, to be able to preempt other federal law and executive branch policy on ownership of inventions made in work receiving federal support has to be as broad as any of those laws and regulations.
To be this broad, Bayh-Dole had to take into account Mine Safety Appliances (1966). In this case, a court determined that a helmet design invented at the University of Southern California “on the side” as it were to a US Navy research contract directed doing some physiology testing as part of a broader Navy project to design better head protection for pilots was within the scope of the research contract in determining that the federal government was entitled to a license to the invention. The new helmet design was then “conceived” within the scope of the federal contract even though USC insisted that no federal money was used in developing the helmet.
One can see how a court could reason this way–if whenever an inventor got an inventive idea while working on a federal contract, he or she could simply shift any work on the new idea out of the federally paid for work and prevent the government from having any right to the invention. This would all the more be the case for investigator-proposed scientific research, in which the deliverable is the performance of the work that’s described, which necessarily does not include what hasn’t been invented yet. But if the exchange is that the federal government should have a license to anything invented in work receiving federal support, then it should not matter whether the federal government funds the inventive work directly. So long as the government provides some funding, somewhere in the project, at some point in time, anything inventive in the project should be subject to a government expectation of at least a license. Thus, if one “conceives” an invention in a project that otherwise receives federal funding, then if the invention is owned by a federal contractor, then Bayh-Dole applies and the invention is a subject invention.
There’s a second case, coming after Bayh-Dole was passed and revised–in re Eddie L. King (1987)–that helps to understand the “first actual reduction to practice.” Actual reduction to practice takes place when testing is necessary to demonstrate that an invention performs in each of its elements as conceived by the inventor. In the King situation, Eddie King, a civilian employee working for the Air Force figured out (in 1969) a better design for shipping pallet couplers. He built a prototype, and prepared a patent application, but didn’t file it until after the Air Force had tested the prototype in its applications. When the tests proved positive, King filed his patent application–which did not incorporate any results from the Air Force testing. A patent was issued to King (1972) and later (1983) he brought an action for compensation for the Air Force’s use of his inventive pallet coupler. The Commissioner of Patents decided that the first actual reduction to practice took place by the Air Force before any patent application had been filed. Even though King had done all his development on his own time and with his own resources, the first actual reduction to practice had been done with federal resources. Thus, the government has its license.
The in re Eddie L. King case was decided under EO 10096 applicable to federal employees, not Bayh-Dole, but the decision does set out what “made” means in the context of inventing and “conception” and “first actual reduction to practice.” Perhaps of most importance, the decision hinged on whether testing was required to demonstrate that the invention would work as conceived. The Commissioner reasoned that the Air Force had tried numerous pallet coupler designs and none had worked as expected. Thus, so the reasoning went, testing was necessary in the case of King’s coupler design. Building a prototype was not sufficient. Nor was a full specification of the invention in the form of a draft patent application.
Now make the jump, then, to pharmaceutical inventions–Bayh-Dole’s ground zero for policy–where these conditions also apply: discoveries or inventions of compounds or methods with claims of therapeutic value may be reduced to practice in the form of prototypes and specifications but until there is testing in humans that demonstrates safety and efficacy, the invention cannot be demonstrated to function for its intended purpose–it has not been first actually reduced to practice.
The “first actual reduction to practice” may well a necessary step for proving that a given invention functions as conceived–even if otherwise patentable and the data from a clinical trial is not included in a patent application. In addition, then, to statements of work for federally funded research or development that specify that a prototype will be constructed or a prototype will be tested, the scope of “first actually reduced to practice” includes work that requires an additional set of testing to demonstrate that the invention performs in its intended environment. In the case of compounds identified as potential medicines for humans, until there’s testing in humans, there’s not first actual reduction to practice. That this matters will be evident from the discussion of the next bit of Bayh-Dole’s definition of subject invention.
in the performance of work under a funding agreement
Bayh-Dole substantially changes the FPR’s “in the course of or under this contract.” “Under” a contract means something along the lines of “specified” by the contract. One looks to a statement of work. “In the course of” means “what was done to do what was specified.” There is an analog in liability arising in master-servant treatments of employment. A master directs a servant to accomplish a task. That’s the “under” portion. The servant decides the means to do so. That’s the “course of” part. “Bring us pizza!” Employee calls in an order, and drives his car to pick it up. Under and in the course of covers both elements. If the employee is involved in a car accident on the way back with the pizza, even though use of the car was not specified in the employer’s specification of work, the employer is still (arguably) also liable for the accident, as it arises in the course of the employee’s work.
In the case of procurement contracts, something similar is at work. The commissioner of research and development seeks not only the deliverables that are specified but also what has been discovered or invented or gathered as the contractor goes about the work. Indeed, in contracts for the procurement of research services, the only deliverable that can be specified is the conduct of the research–everything else that matters arises in the course of what the investigators actually do.
To get at this issues, Bayh-Dole defines a “funding agreement” and directs this definition to a “work” (activity, project) that receives at least partial funding from the federal government. This is *way* different from the FPR’s approach. Bayh-Dole’s definition expands the “funding agreement” to include anyone that a federal contractor adds as a party to the funding agreement. Thus, people who receive no federal funding may, if they have agreed to be added as parties to a funding agreement, may make subject inventions, even though they personally do not have access to or use any federal funds. What matters is that the “work” of which their activity is a part receives federal support.
The “performance of work under a funding agreement” therefore is much broader than “in the course of or under this contract.” The “work” under the funding agreement is the “experimental, developmental, or research work funded in whole or in part by the Federal Government.” Think about it this way. Establish a project–the “work.” Now ask if any part of the project has received federal research, experimental, or development funding. If so, any invention made in the project–when owned by a party to the funding agreement–is a subject invention. It does not matter whether the federal funds are directly accounted for in the making of any given invention–what matters is that the invention was made somewhere in the project and the project somewhere received federal research or development funding. See 37 CFR 401.1 for an awkward explanation that stumbles its way to this conclusion.
It is misleading, therefore to write “federally supported invention” or “invention made with federal funding” with the implied meaning “what federal money has been used to make.” Rather, Bayh-Dole requires “invention made in work receiving at least in part federal funding.” University disclosure forms that ask “inventors” (or “creators” or whatever) to identify the source of funds used to make the invention are utterly skew from Bayh-Dole’s definition of subject invention. What is necessary to know is the scope of the project in which the invention has been made, and whether that project has received federal funding, even in part. Even then, things may be open ended, since at any time a contractor can add other contractors.
For instance, a non-profit contractor can assign an subject invention, and in doing so the receiving party is required by Bayh-Dole to accept the nonprofit’s patent rights clause–making the receiving party a party to the funding agreement, and therefore a contractor, and therefor inventions made by this new contractor, come within the scope of funding agreement–and therefore of the project.
Another instance. A university states in its written policies and descriptions of its invention licensing office that it will seek to develop inventions it owns into commercial products, such as through licensing. Those statements then necessarily extend the scope of any project that receives federal support. The project that the university has undertaken involves both research and commercial development of inventions as products. Federal funding comes to the university to support the research portion. Since the research portion is but one part of the project that the university has committed itself to undertaking, federal claims on inventions extend to inventions that the university makes and inventions made by anyone the university makes parties to the funding agreement by any assignment, substitution of parties, or subcontract of any type. It’s all the more clear that this expansion is the case if the university in its written materials insists that Bayh-Dole requires the university to create projects of this scope, tying research discovery to commercial product development.
Any patentable invention made in the commercial development of a subject invention by employees of a company that has been made a party to the funding agreement, when owned by a party to that funding agreement, is also a subject invention.
Let that bit sink in.