Before the Bayh-Dole bill came together, there were precursors. One of these is a draft bill produced by Senator Dole. It differs in a number of ways from the bill that became Bayh-Dole, and the differences provide some insight into both the eventual architecture of the final bill and the problems that were introduced by changes that weren’t thought out and weren’t followed for effects throughout the bill.
Let’s look at the Dole bill’s use of “subject inventor” and “subject invention”:
There is no “subject inventor” in Bayh-Dole, though perhaps there should be. In the Thornton draft bill (HR 8596) and in Bayh-Dole, eventually, the term used is “contractor.” A contractor becomes “any person (as defined in section 1 of title 1, United States Code) that is a party to the contract” (HR 8596) or “Any person, small business firm, or nonprofit organization that is a party to a funding agreement” (Bayh-Dole).
Though the wording in these three versions appears similar, it actually expresses substantially different concepts. The Thornton draft relies on a standard definition of person (individuals and corporate “persons” of various kinds) and therefore leaves out instrumentalities of state government that are not corporations, companies, societies, and the like. Some universities, say, might be left out because they are chartered under state law and not organized as corporations or the like. Bayh-Dole deals with this by defining both “small business firm” and “nonprofit organization” and adding these to the standard 1 USC 1 definition of person:
In determining the meaning of any Act of Congress, unless the context indicates otherwise— . . .
the words “person” and “whoever” include corporations, companies, associations, firms, partnerships, societies, and joint stock companies, as well as individuals;
So, “person” means individual or company. Federal agencies almost never fund individuals to conduct research–there’s a story there, too. In practice, the core meaning of “contractor” is “company.” One has to stretch to get to individuals, because individuals just don’t contract with the federal government for research or development.
By contrast, the Dole draft uses “subject inventor” and defines this subject inventor as any person (there’s that 1 USC 1 definition again) that is a party to a funding agreement. In a technical sense, we arrive at the same definitional place. But the Dole draft makes a huge difference for inventions. Why? Because when an invention is invented, the inventor is always an individual. “Subject inventor” captures this fact and keeps it present throughout the bill. When we turn to “subject invention,” this use of “subject inventor” then makes perfect sense–a subject invention is an invention of a subject inventor. Where the subject inventor is an individual, the subject invention is an “invention of an inventor.” This works. This is federal patent law. An inventor owns what the inventor has invented.
Substituting “contractor” for “subject inventor,” though, makes a mess of things–a subject invention is now an “invention of the contractor” not an “invention of the subject inventor.” And there’s where the Supreme Court in Stanford v Roche made clear that “of the contractor” means an invention owned by the contractor. If the contractor is an organization, then the contractor has to have gone out and acquired the invention–the invention does not just fall into an organization’s lap, as it were.
The use of “subject inventor” then provides context for the definition of “practical application.” Here, look:
This definition differs from Bayh-Dole’s definition. The first part is the same, but the last part adds “on reasonable terms from the subject inventor or licensee or assignee of the subject inventor.” Again, if we start with the idea that inventors own their inventions until they barter that ownership away, then here in the definition of practical application we see that it is the subject inventor (any party to a funding agreement, but starting with a presumption of an individual being involved) that must provide the benefits of the invention to the public on reasonable terms, and if not the subject inventor, then some other that isn’t a subject inventor–a licensee or assignee. Bayh-Dole eliminates this reference to subject inventor and doesn’t substitute contractor. Practical application in Bayh-Dole is restricted to one form of march-in–35 USC 203(a)(1). Otherwise, utilization is used, oddly, where practical application could have been used, but isn’t.
The obvious issue, then, that the Dole usage raises is why would a law–an amendment to patent law, no less–that depends on inventors owning their inventions then go on to ignore making inventors parties to the funding agreement?
You see the deal–there’s no point in using “subject inventor” without making inventors parties to the funding agreement and the inventions they make subject to the provisions of the law. But the Dole bill nowhere makes inventors parties to the funding agreement or specifies that federal agencies should contract directly with individuals. No matter how one cuts it, there is a fundamental gap between the organizations that federal agencies contract with, and which cannot be inventors “themselves,” and the individuals who have those sparks of insight that cannot be specified in advance, and who own the right to seek patents on their inventions unless they barter that right away.
In one sense, then, it doesn’t much matter whether the law uses “subject inventor” or “contractor.” Either way, individuals are not part of the law. They invent in the usual way, and own their inventions in the usual way, and may contract to assign or license their rights in the usual way.
It might seem obvious that substituting contractor for subject inventor doesn’t change anything, especially if the subject of the definition is just a token and could be anything, really–“bacon” or “flatulence.” “Flatulence means any party to a funding agreement.” Sure, why not? But subject inventor does make a difference, because the context makes clear that the Dole bill intends to start with inventors, and inventors must be individuals. There is no corporate inventor concept in federal patent law. Any corporation, company, and the like that wants to get patent rights to an invention must do so by obtaining an assignment of the desired invention. A corporation does not invent. Inventions are not made by corporations or by universities. Mere employment by a corporation or university does not turn an inventor’s invention into a corporate or university invention. This is fundamental stuff.
The problem for the Dole bill, as for Bayh-Dole, is that federal patent law does not provide an easy way to force inventors to use the patent system. There’s nothing in the Constitution that provides Congress with the power to force inventors to accept monopolies on their inventions, let alone force inventors to accept those monopolies and then force them to assign those monopolies to the organizations that the federal government has chosen to host the inventive work. But this is what Bayh-Dole purports to do, and what university administrators want Bayh-Dole to do, and what federal officials at NIST, NASA, and the NIH, at least, want Bayh-Dole to do. And this is what the Supreme Court expressly held that Bayh-Dole does not do, and would be “deeply troubling” if it did–that is, if it did these things, the law would be unconstitutional. By extension, then, if federal agencies do these things anyway, their actions are unconstitutional.
Thus, there is this gap between the inventors and federal contracting with organizations. Latker tried to backfill the gap in the patent rights clause by including a “written agreement” requirement that itself has no authority in Bayh-Dole. The written agreement requirement in effect requires organizations to flow down–subcontract–as it were, with employees, to make them parties to the funding agreement, so that their inventions come within scope of Bayh-Dole without requiring the ownership of those inventions to vest with organizations or force them to assign their inventions to organizations. But even then, those inventors-now-parties-now contractors-now subject inventors have their own standing under the law to elect to retain title to their inventions.
Really, there’s no way for federal law to force inventors to use the patent system, or to force inventors to give up their inventions to organizations that federal agencies have chosen to carry out various research or development tasks. State law apparently may do such things–see Ohio, for instance. But not federal law.
The Federal Procurement Regulation–which Latker also worked on–handled things somewhat differently. In the standard clause, each contractor is required to agree to assign inventions made under contract to the government unless the government agrees to allow rights to remain with the contractor (1-9.107-5(b)(1)) or with the inventor (b((2)). The contractor then is also required to obtain patent agreements from employees to “effectuate the provisions of this clause” ((e)(3)). Thus, in this federal contracting instrument codifying the Nixon patent policy (and not amending federal patent law), contractors must have agreements with inventors under which the inventors assign to the contractor or assign to the government, unless the government lets the contractor or inventor retain rights.
See the logic? In the FPR, contractors must make inventors agree to assign to the contractor only so that the contractor may assign to the federal government–it’s not so that the contractor can obtain rights from the inventor for the contractor’s own exploitation. That relationship–whether a contractor wants invention rights or not–is outside the FPR, and following the Supreme Court decision in Stanford v Roche, outside Bayh-Dole as well.
The written agreement requirement in Bayh-Dole sets about to do something similar, but fails. There’s no authority for it in Bayh-Dole, and as statute Bayh-Dole controls. The written agreement requirement shows up as more executive branch fuss, but now without any basis in an executive order. Latker just makes it up and no one apparently notices.
“Subject inventor” would work in Bayh-Dole if federal agencies contracted with individuals or if Bayh-Dole required contractors to make employees parties to the funding agreement or required inventors to assign to the federal government on request. For that, a “subject invention” would be any invention made within the scope of a funding agreement–initial ownership of the invention and any subsequent assignment of the invention would not take an invention out of scope. But Bayh-Dole does none of these things.
Dole’s bill, with subject inventor, makes more sense, even if it, too, fails to connect inventors with funding agreements.