I found a passage quoted from an article from 1992 in Wisconsin Law Review–“Faculty Generated Inventions: Who Owns the Golden Egg?” by Pat K. Chew, a distinguished law professor now at the University of Pittsburgh. Chew describes how Bayh-Dole is “intended” to operate:
Bayh-Dole is intended to work in the following way: the government agency sponsors research conducted by the faculty with the university acting as the contractor. If faculty develop an invention arising from the research, they follow the disclosure provisions outlined under the law. The university can then elect title to the invention and work with the faculty members to apply for a patent and to market the invention. The government receives a nonexclusive, nontransferable, irrevocable, paid-up license (shop-right) to the invention.
Just about everything here is twistedly wrong, but overall, the passage does track what university officials–and even Senator Bayh, in his amicus brief in Stanford v Roche–have claimed for the law. In that sense, the passage is accurate. Even though Stanford v Roche shows that this view of the law is not correct, it persists. Worse, people still openly use the tactic of claiming special knowledge of the intention for the law. Let’s work through this passage, as it gives us an opportunity to understand the law as it is, without any claims of special knowledge with regard to intentions.
Bayh-Dole is intended to work
The opening premise is strange. Who is it who intends Bayh-Dole to work in some way? Why the passive voice to hide the responsible agent? Consider an alternative opening: “Here is how Bayh-Dole works.” That would be direct. That’s apparently what the article is to get at (and as it turns out, gets wrong). But here we start with an unattached intention about the law as the guiding force for interpretation. We might posit that university administrators and law professors think this is how Bayh-Dole operates, and we might be led to believe that the law should be what they think, regardless of how the law reads.
What is the burden of “intended”? It would appear to mean, “Bayh-Dole is intended to work this way, but in practice it may work some other way.” That would make sense as a gambit in an article discussing the legal issues around the ownership of “faculty generated” inventions–even the title implies that faculty merely “generate” inventions but don’t necessarily own those inventions. Of course, if one opened with “Here is how Bayh-Dole works” and it turns out that no one works Bayh-Dole that way, then we are left with either a silly commentator or the allegation that there is rampant disregard for the law–or perhaps both.
Consider, by contrast, if we started with a title of “Faculty-owned inventions: Do universities misrepresent federal law to take ownership of the golden eggs?” We would now ask whether it was “intended” that university officials could use federal law to take inventions from inventors. We could then write, “university officials insist Bayh-Dole works this way” or “university officials insist Bayh-Dole is intended to work this way and so they argue that all interpretations of Bayh-Dole must support their claim for intention.” But why are we even talking about unattached intention with regard to how Bayh-Dole works?
Even discussing the intention of Congress leads us into a strange fiction in which assemblies can have intentions though they do not have hearts or minds. We should go there only if the law is provides multiple possible interpretations and the proper interpretation can be conclusively resolved by reference to legislative history. So why not establish first what the law says, and appeal to intention only where the law is uncertain? But even this only gets us to what we are constrained by disciplined interpretation to say that the law requires. We haven’t even got to what the law ought to require–that is, what would be best for the disposition of inventions arising in research or development projects receiving at least some support from the federal government.
We need not fault our author here, but claims of “intention” continue to plague discussion of Bayh-Dole. Presently we are faced with articles claiming that it was never intended for Bayh-Dole (or its march-in provisions to be used) to preclude outrageous monopoly-supported pricing for medicines discovered and developed in projects receiving federal support. We never find out who this “it” is who has done the intending. Perhaps “it” really is a horror-inducing clown. By calling forth intentions, commentators on the law asks us to ignore the law and instead start with an assertion about the intention of the law. Apparently once we accept this extra-legal intention, then we can pick and choose what we want from the law and ignore the rest, filling in any ambiguities, contradictions, and gaps with whatever sounds good in light of the “intentions.”
the government agency sponsors research conducted by the faculty with the university acting as the contractor
In Bayh-Dole, “contractor” is defined as “any person, small business firm, or nonprofit organization that is party to a funding agreement” (see 35 USC 201(c)). The university does not “act as the contractor”–the university is one party to the funding agreement. The faculty member designs the research and chooses who should be asked to sponsor the work. The university approves the faculty member to undertake the research and releases the faculty member from official duties to the university to do so. The federal agency then decides to sponsor the research and the university provides administrative services.
Even the use by Bayh-Dole of “contractor” as a term of convenience creates a misleading cognitive anchor in the law. For procurement arrangements, “contractor” makes perfect sense. But Bayh-Dole is targeted at nonprofits–and the primary form of funding agreement there is the grant, a subvention rather than a procurement, a conditional gift as it were, not a contract.
In the implementing regulations, we are told that federal agencies may replace italicized words with appropriate terms (37 CFR 401.5):
Agencies may replace the italicized words and phrases in the clause at §401.14 with those appropriate to the particular funding agreement. For example, “contractor” could be replaced by “grantee.” Depending on its use, “agency” or “Federal agency” can be replaced either by the identification of the agency or by the specification of the particular office or official within the agency.
Thus, we could have just as easily written, in the context of faculty-originated research, that universities act as “grantees.” Doing so takes a lot of wind out of reasoning from “contractor” to “contract” to “employee” to “employer owning inventions generated by employees under federal contract.” Instead we have “grantee” to “grant” to “faculty beneficiary” to “grantee support for inventions made by faculty beneficiaries of federal subventions.” In a sense, replacing abstract terms with terms appropriate to particular situations, such as when faculty are receiving federal grants for the work they propose to do is how Bayh-Dole works–there it is, in the implementing regulations.
What’s left out by the account of how Bayh-Dole is intended to work is whether faculty and other participants in the research are also parties to the funding agreement. Bayh-Dole does not require faculty to be parties to the funding agreement. Nothing in 35 USC 202(c), where the provisions for the standard patent rights clause are laid out, requires any inventors to be made parties to the funding agreement. All we have is 35 USC 202(d), which conditionally permits inventors to retain ownership of their inventions:
If a contractor does not elect to retain title to a subject invention in cases subject to this section, the Federal agency may consider and after consultation with the contractor grant requests for retention of rights by the inventor subject to the provisions of this Act and regulations promulgated hereunder.
After Stanford v Roche we must read this provision as follows: If a grantee has not received assignment of an invention made in the performance of work within the scope of a funding agreement, and having not acquired ownership cannot then elect to retain ownership, then the federal agency may consider and after consultation with the grantee grant requests for the inventor to retain ownership, subject to Bayh-Dole and its regulations.
Or, bluntly, a federal agency does not have to demand ownership of subject inventions owned by inventors, but before releasing any claim the agency may have, the agency must check with the grantee to make sure that the grantee had notice of the invention and indeed did decline to acquire ownership.
And for this even to work, somehow the invention has to become an invention “of the grantee” or “of a party to the funding agreement” for it to be a subject invention. The first part cannot operate. If an invention becomes a subject invention because the grantee has acquired it, then the inventor cannot have the benefit of “retention of rights”–the inventor would have no rights to retain, having assigned those rights to the grantee, who then decides it doesn’t want the rights.
The Supreme Court in Stanford v Roche made it clear that “retention” in 35 USC 202(d) implied that inventors had not given up their rights–they could retain those rights because they already had those rights:
By using the word “retention,” §202(d) assumes that the inventor had rights in the subject invention at some point, undermining the notion that the Act automatically vests title to federally funded inventions in federal contractors.
And in a footnote:
But there may be situations where an inventor, by the terms of an assignment, has subsidiary rights in an invention to which a contractor has title, as §202(d) suggests. Compare §202(d) (“retention of rights”) with §202(a) (“retain title”) (emphasis added).
The idea then becomes that 202(d) concerns not title–which may pass from inventor to grantee by assignment–but rights that remain with then inventor, such as rights that make the assignment conditional. “If the university decides not to file a patent application or pursue licensing of the invention, then it will return title in the invention to the inventor.” In such a case, the invention becomes a subject invention upon assignment to the grantee, but the grantee, if it chooses not to retain title (for whatever reason), is not constrained to “offer” title to the federal government, nor is a federal agency required to request title from the grantee. Instead, if the grantee or inventor requests retention of rights, then the federal agency may grant the request. It’s a strange but reasonable result. It’s not that anyone requires the law to be competently drafted–just that there is a reasonable interpretation available–and that reasonable interpretation takes precedent over any interpretation, no matter how much someone wants it, that does not account for all the words of the law. It just isn’t good interpretative practice to dismiss words as meaningless, or as mere synonyms thrown about for stylistic effect, or surplusage repeating what is already understood rather than qualifying.
(Here is an example of surplusage in the standard patent rights clause (f)(2): “each subject invention made under contract.” There can be no subject invention that’s not made under contract; the definition of subject invention requires the invention to be made under contract. Clearly here the drafters of the standard patent rights clause did not choose to understand the definition of subject invention–they may have intended something else, but the law they drafted doesn’t support their private intention for the meaning of the law).
The second part is more straightforward. Inventors are made parties to the funding agreement, and thus are themselves also grantees with responsibilities directed at the disposition of inventions. Then 35 USC 202(d) works–inventors can retain their rights, including title, and federal agencies must consult, as a courtesy, with any other grantees to make sure that everyone is on the same page with regard to any private obligations that may be involved–employers with employees, say, or one grantee in its dealings with another grantee (grantees now referring to persons as well as organizations).
Bayh-Dole’s standard patent rights clause at 37 CFR 401.14(f)(2) requires organizations to require their employees to make written agreements–the requirement is in a section headed “Contractor action to protect the government’s interest.” The written agreements that the organization must require are a direct consequence of the patent rights clause in the funding agreement. In effect, the written agreement is a substitution of parties–the employees act in place of the organization. They are the “persons” included in the list of parties to the funding agreement.
One might even go so far as to assert that the (f)(2) requirement of a written agreement to protect the federal government’s interest in subject inventions has the force of federal law, and even if a grantee organization does not expressly comply with the requirement to require written agreements, the patent rights clause itself, as a thing of federal regulation, does so anyway–it does not directly compel inventors to agree, but rather compels the grantee to compel the inventors to make the written agreement. The grantee–the university–must act to release and delegate responsibilities to its employee-inventors so that those inventors become parties to the funding agreement and their inventions therefore, all other conditions met, become subject inventions.
The definition of funding agreement makes it clear that there can be multiple parties. The second sentence of the definition: “Such term includes any assignment, substitution of parties, or subcontract of any type entered into for the performance of experimental, developmental, or research work under a funding agreement as herein defined.” An initial party to a funding agreement may make others parties to the funding agreement by any of the actions identified–assignment, substitution, or subcontract.
We might press it further. Bayh-Dole’s written agreement requirement extends only to employees involved in a given research project. Employees are people who act under the direction of a master. University faculty, for research, unless they expressly give up their freedom of research and publication, are not employees within the meaning of Bayh-Dole. They are not assigned to conduct the proposed research–they are rather released from assignments to do so. The university has no right to direct or review or approve their research. Faculty are not employees for their research activities, no matter the legal fictions that are maintained to handle the bookkeeping. If a university faculty member has been hired to invent–then that agreement has to be express and override university policy commitments to faculty freedom of research and be the clear intent of both university and faculty member.
What have we then? Bayh-Dole’s patent rights clause requires organizations to make employees parties to the funding agreement. Those employees, then, are also “contractors” or “grantees.” They have a prior right over others, unless they have agreed otherwise, to inventions that they “generate” under a federal funding agreement. They have the right to retain title to such inventions upon disclosure and notice, as provided by 35 USC 202(a). And they may have rights if their assignment of a subject invention to another party is conditional, such as on the condition that the new assignee will file a patent application or use the invention or license the invention. That much is “intended.” It’s a tough road to argue that the words are in the law or the patent rights clause but “that’s not what was intended.” A pretend intention is nothing but a device to deceive. All the worse when the object of interpretation is law.
The university, then, acts as a contractor or grantee–as a party to the funding agreement–but not the only contractor or grantee possible–and if a university complies with the patent rights clause in the funding agreement, every employee is also a party to the funding agreement–a contractor, a grantee. If a university refuses to comply with the patent rights clause and does not make its employees parties to the funding agreement, then it is in breach of its obligations as a contractor or grantee. Everything else that follows, then, that the university might claim as a right would be challengeable as invalid or perhaps quasi-magically the inventor is joined as a party to the funding agreement by operation of federal regulation.