The only law that Bayh-Dole does not expressly supersede is the Stevenson-Wydler Act. It’s worth comparing language in the two laws, as Bayh-Dole draws from Stevenson-Wydler, but then works its own magic. Consider the definition cascade on inventions.
(7) ‘‘Invention’’ means any invention or discovery which is or may be patentable or otherwise protected under title 35, United States Code, or any novel variety of plant which is or may be protectable under the Plant Variety Protection Act (7 U.S.C. 2321 et seq.).
Here we have the same extension of invention to include plant varieties that Bayh-Dole makes, but Bayh-Dole does so as part of federal patent law. That is, Stevenson-Wydler refers to patent law for part of its scope, but Bayh-Dole amends federal patent law. Absent in Stevenson-Wydler are two key elements of the definition as used in Bayh-Dole. There is no “of the contractor” to force the definition to operate only when a contractor has acquired ownership of a given invention. And there is no “in the performance of work” requirement that opens the definition up to issues pertaining to the scope of the “performance” that a given contract supports at least in part.
(8) ‘‘Made’’ when used in conjunction with any invention means the conception or first actual reduction to practice of such invention.
The definition of “made” is directed at the earliest date upon which the invention might be claimed and has nothing to do with whether the making was done under contract. Further, there’s nothing additional about the PPVA. Under Stevenson-Wydler, one has to match conception and first actual reduction to practice to the PPVA.
Bayh-Dole, however, uses two definitions–invention and subject invention:
(d) The term “invention” means any invention or discovery which is or may be patentable or otherwise protectable under this title or any novel variety of plant which is or may be protectable under the Plant Variety Protection Act (7 U.S.C. 2321 et seq.).
“may be . . . otherwise protected” becomes “protectable.” “title 35” becomes “this title” (since Bayh-Dole is added to federal patent law, while Stevenson-Wydler just refers to patent law). So far, not much change. Now for Bayh-Dole’s definition of “subject invention”–an attempt to bring into patent law the definition of the IPA template agreement:
(e) The term “subject invention” means any invention of the contractor conceived or first actually reduced to practice in the performance of work under a funding agreement: Provided, That in the case of a variety of plant, the date of determination (as defined in section 41(d) of the Plant Variety Protection Act (7 U.S.C. 2401(d))) must also occur during the period of contract performance
I have highlighted the primary changes relative to Stevenson-Wydler. Rather than dealing with inventions, we get both the “made” element and a scope element. A subject invention is more restricted than just any invention, but the definition here is also expansive in ways that aren’t necessarily obvious.
On the surface, “in the performance of work under a funding agreement” suggests a delivery expectation. If an invention is made in a federal contract, then it is a deliverable, a subject invention. The idea must have been that Bayh-Dole would then intercept this deliverable to the government and re-route it as a deliverable to the organization that hosts the research. Except Bayh-Dole doesn’t actually accomplish the re-routing, as the Supreme Court decision in Stanford v Roche made clear. Hence all the fuss at universities about vesting, led even by Senator Bayh, whose argument the Supreme Court rejected.
But if we track through the definition of “funding agreement,” we start to see the expansive nature of the definition of subject invention. Here’s Bayh-Dole’s definition of “funding agreement”:
(b) The term “funding agreement” means any contract, grant, or cooperative agreement entered into between any Federal agency, other than the Tennessee Valley Authority, and any contractor for the performance of experimental, developmental, or research work funded in whole or in part by the Federal Government. 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
Two sentences. First sentence reduces (setting aside the various qualifications) to “any agreement for the performance of work funded in whole or in part by the Federal government.” The “in part” changes significantly the import of “performance of work.” Without the “in part,” we would have a simple delivery scope–“performance of work.” The funding agreement would specify the work to be performed. That would be straightforward. But that’s not what this definition does. Instead, the definition asserts that the “performance of work” may be broader than the part funded by the government. The government may fund only a part of the work to be performed but have rights to inventions made in the work generally. This claim is entirely consistent with Bayh-Dole’s general statement of policy: “inventions arising from federally supported research or development.” Not “arising in” but “arising from.” Prepositions matter.
A Bayh-Dole funding agreement does not limit its scope to work funded only and directly by the government. The breadth of the definition is: “performance of work funded in part by the Federal Government.” What’s left entirely unspecified, then, is what the full “work” might be.
But there’s more. Second sentence.
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
The definition of funding agreement includes any assignment, substitution of parties, or subcontract of any type for the performance of work under a funding agreement. We have just seen that “performance of work” can be anything that the federal government funds at least in part. Thus, “under a funding agreement as herein defined” here cannot mean “as specified by” but rather must mean “at least in part funded by the federal government.” A party to a funding agreement extends the funding agreement by assignment, substitution of parties, and subcontract. These extensions all point back to the idea that the “work” is generally greater than the part of the work supported by federal funds.
If an invention is made in the “work,” then according to the definition of subject invention, that invention becomes a subject invention if the invention is patentable and is acquired by a contractor (defined as any person, small business, or nonprofit that is a party to a funding agreement). The invention does not have to be directly funded by the federal government, does not have to be specified in a given federal funding agreement–the invention only has to be made in “work” that receives at least some support from federal funding. That’s the necessary implication of “made in the performance of work funded at least in part by the Federal Government.” The government funds at least part of the work, and therefore has a scope of interest in any invention made anywhere in that work.
It is therefore essential in dealing with Bayh-Dole to determine what constitutes the “work” for which a party contracts with the federal government. It is of the form “we are committed to do this work, and the federal government can assist by funding this part of the work, and in return, the federal government will receive an interest in any invention made anywhere in the work, including any work we assign to others or subcontract to others or any work in which others stand in for us.”
Now we can return to Bayh-Dole’s definition of subject invention:
(e) The term “subject invention” means any invention of the contractor conceived or first actually reduced to practice in the performance of work under a funding agreement
We can see now that this is an expansive definition that depends on the specification of the work, some of which is to be performed with federal funding. One thought is that the drafting of Bayh-Dole aimed for such a broad scope to provide the government with the broadest possible rights to inventions made with federal support–either directly, or because the work funded directly by the government supports other work in which inventions are made. Another thought is that the drafting of Bayh-Dole casts such a broad scope to allow the institutions that host research to enjoy a broader claim to the right to “elect to retain title” in this broader set of inventions. What the government enjoys by way of a right of delivery (of title or a non-exclusive license plus a right of oversight), the contractor hosting the research obtains as a right to prevent the delivery of title to the federal government. This common interest in a broad scope, especially if the contractor’s right is thought of as “intercepting” the federal government’s right of invention delivery, leads to the university practice of claiming subject inventions broadly (in order to “take” title to the inventions, even if university patent and research policies would provide otherwise)–it’s just that the universities often get the broad scope wrong.
Where might we look for the specification of the broader work implied by Bayh-Dole’s definitions? It could be established by a federal call for proposals–“The government intends to create a tomato picking machine and seeks applicants to develop the robotic arm.” In such a case, if a contractor comes up with an invention directed to stabilizing the wheels of the picking machine to better support the robotic arm’s picking capability, that stabilizing invention is within scope of the government’s interest.
Or the work could be set out in a proposal for research submitted to the federal government. “We intend to develop a commercially useful tomato picking machine, and we request funding from the federal government to develop the robotic arm.” What’s the point of the federal government getting only a license to robotic arm inventions when it has been asked to support in part the development of a commercially useful tomato picking machine? What will the federal government do with all its own tomatoes, if it only has rights to an arm and not also to stable wheels?
Or the work could be set out in statements and any assignments or subcontracts (of any type) made by the institutional host for the research: “although the proposed research concerns developing robotic arms for tomato picking machines, any time we identify an invention, our formal policy is that we attempt to commercialize it.” Thus, the work stipulated by the grant recipient necessarily includes not just the robotic arm but the commercial implementation of that arm undertaken by the grant recipient or by any assignee (say, of the invention) or subcontractor (as, by means of an exclusive license to prepare a commercial version of the robotic arm as part of a tomato picking machine).
A university or other research host could limit the scope of the “work” indicated by the definition of “funding agreement” by not asserting commercialization as a policy goal, not assigning a subject invention to a company for commercial develop (such as by using an exclusive license that grants all substantial rights in an invention–to make, use, and sell), or by not taking ownership of an invention made in federally supported work, and thus restricting the work supported to whatever the federal agency and the investigators have asserted is their personal intent with regard to the “work.”
Now let’s deal with one more aspect of this definition of subject invention–the conception or first actual reduction to practice part. In Stevenson-Wydler, this is the “made” definition:
(8) ‘‘Made’’ when used in conjunction with any invention means the conception or first actual reduction to practice of such invention.
The “conception” and “first actual reduction to practice” are elements used to determine, in the case of competing patent applications, which inventor made the invention first. To make an invention, in such a context is to start the inventive process (conception) and reduce that conception to practice (either by filing a patent application–constructive reduction to practice or building and testing a prototype of the invention–actual reduction to practice). If someone first conceives, but then does not follow through to reduce to practice, then they lost (under US patent law before AIA’s first-to-file) their priority as an inventor.
Here is how Stevenson-Wydler uses its definitions of invention and made, in connection with setting up NSF cooperative research centers:
(3) the host university or other nonprofit institution has a plan for the management and evaluation of the activities proposed within the particular Center, including:
(A) the agreement between the parties as to the allocation of patent rights on a nonexclusive, partially exclusive, or exclusive license basis to and [sic] inventions conceived or made under the auspices of the Center;
Two points–(1) “conceived” is redundant, given that “made” is defined to include conceived. (2) “under the auspices of the Center” is amazingly broad. An “auspice”–protection, patronage, omen. This scope is rather different from “performance of work.” What’s interesting is that all this reduces to an agreement on inventions among a Center’s participants. They decide the scope of the agreement with regard to inventions and auspices. They can agree, for instance, that they have no rights to an invention unless it was made in a project expressly approved and funded by the Center. There’s nothing here that requires anything with regard to ownership as a matter of law–just that the Center has an agreement allocating patent rights to inventions.
Any invention “made” under a contract becomes a potential deliverable–either ownership or a government license. Compare the definition of “made” in the Kennedy executive branch patent policy (1963):
(2) Made–when used in relation to any invention or discovery means the conception or first actual reduction to practice of such invention in the course of or under the contract
Here, the definition of made is expressly connected to the contract–“in the course of or under,” or “done to achieve the purposes of the contract or specified by the contract as an obligation.” Any invention that claims a priority date in the contract period becomes a deliverable.
Now look at Bayh-Dole again:
(e) The term “subject invention” means any invention of the contractor conceived or first actually reduced to practice in the performance of work under a funding agreement:
On the face of it, Bayh-Dole’s definition of subject invention appears to do just what the Kennedy patent policy does. But because “funding agreement” also has a specialized definition in Bayh-Dole, the definitions are in effect quite different. What is shared is anchoring the invention in a first-to-invent framework. Bayh-Dole makes this framework clear in its second clause of the definition:
Provided, That in the case of a variety of plant, the date of determination (as defined in section 41(d) of the Plant Variety Protection Act (7 U.S.C. 2401(d))) must also occur during the period of contract performance.
Clearly (as if anything in Bayh-Dole is actually clear), this added explanation shows that the issue is one of timing–for plant varieties, the “date of determination” decides whether a plant variety “invention” is within the scope of the definition of subject invention.
Now, section 41(d) has been reorganized in amendments to the PPVA, but here’s the definition from the original:
The term “date of determination” means the date when there has been at least tentative determination that the variety has been sexually reproduced with recognized characteristics, whether or not the novelty of those characteristics has been determined.
The definition was removed from the PPVA in 1994, but no one bothered to tell Bayh-Dole. Now PPVA is “first applicant to meet the requirements” to establish that a sexually reproduced plant variety is new, distinct, uniform, and stable. But the definition of “date of determination” provides a help in understanding that “conceived or first actually reduced to practice” is a parallel construction–it is an effort to establish a date of the invention, just as “date of determination” attempts “tentatively” to establish an earliest date for a plant variety to receive priority for protection. We might say that a subject invention is one for which there is at least a tentative determination that an invention that is or may be patentable where that determination (conception, first actual reduction to practice) takes place during the period of contract performance.
If we let this settle for a moment, we can see that the scope of the definition of subject invention is not bounded by the period of performance for the federal contract. Only for plant varieties does the definition require the date of determination to fall within the period of performance. That’s the point of the “must also” in the second clause regarding the PPVA. A subject invention is defined by the scope of the “work” to which federal funding contributes at least in part, not by the period of contract performance for the particular part of the funding agreement in which federal funds are expended. It does not matter whether “even $1 of federal funding has touched” the invention–that’s simply not relevant. What matters is whether the invention was made in a project (“work”) for which the federal government provided some funding under a funding agreement. No federal money at all may have been used to “make” the invention and yet the invention meets the definition of “subject invention” because it is patentable, owned by a party to the funding agreement (if only a party by any assignment, substitution of parties, or subcontract of any type), and made in the broader project.
This is complicated stuff–mostly as a result of the drafting of Bayh-Dole, which is truly is awful. At times I wonder if the awful drafting was a deliberate strategy to create ambiguities that could later be “clarified” with amendments that changed the meaning of passages substantially from the intent understood by those approving the original bill, such as by re-interpretation in the Code of Federal Regulations that implement the law and establish the standard patent rights clauses and procedures authorized by Bayh-Dole. That would be a reasonable strategy for folks desperate to get anything and hoping to fix the problems later, in implementation.