Some folks who claim to advocate “for” Bayh-Dole, but actually are the law’s worst enemies, argue using the line “Bayh-Dole was never intended . . . .” They leave out the part about who they mean has been doing the intending. Norman Latker? Bob Dole? Joe Allen? A bunch of university IP managers who have never read or grasped the law? Their often poorly trained in IP lawyer buddies? One would think that the intention that counts is that of Congress–Bayh-Dole is a statute, after all.
When the Supreme Court was asked to interpret Bayh-Dole in Stanford v Roche, the court set aside all those university attorneys’ opinions of what they thought “was intended”–and even set aside Senator Bayh’s argument about what he “intended.” Instead, the justices looked at the common meaning of words that carried no special definition, looked at the words that Congress had gone out of its way to define, such as “subject invention,” and then looked for Congressional intention. They did not need Sen. Bayh telling them what to think about the law. The court concluded that if Congress had intended such a sweeping change in the law as the university folks and Sen. Bayh claimed, then Congress would have gone out of its way to make that clear. Congress would have marked its intention to make a change:
It would be noteworthy enough for Congress to supplant one of the fundamental precepts of patent law and deprive inventors of rights in their own inventions. To do so under such unusual terms would be truly surprising. We are confident that if Congress had intended such a sea change in intellectual property rights it would have said so clearly—not obliquely through an ambiguous definition of “subject invention” and an idiosyncratic use of the word “retain.”
It’s clear that what the Bayh-Dole enemies mean when they assert “was never intended” is something very different from what Congress intended. They mean–though they don’t say this–that some unnamed someone never intended for the intention behind the law to be that of Congress. Senators Bayh and Dole, say, or Norman Latker, or Howard Bremer. These folks, apparently, never intended Bayh-Dole to reflect the intention of Congress, but rather, apparently, their personal intention, not expressed in the words of the law, but rather in regulatory rewriting of the law, statements they made later about the law, claims they made about the history of the law. In short, the enemies of Bayh-Dole argue that Bayh-Dole should be a law of personal caprice–a coterie of unelected officials should dictate to the executive branch and to the courts what they want the law to mean. “It was never intended that the law we claim to support should be read in the manner of standard federal law,” but rather Bayh-Dole must be interpreted according to the statements later of what people involved in drafting the law claim that they thought they were doing.
Well, poop on that, and on them. Let’s consider how all this plays out with a simple approach to a fundamental policy issue. I’ll point out that I am going to be coarsely simple, and that won’t sit well with those readers who demand nuance and qualification and sophistication. There’s a time for that, of course, but there’s also value in being, well, coarsely simple.
Here we go.
Coarsely simple. Bayh-Dole creates for federal contractors a duty to provide all information in the possession of any contractor necessary for the government to practice or file patent applications on any subject invention.
In Bayh-Dole, there’s a requirement for a government license 35 USC 202(c)(4). When a contractor who has acquired title to an invention arising in federally supported work decides to keep that title, the contractor must grant to the federal government a non-exclusive license to practice and have practiced the subject invention. “Practice” in executive branch patent policy has a long history of meaning “to make, use, and sell.” That is, the government has the contractor’s permission to practice all the substantial rights in the invention. That license requirement is so important that Bayh-Dole 35 USC 210(c) requires that the provision be included in all funding agreements, not just the ones for nonprofits and small businesses.
This license to practice is entirely consistent with a fundamental policy requirement of Bayh-Dole, at 35 USC 200, that the patent system be used to promote the utilization of such inventions. As it were, Bayh-Dole divides up the responsibilities for that use between the contractor and the federal government. As a result of the government license, both the contractor and government have a policy mandate to use the inventions that arise in federally supported work. Call it a “working requirement.” Many countries have working requirements in their patent laws. The United States does not, but for this introduction of a working requirement for inventions arising in federally supported work.
Now, once we have split up the responsibility to work the invention, there’s an asymmetry: the contractor knows everything about the invention and the federal government knows nothing. A license to the invention means little if the government doesn’t know what it has permission to. There’s a way to read all this to mean: “If you, government, stumble upon the things that we say are an invention, then we agree not to sue you for infringement or bring a claim for compensation in the Court of Federal Claims.” Nice, but pretty empty. We might then ask, is this what Congress intended?
Consider a variation that NIST is actively considering: “If you, government, use our invention in your own research and non-commercial activities, then we agree not to sue you for infringement and stuff.” This variation suffers from the problem of what the definition of “practice” must be. It’s as if Congress adopted the very words of the Kennedy executive branch patent policy for Bayh-Dole but secretly intended a different meaning for those words, so that “practice” can be made to mean anything that NIST decides now to say it means, like “glory” means a “nice, knock-down argument” to Humpty Dumpty. You might think about the Supreme Court’s reasoning with regard to the word “retain.”
Bayh-Dole goes out of its way to require a contractor to inform the government about the inventions that it has acquired, 35 USC 202(c)(1):
that the contractor disclose each subject invention to the Federal agency within a reasonable time after it becomes known to contractor personnel responsible for the administration of patent matters.
This requirement directly echoes the wording of the primary exchange in Bayh-Dole, 35 USC 202(a):
Each nonprofit organization or small business firm may, within a reasonable time after disclosure as required by paragraph (c)(1) of this section, elect to retain title to any subject invention
Disclosure of the invention is the one thing called out in the law as a precondition to choosing to retain title to an invention that a contractor has previously acquired title to. You’d think then that Congress intended that contractors disclose the inventions that they acquire. That part surely is coarsely simple enough.
Again, we might read “disclosure” to mean “notice” or “report,” as in “Hey, government, we acquired an invention, and now you know and we’re all good, bye.” But would that reading fairly reflect the intention of Congress? Why require disclosure of a piddling bit of information–an invention exists! it’s ours, not yours!–and also require a license to make, use, and sell, and authorize others to do so as well on behalf of the government? You’d think, if you think coarsely and simply about it, that Congress intended that the disclosure of the invention would provide the government with all the information that the contractor had that would be needed for the government to practice the invention.
The disclosure requirement and the government license are both anchored in the idea of an invention, and Bayh-Dole goes out of its way to define invention, and this is something new to federal patent law, as there’s otherwise no statutory definition in patent law of “invention,” odd as that may seem. But here’s Bayh-Dole, burbling as it comes, defining invention:
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.).
There is a lot that’s stupid about this definition–we are being coarsely simple here, so that’s how it is. First, the definition just repeats itself–an “invention is any invention”–oh, that’s wonderful! Wait, there’s more–an “invention is any invention or discovery.” Well, then invention and discovery are distinct things, and an invention can be an invention or it can be something that’s not an invention, a discovery. Nice, but why? Federal patent law ignores terms for things and talks about that which is eligible for a patent. But the definition isn’t done. It’s “an invention is any invention or discovery or novel variety of plant.” Now we are into “an invention is an invention, a not-an-invention of kind one, or a not-an-invention of kind 2.” And we are still not done. For each of these invention candidates, we have a qualifier: “which is or may be patentable” for inventions and discoveries, and “which is or may be protectable” for plant varieties. (Some plants are patentable subject matter, and others aren’t–but those others may have “protection” under the Plant Variety Protection Act).
Well, this qualifier wording is strange. For something to meet the “is” patentable, the Patent Office has to agree. It’s like an ump’s call in baseball. For someone to be “safe,” an ump has to call them safe. If someone “may be safe,” then an ump hasn’t yet made any call. (We skip the complications of video review, oh please.) An invention “is patentable” when the Patent Office says it is–meaning after a patent application has been filed, reviewed, and found to meet the requirements for obtaining a patent. Now about the “may be patentable.” This bit is much more loose. Who determines what “may be patentable”? Is that the inventor thinking “this may be patentable!” or is it a person “responsible for the administration of patent matters” such as a patent attorney or whomever else that a contractor might designate?
Obviously, Congress didn’t think it was necessary to elaborate. Something that “may be patentable” also necessarily “may not be patentable.” “May” encompasses both “has the standing to be” and “there’s an argument it ought to be.” Congress didn’t see the need to sort this point out. Nor has anyone writing the Bayh-Dole regulations. The definition of invention gets repeated at 37 CFR 401.2(c) and again at 37 CFR 401.14(a)(1), with only minor alterations to reflect that the regulations are not in Chapter 18 of Title 35.
Let’s be coarsely simple then. If it is plausible that any reasonable person being reasonable would think that an invention might be patentable, then that invention comes within Bayh-Dole’s definition of “invention.” Really, the “definition” here is not really a definition–it’s a statement of scope. Anything for which a patent might be obtained or a plant variety certificate issued is subject to the requirement for disclosure, a contractor’s option to retain, if a contractor has gone to the bother to acquire, and a government license.
Now we have go slowly to keep it simple. Bayh-Dole is directed at these “inventions.” It is broader than patent law’s interest in inventions that “are” patentable. There’s that “may be” for Bayh-Dole that’s not elsewhere. If an invention “may be” patentable, then it’s just as subject to Bayh-Dole’s patent rights clause as an invention that “is” patentable. Put another way, such inventions are subject to Bayh-Dole whether a contractor files a patent application or doesn’t, and are subject to Bayh-Dole until the Patent Office rules, finally, that a given invention is not patentable. In such an instance, to compound the utter strangeness that is Bayh-Dole, such a, a, er, a thing is no longer, by definition, an “invention.” An invention is or may be patentable. When the Patent Office rules some thing is not patentable, then that thing can no longer be “is or may be patentable.” And thus, cannot be, for Bayh-Dole, an invention and therefore not a subject invention.
The upshot of all this is that a Bayh-Dole invention is not a thing, but a statement of scope about things. As long as it appears that a thing may be patentable, then that thing is an invention, and if a contractor acquires title to it, then it is a subject invention. Gosh, that’s simple, though a bit weird.
Now, this upshot does not depend at all on what a contractor might include in any given patent application. In a patent application, an applicant must specify the claimed invention. A patent application consists of a set of claims that set out what the boundaries of the invention are, and a specification section that teaches how to practice the invention. Here’s 35 USC 112(a) and (b):
The specification shall contain a 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 the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
For Bayh-Dole, there’s a hitch. Bayh-Dole’s definition of invention is broader than what a contractor might, in a given patent application, specify and claim. Bayh-Dole doesn’t equate an invention with what a contractor applies for and claims. It could have done this, but didn’t. Bayh-Dole could have done something along these lines: “If a contractor applies for a patent on an invention made in federally supported work, then the contractor must notify the federal agency that it has done so, and if a patent issues to the contractor, then the contractor grants a license to the federal government to practice and have practiced the patented invention.”
But that part wouldn’t work so well for Bayh-Dole, since the law also authorizes federal agencies to grant exclusive licenses–essentially to threaten to exclude U.S. citizens from practicing inventions made in federally supported work, and to set up private parties to do the threatening themselves, backed by the U.S. government to do so. To make that happen, Bayh-Dole also takes an interest in the inventions that a contractor does not choose to “retain” and does not timely file patent applications on. For such inventions, Bayh-Dole provides that the federal government can request title upon failure of the contractor to retain title or file patent applications. Thus, Bayh-Dole ends up scoped to inventions before any patent work has determined that they are patentable or not inventions at all.
Again, oddly, Bayh-Dole takes no interest in inventions made in federally supported work that a contractor does not acquire title to. Put differently, Congress shows no intent to have an interest in inventions that a contractor does not acquire title to. Folks have been stumped by this bit of Congressional direction. There’s no point in constructing a fantasy secret intent for Congress, as if Congress really did want to cover all inventions made in federally funded work, but through an oversight of drafting (ahem, Latker) didn’t manage to express that intent, even though the template for doing so was there for the copying in the NIH IPA program master agreement based on the Kennedy patent policy–and which Latker claimed was the model for Bayh-Dole.
We may argue all we want about what Norman Latker intended. Maybe he intended Bayh-Dole to be a vesting statute that disenfranchised inventors in favor of what Latker took to be (so we fantasize) a more effective owner of such inventions–universities and their non-profit patenting agents. But what Latker intended really has nothing to do with what Congress intended. Even if Sens. Bayh and Dole intended (so we continue to fantasize) that whatever Latker said he intended was what they intended the law was to do, that still does not get at what Congress intended.
Let’s return, then, to the government license. It, too, is disengaged from whatever is patentable to include what may be patentable. As long as something is not determined not to be patentable (say, by the Patent Office), then it remains within scope of Bayh-Dole. For such somethings, the government gets a license to practice and have practiced. Such somethings must be disclosed to the government. The disclosure requirement and the government license are broader than what a contractor decides to claim in any given patent application. When a contractor grants the required license, then, that license is not merely made under the contractor’s actual patent rights, but under the right of ownership–title–to the invention broadly, the invention that “may be” patentable even if the contractor has not claimed and specified all the bits of that invention in a patent application. The invention defined by Bayh-Dole is in general broader than any invention claimed by a contractor in an issued patent.
Now we can come back around to that government license. Let’s say that Congress intends the disclosure requirement to ensure that the federal government has the broadest reasonable right to practice and have practiced any subject invention, and has a reasonable opportunity to file patent applications on anything that a contractor chooses not to retain title to or not to seek a patent for. The obligation to disclose, then, would also be in general broader than what a contractor discloses in any patent application. Put another way, a contractor cannot divide up a Bayh-Dole invention into a patentable part and a trade secret part. The government gets a license to the entire invention, prior to any patenting activity.
Now, we can ask: did Congress intend that the federal government would get a license to each subject invention but not intend for the federal government to receive from each contractor a full disclosure of that invention?
It would seem reasonable that Congress intended for the federal government to get both a full disclosure consistent with its broad license to practice and have practiced. It is not then just for some claimed invention that the contractor has to describe the “making and using” so that “any person skilled in the art” can make and use the invention. The disclosure requirement is directed to the full “may be patentable” invention. And for any bit of the invention not claimed in a patent application, that bit remains “may be patentable” until there’s a determination that it is not patentable.
Now, for the simple. Congress intended the disclosure requirement to be that which is necessary for the government to practice and have practiced the invention under its license and to decide what bits should be the subject of a patent application if the contractor fails to choose to keep title or to file patent applications claiming everything that could be claimed.
Coarsely: Contractors have a duty to disclose to the government or to those government authorizes the information necessary to practice each subject invention.
This duty has nothing to do with what’s in a patent application. If Congress had intended to limit things this way, it could have done so, easily–just read the issued patent, dodos. An issued patent establishes that a patent application has been examined and found to meet the specification requirement of 35 USC 112. There would be no need for a disclosure requirement at all unless a contractor chose not to retain title.
And this disclosure obligation is on-going to the extent that new information is required for, say, regulatory approval of a new drug. If the government is to have a meaningful right to practice and have practiced, it must have the right as well to receive all the information that a contractor has that supports the contractor’s own practice of the invention–including what the contractor supplies to the FDA for regulatory approval, but also what the contractor retains as parts of the broader invention that the contractor chooses not to include in its patent applications.
When people then talk about how the patents are not the issue–that it’s the technology transfer–they make an important point. However, when they also assert that the technology transfer lies outside Bayh-Dole, they make an implicit claim that Congress went to all the trouble to construct Bayh-Dole so that the government would get an empty license–to practice, but without receiving the information from the contractor necessary to practice when the contractor has just such information.
We might say, coarsely and simply, that Congress intended that contractors will provide to the government as part of their disclosure obligation all information necessary for the government to practice and have practiced each subject invention and to determine what the government might choose to file (and therefore publish) if the government acquired the right to file patent applications on the invention. This contractor information is not just what a contractor would provide to the Patent Office in conjunction with any patent application.
Ever more simply: Each contractor has a continuing duty to transfer to the federal government the capability, as the contractor has developed it, to practice each subject invention.
We can chase down the details of how Bayh-Dole regulations deal with disclosure. We can consider how Bayh-Dole extends the scope of contractors through its definition of funding agreement to include anyone accepting an assignment, substitution of parties, or subcontract from a federal contractor–and so extends the obligation to disclose information pertaining to “practical application” of the invention to include exclusive licensees (such as university “commercialization partners”) that obtain all substantial rights in the deal. We will still come back around to the reasonable observation that Congress intended the federal government to receive from all the contractors involved the information necessary for the government to practice each subject invention–make, use, and sell.
Congress did not intend for contractors to be able to hold back information, cleverly, and so provide the federal government with an empty, mostly useless license to practice something that the contractor knows how to practice but declines not to tell the government, even when the government has provided financial support to the work, is even as it were an investor in that work, a co-owner of that work.
To someone who insists that a contractor can hold trade secrets in how to practice an invention–how to “practicably apply” that invention–and not disclose exactly such information to the federal government, we then might ask: show how it is reasonable that Congress would intend such a result. If Congress really did intend to cock itself up like this, why don’t we find anything in Bayh-Dole to indicate that Congress intended that the required license may well be useless, and that the government would have to fund all over again the research to ascertain how to practice an invention that the government already funded a share of the work to produce. No, it just doesn’t work out. We might believe that Congress does dumb things, but when it comes to establishing what Congress intended, we don’t start with the idea that Congress intended something it didn’t enact, or that it enacted something that left it with nothing useful despite any number of indications that Congress expected to acquire something actually useful to the government.