During the kerfluffle known as Stanford v Roche, one of the big advocates for Bayh-Dole as a vesting statute was MIT. The MIT amicus brief is here. It’s in this amicus brief that the idea that a present assignment trumping a promise to assign gets big play. Perhaps this is the source of the present circulation of the “you must use present assignment language” advice. The brief gets a lot of things wrong. Bayh-Dole does not provide a “statutorily prescribed right” to the government for federally supported inventions–rather, it requires agencies to obtain this right as part of the bargain in any funding agreement, unless exceptional circumstances arise, in which case it can get what it needs. Bayh-Dole does not vest title to inventions in universities, but rather stipulates restrictions to be placed on universities if they do obtain title to federally supported inventions–and those restrictions are greater than on for-profit contractors and on inventors acting independent of their university employers.
The MIT brief argues:
Section 202(d), and indeed the entire Act, are premised on universities and other contractors obtaining title to subject inventions from researchers and others who perform the underlying research. Otherwise Bayh-Dole makes no sense in giving a university the right to “elect to retain title to any subject invention[,]” id. § 202(a), or, conversely, the right “not [to] elect to retain title to a subject invention in cases subject to” the Act. Id. § 202(d).
You have to appreciate the logic of these two sentences, written as advice to the Supreme Court. Bayh-Dole has to mean what we says it means, otherwise it makes no sense. That is, we denounce any other reading of the Act, because we have to be right. There is no demonstration that any other reading makes no sense. Clearly, there were other readings, and better ones, and sensible ones. The problem is that in allowing a university the right to “retain” title, the standard patent rights clause does not “give” a right to that title. The “right” a university obtains, if there is one, is not to give up title easily if it obtains title. The “right” being claimed by the university is merely the release from a previously expected obligation. It’s amazing how difficult it is for some folks to recognize freedom, and when they find it, it is to be stamped out.
The MIT reading of Bayh-Dole then turns the prohibition on assignment of rights in 37 CFR401.14(a)(k)(1) into a prohibition on inventors rather than a conditional limitation on assignment on nonprofit organizations: “That prohibition necessarily assumes that the university in the ordinary course obtains and retains title.” (k) concerns only nonprofits, the approval is required only if the assignee does not have invention management as a primary function, and (k) makes perfect sense in the context of not trusting a university, once it has got title, to do the right thing with it. (Keep in mind, (k) is replacing agency review of each university request to manage an invention–and so these are terms agencies would think to impose on universities anyway–and wouldn’t necessarily impose on, say, small businesses.) (k) does assume that a nonprofit has got title. It doesn’t assume that a nonprofit has to obtain title. Of course, for (k) to “make sense”, a nonprofit has to have title, but nothing requires (k) to operate. (k) is not the essence of the Act.
The brief then argues that it makes no sense for the Act to allow inventors to retain title “subject to the provisions of this Act”:
If the Act assumed that the inventor had title to subject inventions, then the restriction on the inventor’s exercise of retained rights would be a non sequitur; an inventor who had title in the first instance would be able to exercise those rights free and clear of any obligation to comply with Bayh-Dole or implementing regulations. But that is not what the Act provides.
Here we encounter a strategy of interpretation that goes something like this: if we only read part of the text, then to make sense of it we have to misread it a bit, and then we assert that any other readings don’t make logical sense. The MIT misreading starts by not recognizing the role of the 37 CFR 401.14(a)(f)(2) agreement that MIT and other universities agree to require of their employees. That agreement is necessary to ensure that inventors, with their personally held patent rights, owe to the government whatever rights the government has bargained for in the funding agreement, even if the contractor does not seek to obtain title or royally screws up. Further, Bayh-Dole provides in 35 USC 202(d) that an agency can allow the inventors to keep their title–reflected in the regulation at 37 CFR 401.9. The fundamental misreading by MIT, then, is that they don’t understand Bayh-Dole at its most basic, overt level. The Act is a law pertaining to federal agencies. It places requirements on their funding agreements to normalize these agreements with regard to inventions, and to normalize how the agencies may vary from the normals. The funding agreements use these normals. Universities agree to the terms of the funding agreements, however these are deployed (typically via 2 CFR 215 or the FARs). Under these funding agreements, universities–not the law, not federal agencies–require those participating in research to make commitments to the government with regard to establishing the government’s interest (not the university’s interest) in inventions.
MIT reads the Act as applying to universities. It reads the standard patent rights clause in 37 CFR 401.14(a) as *the* Act. Thus, these things required of agencies, assembled by the Department of Commerce into standard clauses, become instead law, not regulation, not federal contract, not practice with the possibility of breach or waiver. Thus, they ignore (f)(2) as if it does not require actual performance by contractors. They ignore 37 CFR 401.9 because it’s never supposed to happen, or they cannot make sense of it.
Thus, it is not just an isolated misreading–it is a complex of misreadings. “Electing to retain title” becomes “electing title”; “retaining title” demands that the university already must have got title, and since the Act assumes a university has title, it must have automagically somewhere already vested that title. The Act becomes a federal mandate to strip rights. It becomes the way to overcome university policies that left invention rights with inventors except when otherwise required by the terms of sponsored research. Here was sponsored research with a vengeance, amped up with the force of federal law to require university ownership and therefore supersede the wishy-washy gestures to freedom, academic governance, and expectations of faculty public service. Here, with a convenient misreading of Bayh-Dole, folks could, for a key bit of the results of research, assert that the state steps in and says, “enough is enough,” and hands invention rights to university administrators.
The writers of this brief don’t give evidence of understanding that the Act is premised on federal agencies not treating each invention made with federal support as a service invention owed to the government. Inventions under Bayh-Dole become free inventions with regard to the government. Inventors have the opportunity to present the inventions to whomever they think may do the best to develop the invention, including the government. But universities are having none of it. They turn the opportunity for a university to hold title (since the federal agency will not contest that ownership as a matter of contract deliverables) as a mandate to hold title. Clearly, it is nothing of the sort.
In my discussions with folks who want Bayh-Dole to be a vesting statute, or at least an inventor’s rights stripping statute, it’s clear that they don’t want to work through the text of the Act. They don’t want to reason it. They want to argue a point–vesting–and whatever argument they pose is good if it makes anyone with a contrary argument stand down. They are not reasoning, they are bullying. Pointing out the flaws in their argument does not send them back to school to figure things out, but rather causes them to double down to assert their priority. When you are at a school with the reputation such as MIT has, it’s easy to trade on that reputation to beat down anyone one wants. This, even more than the misreading of Bayh-Dole, concerns me. The manner of the discussion is one of politics, not reason. And the politics is about institutional control of invention assets by any available means–federal vesting is perfect. It is a totalitarian mindset. It should give one pause to think that across scores of major research universities, the technology transfer programs are given over to folks fully committed to totalitarian state control of research inventions while avoiding accountability for their own actions by not reporting publicly the actual data for each invention that they do come to own.
One might think, this is just a misunderstanding regarding a complicated set of regulations. But it’s not. Even if it were a misunderstanding, the tech transfer officers *want* their misunderstanding to be true. They are not complaining to the government that they are being forced to do something against their principles, antagonistic to academic freedom or the way in which university inventions flourished pre-Bayh-Dole. Just the opposite. They want the misreading. It’s fundamental. They want the federal government to strip private rights of citizens, and hand those rights to universities. This, not getting federal agencies to stop taking title, is why these folks think Bayh-Dole is “inspired” legislation. Heady stuff.
Even still, it is clearly a misunderstanding, it happened early in the life of Bayh-Dole, and it happened that way at places like MIT and Wisconsin, and once it was put into policy statements and into AUTM “training” courses led by the folks from these schools, it was everywhere.
If we look at MIT’s intellectual property policy, we see the vesting idea built in:
2.1.1 SPONSORED RESEARCH AND OTHER AGREEMENTS
PATENTS: Research contracts sponsored by the Federal Government are subject to statutes and regulations under which M.I.T. acquires title in inventions conceived or first reduced to practice in the performance of the research. M.I.T.’s ownership is subject to a nonexclusive license to the government and the requirement that M.I.T. retain title and take effective steps to develop the practical applications of the invention by
licensing and other means.
Further on, in Section 2.4, we get this:
When it has the right to do so, M.I.T. may, if requested by the inventors or copyright authors and at M.I.T.’s discretion, “stand aside” in those situations where M.I.T. believes that it would enhance the transfer of technology to the public, is consistent with M.I.T.’s obligations to third parties, and does not
involve a conflict of interest as set forth below. By “standing aside”, M.I.T. agrees not to exercise its contractual rights to the technology, clearing the way for the M.I.T. inventors and authors to seek ownership. Inventors and authors may request that M.I.T. “stand aside” by submitting the letter that is Form 1 in Appendix A
In the case of Federal agency sponsorship, any “stand aside” by M.I.T. for a patentable invention must be made by releasing the invention to the Federal government, following which the inventor may directly petition the agency for a release of rights to himself or herself. Federal research agreements are
generally subject to a uniform patent law which provides that universities take title to resulting inventions subject to certain obligations concerning the exploitation in the public interest, Federal approval of any assignment of ownership, preferences for licensing, the retention by the Federal government of certain license rights, and march-in rights. Decisions by the Federal sponsors to permit individual inventors to acquire ownership are generally made on a case-by-case basis with the Federal Government retaining for itself those rights previously discussed.
With the discussion above, you should be able to see clearly how very wrong the MIT policy is with regard to Bayh-Dole, and how this wrongness plays nicely into an institutional pitch for ownership of faculty work. It’s still defective in that if Bayh-Dole vests ownership with MIT, then it’s not a matter of “contractual rights to the technology”, but of law. Furthermore, the policy defines “technology” as “technical innovations, inventions, and discoveries, as well as writings and other information in various forms, including computer software.” That’s way beyond the scope of Bayh-Dole’s definition of “subject invention”–even beyond the scope of 2 CFR 215.36’s interest in copyrights, patents, and data. As can be seen, MIT then states just what it has in its amicus brief–that Bayh-Dole “provides that universities take title” and that the government “must approve any assignment of ownership.” That is, Bayh-Dole vests title with MIT, and for MIT to let the inventors own by “standing aside”, the federal government has to approve the transfer of title from MIT to the inventors.
This way of dealing just isn’t so. The Supreme Court made that perfectly clear. One might think that MIT would revise its policy statements, perhaps with an explanation that they had it wrong, if they were really big-souled about it. As the Supreme Court confirmed, (much to the relief of those of us who can read, and value our freedom), Bayh-Dole does not vest title with contractors. They get title, if they get it, the conventional way, by written assignment. Promises to assign are not assignments, and their conditions can be waived, and in approved consulting, this can happen, and in the Cetus case, did. Bayh-Dole doesn’t void such deals, or make up for lapses and mistakes. Furthermore, MIT’s policy misstates the assignment provisions in (k) and ignores 35 USC 202(d) as if it doesn’t exist. MIT imagines that for inventors to acquire title, MIT has to transfer ownership to the government (or perhaps by waiving an interest, title vests magically with the government, without any paperwork) and then the government has to transfer title to the inventors. But that’s not what 202(d) says, nor what 37 CFR 401.9 implements. Title is with the inventors until someone manages to pry it away from them. The public policy question is, why would universities want to do that? A place to look for the rationale is Archie Palmer’s patent policy survey of 1948, long before this muddle of Bayh-Dole misreadings.
So MIT gets it wrong in its policy and in its amicus brief. So what? One, an elite university can’t read Bayh-Dole. Two, it embeds a misreading into its policy that then it enforces on its employees. Even if there are other, better, accurate interpretations of Bayh-Dole, MIT employees are required to conform to MIT’s own special reading. Three, this special reading evidences the huge institutional conflict of interest at work, asserting that federal law requires MIT to take ownership of federally supported inventions. That’s an institutional claim up against academic freedom and public policy, and despite the protestations in MIT’s patent policy that technology transfer is “subordinate to education and research” (also a strange claim–but not for now), federal law (interpreted specially by MIT administrators) requires MIT to own and manage federally supported inventions.
It may be that the faculty at MIT really do want the OTL to manage their inventions, federally supported or otherwise. Maybe it’s a sweet relief for the majority that the anarchy of private ownership of inventions is conveniently ended at MIT by federal law. Maybe uniform ownership by MIT makes for contented faculty, relieved of the responsibility, and complexity, and pressure of dealing with the outcomes of their research. “OMG!” they might cry like the T-Rex in Toy Story, “it’s a really significant discovery or invention–I know I just can’t handle this sort of thing, and it’s a relief that there are administrators who can!” Maybe it’s that way at MIT. I’m not buying it, but that’s what the policy suggests is the administratively preferred world.
The fourth outcome of all this MIT misreading is that it underscores an urge for state control of inventions for the benefit of university administrators. From that, the administrators can argue that they have a federal mandate to commercialize, that whatever it is they inflict on industry is in the public interest, and from there blossoms the smelly flower that is AUTM.
I hesitate to suggest that the amicus brief was written to support MIT’s policy statements, and all of the MIT OTL’s advocacy of their misreading around the country. Surely getting it so wrong in such a public, self-assured way over many years and for many universities and many inventions would be more than embarrassing. Needless to say, it is dismaying to see a top university get Bayh-Dole so very wrong in so many ways. As one dean of engineering quipped to me, when I argued that there was a tech transfer office that had some very good people in it, “why would I want my faculty to be working with the top 5% of the bottom quartile?” The quip has haunted me–at first because I thought it so misjudged and disparaged the abilities of the folks doing tech transfer at his institution and elsewhere–but now, later, I’m bothered by it because, in terms of the ability to read the foundational law on which most university technology transfer is now built, it’s apparent that the bottom quartile has been in the majority in writing and now re-writing the university patent policies.
It is those policies that have to be held accountable for what’s happening with American research enterprise. Before folks get all sophisticated in their efforts to make tech transfer more efficient, with template agreements and better training and more funding for gaps and CLPs, it would be good for them to get back to basics, and address the “The Patent Policy Problem” in universities. After Stanford v. Roche, there is no excuse for MIT and other universities to keep to the present wording in their policies–wording that never was correct as to the law, and never was right as to public policy the law represented, and wording that has given rise to a nicely articulated but fundamentally limited, if not flawed, way of dealing with research inventions, and with research enterprise.