Adrian Johns, in Piracy: The Intellectual Property Wars From Gutenberg to Gates, describes the difference in approach to literary property between the rival cities London and Dublin. For London publishers, literary property was held to be perpetual. For Dublin publishers, there was essentially no such thing as literary property. The Dublin book trade thrived–anyone could print anything, sell copies wherever and however, and the competition was who could get to market most quickly, with book product people would buy. In London, titles were controlled by the major publishers. Maybe something would be printed, maybe not. And if so, at prices set by the publishers, not by competition for readers.
Here’s Johns’s paragraph:
According to leading authorities across Europe, some form of literary property, however defined, was the bedrock on which public reason had to rest. print’s role in progress depended on fidelity and security of authorship, and those could not be guaranteed without a regime of some kind. There could certainly be dispute about the proper form of such ownership, about where its boundaries lay, and about who should possess it; but little scope existed to deny the need for some such principle. Yet it was precisely this that Ireland scandalously lacked. And it seemed to relish lacking it. If literary property and rules of authorship were so central to enlightenment, why did the Irish model not collapse into chaos and ignorance? Why, on the contrary, did it seem to thrive as never before? (147)
Johns follows this observation up with a statement of relevance:
That is a question that merits being asked in the present tense too.
I’m not ready to argue that IP should be abandoned entirely, though I see some merit in doing so. But let’s push this into university IP as a specific area of interest. There’s no question that universities have turned first to establishing patent policies, under the influence of Archie Palmer and the National Research Council, which engaged in a three-decade effort to get universities to adopt patent policies. As the abstract for a 2007 article by Jane Robbins puts it, “Beginning in the 1930s, the National Research Council waged an organized campaign to encourage academic patenting and centralize it within one organization.” Once universities had patent policies, it took much less effort for those policies to be expanded to “IP” policies, to change voluntary invention reporting into first limited university claims to invention ownership, and then to broad claims to invention ownership–even to ownership of inventions whether or not they were inventions. Well now.
In a sense, then, universities went from being Dublin to being London, from a thriving research base in a world of wide open possibility to a controlled, regulated, paperwork, bureaucratic process world constrained “for the public benefit” (i.e., for the benefit of universities) by London-style controls over research property. Rather than thriving under this new regime, research innovation has been stifled. Universities have obtained over 140,00 U.S. patent since about 1981 and taken control of perhaps 250,000 research developments on about $1 trillion in sponsored research funding. Most of those patents are unlicensed, and even fewer of the underlying inventions–maybe less than 2% have ever achieved “commercial” success, and far fewer what Bayh-Dole calls “practical application”–use of the invention such that the benefits of that use are available to the public on reasonable terms–where “reasonable terms” means, for Bayh-Dole, terms that would reasonably be offered if there were competition, even if a patent otherwise prevented that competition.
While there may be reasons for intellectual property–I can see trademark, for instance, as useful–and even for trade secrecy–universities ought to be different when it comes to attempts to regulate patents and copyrights. Universities ought to be more like Dublin than London. Universities ought to be islands of activity in a sea of IP claims. The aim of universities should not be to demand monopoly control over research developments, and then to try to find monopolists who will pay to take over that control, under the premise that only monopolists will use research findings. The actual university argument is that if a university does not assert monopoly control, then no investor or company that makes a habit of monopoly control will be attracted to pay. That would seem to be pretty safe, but it is not at all the same as the argument that without monopoly control taken by a university, no one will use any invention or other research finding. As Niels Reimers commented on the Cohen-Boyer gene splicing invention that he managed for Stanford and the University of California, the technology was going to be used no matter what, and that asserting a patent position on behalf of the universities was nothing more than a “tax” on industry. Of course the invention would be used, and I rather think Reimers is wrong about the role of the patents in a non-exclusive licensing effort, but that’s just because Reimers simply cannot conceive of any uses for patents besides creating product monopolies that jack pricing, and licensing revenues, by 100x. How that’s not also a “tax” is beyond me.
In the world of research noodling resulting in discoveries and inventions, among other things (like, say, software or collections of rocks, beetles, images, and technical data) the assets that matter are the NIPIA–the non-IP intangible assets. We are talking about personal dynamism, grad students who are articulate, formation of social networks that convey new ideas and technology, circulation of research tools, interoperability, and formation of ad hoc standards. IP can be used to augment NIPIA, or can be used to stifle NIPIA. To be equitable, universities use their IP ownership to stifle both IP and NIPIA.
It may well be that the best university IP policy is to have no IP policy. The University of Wisconsin held out perhaps the longest among American universities. Faculty researchers negotiated their own IP clauses in research contracts. It took shenanigans by the Wisconsin Alumni Research Foundation and the NIH to work out a scheme where the university was required, as a condition of federal funding, to take ownership of any NIH-funded invention that the university (via WARF) wanted to patent. Norman Latker then pushed for all federal agencies to adopt the scheme, and when that was blocked, he tried with the Thorton bill, and when that failed he tried again with what became Bayh-Dole, which in its way succeeded (in that it was passed with ambiguous and incomplete language that read one way to legislators but could be changed by means of regulations, which is where Latker then went to work).
To my mind, this would be a good time to address the Palmer/WARF/Latker scheme to turn universities from thriving Dublin to dour London.
Rescind Reagan’s (Latker drafted) 1983 Memorandum on executive branch patent policy, revert to the 1963 Kennedy patent policy with the 1947 AG’s report as backing reasoning–the government may own patents (i.e., issue them to itself or acquire them from others) but it cannot (as a matter of public policy) grant exclusive licenses, seek royalties, or enforce its patent rights against US citizens using inventions subject to such government-owned patents. Revise Reagan’s 1987 Executive Order 12591 to remove confusticating language (including added confustications from Reagan’s Executive Order 12618). Then revise 37 CFR 404 to remove authorization to grant “co-exclusive, partially exclusive, or exclusive licenses”(404.4, and including the entire section at 404.7) and remove authorization to receive royalties. Remove the crazy restrictions on licensing, such as the requirement that all candidate licensees must submit a development or marketing plan (404.5(a)(1)) or make a complicated “application for a license” (404.8), as well as the U.S. manufacturing requirement (which is Bayh-Dole (35 USC 204) is restricted to contractor-granted exclusive licenses). Eliminate all authorizations for federal agencies to enforce patent rights, or for licensees to enforce those rights as proxies for federal agencies (such as 404.5(b)(2) and 404.12).
So much for the federal government side of Bayh-Dole. While Congress may have authorized the executive branch agencies to obtain patents, grant exclusive licenses, and to enforce those patents, including by taking a royalty interest, there’s nothing that requires the agencies to do any such things. The agencies (and the executive branch) would appear to be free to choose not to patent, or if they do request the executive branch to issue patents to the executive branch, they are free to license non-exclusively, royalty free, and not seek to enforce any such patents against use by U.S. citizens and companies. So, formalize such practices, per the 1947 A.G.’s report (overseen by David Lloyd Kreeger).
Now for the contractor side. Bayh-Dole’s implementing regulations set up three classes of contractor. Most people don’t get this, don’t want to get it, and obfuscate to prevent others from getting it. But it is there plain and clear in the statute. Those classes are small businesses, non-profits, and inventors who are made parties to any federal funding agreement. Each class has its own standard patent rights clause (though again NIST, designated by the Department of Commerce to administrate Bayh-Dole regulations, has done its best to obscure these patent rights clauses.) The most stringent clause applies to non-profits (funny that, eh?). The most liberal applies to inventors who become parties to a funding agreement (and therefore, under Bayh-Dole’s definitions, contractors). Small businesses are somewhere in between.
The key concept here is that Bayh-Dole applies only to “subject inventions”–inventions that a contractor has acquired. If a contractor does not acquire an invention, even if made in work that receives federal support through a funding agreement, then that invention is not a subject invention, and is not subject to Bayh-Dole. The invention might be subject to other federal requirements (if the federal government thought to make those requirements–such as were in the old Federal Procurement Regulations (which required contractors to require their research employees to agree to the same requirements as the contractors–that is, become parties to the funding agreement patent rights clause and regulations). That requirement is not in Bayh-Dole. The Supreme Court in Stanford v Roche looked for it, and didn’t find it. We have to accept that the omission reflected the intent of Congress. Contractors must follow the patent rights clause, but contractors are not forced by the statute to make inventors assign inventions to the contractors. The inventors are free of federal obligations regarding their inventions unless they are made to be contractors. And then, the inventors enjoy the benefits of the patent rights clause specific to them (37 CFR 401.9). That’s getting darn close to Dublin.
We just need an Executive Order that makes clear that the NIST-stupid assignment clause at 37 CFR 401.14(f)(2) operates only for inventions that a non-profit or small business obtains by means of equitable title (that is, when the employer directs an employee to invent–which is not the case for university faculty or students or even university staff assigned to assist faculty) or an express patent agreement (not merely an employment contract). What’s odd is that the (f)(2) agreement effectively displaces, for federal funding agreements, any other employer agreement as to ownership of inventions made in work receiving federal support. An employer contractor must receive assignment of inventions that the employer contractor already owns (that equitable title thing again) but has to agree not to claim ownership of inventions that the employer contractor has no right to claim title to. Yeah, it’s a crazy, mixed up world here. But the fundamental is that Bayh-Dole supports Dublin, while Latker, NIST, and university monopolist IP licensing operations cry London.
And there’s at least one more push. Bayh-Dole provides that federal agencies can alter the standard patent rights clauses when “exceptional” circumstances are determined. Under the Kennedy patent policy, there were four such exceptional circumstances: (1) when a federal agency decides to develop an invention to the point of practical application (such as using multiple contractors) and then releasing the developed invention for use (as, with a tomato-picking machine); (2) in research directed at public health or welfare; (3) where the federal government is the principal user or funder of the work (as with nuclear bombs or space technology); and (4) when the contractor’s role is to supervise the work of other contractors (as in the management of a federal laboratory). “Exceptional” circumstances are not necessarily “rare” or “extraordinary”–they are simply exceptions to a stated default. The Department of Energy fought tooth and nail for exception (3) in Kennedy, and you will see signs of that in the Bayh-Dole patent rights clauses (now merged) for non-profits and small businesses.
The primary target of Latker, the rogue NIH patent counsel, was exception (2)–for health research. There, without coming out and saying as much, Latker aimed to permit contractors to take patent monopoly positions, and pass these positions to pharma companies on behalf of the NIH, which under the prevailing guidance of the 1947 A.G.’s report, could not pass patent monopolies itself directly to favored companies under exclusive licenses. But public health is every much an exceptional circumstance under Kennedy as making nuclear bombs or popping a nuclear reactor into a submarine. Nothing has changed but the appearance of a “reversal” of the expectation of title to inventions made in federally supported work. That is, this “reversal” is really just a need to re-establish an exceptional circumstance for Kennedy’s (2) exception for public health and welfare research. In short, an Executive Order should be sufficient, even with all the bombast regulatory certification put up by 35 USC 202(b). Boils down to–get this–the beliefs of the Secretary of Commerce and the determinations of the Administrator of the Office of Federal Procurement Policy. If contractors don’t like it, they can start a long appeals process designed to prevent “march-in” (at 35 USC 203(b)) which ends up in the Court of Federal Claims. Basically, public policy stymied by lawfare, if it goes that way.
Anyhow, that’s a good start to how the executive branch could break the London-style monopoly on federal research inventions now held by universities and other nonprofits. Their London calling has been a total failure. Even the few inventions that have seen “commercialization” have got there *despite* the university involvement, have for the most part failed the Bayh-Dole standard of practical application, and even those few successes have come at the cost of tens of thousands of research findings suppressed by university patents and other proprietary and policy claims. It’s a stupendous failure, wrapped in public virtue signaling about the need for bureaucratic thumbs in every innovation pie with worthless paperwork, institutional claims, slow nonsense procedures, and a fanatical devotion to patent monopoly licensing or nothing at all. Tear it all down would be good, but a few Executive Orders that rescind or modify existing Executive Orders and the regulations that these orders spawned would get at much of it, short of an Act of Congress, which these days seems that twerking would be a more likely act, But then again, it’s strange times.
If we wanted to see how terribly bad university patent management has been, audit the universities. It won’t be pretty. Audit for compliance with Bayh-Dole. Audit for successful practical application. Audit for claiming more inventions than due (er, denying rights under the color of law, say). Audit for being truthful in invention development reports.
The most important “rights” created in publicly spirited research are NIPIA rights. IP can be used to create NIPIA, such as patents dedicated to a standard. The standard is not itself IP but rather is NIPIA. Similarly, the movement of technology creates social networks, and these networks are not IP but rather NIPIA. A network may also become a qualified lead list–also not IP but rather NIPIA. Universities, and Bayh-Dole, suppress such NIPIA in pursuit of high paying monopoly patent relationships, and those are hard to come by, and when got do not address the public benefit (except as crumbs on the way to private benefit). There are ways to redeem Bayh-Dole practice, but the universities will go to the mat to prevent it, despite their horrible record. Becoming Moloch means never ever saying you are sorry.