How has the University of Missouri gone from a remarkably sensible and readable policy in 1956 to a grasping, scheming, convoluted, and at times incomprehensibly bombastic policy now? And why is it that university administrators insist that their policies give them ownership of “all intellectual property created utilizing University facilities” by university employees, when neither the patent policy nor the copyright policy do any such thing?
I don’t have sufficient insight to know the causes of such administrative shenanigans, but there are some tell-tale signs that the Bayh-Dole Act has something to do with it all. I will show you the evidence, and you can decide.
The Bayh-Dole Act came into effect in 1981, and the University of Missouri next modified its patent policy in 1992, then again in 1996. By 1996, the policy was in nearly its present form. In the 1996 version, the policy has been changed from the 1956 original in some telltale ways. First, a definition of “invention” has been added, together with a couple of odd footnotes that make a jumble of patent law rather than just reciting it or stating things simply.
In the 1956 patent policy, “invention” referred to a patentable invention. There was no need for any added definition or explanation. The policy expressly disclaimed an interest in inventions that were “obviously” not patentable. By the time of the 1996 patent policy, however, we have an invention definition that is expanded to include plant variety certificates. No doubt plant variety protection is important to a university with a strong agricultural program, but plant variety certificates are not a matter of patentable subject matter. They aren’t “inventions” in a patent sense, and the law treating them isn’t patent law. There are plant patents, but plant patents are distinct from plant variety certificates. Plant patents deal with asexually propagated plants, while plant variety certificates deal with sexually propagated plants and tubers. So why all of a sudden to plant variety certificates show up in a patent policy? Why not in a “plant varieties policy”? After all, the university has a copyright policy to deal with copyright, which is distinct from patent.
The reason is Bayh-Dole. The Bayh-Dole Act defines an invention to be:
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 the Plant Variety Protection Act, stuffed into a definition of invention in patent law, even though PVPA is not patent law. It’s a wonder Bayh-Dole did not include mask works (a copyright law matter) as well. Why should the PVPA get added to the definition of invention in Bayh-Dole? I don’t know. Apparently, however, universities wanted to control plant varieties under the PVPA in the same way that they wanted to control patentable inventions.
It is worth observing that Bayh-Dole defines an invention to be one that “is or may be patentable” or otherwise protectable under patent law or the PVPA. “Protectable” here means subject to ownership as property under a statute (either patent law or the PVPA). Many university patent policies define invention more broadly, to mean whatever administrators might call an “invention”–including unpatentable work, know-how, expertise, software, “technology,” whatever. The University of Missouri follows this pattern as well in introducing a definition of invention. It would be easy enough to simply repeat the Bayh-Dole definition, but that wouldn’t do. Here’s the Missouri definition of invention (100.020.C.7):
“Invention” is comprised of:
Conception of the idea (mental element); and
Reduction to practice of the inventive concept (physical element).
This is a rather strange way to go about defining “invention,” as conception and reduction to practice come up in patent law primarily because in a first-to-invent system, the question arises when an invention has been made, and what constitutes being “first” to invent. For this purpose, conception marks the beginning of an invention for patenting purposes and reduction to practice marks the completion. One might say that conception is the “mental” part, but reduction to practice is not the “physical” part. A better account would be that reduction to practice is the “demonstrative” or “evidentiary” part. Reduction to practice may take the form of making or demonstrating an invention so that it can be tested and evaluated to show that it works as conceived. But reduction to practice may also take the form of a written account of the invention in sufficient detail that anyone with ordinary skill can practice the invention without undue experimentation. A patent application requires constructive reduction to practice–that’s the standard for publication of the invention to the public, for which an inventor is awarded exclusive rights for a limited time–a patent.
One can see already in this definition of an invention a basic confusion about inventions and what needs to be done in policy. Look at how the definition wallows for nouns:
conception of the idea
reduction to practice of the inventive concept
The problem is, that what one conceives is the invention, and what one reduces to practice is the invention. Conception is the invention held in the mind; reduction to practice is the invention demonstrated to perform as conceived or described in detail as conceived. Do both and one has completed the invention for the purpose of establishing a first-to-invent priority date. We haven’t been under first to invent, however, for four years now, so none of this matters for patenting, but even at the time–1996 or so–this definition of invention was a poor way to go about definition.
It’s worth noting that patent law does not define invention, nor does the U.S. Constitution in giving the federal government power to give inventors exclusive rights for limited times to their inventions and discoveries. Why then does a university patent policy have to define invention? Why not just focus on what is or may be patentable?
Here’s 35 USC 101:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Patent law then does not get tangled up in definitions of process, machine, manufacture, or composition of matter–though case law has. Are software algorithms processes or mathematical laws? Are life forms compositions of matter or bits of divine spark? The key issue is whether some thing that affects the material world is new and useful. New, in turn, requires that something is novel (new to the world) and non-obvious (something that takes an inventive step, not something that just anyone could do, knowing what just anyones typically know). But even here, a university patent policy is best off delegating authority, stating claims, and establishing procedures that pertain to inventions that are or may be patentable. One does not deal in ideas or expertise or improvements or know-how or technology or software. One deals in patentable inventions.
The Missouri definition of invention, then, is actually a partial account of what makes an invention complete for the purposes of determining when the invention was “made.” In a first-to-invent system, “made” means something like “the earliest date the invention existed in an evidentiary form that was completed with diligence to reduction to practice.” In a first-to-file system, “made” means rather “the earliest date on which someone files a patent application on the invention.” An invention can be made, obviously, before a patent application is filed, but with first-to-file, all that matters is the constructive reduction to practice evidenced by a patent application. The Missouri definition of invention leaves out all sorts of things. An invention, even if complete, requires “contemporaneous recognition and appreciation” of the invention (to use the MPEP’s heading on the matter). That is, the invention does not get a priority date simply by being conceived and reduced to practice if no one recognized the invention as an invention.
For a university patent policy, recognition of something as inventive is way more important than the arcanity of establishing a priority date based on conception and reduction to practice. Recognition of inventive work matters even in a first-to-file system. That’s one service that in-house patent personnel can perform–helping researchers learn to recognize new stuff that is inventive. Often research personnel have way more than ordinary skill in a given area of practice, and so it might come to them as a surprise that their work involves inventive steps–stuff that isn’t obvious to the ordinary person with training in the field. One might think that a university patent policy would do well to encourage researchers and other creative types to talk about their work with university patent personnel, if only to become better aware of when something might be inventive.
And this gets us to what we might call the Feynman Point, after Richard Feynman the physicist. Feynman recognized that he could “invent” all day, every day. He could come up with invention after invention. In an interview, he describes how he “invented” a rocket-propelled airplane. “There are so many ideas about nuclear energy that are so perfectly obvious, that I’d be here all day telling you stuff,” says Feynman to the Los Alamos patent attorney looking for inventions. “There’s a million ideas!” The patent attorney comes back with patent applications for three of the ideas Feynman has rattled off.
I have worked with scientists who I have asked to review the patent literature on a particular area of invention. They skim through patent after patent going, “that’s Frog’s (or Toad’s) work–gawd, they patented that?” The reality is that there is an oort cloud of inventions circling any research program. It’s just that most research scientists don’t have time to start listing them all. They would never get anything done. Any engineering design project worth its salt can produce thirty invention disclosures at any given time. It’s just that it’s rare that anyone wants to stop work long enough to do that.
The fact is, harvesting the world of ideas for patentable inventions is a meaningless, if not worthless, undertaking. It’s not that most such inventions are meaningless or worthless–it’s that patents on them are. If there’s one management principle that matters most with regard to patents it is: You have to be selective.
If a university wanted a great patent policy, the first thing they would do is figure a way to be selective about what patents the university should acquire–if any. One way–the traditional way–was to wait until someone asked the university to get involved, and for what purpose. A university might get involved to hold patent rights on behalf of everyone, creating a commons, just as the federal government did before Bayh-Dole wrecked even that. A university might get involved to manage disagreements among academic inventors. That’s what the University of Toronto did with patents on insulin–licensed everyone who was capable and agreed not to cut corners and produce tainted insulin.
But even if a university wants to make money from patents to fund ever more research, the way that has been proven to work is to wait until someone has an invention that might make money for a patent owner and wants the university to make money from owning it. That’s right–acquire patents when someone wants you to profit from them. All the better if that someone is “industry.” That was the case, for instance, with the Cohen-Boyer patents. Industry agreed that Stanford ought to pursue patenting, and Stanford kept the patent prosecution open, so everyone could see what they were doing. Similarly, that’s how Research Corporation operated, with a board of directors drawn from industry, looking out after industry. When Research Corporation agreed to manage an invention, the overarching idea was “industry will want the university and inventor to make some money from this invention to reward them for their effort to make it available to us.” Research Corporation’s selectivity meant that it was industry marketing to industry inventions made at universities.
University administrators, however, are fixated on being just the opposite: non-selective. They define “invention” so broadly it can mean most anything–whatever an administrator says is an “invention,” regardless of whether that thing is inventive, patentable, or meaningful as patented. They demand everything they can and create a situation in which most everyone is out of compliance with their demands. They can then use that fact to threaten people when they end up locking horns with someone over an “invention” or anything else that bothers them. The result is that universities trawl the bottom of the invention sea and haul up everything, destroying most of it to get control of maybe one thing a decade that matters. Even that one thing likely would have mattered without university ownership. The university patenting and licensing program generally does nothing for such “big hit” inventions. The thing most people don’t get is that for every “success” where the university’s compulsory ownership policy was invoked–again, one or so a decade–maybe five other opportunities were destroyed because inventors did not want to work with the university, did not want to deal in a monopoly over academic work, did not want university administrators approving who could and could not use an invention.