In reviewing university IP policies while I consider how a policy can come to transfer, as if by magic, IP from an author or inventor to a university merely by the assertion of university ownership, I came upon the following Form HR36100 at the University of Michigan titled “Supplemental Appointment Information.” After a section in which the appointee provides name and social security number, we have this:
It is apparent that university has implemented the “present assignment” intervention that has circulated in the tech transfer community by folks who don’t understand the Bayh-Dole Act and refuse to accept the Supreme Court decision in Stanford v Roche. Thus, we have “hereby assign” language upfront, before any invention has ever been made. The clever idea is that when any invention is made, it automatically becomes owned by the university by operation of this assignment document. The problem is not how the assignment is made, however, my clever children; rather, the problem is how the assignment is scoped–the definitions that operate, the inventions that come within that scope, and a reasonable interpretation of the assignment that has been made.
There is plenty not to like in how this assignment document is framed. First, take a look at Regents Bylaw 3.10:
Bylaw 3.10 dates from at least the 1970s–I have been unable to trace its complete history. The first paragraph, however, is clearly based on the university patent policy from 1944, which was still in effect in 1962 when Archie Palmer compiled his second set of university patent policies. The 1944 patent policy is so short I quote it in full:
The 1944 policy presents two situations with parallel language. The first paragraph deals with properties the university acquires in connection with research projects. The second paragraph deals with patents “issued” in connection with research projects. This second paragraph claims ownership of the patents and any income from the patents.
“Properties” has been changed to “patents and copyrights”
“acquired” becomes “issued or acquired”
“in connection with and for the prosecution of ” is “as the result of or in connection with”
“research projects” is “administration, research, or other educational activities”
“carried on by” now is “conducted by”
“shall belong to the university” becomes “shall be the property of the university”
In addition, 3.10 inserts a qualification regarding the nature of the projects:
supported directly or indirectly (e.g., through the use of university resources or facilities) by funds administered by the university, regardless of the source of such funds
The 1944 policy’s second paragraph claim on royalties is rolled into the new 3.10 first paragraph, and now made to refer not only to patents and copyrights acquired but also ones that are “issued” (whatever “issued” means for copyrights).
The first paragraph concerns incoming patent rights. It makes sense that if the university acquires a patent to assist in the conduct of a research project, the university should be the owner of what it has acquired. The second paragraph concerns patents that “issue” in connection with research. What did this claim actually mean? It is clear from the Proceedings of the Board of Regents that faculty were expected to own their inventions unless a research contract required otherwise or the inventors voluntarily agreed otherwise. For instance, in 1960, the Board of Regents entered into an invention management agreement with the Michigan Research Foundation. The agreement provides that the university will “recommend” to faculty, associates, and employees (note the distinction between faculty and employees here) that they assign inventions to the foundation:
So the university may recommend at its discretion that inventors assign inventions to the foundation, and the university agrees that inventors can help the foundation with patenting without the foundation having to compensate the university for the inventors’ time. If the university expected to own inventions made by faculty in their research, there would be no need for any of this–the university would simply take assignment directly and assign to the foundation (or license to, or task the foundation as an agent) to patent and market the inventions. Other elements of the invention management agreement make all this clear. Article II, paragraph 10 requires the foundation to assign inventions, patent applications, and patents to the university on termination of the agreement:
Importantly, the assignment to the university is conditioned on any “letter-agreements” between the inventors and the university. That is, arrangements between the inventors and the foundation take precedence, as to any other licenses or contracts. The assignment to the university is the place where all inventions go that are not otherwise provided for by agreements between the foundation and inventors, and between the foundation and others acquiring an interest in certain inventions. Again, if the university expected to own inventions outright as a matter of policy, there would be no reason to contemplate that inventors might have private agreements with the foundation regarding the disposition of ownership of the university’s inventions.
Attached to the agreement as Exhibit A is a template letter-agreement by which the foundation is to request assignment of invention rights, negotiate sharing licensing income, and obtain an assurance of the inventor’s assistance. It’s clear that this letter-agreement sets out the terms of the transaction–this is an invention management agreement between the foundation and the inventor, with a request that the inventor execute an enclosed assignment of the invention to the foundation. It is also clear that the inventor has an opportunity to negotiate other terms, if desired.
In the 1944 patent policy, what then does it mean that “patents issued in connection with research projects” will belong to the university? Expressly, this language is not a claim on all inventions made by university staff–it only concerns patents. Thus, if an invention were to be made and not patented, the policy makes no ownership claim. There is nothing here to force university inventors to patent their inventions. Further, the language is specific to research projects. Inventions made outside “projects”–and the language here indicates projects funded by external sponsors–are also not a concern of the policy. The policy appears to be concerned with inventions made in research funded by external sponsors in which the sponsors desired patents to issue on inventions made in a project. “In connection with” is not “arising in” or “as a result of”–there is something more specific indicated by “in connection with.” My read is that what was meant was a situation in which patenting of inventions was specified in the statement of work, as an objective or deliverable of the contract. There is nothing to indicate that the policy aimed to claim all inventions made in the context of a research project. If that had been the objective of the policy, then that claim would have been included in the first paragraph of the 1944 policy, where faculty research is considered with only university support.
Sometime after 1960 and 1976, the Regents introduce Bylaw 3.10, revising the patent policy language. The revision reduces the two paragraphs to one, conflating incoming and new patent management.The revision adds copyright to patent (so seductive, so foolish), expands research to “administration, research, and other,” changes “projects” to “activities,” adds a restriction that the activities must be “supported” by university-administered “funds,” and then tries to limit the restriction with expansions on support–direct or indirect–and funds–regardless of source.
Finally, the revision adds “as a result of” to “in connection with,” making it clear, at least, that “in connection with” was not sufficiently broad. Whatever “in connection with” meant, it didn’t mean “as a result of” or there would have been no need to add the extra language. If “in connection with” meant “as provided for in a contract with a sponsor,” then “as a result of” must mean something else–such as simply that an invention is made, even if the sponsor and the inventors *don’t* want to pursue a patent. That’s a substantial change!
Even with the expansion of this expansion of scope, there is a difference between an invention arising in research (or a funded activity) and a patent being acquired or issuing as a result of the research or funded activity. If words mean things, then differences matter. If this is all just handwaving for the university getting whatever it wants, then there’s no point in reading carefully. Bylaw 3.10 does not claim inventions–it claims patents and connects these patents to activities supported by university funds or through the use of “resources or facilities.” There’s a big gap in there: if the university claims patents, not inventions, then inventors may not have to assign all their inventions to the university–only those inventions that they intend to patent, and then only those inventions that have been supported by university funding.
Let’s stop to consider the problem of how funding connects with inventing. A person can work on a project a long time, and then, later, have a realization about how something might work, or how to get around a problem, or how to apply a result. The inventive step has nothing to do with the project, which may have ended years previously, but the invention benefits from that project. For all that, one could study someone else’s project and come to an inventive realization. Is that invention “a result” of the other investigator’s project? One could run far afield if so–anything that anyone invents that could be traced back to a benefit coming from a university “activity” could be claimed by the policy. But such stuff is beyond any reasonable interpretation. Even if an invention is made during a project–at the same time as the project is being conducted–it’s still the same analysis: the invention might benefit from the project activity, but in what sense is the invention (and much less, any patent) supported by university funding? It’s not, unless there’s evidence to show that the invention arises as an objective of the project.
This is a problem, then, with the move from “project” to “activity.” No doubt the drafters were aiming to broaden the claim, but what they did was muddy it. A project, in the 1944 policy, carries with it the idea of a statement of work, a written agreement setting the objectives and plan for a piece of research. It is clear how a patent might issue from a defined project, and how that patent might have reasonable claims placed on it. But “activity” has no such boundaries. “Activity” is like “behavior”–any action is in some sense an “activity.” There’s no indication of boundaries, of definition, of a written objective and plan–anything might be an “activity.”
The implementing regulations for the Bayh-Dole Act start with this problem of scope, right there in 37 CFR 401.1. First thing, before definitions, even. The law restricts the scope of what federal agencies may count as a subject invention–even before we get to what subject inventions a federal agency might claim the federal government has a right to own. A subject invention is one made “in the performance of work under a funding agreement.” The scoping statement in 37 CFR 401.1 takes pains to provide an interpretation for what this language means. It does not mean “during” a project. It does not mean anything closely related. It does not mean anything arising from work that was not “planned and committed,” so long as the work does not “diminish or distract” from the performance of work under the funding agreement. We are left with a narrow scope–look to the grant sourcing documents such as the call for proposals, look at the proposal submitted, look at the requirements of the funding agreement (typically but not always 2 CFR 215), and determine what activities in the federally funded project were “planned and committed.” If the invention arises in these activities (and if the (f)(2) agreement has been executed, or the inventors assign to their university anyway), then the invention is a subject invention–that is, it is within scope of the requirements of the standard patent rights clause on disclosure and retention of rights.
But in the Michigan case, we find the university making the scope less clear rather than more definite. Activities within a defined project funded by the university may give rise to an invention that the university might claim. But an activity without a project could be anything. Similarly for the idea of “indirect” funding through the use of “facilities” (generally, buildings, but bad policy drafters seem to love abstractions, thinking abstractions are clearer, or at least more authority-sounding). If a faculty member goes to the library regularly to think creatively, is that an “activity” that’s “indirectly funded” through the use of a university “facility”? One might think that even the provision of a salary is sufficient to capture “indirect funding” of any activity, if “resource” (another grossly abstract term) can include “human capital”–the inventor herself. But going this far does not hold up, because Paragraph 4 distinguishes between university “support” and university “employment.” Even though Paragraph 4 is ineptly drafted to conflate “supported by funds administered by the university” with “university support”–different things, it still manages to make clear that employment is not the standard on which 3.10’s claim is built. There has to be money other than salary at play, and–by implication–use of resources or facilities that are allocated to an activity other than an activity of employment. In the old-style university patent policies, before Bayh-Dole, a university had to allocate support expressly–in a budget, in an agreement to allow special use of facilities–to have a claim to an equitable share of proceeds from the exploitation of a patent. That distinction still exists in 3.10, but it appears that the drafters were at some pains to bury it.
This is a typical strategy. Claim the revisions are to add clarity and refinement, but do them in such a way as to permit a fundamentally new interpretation once the revisions have been formally approved. The words don’t appear to change much, but the later interpretations are often hugely different. The wording was chosen, no doubt, to be read by administrators as broadly as they chose to read it. If so, then the actual policy statement reduces to “the university owns whatever its administrators say they want.” In fact, this is exactly what a university administrator reported to the Regents in 1985:
The university just owns everything, and so does most everyone else. This, despite a policy history that does not provide for that at all. From the words in the policy, a reasonable interpretation–as distinct from the late Dr. Sussman’s assessment–is that there must be some showing of university support–a budget, a written agreement, a special allocation of resources, a contract with a sponsor, a donor’s restrictions–that provides a definite bound for an “activity” to be one supported by university funds and thus within the scope of the policy claim. Otherwise, why write a policy this way, other than to bamboozle faculty?
The 1944 policy claimed patents issued in connection with research projects. The later Bylaw 3.10 claimed patents acquired or issued as a result of or in connection with research and with administration and with any other educational activity. The “other” perhaps is intended to make “research” and “administration” merely illustrative, and adding “educational” is not supposed to indicate “instruction” but rather “higher educational”–anything that is remotely associated with the university, anything that the university might take credit for.
But lists don’t necessarily expand claims–they may actually reduce them. The difference is between what an administrator wants to use the words to mean and what a judge, considering what a reasonable party dealing with what amounts to an adhesion contract, might construct as the sense of the statement. In copyright practice, one convention is that if a party provides a list, say, of media, even if there’s some tag such as “including but not limited to,” the list is all one gets, not the list and everything unstated that is “just like the list” or “obviously in the same set as the list.” Et cetera is not a term of art with which to provide clarity.
Here in 3.10, “other educational activities” aims to make the policy claim “any educational activity” but one might well find a judge willing to restrict the policy to “research activities” and “administrative activities”–and only research activities meaningfully result in inventions. It’s hard to see how a clerical worker could be construed to be hired to invent, or have invention within the scope of his or her duties. Even the standard patent rights clause authorized by Bayh-Dole excludes clerical and non-technical workers. That’s done for public policy reasons, not as an oversight or out of generosity. Universities are subject to the same public policy arguments. It’s strange that the university patent administrators turn out as a bunch mostly not to care about public policy, while giving lip service to public mission and public benefit and all the help and incentives they aim to provide to inventors.
I have come to believe that a number of university tech transfer professionals (both managers and lawyers)–people who should be expected to be careful with their drafting–are barely literate when it comes to university policy documents. They desperately want the documents to claim everything possible, as expeditiously as possible, and free themselves from any accountability to inventors for anything they do–that is, in some dark part of their souls, they want to be institutional invention fascists. The last thing these people want is to be agents for inventors or colleagues ready to collaborate on an equal basis. They want the authority, the stream of disclosures, and the salary, but they don’t want to be careful with policy or accountable for their decisions and the outcomes of those decisions. Worse, by far, than the invention management companies that advertise late at night.
You have been brave to read this far. Now let’s back to that present assignment agreement, to show how it garbles 3.10, which already has made a hash of the long-standing 1944 patent policy.
As part of my obligations under Bylaw 3.10, I hereby assign to the Regents of the University of Michigan (“University”) all right, title, and interest in patent rights in inventions made (e.g., conceived or first reduced to practice) by me as a University employee with the direct or indirect support of funds administered by the University (e.g., through the use of University resources or facilities) regardless of the source of such funds.
I don’t see how Bylaw 3.10 creates an obligation for university employees. It establishes an expectation within the Board of Regents, a corporation established to operate the university. University officials may have an obligation to give effect to the Bylaw by taking actions to secure patents that the Board of Regents intends to claim, but the obligation gets to employees not by a notice regarding the desires of the corporation, but by actual agreements made between employees and the university. The Supplemental Appointment Information form is mislabled. It is actually an assignment form tied to an employment agreement, except it appears to come afterwards, and be subordinate to the appointment, not a precondition. So “As part of my obligations” is actually “The Board of Regents requires you to make the following assignment.”
Now let’s clean up the garbage. The assignment is limited to “patent rights in inventions.” So far so good–that’s much like “patents acquired or issued”–except, of course, it isn’t. 3.10 operates on patents, not on inventions, and so operates only where patents have come into existence. Yet this assignment is directed at patent rights in a class of inventions, regardless of whether patents are desired, applied for, or obtained. That’s an expansion of claim beyond what 3.10 authorizes. One might think it is abuse of authority. But perhaps it is just incompetence. If incompetence, it is a convenient one for attempting to broaden the claim.
More: “inventions made (e.g., conceived or first reduced to practice).” This is filled with delightful nonsense. First, “e.g.” means “for example” not “that is.” The parenthetical does not explain what “made” means, but rather gives examples of what “made” means. And yet, for federal patent law, an invention is not “made” until it is *both* conceived and reduced to practice. Reduction to practice may be constructive (describe the thing) or actual (make or do it). Without both, no invention is “made.” But here we don’t have an and, we have an or. For this assignment, an invention is made when it is either conceived or first reduced to practice. That’s garble. If an invention is made when it is conceived, there are no patent rights in it to assign, so the assignment does not operate. Same for an invention first reduced to practice, but not conceived. Patent law is clear that an inventor must recognize an invention as inventive to be an inventor. One can build something and not realize the significance of the work. Again, nothing operates in the assignment. What is left out–and totally ambiguous–is the condition that matters the most, when an invention is both conceived and reduced to practice. Then it is potentially patentable; then the assignment might operate.
Where does this “or” construction come from? The likely source is the Bayh-Dole Act and the definition of subject invention. “Subject invention” is not a definition of an invention, but of the scope of a federal agency’s interest in inventions made with federal support. The difference is substantial between what constitutes a patentable invention and what constitutes an invention within the scope of a contractual claim. Bayh-Dole defines a subject invention as an invention (an invention that is or may be patentable) “of the contractor” (that is, owned by a party to the funding agreement–and the inventors are to be parties to the funding agreement) that is “conceived or first actually reduced to practice in the performance of work under a funding agreement.” The scoping statement does not define invention, but sets out what one has to find in the statement of work, the “planned and committed activities,” to justify a finding of a subject invention. Specifically, one has to find that the planned and committed activities include either conceiving the invention (solving a specific problem, say) or first using the invention for its intended purpose–actually reducing it to practice.
The definition of subject invention then catches certain potentially patentable inventions. But for the scope to operate, the invention must first be potentially patentable. That is, it must be both conceived *and* reduced to practice. That’s what the definition of “invention” in Bayh-Dole requires. Once you have an invention that satisfies both elements, then one looks to see whether either element, or both, were “planned and committed” in the funding agreement statement of work. If either element is present in the statement of work, then the invention (if it is owned by a party to the funding agreement) is a subject invention. Otherwise, it is outside the scope of the definition of subject invention and not within what a federal agency has contracted. It’s not that the agency has a claim outside of Bayh-Dole and the standard patent rights clause–it’s that the agency has no claim whatsoever, other than if it were to invoke eminent domain or national security (another way to deal with inventions that an agency wants to have live in perpetual darkness–I’ve seen this happen).
Again, first there’s a patentable invention. Then there’s a check with the statement of work. If either conception or actual build/use is planned for and committed, then the invention is subject to a patent rights clause–typically 37 CFR 401.14(a) or 37 CFR 401.9. The use of “or” in the scoping agreement, in contrast to the “and” in the definition of a patentable invention, covers the three possibilities that may arise in a statement of work:(i) conception of an invention is part of the planned and committed activities, but actual reduction to practice is not; (ii) actual reduction to practice is part of the planned and committed activities, but conception (solving a problem, say) is not; or, obviously, (iii) both elements may be found in the statement of work.
So you can see how the Michigan present assignment statement clucks this all fup. It defines an invention as either element, rather than requiring at least one element to be within scope for an invention that otherwise already is potentially patentable. Then it requires the prospective–future patent rights to either of these elements. Of course, no inventions have been made at the time of assignment. But further, the scope of when the assignment operates ends up being greatly expanded over the authority for ownership established in Bylaw 3.10. The patent rights get assigned not only before there is any invention, but also when only part of an invention gets made–conception or reduction to practice–so the future patent rights are assigned upfront for what amount to be non-patentable inventions. A mess.
Not done. There’s a set of restrictions on the scope of assignment.
“as an employee”
“with the direct or indirect support of with the direct or indirect support of funds administered by the University (e.g., through the use of University resources or facilities) regardless of the source of such funds”
Notice that the parenthetical has been moved from immediately after “indirect” in 3.10 to after the entire clause. In 3.10, the parenthetical modifies the meaning of “indirect” funds used to support “educational activities,” while in the present assignment clause, the parenthetical modifies the entire clause–both direct and indirect support. In the assignment, this shift could reasonably be read as limiting the assignment (in the future) to patent rights in *inventions* that have received support, not in *activities* that have received support. The switch is important. Bylaw 3.10 asks whether an activity has been supported by university funds. If so, any patent “resulting from” or “in connection with” is claimed. But in the present assignment, it is patent *rights* in any *invention* that is funded by the university–the review is whether the invention was funded, not the activity.
The present assignment also adds a further restriction, one that carries its own consternation. The invention must be made by the inventor “as an employee.” This is not a qualification of Bylaw 3.10. All that matters to 3.10 is that an educational activity was funded by the university. In the present assignment, what matters is that the inventor made the invention “as an employee” AND the *invention* was funded by the university. We have discussed the distinction Bylaw 3.10 makes between “support by funding” and the exclusion of limitations that might otherwise arise from “university employment.”
What does this mean? It means that the Bylaw 3.10 does not claim patents made as a result of the activity of *employment*–but only those that are connected with “educational activities,” if such activities can be anything other than “research projects” that receive special funding or provision of lab resources–the focus of the 1944 policy, which is looking better all the time. But the present assignment expressly limits the inventions to those made “as an employee.” This claim appears to be entirely outside the authority of Bylaw 3.10, which focuses on specially supported activities, not those that arise from mere employment.
There’s another huge problem with the use of “as an employee” in the present assignment. As the US Supreme Court majority opinion put it in Stanford v Roche:
But, as noted, patent law has always been different: We have rejected the idea that mere employment is sufficient to vest title to an employee’s invention in the employer….
“Employment” means something specific in IP contexts, not mere “payment of money” to support faculty. For patents, there has to be some form of patent agreement that supports a claim by an employer to a patent right. Some states have codified this condition in state law, to make the public policy explicit. For instance, California Labor Code 28270-2872 limits claims made by an employer to those that
(1) Relate at the time of conception or reduction to practice of the invention to the employer’s business, or actual or demonstrably anticipated research or development of the employer; or
(2) Result from any work performed by the employee for the employer.
I have bolded the key repeated emphasis–the invention has to relate to the *employer’s business*, to research *of the employer*, *for the employer*. It is not correct to substitute “university” for “employer” and “at” for “of” and think it’s all the same. It’s not. The use of “employer” in the California Labor Code here makes clear that the employer is specifying the work, has a business interest in the result, and directs the employee to perform work for the employer. Mere employment is not sufficient. Mere payment of a salary to support personally directed activity is not enough. And the state law makes it clear that it is against public policy, at least in California, Washington, Illinois, and other states that have implemented similarly worded laws, that simply changing an employment agreement to require assignment of inventions for whatever an employee might do while employed is not a legal, enforceable work-around.
Copyright law, which is drawn in by Bylaw 3.10, works in a similar way. While a “work made for hire” is defined by federal law as a work “prepared by an employee within the scope of his or her employment,” for the employer to be the author, the employer has to demonstrate control over the work being prepared. The US Supreme Court in Reid v CCNV argues that “scope of employment” is adapted from the common law of agency, and that common law must take precedence over any state law:
the general common law of agency must be relied on, rather than the law of any particular State, since the Act is expressly intended to create a federal law of uniform, nationwide application by broadly pre-empting state statutory and common-law copyright regulation.
The Supreme Court then examines four interpretations of what “within the scope of employment” might mean:
1) “a work is prepared by an employee whenever the hiring party retains the right to control the product”
2) “when the hiring party has actually wielded control with respect to the creation of a particular work”
3) “the term “employee” within 101(1) carries its common-law agency law meaning”
4) term “employee” only refers to “formal, salaried” employees”
The Court sets aside the first two approaches, arguing that control is not the determining factor. “Congress,” the Court argued, “meant to refer to a hired party in a conventional employment relationship.” We pause to note that university faculty are generally *not* in a conventional employment relationship, there is not the usual “master-servant” relationship in such agency. The faculty member does not do the bidding of the master. The faculty member is called a “member” with an “appointment” for a reason that goes way beyond the appearance of old language. The old rights to control work still live on.
The Supreme Court then lays out a number of factors to consider in determining whether there is an employer-employee relationship for the production of any given work (citations removed, formatted as a list):
In determining whether a hired party is an employee under the general common law of agency, we consider the hiring party’s right to control the manner and means by which the product is accomplished. Among the other factors relevant to this inquiry are
the skill required;
the source of the instrumentalities and tools;
the location of the work;
the duration of the relationship between the parties;
whether the hiring party has the right to assign additional projects to the hired party;
the extent of the hired party’s discretion over when and how long to work;
the method of payment;
the hired party’s role in hiring and paying assistants;
whether the work is part of the regular business of the hiring party;
whether the hiring party is in business;
the provision of employee benefits; and
the tax treatment of the hired party.
For university faculty, the skill, source of tools, location, right to assign, discretion, use of assistants, and regular business tests are regularly in their favor. Faculty are not required to produce any particular scholarly works, and the university does not assign or review those works. Indeed, the university as an “employer” is decidedly disinterested in the topic or substance of faculty work, other than that it is scholarly and notorious. The principles of tenure and academic freedom, built into nearly every faculty appointment, push these agency tests even further. The hiring party–the university–expects faculty to stay and once a faculty member has been granted tenure, the university declines to fire the member for failure to produce scholarship or seek grants. The penalty, if there is one, is that the faculty member might be denied promotion or access to special resources and may be assigned to undesirable office space or teaching schedules. There is no conventional master-servant relationship between a university and its faculty members over nearly the entirety of their scholarly production. Nor are they “independent contractors” commissioned to produce works, unless there’s some express written agreement for them to do so–as many university copyright policies still acknowledge.
Employment for copyright purposes must entail for each work a review of the circumstances–what is there that evidences that the employer has standing to be considered the author of the work–not simply having the right to control the work or actually controlling the work (neither of which a university normally exercises over its faculty)–but is the faculty member acting as a servant to the university, as an agent of the university, in producing any particular work? For most all faculty work, the answer must be “absolutely not”–not merely as a technicality, not as an oversight of paperwork or wording, but as a matter of public policy, as expressly laid out in the concepts of “member” and “appointment,” of tenure and academic freedom. When we touch on the second weird assignment in the Supplemental Information Form, we will come back around to this point.
Returning once again to the matter of employment for patents, we see that there cannot be two different versions of employment operating in one document concerned with patents and copyrights. Even if there were, how would anyone reading the document think to understand such a thing? Furthermore, it is a failure to argue that everyone can see what the university “intends” with its wording–that it must want to own everything it chooses to own. As a California court stated in a case involving the University of California’s patent policy:
The true intent of a contracting party is irrelevant if it remains unexpressed.
It does not matter if Dr. Sussman tells the Regents that the university owns faculty work. What matters is what is expressed in the agreement that is put in front of the faculty. Words matter. Words mean what they mean as a meeting of minds, not as a secret wish list of university administrators. That’s why it’s so pernicious that university patent administrators write policies as if they are contracts imposed on the faculty, and then include language that the university is the sole arbiter of the language they have written. The moment the University of Michigan puts the Supplemental Information Form in front of a faculty member to sign, it is creating an express contract, drawing in the Bylaws and patent policy. And as the Shaw court held:
We find no merit in the University’s suggestion that, as a public employee who is employed pursuant to statute, not contract, Shaw has no vested contractual right in his terms of employment, such terms being subject to change by the University.
If the university can change the terms, it’s not a contract. If it’s a contract, then the expressed intent is what matters, not the administrative desire to control the interpretation of the document. That’s why we read Bylaw 3.10 and the present assignment document carefully. Words matter.
Thus, if the present assignment restricts the scope of assignment to inventions made “as an employee,” that restriction means something–and it’s not “anything that someone might make while receiving a salary from the university.” That’s not patent law. That’s not copyright law. If the invention is made as “an employee” rather than “while also an employee,” then the university will have to show that it has directed the invention, has business involving the invention, that the invention was made for the university. Absent such a showing, an invention, by the words of the present assignment, is out of scope.
So where have we got to? That the present assignment document lays a claim that is outside the authority of Bylaw 3.10. That the present assignment restricts the scope of assignment to inventions made “as an employee” rather than, say, as an “appointee.” That the present assignment purports to claim patent rights in a novel form of invention–one that is not even potentially patentable because it may involve merely conception or reduction to practice. That the present assignment fails to obtain a swath of patent rights that Bylaw 3.10 claims. A strange, convoluted, inconsistent, ambiguous, goofy, incompetent instrument. There’s almost no way to establish what the university intends to have its personnel assign–there’s no clear definition of the inventions that are anticipated, no statement of work, no requirement for written agreements, nothing. It’s an assignment into the empty space of the future. Unlike a similar document used by a business, that has a line of business, that controls its employees, that expects employees to perform work as agents of the business, in a master-servant relationship, the university has nothing of this. The present assignment is legal-looking form that does next to nothing. What foolishness. Yet, the university has access to virtually unlimited legal budget to stick its interpretation to anyone who might challenge it. As one attorney who defends faculty told me, and I paraphrase, such cases are winnable–it just takes a couple of hundred thousand dollars to prove it on the university.
All this, and we have yet to look at the University of Michigan Technology Transfer Policy, which introduces all new muddle. At least, we can be relieved to know, Michigan is big on consistency in producing legal-sounding silliness. I just wish it was silly-funny and not silly-fascist. There’s absolutely no humor or goodwill operating in these policy documents.
As an added bonus, in a later article, I will look at the weird copyright reverse assignment in this same present assignment document.