The University of Michigan’s Mess of a Copyright Policy, Part III

Part I of this series looked at the University of Michigan 1944 patent policy and its transmogrified afterlife as Regents Bylaw 3.10, and the strange Supplemental Appointment Information invention present assignment document that claims to derive from Bylaw 3.10.

Part II moved to the Technology Transfer Policy, current from 2009, and worked through key paragraphs on purpose and ownership, showing that the morass of language reduces to the core of the 1944 patent policy, with all the bombast text–nigh unto unreadable–doing little more than adding ambiguity and asserting new powers for the Office of the Vice President for Research, expanding the scope of university claims on intellectual property (and non-intellectual property), and further asserting that the Office of the Vice President for Research does not have to follow its own policy. Nor does any other unit of the university that “in good faith” contracts with anyone for any other terms. Thus, HR can “in good faith” require a present assignment of invention rights from incoming faculty and staff, and ignore the Technology Transfer Policy.

The Technology Transfer Policy asserts that the fundamental claim of ownership, set forth in the first sentence of the policy’s discussion of ownership and derived from Bylaw 3.10, is expressly stated to be subordinate to anything else in the rest of the policy, and to any other policy of the university, including the copyright policy. The Technology Transfer Policy subordinates Bylaw 3.10 to the whims of administrators. Cheeky.

In this Part III, we will take up the University of Michigan Copyright Policy. Bylaw 3.10 adds a reference to copyright when it adapted the 1944 patent policy sometime after 1962 and before 1975. A reasonable reading of the 1944 policy, and Bylaw 3.10, is that the university claims ownership of patents and copyrights only when the patents and copyrights are acquired or issued (meaning “registered”?–subject to a formal governmental action–for copyright) by the university “as a result of or in connection with” an educational activity that uses university funds other than funds stemming from employment. That is, look for spending from an account other than payroll, then ask if the spending is for an educational activity, and then look whether patents or copyrights were acquired or issued in connection with that activity. There you have the Regents’ authorization for the university to own and manage such patents and copyrights.

The Regents’ Bylaws permit units of the university to deal with situations in which the Bylaws do not authorize an action at the level of the Bylaws and to deal with issues “of a less important nature” than those addressed by the Bylaws (see Section 14.01). The Technology Transfer Policy, however, asserts that it only need notify the Regents of changes, without distinguishing changes that are at the level of the Bylaws, such as ownership of patents and copyrights, and changes that are of less importance, such as the definition of Tangible Materials. Lazy? Clever? Incompetent? Why do we have to choose only one?

It’s important to recognize what happens when the Regents add copyright to the structure of the 1944 patent policy and elevate everything to a Bylaw. The university’s interest in copyright, whatever it may have been prior to the adoption of Bylaw 3.10, is restricted to copyrights acquired or issued in connection with educational activities supported by university funds and expressly not with matters of employment. “Educational activities” are not “anything within the scope of employment.” “Supported by University-administrated funds” is not “supported by University salary or other compensation for services.” This is a Regents-level policy, and it reads as a restriction on what otherwise might be claimed, were there no policy directive in place.

We have seen (Part I), that employment for patents and copyrights has specific implications. One cannot use “employment” loosely and generally and then assume that patents and copyrights fall within the expectations of this loose wording. That’s a convenient ploy for administrators, but it does not hold up in IP practice. The Supreme Court has provided guidance on both patents and copyrights. For patents, employment is not a sufficient condition for the employer to assert a claim of ownership. Per federal patent law, there is no “invent for hire” condition.

Nothing in a master-servant relationship (“employment”) provides the master with a legal right to own the inventions of the servant. There has to be an agreement on patents distinct from employment. It may be that such an agreement is made a condition of employment–that is, if one does not agree to the condition, then the master may release the servant from employment–but “condition of employment” cannot mean–it violates federal law to mean–that despite employment not being a sufficient condition, the master demands that the servant pretend that employment is a sufficient condition. Law forbids Uni from taking Proffy’s work, but Uni creates a policy that claims Proffy agrees to ignore the law.

The gesture in these university policies, as they create obligations for employees, whether contractual or not, is that private parties can contract to ignore federal law, or in the alternative, a state university can impose obligations that require its employees (and not employees) to ignore federal law. Adding back in something like “to the extent permitted by law” to a contract whose fundamental purpose is to ignore the law can only be to deceive the employee and to provide, if called out, a defense that the university didn’t really, technically, actually mean that the employee had an obligation to ignore the law, and that everything that was required was really something else, not expressed, but to be constructed by the court to make everything right.

If a university patent policy is a contract, then the Shaw case guidance is relevant:

  • a policy may be incorporated into a contract
  • employees have an interest in such a contract for employment
  • the university cannot change the contract at whim
  • the university’s true intent, if unexpressed, is irrelevant

Of course, if the patent policy does not form a contract, then all the talk about obligation to assign does not depend on agreement. The university may not even be able to fire an employee who refuses to assign, depending on other policies, such as on tenure, and state laws regarding employment. That is, any retaliation by university officials against an employee who resists a demand that has not been established in contract with the employee to assign his or her patent or copyright to university may also give rise to a claim for damages, if not sanctions against the officials personally. It’s a pity that university faculty are not paid better, so they have the financial resources to resist university officials who abuse their powers. It’s also a pity that state attorney general officers do not take an interest in administrative abuse of power and, at least with regard to intellectual property matters, appear to treat the administrators as clients, and other state employees as potential adversaries in court. Universities would be different places if law weren’t a chapman, with state attorneys taking the side of administrative money and power rather than justice.

For copyright, employment also has special meaning. Actually, employment has its common law meaning–that is, the Supreme Court in CCNV v Reid made clear that federal law supersedes state law (and any contracts formed under state law) and imposes common law rules of agency on the definition of “employment” for the purpose of construing the definition of a “work made for hire”–“prepared by an employee within the scope of his or her employment” (there is a second prong to the definition, for commissioned works prepared by an independent contractor–more on that later). One cannot merely ask, “Does the university define these people as ’employees'” and if so then apply work for hire conditions. Instead, one has to look to the common law of agency and a multi-point evaluation of whether the employer exercises control over both the work and the workplace.

It is not a sufficient analysis to find that an employer exercised control or had the right to exercise control; nor is it sufficient to find that the employer has a long-term, continuing relationship with one who produces work, and pays employment taxes on compensation. “Employment” for copyright requires common law of agency–is there a master-servant relationship for the purpose of creating the work in question? Is, for the creation of the work, the servant the agent of the master? Look at the circumstances, not assertions. Deal in specifics. Do not create a general case and impose general expectations on specific actions. Do not reason from the species to the individual. The logic must go the other way.

Similarly, “scope” is an important part of the definition of work made for hire. It is not merely that an individual is “employed”–the work must be within the scope of that employment. Where “scope of employment” gets a lot of attention is in civil liability. There, “respondeat superior” comes into play–the master answers for the actions of the servant. Liability law draws scope broadly. As an on-line legal dictionary puts it:

The scope of employment includes all acts reasonably necessary or incident to the performance of work, including matters of personal convenience and comfort that do not conflict with specific instructions.

Or, as Black’s Legal Dictionary puts it,

The reasonable and foreseeable activities that an employee engages in while carrying out the employer’s business.

The “scope of employment” for liability purposes, however, is not the “scope of employment” for copyright purposes. The “employment” in copyright is directed to the circumstances surrounding the work produced, not a general condition of being employed (for some work, not for all activities), and for being engaged in work to advance the employer’s business. Just as one cannot ask if a faculty member is “employed” by a university and then deduce that the faculty member is also “employed” to produce every bit of scholarly writing (or other such work); so also, one cannot ask if a university might be found liable for the actions of a faculty member, and if so then deduce that all works prepared by that faculty member must be within the scope of employment for copyright.

“Employment” is a general term. “Employee” is equally abstract, describing an individual in a state of employment. An individual may be an employee and also engage in other activities, even activities that are similar to those for which the individual is an employee. One may be employed, say, as a cook, and also cook at home or run a catering business on the side. There may be an issue of non-competition with the individual’s employer, but the employer cannot arbitrarily expand the “scope of employment” to include any cooking the individual does other than the cooking assigned for the benefit of the employer. Alternatively, if the employer does not want to have an interest in, or liability for, certain cooking that the employee might do, such as at company picnics, the employer and employee can agree to restrict the scope of employment to exclude those cooking activities.

The definition of work made for hire is a matter of federal law. One cannot contract to change the definition, to add works to the definition, or to avoid the definition when it applies. One can, however, shape employment and the scope of employment to place works within the definition (if otherwise by an employee) or outside of the definition. Employment may be negotiated. Scope may be negotiated. The definition of work made for hire is not.

Now let’s return to the second assignment in the Supplemental Appointment Information Form, HR 36100, [revised 9/2016, copy here–the IP language is the same] which has the following odd assignment statement, that the newly hired faculty or staff member is required to sign:

I hereby accept the transfer of copyright—from the University to me—of the scholarly works I author in accordance with SPG 601.28 and as it may be changed from time to time.

This is truly an odd statement. It is in the form of a present assignment from the University to the individual of copyrights in the individual’s “scholarly works.” Let’s parse it for a moment. First, assignments of copyright must be written instruments signed by the owner of the copyright.

A transfer of copyright ownership, other than by operation of law, is not valid unless an instrument of conveyance, or a note or memorandum of the transfer, is in writing and signed by the owner of the rights conveyed or such owner’s duly authorized agent.

Here, it is the recipient of assignment who is asked to sign. There is no signature by a representative of the university. Whatever the employee’s signature does, it does not effect the transfer of copyright from the university to the individual.

Second, the purported transfer is only of “scholarly” works. There is no guidance here what constitutes a “scholarly” work. In US copyright law, “scholarly work” is not a defined category of work, though “scholarship” appears as a purpose in discussions of fair uses of work protected by copyright. The implication, however, of the use of “scholarly” here is that all other forms of work are not transferred by the university. One might imagine all sorts of such other categories–works of humor and entertainment, technical documentation, software programs, artistic works and musical compositions, newsletter copy–all might be described as something other than “scholarly.”

Third, the transfer uses “author”–“in the scholarly works I author.” “Author” is a term used in copyright law–as a noun, not a verb. Though US copyright law does not provide a formal definition of “author” (thereby creating its own teapot of storms), an author  is generally one who prepares an original work of authorship and fixes it, or has it fixed under his or her authority in a tangible medium of expression. Writing loosely (as is apparently the University of Michigan practice in policy matters), one then would have that faculty and staff author certain works–they are the authors of those works. To the extent those are original works of authorship–then under US copyright law, they own the copyright in those works. This is true even if the authors themselves do not fix the works in tangible media of expression.

A sculptor may use assistants to prepare a work–instructing each in what to do, reviewing the work of each to ensure that it meets the sculptor’s idea of the work. The sculptor is the author of the work.

A writer may dictate her work to an assistant, who writes the work down and thus is the one who fixes it in a tangible medium. But the assistant’s own work product, the writing, is not original–the writer’s work is original. The writer is the author.

Similarly, a faculty member may instruct a staff programmer working for a university how to create a program for learning instruction, may review that program and determine whether it does what the faculty member has determined it should do. The faculty member is the author, not the programmer (or, there may be coauthorship, if circumstances so present). See Lindsay v Titanic for an instructive analysis of such situations (a videographer controls dive sequence photography taken of the wreck of the Titanic; the cameras are operated by employees working for hire for a company; despite the work for hire arrangements, the court finds that the camera operators are not preparing original work, nor is the company that hired them the author or owner of copyright in the video sequences–the videographer is, though never on a submarine, never operating a camera.

Authors author. When they author original works of authorship which are fixed in a tangible medium of expression, then they also become the owners of copyrights in those works. That is, for the Supplemental Appointment Information form, if a person authors scholarly works, then the only way the university comes to own copyright in those works is if the author, the scholar, has previously assigned those works to the university by means of a signed, written instrument. The apparent effect of the transfer offered in the sentence of the Supplemental Appointment Information form is that any scholarly work assigned to the university is immediately, by operation of this sentence, assigned back to the individual. Of course, because the university signs nothing here, nothing is actually assigned back, but one might argue that it creates for the individual an equitable expectation of ownership that might be enforced on a university, should it come to that. From another perspective, one could suggest the university means to quit claim an interest in scholarly works, and uses a vague and inoperative transfer mechanism to do that.

And all of this is modified by an ambiguous terminal constituent: “I author in accordance with SPG 601.28 and as it may be changed from time to time.” There are three readings. In one, the “in accordance” clause modifies the verb “author”–one authors in accordance with policy. In a second reading, the “in accordance” clause modifies “transfers”–the university transfers copyrights in accordance with policy. In the third reading, the clause modifies “accepts”–the employee accepts in accordance with policy the university’s transfer of copyright. That is, the employee, to comply with policy, is required to accept the transfer, not that the university is required to make it, or that the employee is limited in receiving the transfer of copyright in only those scholarly works that conform to the stated policy. A standard of interpretation is that modifiers modify those elements closest to them unless there is a clear indication otherwise. That would favor the first reading, which is the strangest. Of the other two, one can only guess at the intent. Folks appear not to care how one might read what they write, even when they are constructing what they hope will be an assignment clause covering the scholarly work of a major American public university.

What to do with “and as it may be changed from time to time”? If the policy can be changed at will, this is nothing more than an agreement to agree. What does that do to what’s purportedly “transferred”? Can the policy change so that thousands, nay, hundreds of thousands, nay, millions of copyrights pass to and fro at whim of administrative changes to a copyright policy? Apparently. How secure is a copyright transferred under this statement, made in this “supplemental” form–clearly not at all, given there is no signature on behalf of the university causing it to assign copyright and further given that the university reserves the right to change the policy and therefore potentially what the university does and does not assign. “Hereby transfers” claims present action on future works, but “as it may be changed from time to time” leaves everything uncertain about the future. Not giving something is simply selfish. Gesturing to give something but setting up clever language to cast the gift into doubt or cast the act of giving into doubt–that’s bad faith. Of course, if none of this is intended, one might ask, wonderingly, just what the heck the university does intend, and is anyone there competent to express that intent using language that might be understood by ordinary mortals?

So it is time to turn to SPG 601.28, and see what it has to say about copyright that prompts this sad, strange little sentence that all new faculty and staff hires are forced to accept.

The burden of SPG 601.28 is ownership of copyrights “by establishing a framework for who holds copyright at the University.” So far, so good. One might expect that the policy would then clarify that for faculty members, their scope of employment does not include the production of any original works of authorship unless the university and faculty member expressly agree in writing to so expand the scope of employment for a particular project. Many university copyright policies have some gesture in this direction. But no, we encounter a reference to a “legal default”–meaning not a failure (which, literally, it will turn out to be) but rather a requirement. Here it is (bold in original):

A. The Default: Under U.S. copyright law, the University holds the copyright (as “works made for hire”) in copyrighted works authored by its EMPLOYEES who are acting within the scope of their employment. Otherwise, the University does not hold copyright in a work, unless the copyright has been transferred legally to it by written assignment or other process of law.

Let’s work through this. The definition of “EMPLOYEES” is of no help, being circular:

EMPLOYEES means any people employed by the University of Michigan in any capacity, whether they are faculty, staff, administration, or students and whether they are employed full-time, part-time, or in a temporary capacity.

So, “employees” means anyone “employed.” All the definition does is provide a meditation on the various categories by which someone may be employed.

Our discussion above has prepared us to consider the statement here regarding work made for hire. It is true that a work made for hire is one that is prepared by an employee within the scope of his or her employment. It is true that the author of such a work is, by federal law, the employer. It is furthermore true that the author of a work is the owner of the copyright. The university, as author of works made for hire, owns copyright in those works. And yes, the university can obtain ownership of any other work only by operation of law or a written instrument signed by the copyright owner.

Now let’s look at the words used by the policy. The policy claims that the university owns works authored by employees. That’s not “prepared” by an employee–the language of the federal copyright law. If a university employee authors a work, that work is *not* work made for hire, because the author of a work made for hire is the university. The university authors the work. The university owns the work. But if the employee authors the work, it cannot be within the scope of employment or it would meet the definition of work made for hire and would be prepared, not authored, by the employee. Sigh.

The definition of work made for hire in the law is a “work prepared … within the scope of … employment.” “Within the scope” is adverbial, modifying “prepared”–the point at issue is whether the preparation of the work is within the scope of employment, not whether the employee is “acting” within the scope of employment. One can act within the scope of employment and prepare a work that is not within the scope of employment. One could be hired to write, for instance, but not to write a novel. One could then be acting within the scope of employment–to write–but not prepare a work within the scope of employment–no novels, please, we lack imagination. Perhaps it is a point too subtle to live. But why not just use the language of the statute? Is it sloppiness? Or do words mean things in formal documents?

The copyright policy then moves to offer a conditional present assignment (using the verb “transfer”) of copyright in scholarly works, and excludes some scholarly works from this transfer. So for works that are not within the scope of employment, the university has no ownership position but for some later written assignment. The worry of the policy is for works made for hire that are also “scholarly.” That is, somehow works that the university-as-employer has a master-servant relationship with employees but which also are “SCHOLARLY” works, the university wants to give copyright to the employees who prepared those works.

Much to do here. Let’s look at the definition of “SCHOLARLY WORKS” first. Things get bad. Brace yourself.

SCHOLARLY WORKS means works authored by FACULTY within the scope of their employment as part of or in connection with their teaching, research, or scholarship. Common examples of SCHOLARLY WORKS include: lecture notes, case examples, course materials, textbooks, works of nonfiction, novels, lyrics, musical compositions/arrangements and recordings, journal articles, scholarly papers, poems, architectural drawings, software, visual works of art, sculpture, and other artistic creations, among others, regardless of the medium in which those works are fixed or disseminated.

The definition of “scholarly works” is restricted only to faculty (a defined term–meaning faculty and some non-faculty–at least the university is consistent in screwing up definitions so words do not have their common meanings); students cannot write scholarly works, apparently, nor can post-doc researchers. We encounter again the verb “authored” rather than “prepared”–so the definition introduces its own idiosyncratic meaning rather than following the definition of federal copyright law. These works are ones that meet two tests–(i) the works are “authored” within the scope of employment (ii) “as part of or in connection with their teaching, research, or scholarship.” So scholarly works involve scholarship and non-scholarship (teaching and research being distinguished from scholarship–it’s not even “teaching, research, or other scholarship”).

The definition is of little help. So far we have that faculty can produce works of authorship within the scope of their employment and outside the scope of their employment. The university’s worry is that copyright in some of the scholarly works prepared by faculty within the scope of their employment should be owned by those faculty members and not by the university. Because employment for copyright purposes involves a multi-factor test for common law agency, it should be clear that university faculty–and students–and even some staff–are not working within the scope of employment when they teach, conduct research, and assist the public.

They are *employed* by the university for what are often called their “official duties”–they are not *employed* for the rest of their activities. Official duties are ones that the university directs, reviews, controls. Anything else is independently undertaken. Maybe within the scope of one’s “appointment”–something one is not prevented from doing, something that one may be expected to do, but not something that the university expressly contracts with a person to do. The person is an “employee” of the university for official duties, but receives compensation not just to do those things, but as a stipend to permit one to do other things as well, and not have to grub for money through a second job.

Again, imagine a state legislator. A state legislator is paid a stipend to perform official duties for the state. But the legislator’s writings are not work made for hire. The legislator may be described as “employed” because he or she receives compensation from the state for providing services to the state, and communications he or she sends and receives may be public records, but none of this requires that the state owns copyright in the legislature’s expression simply because that expression is connected with the work of representing the interests of the public in state service. Faculty operate in a similar way, as do students in their capacity as students.  Very little of what any faculty member does is within the scope of employment for purposes of master-servant relationships. The university is no master of the faculty member’s expression. The faculty member is no agent of the university’s corporate body. The faculty member is a “member” for these purposes, not an “employee.”

Now here’s the rub. The policy then goes on to provide an extensive list of examples of
“SCHOLARLY WORKS”–not “scholarly” works–that is, works that might be works for hire and also having scholarly attributes. The policy uses the defined term instead. Rather than using words with their common meaning to *exclude* these works from the definition, the list serves to *include* these forms of work within the defined term, “SCHOLARLY WORKS.” That is, it makes it appear that these works are always works for hire, ones made within the scope of employment, when the reality of the working relationship with faculty (and others) is that they are not within the scope of employment, within the scope of common law agency. The whole point of the faculty relationship with the university is that it is not a matter of common law agency!

Think of it this way. Faculty are great blue herons. The university provides a wetlands to attract them, where they hunt for food, rear their young, and stand majestically, for the awe and delight of the general public. It is just that with faculty, who are people, not birds, what they do is valuable to the public–teaching, research, assistance–and what they use of the university’s resources is a matter of negotiation. There is nothing, however, that requires a university to construe the contribution of resources–including funding or “pay” to be a matter of “employment” forming a master-servant relationship, so that the university is directing the faculty member’s work and that work is done for the university and not self-directed by, and for the benefit of the public, by each faculty member.

Or put it another way. Faculty are both employees and independent contractors and private parties. For anything that the university can assign them to do, and direct them in doing, and approve or reject what they are to do, and dictate where and how they do it, the university is the employer, and they work within the scope of their employment. For other things, faculty are independent contractors. For these things, the university and the faculty member enter into written agreements that specify what will be done, and who gets what as a result.

These are the funding arrangements anticipated by the 1944 patent policy, garbled but still extant in the Regents Bylaw 3.10, and that persist in many university copyright policies (but not any more at Michigan). And faculty members are of course also private citizens, provided with the freedom of the university. Where they do not serve the university as a master for official duties, and they are not working under a written agreement for a specific project, then they represent themselves, with no obligation to the university with regard to their work product.

This is the magic of the university–this freedom of capable people from control by the state, by any establishment, with regard to their work and their work products–call them scholarly, creative, inventive; call them workaday, utilitarian, crank, goofy. Doesn’t matter–the purpose is to assemble people who can teach, study, and help–and let them do what they do, with minimal constraint. This is not idealism. It is the condition that faculty (most of them anyway, I am willing to bet) understand. This is what is set out in university academic codes, in the governance of the faculty senate, in statements of academic freedom and tenure. It should take more than incompetent drafting of patent and copyright policies combined with administrative love of power without accountability to overturn a few hundred years of successful cultivation of faculty. But there is always the possibility of a generation that destroys what has benefited them. It may be that Bayh-Dole will be seen as an agent that furthered such a destruction, by attacking the core of faculty freedom, allowing the expansion of university claims to invention ownership well beyond federal funding and beyond patents to “everything”–just as Dr. Sussman imagined. Such a dark, dismal thought, and yet apparently, from the records, no Regent objected. They must have wanted such an outcome.

So look at what this list in the definition of “SCHOLARLY WORKS” does. It makes it appear that all these forms of works and uses of works are works made for hire, not that it is possible that in some circumstances the university could direct the production of a work and yet not want to own it as a work made for hire, though technically the relationship crosses over into that of a master and servant.

For some of the items in the list, it’s impossible to understand–“course materials” for instance. Materials introduced into a course may come from any direction–musical works, articles, art–what makes them “course” materials is merely that they are used for instruction in a course. Such materials may be subject to a fair use analysis, but as an illustration of the scope of work involved in a work made for hire claim, it’s hopeless vague and not illustrative of much of anything. It amounts to the claim that if something is used in a course, it must have been created as a work made for hire. I write a letter that I then bring into the classroom. It is now “course material.”

The effect of this policy statement is to make it appear that the scope of employment at the University of Michigan is broad enough to include all manner of material that would otherwise be authored entirely outside the scope of employment. The stated purpose of policy is to somehow respect academic freedom. But the action of the policy is to create exactly the circumstances it deplores in “The Default.” The policy creates a presumption of scope that is entirely opposed to the relationship that the faculty have with the university, or should have, but for such policy statements.

This apparatus of worry, default, definition of work to bring (or purport to bring) everything within the default, only to transfer it back is subject to conditions and exceptions. This is the meat of the policy, once the apparatus has done its horrific job.

The university doesn’t transfer the copyright back. It retains a broad right to “use” works. “Use” is a term one avoids using in an undefined way in copyright licensing. “Fair use” is given meaning in copyright law. That’s one thing. But to license another to “use” a work is nigh to a blank check to do anything–make copies, make derivative works, distribute, whatever. For all practical purposes, the transfer of copyright to the faculty is worse than offering joint ownership, since at least with joint ownership, the university would have an obligation to account to the other joint owner, the faculty member, for any profits made by the university in “using” the work. But with a free license to “use,” the university avoids even this joint accounting requirement. So it’s not much of a transfer.

The harsher point of the policy, however, has to do with its restrictions of scope:

a. that are authored as required DELIVERABLES under a sponsored activity agreement;

b. when that would put the University in violation of or conflict with an applicable contract or law;

c. that are specifically commissioned by the University or are created as part of an administrative assignment to, for, or on behalf of the University;

d. that are software under Regents Bylaw 3.10; or

e. that are or have been transferred to the University in a writing (other than the Regents Bylaw 3.10 acknowledgment, which FACULTY sign as a condition of employment).

Paragraph c reflects what would be otherwise the standard condition under which the university might acquire copyright in any work of a scholarly nature. That’s where scope of employment is articulated and negotiated. But you can see how the university has attempted to expand the scope of employment to be anything a faculty member might do, so that the generosity of the copyright policy is to transfer back some remnant.

Paragraph b provides that the university can prevent the transfer back merely be contracting otherwise.

Paragraph a introduces the idea of “DELIVERABLES”–but restricts those to works “that must be authored and delivered” rather than to those works that are discretionary, but if created are also deliverables. That is a big hole, given that the apparatus is set up as an auto-implemented system of present assignments and reverse present assignments. Perhaps paragraph b saves the day, but then why even have paragraph a, since if the sponsored research contract establishes what must be done and delivered, that covers it. Paragraph a presents as a restriction on paragraph b, and at best creates uncertainty, and at worst, fails to provide to the university what administrators might think they are writing words to obtain. Totalitarians think this way, when they have power. But negotiators cannot afford to do so, because they respect the meeting of minds as the foundation for enforceable promises.

Paragraph d misrepresents Bylaw 3.10. Or, one might say, provides an administrative reading that suggests an unexpressed assumption administrators make as to the meaning of Bylaw 3.10. Here’s 3.10 on software:

Computer software created by members of the university staff in connection with administration, research, or other educational activities supported directly or indirectly by funds administered by the university, regardless of the source of such funds, shall be the property of the university. Such computer software may be made available for use on a non-exclusive basis by those who pay appropriate charges to reimburse the university for the costs of development, distribution, and reproduction.

It’s just the first paragraph of 3.10, which concerned patents and copyrights, with “Computer software” replacing patents and copyrights, dropping “as a result of” (and so only “in connection with” and *not* as a result of, apparently) and dropping the misplaced parenthetical giving examples of indirect use of funds. The paragraph then adds a condition on software distribution that prohibits free distribution of software–and thus greatly restricts, by policy, the university’s participation in open source software distribution. Everyone (for instance) appears to ignore this, however, other than perhaps at the Technology Transfer Office. This second sentence also indicates a restriction on exclusive licenses–the Regents, at least, provide that software, if made available, is for a fee, and “on a non-exclusive basis.” One might argue the Regents are thus silent regarding exclusive licensing of software, and that exclusive licensing could be either free or for a fee. Going that direction, the Regents Bylaws mean little more than places where the Regents dabble in the affairs of the university, and actually have little to do with the principles or policies that the university follows.

However one worries the strange uncertainty of the second sentence of this Bylaw 3.10 paragraph on software, the first sentence reads parallel to that of the first paragraph–the claim for ownership is only for software that is made “in connection with” a specially funded activity, and not as a condition of employment. Again, Bylaw 3.10, paragraph 4:

Patents, copyrights, and property rights in computer software resulting from activities which have received no support, direct or indirect, from the university shall be property of the inventor, author, or creator thereof, free of any limitation which might otherwise arise by virtue of university employment.

There’s the scope of employment condition, being *disclaimed* in the Regents Bylaws. The “property rights” in computer software are, essentially patents and copyrights, depending on just what the “software” entails. To have a tangible property right to the media on which the software is recorded would take more than a statement in policy–one would expect to have purchased the media. To have a trade secret right to the software would mean having the power to prevent faculty from publishing, and that drives to abrogating academic freedom and tenure, which does not appear to be an objective that could be accomplished with the apparatus here in paragraph 4. But for the obfuscations of administrators–who show no competence with drafting, intellectual property, or policy formation–it is clear that Bylaw 3.10 concerns itself only with those patents and copyrights (including those in software) that the university has specially funded, but not by “virtue of university employment.” That makes abundant sense.

As an aside, this software paragraph was introduced in 1976, before it was at all clear that software was protectable by either patent or copyright. The Copyright Act of 1976 dealt with software, but it went into effect in 1978. Diamond v Diehr in 1981 is considered a turning point in allowing patents to recite software. So someone wanted the Regents to assert property rights in software where there was neither patent or copyright apparently available. Thus, the separate paragraph, and thus, also the words “property rights” without any apparent idea what these might be, except that these property rights weren’t patent or copyright. What possible “property rights” could they have been? Perhaps someone anticipated a new statute on software, independent of either patent or copyright, and wanted to be sure the university was ready for it, whatever it turned out to be. Given the position of paragraph 4, one might argue that the purpose of paragraph 2 was to ensure both that the university was authorized to acquire property rights in software, and that software was not subject to a claim of ownership by the university any greater extent than works subject to patent or copyright. Again, all of this would make sense. What the university administrators now construe Bylaw 3.10 to mean does not make sense, is inconsistent with the language used by Bylaw 3.10, and is inconsistent with the relationship between faculty and the university indicated by statements on academic freedom.

Here, for instance, is the University of Michigan statement of academic freedom, from the Faculty Handbook (my bold):

Academic freedom is the liberty that faculty members must have if they are to practice their scholarly profession in accordance with the norms of that profession. Academic freedom is not a term or a condition of employment; rather, it is based in the institutional structure of this and other universities and is fundamental to their common mission of promoting inquiry and advancing the sum of human knowledge and understanding. Although some aspects of academic freedom are also protected by the First Amendment to the United States Constitution, academic freedom exists, independent of any external protection, as a basic prerequisite for universities to fulfill their mission to our society. Academic freedom is most commonly vindicated by individual faculty members, but remains first and foremost a professional prerequisite of faculty members as a group.

The statement then enumerates four basic “freedoms,” the first of which is (again, my bold):

freedom of research and publication. Within the broad standards of accountability established by their profession and their individual disciplines, faculty members must enjoy the fullest possible freedom in their research and in circulating and publishing their results. This freedom follows immediately from the university’s basic commitment to advancing knowledge and understanding. Restrictions on research and publication should be minimal and unobtrusive.

The university is in no position to assert either a trade secret ownership of faculty work, nor to claim to own (and therefore control the publication or public use of) such work as a condition of employment or as consideration for employment or access to university resources, not with regard to patents, nor copyrights, nor property rights yet to be enacted into law.

In light of all this, we can then consider paragraph e of the copyright policy’s list of exceptions to the transfer of copyright:

that are or have been transferred to the University in a writing (other than the Regents Bylaw 3.10 acknowledgment, which FACULTY sign as a condition of employment).

The first part is clear enough–if a faculty member assigns a copyright to the university (now or in the past), then that copyright is not automagically transferred back by the policy. But that clarity is really murk, because a copyright in a work for hire never gets assigned to the employer. The employer is the author under the law. The employer owns outright. So the copyrights that would be transferred in a writing would not be copryights works made for hire (at least not the university’s works made for hire–they could be works made for hire in which the faculty member is the author and others have prepared the work under the faculty member’s authority). The upshot is that it makes no sense for this paragraph to exist. There are no works made for hire that could logically or legally exist in this category. There are no copyrights to identify as an exception. If the university required as a condition of employment that faculty assign copyright in all their work over to the university and the university would assign back everything that met some definition of “scholarly” things would be different. But the university doesn’t require such an assignment (and in any event, the assignment would be for things outside the scope of employment) and so this paragraph has no reason to exist.

But exist it does and it leaves its own red mark in the form of a parenthetical:

(other than the Regents Bylaw 3.10 acknowledgment, which FACULTY sign as a condition of employment)

If this statement refers to the Supplemental Appointment Information Form’s present assignment clause, things are wrong. First, that present assignment clause is not an “acknowledgement”–it is an assignment clause–garbled and indefinite and inconsistent with Bylaw 3.10 as it is. The clause does not “acknowledge” anything. It grants ownership up front in things that have not been made, subject to wild fantasies of administrators who cannot draft a policy to save their souls.

Nothing in Bylaw 3.10 makes assignment a condition of employment. Perhaps some other policy authorizes administrators to make assignment a condition of employment. I haven’t found such a policy. To exist, it would have to limit or abrogate the statement on academic freedom. At least as of 2010, university faculty had not got a memo about it. I suggest that administrators have just made up the obligation. It’s just a convenient fiction they have come to believe.

They present Bylaw 3.10 to mean faculty must assign everything made in any education activity, or made with the use of university resources, or in university facilities, whether the support is direct or indirect. All this despite the history of the language, despite the specific wording directed to patents and copyrights, despite the apparatus to define special funding and disclaim employment as the basis, despite the statement of academic freedom, despite the conditions of common law agency that attach to the meaning of “scope of employment” for copyright law, despite the requirement in patent law that “mere employment” (including master-servant relationships that might give rise to work made for hire) doesn’t create an ownership position for the employer in the inventions of the employee.

Despite all this, and despite elaborate, chaotic, garbled policies on technology transfer and copyright, it appears that the Human Resources folks have gone along with the idea that the university is within its authority to induce employees to believe that they must give up their inventions and copyrights and whatever else as a condition of employment, and to give those property rights up upfront, without clear definition, as if the Regents had required it.

If indeed giving up personal property rights in patents and copyrights was a condition of employment, the university might be upfront about it and make it part of offer letters, and not bury it in a document deceptively labeled “Supplemental Appointment Information.” The present assignment is neither supplemental nor information. Or, if the form’s title’s words *mean* something, then on the face of it, as any new faculty or staff member might reasonably be led to believe, the present assignment requested is indeed “supplemental”–ancillary, after the fact, not a condition of employment, just as the Faculty Handbook’s statement of academic freedom provides, just as the Regents Bylaws 3.10 paragraph 4 provides.

What a mess. What a killer of innovation.

In Part IV of this series, I consider the University of Michigan’s 1987 patent policy and arrive at the odd result that despite all the wording, most of the university’s IP management program rests on neither policy obligations nor contracts but on the voluntary choice of the faculty. The reason I propose for that choice I leave for the end of that article. But the result actually is more general that people might guess–most university patent policies are mess, and faculty follow them not because they are legally bound to, but because they choose to. Although the policy regime is claimed to be one of compulsion, the actual practice appears to persist from over fifty years ago–academic freedom and faculty choice.


This entry was posted in Bozonet, Freedom, IP, Policy, Present Assignment. Bookmark the permalink.