University Invention Fascism

Back in the day, “totalitarian” had a meaning more akin to “holistic”–the state should take care of each of its citizens.  But nice gestures are often vectors for dangerous pathogens.  Universities have adopted a similar set of claims with their revised intellectual property policies, claiming more than even for-profit companies are able to claim, and with less in the way of contracts to back up their assertions.  Without contracts, university administrators resort to threats, disciplinary actions, litigation–where they have nearly unlimited resources available, and faculty do not.

One class of university patent policies is just plain wrong about the Bayh-Dole Act. Another class is deliberately ambiguous, reciting language closer to the actual text of the Act, but leaving the reader to come to the intended, but unfounded, conclusion that the Bayh-Dole Act somehow gives university administrators total control over title to patent rights.  This class of policies use “retain” as in “retain title” to appear to mean “take” or “have the right to obtain” rather than “keep” or “resist a federal agency’s requirement that the inventors assign title to the government once the inventors have chosen to assign to the university.”

Consider, for instance, the University of Delaware’s treatment.  Here’s the primary paragraph in their patent policy:

Government-funded research

Inventions or discoveries that result from efforts financed wholly or in part by Federal funds will be treated in accordance with the provisions of Public Law 96-517, “The Patent and Trademark Amendments of 1980” known as the Bayh-Dole Act. This act provides that a contractor (University of Delaware) may retain full title, right, and interest in inventions made under contract with the government, with certain exceptions for unusual circumstances. Except as specifically superseded by provisions of a specific funding agreement, inventions and discoveries covered by this section will be treated as outlined in section B, above.

A proper statement of the Bayh-Dole Act would be something like “This act provides that, other than in exceptional circumstances, if a contractor (such as the University of Delaware) obtains title to an invention made with federal support, the contractor may retain title, provided that the contractor complies with various provisions of the patent rights clause in the federal funding agreement under which the invention was made.”  Or, more simply, “This act provides that if a contractor obtains title to an invention made with federal support, the contractor may choose to retain title, subject to certain requirements.”

The policy statement grows absurd with inclusion of the bolded text.  It is unlikely that any federal funding agreement will “specifically supersede” any given university’s patent policy.  What is true is that a federal contract generally always supersedes a non-federal contract.  What is true is that a valid contract of any sort supersedes a policy statement. Even if a policy statement is somehow incorporated into a contract with an employee, the university’s side of that contract (not the employee’s side) is subordinate to what the university agrees to with the federal government.  In some circumstances, a university may breach its contract with the employee by contracting with the federal government.

The impression left by the Delaware policy statement is that Bayh-Dole gives the university the right to take inventions from inventors.  This is not so.  The “section B” referred to by the policy paragraph above includes these two statements regarding disposition of inventions relative to Bayh-Dole:

Following this review and evaluation, the inventions, developments, or works will then:

. . .

3.  be offered to the Federal funding agency when the University initially elected to retain title under the Bayh-Dole statue but the invention and associated patent application(s) are no longer of interest to the University;

4. be released to the inventor if found not to be of interest to the University or any Federal agency which may have funded the research leading to the invention; or

5. in certain cases referred to brokers or outside entities with whom the University has contracts to assess commercial potential and interest in their filing and attempting commercial development.

Paragraph 3 makes it appear that “initially elected to retain title” has something to do with Bayh-Dole rather than with the standard patent rights clause, which is authorized by Bayh-Dole but contains requirements that are not in Bayh-Dole but are part of the deal that the university agrees to.  This paragraph then attempts to make sense of the conditions under which the government may obtain title.  These are set out in the standard patent rights clause at 37 CFR 401.14(a)(d):

(d) Conditions When the Government May Obtain Title.

The contractor will convey to the Federal agency, upon written request, title to any subject invention —

(1) If the contractor fails to disclose or elect title to the subject invention within the times specified in (c), above, or elects not to retain title; provided that the agency may only request title within 60 days after learning of the failure of the contractor to disclose or elect within the specified times.

(2) In those countries in which the contractor fails to file patent applications within the times specified in (c) above; provided, however, that if the contractor has filed a patent application in a country after the times specified in (c) above, but prior to its receipt of the written request of the Federal agency, the contractor shall continue to retain title in that country.

(3) In any country in which the contractor decides not to continue the prosecution of any application for, to pay the maintenance fees on, or defend in reexamination or opposition proceeding on, a patent on a subject invention.

I have marked some parts in bold.  Details.  It is not the invention that is of concern, but title to the invention–that is, the right to have a patent issue in the name of the owner of the invention.  The process is not one of “offering” the invention to the federal government, but rather the government requests title–which could be at any time–but the contractor agrees to convey title only in the conditions as set forth in paragraph (d).  The government learns of its rights, bar dates, and contractor actions by notice of the contractor.  The contractor reports inventions and bar dates (c)(1), and will notify the government if the contractor decides not continue prosecution of a patent application, pay maintenance fees, or defend a patent in reexamination or opposition (f)(3).   But other matters are not dealt with.  For instance, how would the government know that a contractor has failed to report a subject invention?  How would the government know that a contractor actually has obtained title when the contractor elects to retain title?  How would the government know that a contractor, having elected to retain title, has indeed filed patent applications?  In each of these situations, the situation depends on contractor compliance with the standard patent rights clause.

As with march-in procedures, here is another place in the implementing regulations that Bayh-Dole has been apparently designed to fail.  Rather than requiring notice of filing a patent application, or recording assignment in the patent office with a copy to the funding agency, or requiring inventors to notify the government directly, the standard patent rights clause leaves off all such communications.  It is an odd silence.  No company sponsor of university research would stand for it.  The result is, if a contractor fails to notify the government, and fails to file a patent application, then the government, if it ever learns of the non-compliance, will likely have no rights.  There likely will be no patentable invention, as a result of publication, first-to-file applications by others, or on-sale bars.

In any event, the university has no obligation to offer anything to the government:  it is entirely up to the government to request title in writing.  Why does it matter, in the context of the Delaware patent policy?  First, the Delaware policy posits a situation that is not required by the standard patent rights clause.  One might say, there is no “offering” in Bayh-Dole.  Offering is not compliant.  Second, the management of title under the standard patent rights clause is muddled and conflates two different things–how the federal government might obtain title, and the relationship between the university and its inventors with regard to title.  A university policy statement might reasonably set out what a university will do when it decides not to pursue or maintain a patent; but the policy must be careful with regard to stipulating how the university will comply with federal contracts.  If it gets this wrong, and the policy is part of a contract with an employee, then it breaches the contract with the employee to comply with the federal contract.

If this is all sounding like technical nonsense, then so is any purported contract with the employee.  If a contract’s language is just a welter of official sounding words that lack meaning, definition, or stability, so convoluted and buried in subclauses of clauses of implementations of interpretations–that they are there apparently as aspirations, intentions, or something that sounds important–then how again is there a contract?  Where is the meeting of minds?

The Bayh-Dole Act, implementing regulations, and standard patent rights clause are silent on a number of matters.  What happens if a university elects to retain title and so notifies the government, but actually never has received assignment of the invention?  That is, the invention may be a “subject invention” (by meeting the definition in the standard patent rights clause, combined with the inventors meeting the definition of “contractor” under the second sentence of the definition of “funding agreement” as a party to a funding agreement by substitution, assignment, or subcontract), but the inventor, not the university, may be the contractor.  In such a case, the university has no title to retain.

Now for a more difficult case.  Can a university re-assigns to the inventors a subject invention?  Here is the standard patent rights clause language (k)(1):

If the contractor is a nonprofit organization, it agrees that:

(1) Rights to a subject invention in the United States may not be assigned without the approval of the Federal agency, except where such assignment is made to an organization which has as one of its primary functions the management of inventions, provided that such assignee will be subject to the same provisions as the contractor;

So the short answer is “yes.”  A university can re-assign to the inventors, but doing so requires that the federal funding agency approve the re-assignment, or that the inventors create an organization that has as one of its primary functions the management of inventions.  The odd thing about the standard patent rights clause is that once inventors assign to the university, then any further assignment requires the “same provisions as the contractor”–meaning, apparently, the provisions that apply to the university.  If, however, the university does not receive assignment of the subject invention from the inventor, then when the university chooses not to pursue a patent application, there is nothing for the university to “offer” to the government, and nothing for the government to require of the university:  the invention is a subject invention (perhaps), but the contractor is the inventor, not the university.  The disposition of inventions in such circumstances are controlled by 37 CFR 401.9 as an expression of 35 USC 202(d).  That is, 37 CFR 401.9 functions as a standard patent rights clause specific to inventors.  That clause permits an agency to decide whether to request assignment of title from the inventors, or to permit the inventors to retain ownership.

Take a look again at what the Delware policy does with this situation:

4. be released to the inventor if found not to be of interest to the University or any Federal agency which may have funded the research leading to the invention; or

5. in certain cases referred to brokers or outside entities with whom the University has contracts to assess commercial potential and interest in their filing and attempting commercial development.

The policy has already attempted to address the situation in which the university obtains assignment of a federally supported invention that it no longer wants.  Now the policy turns to a different situation, but giving the impression that this new situation is the next in a succession of steps.  In the Delaware policy university, first the university decides not to pursue patenting, then offers title to the government, and if the government doesn’t want title, then the university comes to paragraphs 4 and 5 and can either “release” the invention to the inventor or “in certain cases” “refer” the invention to brokers for their management.   The sequence under the standard patent rights clause is, if the university does not obtain title, then the invention remains owned by the inventor unless the federal funding agency requests ownership.  There is no paragraph 4 or 5 choice for the university.

The critical piece that is missing in the above paragraph is the requirement in (f)(2) of the standard patent rights clause that the university require research employees to make a written agreement to protect the federal government’s interest in inventions:

The contractor agrees to require, by written agreement, its employees, other than clerical and nontechnical employees, to disclose promptly in writing . . . and to execute all papers necessary to file patent applications on subject inventions and to establish the government’s rights in the subject inventions.

This written agreement is one made under the funding agreement, not in anticipation of it. This written agreement, furthermore, is not directed at assignment of inventions to the university but in creating an obligation by which potential inventors commit to take personal actions to establish the federal government’s rights–that is, either granting a license or assigning title, and assisting in filing patent applications (such as signing the declaration and oath).

It is possible for a university to circumvent this (f)(2) requirement–to make it not function.  A university can demand to own all inventions made by research employees supported by federal funds.  A university can institute present assignments to force all such inventions to university ownership outright.  And a university can demand as a condition of employment that inventors sign all the necessary paperwork to allow patent applications to be filed.  All this a university can do privately and cover a lot of the ground of the (f)(2) agreement.

A lot of the ground, but not all of it.  There are two critical failings with this approach.  The first is that the university, in accepting federal funds, agrees to do (f)(2).  It agrees to do this and not to do other things that might have the same effect.  Having agreed to do (f)(2), it cannot substitute some other set of actions that it prefers, nor can it enforce on its research employees what it has agreed with the federal government to manage by other means.

The first failing may appear to be just one of swapping one action–requiring a written agreement–for a more convenient action–stipulating things in policy and constructing a set of arguments that the policy statement combined with employment creates a contractual obligation.  The second failing, however, is fatal.  The purpose of the (f)(2) agreement requirement is not to ensure that a university get ownership expeditiously of all inventions made with federal support.  The (f)(2) agreement has two purposes.  The first is to require the university as contractor to make potential inventors party to the funding agreement.  When the university, acting in compliance with a federal funding agreement, requires its research employees to commit to perform actions under that federal funding agreement, the university is delegating performance under the federal funding agreement to those employees.  This delegation is not simply threatening them with disciplinary action should they choose not to perform their duties as employees, nor expanding their duties to include whatever the funding agreement requires.  There is no need for a written agreement with employees if something is already covered in their employment agreement.  The writing is not a memorialization of what what has already been agreed to by employees; it is an agreement, with actual standing, that does something.  That is why it is a delegation to individuals not a work assignment to employees.

The scope of the required delegation is specific to a class of employees.  But the delegation itself is not part of the employment agreement.  The federal government is not requiring contractors to obtain assignment of inventions from their employees.  The federal government is requiring contractors to delegate to their employees as potential inventors specific obligations under the federal funding agreement.   This delegation performs two actions critical to the operation of the standard patent rights clause and Bayh-Dole in general.  The first is to enable the definition of “subject invention” from the moment an invention is made.  If a subject invention is an “invention of the contractor” and “contractor” is understood to be only the university, then there appears to be a gaping problem with the definition, since no invention then is a subject invention when it is made, because it is owned by the inventor, not the contractor, and therefore does not have to be disclosed or anything else.  University administrators, dense as mud and with moral compass spinning wildly, see this “gape” as an opportunity to impose “present assignments” so that, according to their plan, whenever an invention is made, it is automatically owned by the university, thus saving the definition in the law from bad drafting.

However, it appears that Lackner also saw the problem in drafting the implementing regulations for the Bayh-Dole Act in the form that Congress approved.   In the former Institutional Patent Agreements, on which the text of Bayh-Dole is largely derived, the government stipulated that inventors must assign to the university.  This requirement is not present in Bayh-Dole.  One reason for this is that universities operated on a variety of policies, and to make uniform a patent rights clause, one either had to force universities to adopt a policy of invention ownership that few had–most designated a patent management agent such as an affiliated foundation or Research Corporation–or one had to back off of the assignment requirement and let universities have their policies in whatever form they had them.  In the IPAs, there is a requirement for approved university patent policy.  In Bayh-Dole, noticeably, there is no such policy requirement.  A university may have no patent policy whatsoever.  There is no requirement in Bayh-Dole that a university have a patent policy.  Thus, to achieve a uniform patent clause, Lackner had to sacrifice a federal demand that universities adopt patent policies and that those policies stipulate that inventors must give over title to inventions to the universities that happened to host their federally supported work.

Thus, the “gape” in the law, which had to be dealt with in the standard patent rights clause itself.  The approach was elegant–add a provision, (f)(2), that does everything that a university patent policy might do, but leave the beneficiary of the performance unstated–it could be the federal government, the university, an invention management organization, whomever.  But once the research employees have made the written agreement, they are parties to the federal funding agreement in its patent rights clause and thus, by operation of the definitions of funding agreement and contractor, they are also contractors.  Thus, the definition of subject invention operates.  With (f)(2) in place, any invention made in the performance of a funding agreement is subject to the patent rights clause, and is a subject invention.  Inventors have an obligation to disclose the invention to their employer.  More importantly, no matter what the employer might do to screw things up, the federal government has the independent personal commitment of each inventor to establish the government’s rights directly, and to assist the government, if it turns out that the government has standing to acquire title to a subject invention or receive a non-exclusive license.

The only way to prevent the need for the operation of (f)(2) is to try to implement a circumvention that prevents anything of the sort from happening–which means, total control of the employee, to force instantaneous assignment and to force signing paperwork.  In the absence of (f)(2)–which is in its way a very liberal approach to the diversity of possible attitudes toward patents that universities (and other nonprofits, and small businesses) might have–universities adopted invention fascism.  To avoid (f)(2), university administrators represent as necessary the total institutional control of creative class personnel, such as faculty, staff, students, and collaborators, with regards to their inventive work.  The administrators represent this invention fascism as necessary to comply with federal law (i.e., Bayh-Dole).  What they do not explain is that invention fascism is only necessary because the universities have chosen to ignore compliance with (f)(2).  Instead of compliance, they institute a self-interested policy requirement they prefer, and ascribe the need for that requirement to the federal law.

It is almost as if the moment the standard patent rights clause was formalized, the word went out to ignore the (f)(2) arrangement and dupe university faculty into accepting universal institutional ownership of inventions as a requirement of Bayh-Dole.  That in turn justifies changing university policies to “reflect the law” and once the policies are changed, it doesn’t really matter what the arguments used were, the policies stand on their own–or so the argument goes.  If one is hoping to get rich dealing in faculty inventions, and the government opens up a whole new realm for speculative play, then the greedy thing to do, the shortcut in the name of efficiency and survival of one’s licensing office franchise, is to demand ownership of all these readily available inventions.  Why wait for faculty inventors to be interested in you, when you can be interested in them?  It’s like a near perfect Trotsky moment.

Returning again to the options of paragraphs 4 and 5 in the Delaware policy, one may now see that it’s all remarkably messed up.  The logic ought to be, if the university has no interest before filing a patent application, that it simply not take title and leave inventors to do what they wish.  Why shop an invention after the fact of university disinterest?  Why not, if one is going to shop for brokers, do that as part of determining university interest? After all, that’s how it was done by a number of universities–only take an interest if a qualified broker takes an interest.  If a broker doesn’t take an interest, then release any claim (to equity in the invention) and only agree to take an interest if the inventor asks the university to take an interest anyway, such as in exchange for specific support. Under the Delaware policy logic, what kind of broker could possibly want an invention that the university does not want and the federal government does not want?  It would seem to be a troll broker aiming to speculate on patent positions, not some honest plowman of a company seeking to invest resources to make a product to benefit the public.  It makes one wonder what “attempting commercial development” possibly could mean in the Delaware policy’s paragraph 5.

What a mess, to make it policy to reserve the opportunity to seek out such folks with faculty inventions, rather than releasing any claim on those inventions.  If something is better in the public domain, for instance, then it is simply cruel to try to find someone who would derive income from keeping that something from the public domain or from faculty inventors, and even nastier to want them to derive the highest income they can from doing so and share this income with the university, and it is offensive to think that the university might enlist faculty cooperation with this approach by sharing some of this same money with the inventors, so they can show off expensive cars and the like.

University Bayh-Dole practice is a mess.  A whole lot of university administrators and attorneys are unwilling to get it right or even to correct their policies after Stanford v Roche showed the policies to be wrong.  Instead, they are pursuing university invention fascism to cover over their wrong-headedness.  In the process, they are willing to attack academic freedom, the public domain, scholarly publication, research collaboration, and innovation.  People who should know better don’t speak out.  Such a shame that there cannot be, even, an open discussion of the matter without good people feeling they will lose their jobs and their careers.   University faculty could go a long way toward addressing the situation by engaging patent licensing personnel in dialogue, and assuring those that do speak out that they will have faculty support.  There needs to be some kind of reconciliation for thirty years of war on faculty inventions, waged by patent speculators against academic freedom and the public values of American universities.

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