Court to Employer: No Paper, No Assignment

By Dennis Crouch

Peregrine Semiconductor v. RF Micro Devices (S.D. Cal. 2014) (peregrine decision)

Interesting decision here involving patent ownership. Peregrine sued RFMD for infringement of several of its patents. However, during the course of the lawsuit the parties figures out that a former Peregrine worker – Robert Benton – should have been a named inventor on the asserted patents. Rather than siding with his former bosses, Benton instead assigned his rights to the defendant RFMD. The question in the case then is whether that assignment is proper or did the patentee already hold equitable title due to Benton’s employment.

Ownership of patent rights generally begin with the notion that the inventor begins establishing patent rights at the moment of conception. “The general rule is that an individual owns the patent rights to the subject matter of which he is an inventor, even though he conceived it or reduced it to practice in the course of his employment.” Banks v. Unisys Corp., 228 F.3d 1357 (Fed.Cir.2000). At the point of conception, the right is inchoate since no patent application has been filed or patent issued. However, courts typically allow assignment of that inchoate right through a written document. That rule could be seen in contrast to the traditional rule regarding other inchoate rights, such as a possibility-of-reverter, that could not be transferred inter vivos. In the 2011 Stanford v. Roche decision, the Supreme Court reiterated the basic law of inventor’s rights and the proposition that – absent an express assignment – an employer might not own its employees inventions even if those were accomplished during the course of the employment. “In most circumstances, an inventor must expressly grant his rights in an invention to his employer if the employer is to obtain those rights.” Bd. of Trustees of Leland Stanford Junior Univ. v. Roche Molecular Sys., Inc., 131 S.Ct. 2188 (2011) (citing United States v. Dubilier Condenser Corp., 289 U.S. 178 (1933)).

Peregrine made two arguments as to why such a transfer occurred and that it held at least equitable title to Benton’s patent rights: (1) a contractual promise to assign; and (2) the hired to invent doctrine.

No Paper à No Express Assignment: Courts have long specifically enforced contracts to assign patent rights – requiring a contracting inventor to follow through with that promise as opposed to requiring only the ordinary contract remedy of expectation damages. Peregrine’s problem is that it has no written evidence of such a contract formation. Peregrine’s business practice is to require all employees to sign an “Employment and Assignment agreement,” but Benton testified that he had no recollection of executing any such written employment agreement or assignment during his time working at Peregrine. Here, the court seemed to be swayed as well by the Patent Act’s statute of frauds that require any assignment of patent rights to be in writing.

A patent owner who seeks to assign his interest in the patent must do so in writing. 35 U.S.C. § 261; Sky Techs. LLC v. SAP AG, 576 F.3d 1374 (Fed.Cir.2009) (citing Akazawa v. Link New Tech. Int’l, Inc., 520 F.3d 1354 (Fed.Cir.2008)). Peregrine admits that it does not have any documentation of an Employment and Assignment agreement or Policy Manual signed by Benton.

Although I suspect that the court made the right decision here, the statute-of-frauds theory has some failings. First, the statute requires that assignments be in writing, but does not expressly require that contracts-promising-assignment be in writing. Further, courts can still enforce a “lost grant” despite the statute of frauds so long as there is sufficient evidence to prove that the grant existed. Here, however, that evidence appeared to be lacking.

What Exactly Was he Hired to Invent?: Peregrine also argued that Benton had a duty to assign his patent rights based upon the (Federal?) common law “hired to invent” doctrine as well as the California Labor Code § 2860. In California, the statutory provisions of § 2860 are seen as coextensive with the common law doctrine and applies when an employee is “hired to invent something or solve a particular problem.” The doctrine focuses on the specificity of the task assigned to the employee. As Don Chisum writes, “[t]he primary factor in finding an employment-to-invent is the specificity of the task assigned to the employee.” Here, the court cited a 1960s California decision distinguishing between work “narrowly directed by the employer towards the resolution of a specific problem” and work that is “generalized within a field.” With only the former creating an obligation to assign. See Banner Metals, Inc. v. Lockwood, 178 Cal.App.2d 643 (Cal.Ct.App.1960).

In this case the court found that Benton’s work appeared to be wholly within the field of semiconductor development, but generalized within that field. Importantly, the court noted that Benton worked on a variety of products during his employment and also spent time on non-inventing activities such as marketing and customer support. As such, the court ruled that Benton is unlikely to be bound by the hired-to-invent doctrine.

Hired-to-invent cases will always be somewhat squirrely since they are ordinarily raised only as a backstop when the employer’s written agreement failed and, as such, there is not likely to be written evidence particularly defining the reason why an individual was hired. This case fits that description.

Preliminary Injunction: Now, this case is only at the preliminary injunction stage. Peregrine had motioned for a preliminary injunction and that injunction was denied based upon the likelihood that Peregrine will not be able to prove its ownership. The denial also offers Peregrine the opportunity to immediately appeal this case to the Federal Circuit.

40 thoughts on “Court to Employer: No Paper, No Assignment

    1. 3.1

      I largely agree. However, there is some ambiguity in the statute itself.
      Section 261 governs transfer of “Applications for patent, patents, or any interest therein.” At what point does something become an application? And, what additional “interest therein” is the statute talking about?

      1. 3.1.1

        Well Dennis, consider a situation where employee develops a very viable invention for the company, but he has not agreed to assign the patent rights to the company. The company, for sake of argument, deems the invention so important that it decides to keep the invention a trade secret. Does the inventor have a right to go off by himself and file a patent application on the invention which would disclose the invention to the public thus destroying the trade secret?

        I think this issue is solely governed by state law because it involves the relationship between the employer and employee, the obligations inherent in maintaining the confidential information of the employer confidential, and trade secret law. I would be very surprised if federal patent law would intervene and say that the employee who does not have an obligation to assign patents to accompany nevertheless has a right to disclose to the public confidential information of the company in his patent application.

        Now regarding assignments, under traditional common law and most state law, until a patent application actually exists there is nothing to assign. An assignment of a future patent application creates an equitable obligation to assign that must be formalized by a legal assignment. Most state laws and the common law would also say a present assignment of a future obligation creates nothing more than an equitable obligation to create a legal assignment in the future.

        If such equitable assignments of future obligations were real legal assignments, they could be recorded and then future patent applications of the employee merely indexed against the prior agreement to assign. But as we know, there are state laws that limit the assignability of a inventions while employed by statute. California is one of those states. Thus not every invention made by an employee while employed can legally be assigned to the employer, by statute.

        At some point in time Federal Circuit has to recognize that state law governs the assignment of inventions and the assignment of patent rights. The US statute only requires that assignments of patents and applications be in writing, and provides for the effect of the recording statute. Everything else is governed by state law.



          You have thrown in an additional wrinkle without noting that you have done so.

          What is now also in the mix is any agreement signed by the employee as far as requirements on treating trade secrets.

          If there are no such agreements in place, your employer will most likely come out on the short end of the stick in trying to make something into a trade secret for which they lacked a prior agreement.

          Of course, as you and I both know (per my directions to you to brush up on Chapter 22 in Chisum), the state-by-state patchwork will make this area of law very very messy.

          Now I do see the possibility of federal law intervening as far as making sure the inventor – with that initial title – and lack of any prior agreements – since patent law is recognized as one of those areas of law that will preempt state law – at the very least at the “First Inventor” point (surely you notice the lack of state by state patent systems,right?).


            anon, even without a signed contract a confidentiality, I would think that the common law would impose upon the employee an obligation to maintain his employers confidential information confidential, including his trade secrets. Do you strongly disagree with this?


              I strongly disagree that such an unsigned provision trumps patent law.

              As may be evident to you though, I think we both see the writing on the wall of such (un)foreseen consequences of the AIA, and the attempted transitions in patent law to give greater power to corporations in an overall weakened patent system. The first step is to so weaken the patent system and switch the power to Trade Secrets. Once that happens, then it is an extremely easy step to change employment contracts to force capitulation to the Large Entity (right from the start). (Most) employees simply do not have leverage to say no. The overall (and not in-coincidentally) trend is to create a guild-like power centralization to large established entities – the very entities most threatened by what Christensen labels as disruptive innovation.


              Okay then, assume a signed agreement that the employee must keep the employers confidential mission confidential, and that the work product of the employee is the exclusive property of the employer.

              1. Who owns the invention?

              2. Can the employee disclose the invention to the public?

              3. Can the employee exploit the invention in a new company in competition with this employer?

              If you can’t do any of these, and I think you would agree that he cannot, what gives him the right to file a patent application and disclose his employer’s trade secrets to the public, to license his employer’s competitors, and to get into competition with his employer?



              You want to move the goalposts away from what I said at post…

              Well, as long as you realize that you have moved the goalposts, let’s take a look at your new situation.

              Your statement of “signed agreement that the employee must keep the employers confidential mission confidential, and that the work product of the employee is the exclusive property of the employerdoes not supply enough detail to answer your questions.

              What is the actual wording involved? As we have seen in the Stanford v. Roche case, the actual wording really does matter.

              What state are we in? I ask because even though patent law will trump state employment and trade secret law in those areas in which preemption reaches, the answers to your questions may very well be impacted by state law (or an absence of state law).

              As we have discussed in the past, the inventor/owner issues already exist in a quagmire of law. That was why I repeatedly stressed to you that you needed to be aware of what Chisum wrote in chapter 22 of his treatise.

              You seem to have a difficult time acknowledging my telling you this.

  1. 2

    From the case: “To state a valid patent infringement claim, all inventors must be listed on the patent. See 35 U.S.C. § 256.”


    May I quote §282(b): DEFENSES. – The following shall be defenses in any action involving the validity or infringement of a patent and shall be pleaded:

    (2) invalidity of the patent or any claim in suit on any grounds specified in Part II as a condition for patentability;

    and (3) that relate to section 112 and 251, and (4) any other factor or act made a defense by this title.

    As we know, paragraph(2) references §102/103 as they are the only sections in Part II that are specified as conditions for patentability.

    Thus, §256 does not state an independent basis for invalidating a patent. It specifies the way one might correct a patent to name an inventor that was omitted. The statutory basis for invalidity where an inventor is not named is §102(f) – which has been repealed by the AIA.But since his patent issued before the AIA was passed, §102(f) still applies and should have been cited by the court.

    1. 2.1

      Don’t forget that § 115 — Oath of Applicant — falls under Part II. Because an omitted co-inventor does not execute an oath, invalidity due to an omitted co-inventor seems to be an applicable defense under § 282.

      Even without the issue of validity, please remember that jurisdiction cannot be established without all owners of patent being included in the suit. Patent owners are necessary parties. Because an inventor is a presumed owner, the alleged infringer holding an assignment of an omitted co-inventor wins because the accuser cannot get all of the owners on his side. See Ethicon v. U.S. Surgical, 135 F.3D 1456 (Fed. Cir. 1998).

      1. 2.1.1

        West Coast Guy, even though §115 might appear in Part II, it is not a condition the patentability. There are only two sections in the statute that are identified as conditions for patentability. They are §102, and §103.

        Thus under the new AIA, with §102(f) repealed, omitting an inventor is not statutory grounds for invalidity. The inventor may file a lawsuit under §256(b) trying to be named as a coinventor, but I would think that he would bear the burden of proof. A smart defendant might seek out such a coinventor, as was the case here, obtain a license in advance, and then file a §256 action themselves, perhaps as a cross-complaint.

        Of course, in this case with the patent owner admitting that the omitted inventor should have been named, there is little doubt that the correction of the inventorship could be done through a proceeding in the patent office as opposed to a court proceeding. §256(a). But with the omitted inventor cooperating with the defendant, and without an assignment from that omitted inventor, the patent owner is up the proverbial sh%t creek without a paddle.


          Looks like a Rule 12(b)7 defense applies. Without all owners (i.e., necessary parties), the patent becomes unenforceable in a federal court.


            West, the omitted inventor does not own anything until he succeeds in getting himself named on the patent per §256.

            However, given that this was a preliminary injunction and given that the patent owner a conceded that the omitted inventor should of been named, the court solve the handwriting on the wall and new that the omitted inventor would be named, that he would become a co-owner, and that therefore the defendant would be licensed.




              Here’s what the Federal Circuit states in Ethicon (p. 1465-66):

              “[E]ach co-inventor presumptively owns a pro rata undivided interest in the entire patent, no matter what their respective contributions.”

              “[A] joint inventor . . . enjoys a presumption of ownership in the entire patent.”

              An omitted co-inventor owns the patent by presumption.



              The court in the instant case has gotten it wrong.

              Had the court considered Ethicon v. U.S. Surgical, 135 F.3D 1456 (Fed. Cir. 1998), it would have had to dismiss — especially since the existence of an omitted co-inventor has been admitted to.


              West Coast Guy, it does help to read the case and its facts before you give opinions that are blatantly wrong.

              In Ethicon , the omitted inventor had intervened and had prevailed on his § 256 action such that he was effectively named on the patent. It was this that gave him coownership. One actually has to go through the procedures or actions involved in §256 in order to be named in order to obtain coownership.



              Sorry. Not wrong.

              Inventorship presumes ownership. Since the parties in the instant case agree in Benton’s inventorship, the parties have agreed that he is a presumed an owner.

              The need to address the issue of ownership by assignment need not be addressed where both parties agree that Benson is an omitted co-inventor.

              The fact that Benson is not a named patentee means nothing with respect to ownership.


              A clarification….

              The court got it right in denying the preliminary amendment.

              The court did not go far enough. It should have gone further by dismissing the case due to the absence of an owner presumed by Peregrine’s concession that Benton was an omitted co-inventor.


              The fact that Benson is not a named patentee means nothing with respect to ownership.

              While I agree with WCG (Ned, you should read the SC case on Stanford again – paying attention to the notion that the inventor is a First Owner), had the fact pattern been slightly different, naming may have become an issue – that is, had the fact pattern involved assignments away from the original owners/inventors, then the facts of those assignments (such was the unnamed-on-the-application inventor nevertheless named on the assignment chain), may have become relevant.



              Peregrine’s concession that Benton is an omitted co-inventor is a concession that Benton is an owner.


              West Coast, this was a preliminary injunction.

              If this were a summary judgment under 256/standing, you would be right.

              But one does not actually own anything until he or she become a named inventor. This requires an action/procedure under 256.



              If one co-invents an invention, he or she owns (presumptively) the invention. regardless of his or her status being named a patentee.



              My answer to your questions is “I don’t know.” Your questions go way beyond the point that I’ve been trying to make.


            “But one does not actually own anything until he or she become a named inventor.”


            When Peregrine conceded Benton’s inventorship, it conceded Benton’s ownership without him being named.


              Conceded, yes. But still the formalities of 256 must be observed.

              There is a difference between legal and equitable title.

              Have you been to law school?



              Yes, I’ve been to law school. Have you?

              Why would a court burden itself with a § 256 hearing in an infringement lawsuit when it can dismiss the infringement claims due to the absence of an owner (i.e., a necessary party)?

              The dismissal of an infringement claim does not depend upon § 256 hearing when the existence of an omitted co-inventor has been conceded.



              A judge could say, “OK, patent owner, if you really want the formality of a § 256 hearing from which I can order the PTO to add Benton as a named inventor, I will schedule the hearing for sometime in June 2014. In the meantime, I am dismissing the infringement claim(s) today due to a lack of a necessary party (the conceded owner/omitted co-inventor).

              A § 256 hearing it not necessary to dismiss the infringement action.


              West, from 256: “The court before which such matter is called in question may order correction of the patent on notice and hearing of all parties concerned and the Director shall issue a certificate accordingly.”

              The court must make the order and the change in title becomes effective upon issuance of the certificate.

              Once he has made the order, he should dismiss the case — or stay it until the certificate has issued, then dismiss it. But for all practical purposes, the case is over, final, and appealable once the order to amend inventorship is issued and summary judgment of no standing granted.


            West Coast Guy, If one co-invents an invention, he or she owns (presumptively) the invention. regardless of his or her status being named a patentee.

            This is where concepts get dicey. I will agree that under the law, that the inventor initially owns the right to file a patent application and that he must assign that right to enable anybody else to own his right to file and obtain ownership of the eventual patent application/patent.

            But if one considers that the so-called invention might actually be embodied in work product, that includes know-how and specifications, prototypes, test data and the like, then I would think that his employer would own these. Would you agree? And if the employer wanted to keep this know-how secret, how could the employee obtain a patent without disclosing his employer’s trade secrets?


              Which came first Ned, the invention or the application of the invention?

              (Notwithstanding your desire to ignore federal preemption)


              The invention is made when it is reduced to practice.

              Commercial embodiments probably come later.



              Reduction to practice is not the only way Ned – Come man you know that a constructive reduction to practice occurs with the application itself.

              If you approached this with a sense of intellectual honesty, you would not have your hand caught in the cookie jar.



          As we have discussed Ned, things such as standing still apply.

          Just ask Stanford.


            Anon, of course they do. But only the named inventors and their assignees own the patent.


              You are tripping over yourself again with your inability to understand the dynamics in the inventor/owner situations.



              I would hearken you to re-read our early falsely-named/pirate not-true inventor discussions.

              As you now seem to implicitly accept my wisdom regarding Chisum’s chapter 22, I think that it would be prudent of you to reassess your views on this entire subject.

  2. 1

    Since the employer could not produce any written employment agreement, my suspicion would be that it might have bee destroyed by one of those often poorly thought out “document retention policies” that so many companies have, by which all documents are destroyed after a fixed term of years or after certain events in respective categories to reduce storage and litigation discovery burdens.
    P.S. Reliance on Stanford v. Roche per se may be risky, since [as one Sup. Ct. justice apparently noted] Stanford had failed to even argue that State law should apply to employment contracts, and that would apparently have given a different result.

    1. 1.1

      The failure to argue was more due to how the existing agreement had been penned, Paul, and less due to the state law itself.

    2. 1.2

      Everyone treats document retention policies as a giant inconvenience until the shredder truck, long gone, has destroyed the path to the employment agreement, the assignment agreement, and a host of other documents that shouldn’t be destroyed during the enforceable life of a patent but records management is so….boring and dreary.

      1. 1.2.1

        Yes, so boring that patent attorneys often leave document retention policies up to G.P. attorneys having no clue how many patent-related documents may be vitally needed for 20 or more years. Or, patent attorneys are not even consulted in the policymaking.
        [Also, some patent attorneys are still obsessed with laboratory notebook retentions, which, even when they still exist, are surprisingly infrequently the decisive factor in winning patent disputes, even in pre-AIA cases.]

Comments are closed.