Summary Contents
What is an intellectual property (IP) license? The answer to this seemingly straightforward question is far from obvious, and it has engendered no small amount of judicial hand-wringing and scholarly debate over the years. We are all familiar, of course, with the licensing agreement. A licensing agreement is a contract, whether oral or written, whether signed with a pen, affirmed by a handshake or assented with a click. And, as such, the rules of contract law apply – rules that have been developed over centuries of common law.
But a licensing agreement, according to some, is more than an ordinary contract, just as a rental agreement for an apartment is more than a mere contract. It conveys an interest in a property right. Thus, while a rental agreement is a contract, interpreted in accordance with the laws of contract, it also conveys a leasehold interest, a property interest that has an existence that is both dependent on, but also independent of, the contract that created it. That is, there are aspects of a leasehold that need not be written into a rental agreement, but which exist nonetheless – the result of even more centuries of common law development.
Similarly, we can talk about licenses separately from licensing agreements. While these two legal creations are often inextricably linked, they have separate qualities as well. In contrast to a licensing agreement, a license is an authorization to exploit some exclusive right that the law confers on the owner of IP. For example, under the U.S. Patent Act, “whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United States or imports into the United States any patented invention during the term of the patent therefor, infringes the patent” (35 U.S.C. § 271(a) (emphasis added)). The “authority” referenced in the Act is typically referred to as a “license” to practice the patent. A license “[i]n its simplest form … means only leave to do a thing which the licensor otherwise would have a right to prevent.”Footnote 1
The importance of treating a licensing agreement as distinct from a license is illustrated by the following example: Under a particular licensing agreement the licensor may grant a license under a copyright and under a patent. If the copyright license terminates for some reason (e.g., nonpayment of the royalty), the overall licensing agreement may continue, as may the patent license. Likewise, an agreement may terminate, yet a license granted under the agreement may be specified to continue in perpetuity after that termination.Footnote 2 The duration of a license and the licensing agreement under which it is granted need not be concurrent or identical.
But is an IP license a “property” interest, like a leasehold? The answer to that question depends on an even deeper question, which is the degree to which intellectual property itself constitutes “property.” That, too, is the subject of significant debate, and the answer may vary depending on the type of IP involved.Footnote 3
In this chapter we will explore some of the metaphysical issues surrounding the nature of an IP license – issues that sometimes have a very real effect on parties and transactions.
3.1 License Versus Ownership of IP
Perhaps the easiest way to begin to think about the nature of an IP license is to compare it to its counterpart – IP ownership (discussed in Chapter 2). Just as every lease requires a lessor, every license requires a licensor. The licensor of an IP right may be its owner, or it may simply be another licensee who is sublicensing certain of its rights (just as a lessee of real property may sublease the leased premises). But for our purposes, it is useful to compare the rights that an IP licensee has with those possessed by an IP owner – one who has come into possession of legal title to IP through creation or assignment.
Professor Ray Nimmer explained this distinction as follows:
Licenses are often contrasted with assignments of rights in information. The novice can think of an “assignment” as the equivalent of a sale of all rights in the intangible subject matter and not be far wrong. Commercial practice, however, frequently blurs the line between a license and an assignment. The fundamental difference is that, while licenses and assignments both focus on rights in, or use of, information, in an assignment the original rights owner tends to divest itself of rights in the subject matter, while in a license the transferor (“licensor”) retains more rights in the subject matter of the license. It can do this not only because the parties have agreed to a transaction that enables a split of ownership and use rights in the information, but because unlike hard goods, information can be both transferred and retained.Footnote 4
Table 3.1 offers a quick summary of the differences between the rights held by IP owners and licensees (both exclusive and nonexclusive).
Table 3.1 Rights in an item of IP (X) after a transaction between A and B
| After the transaction, which party (A or B): | Assignment | License (exclusive) | License (nonexclusive) |
|---|---|---|---|
| Owns X? | B | A | A |
| Has the right to exploit/use X | B | B | B/A |
| Has the right to prevent others from using X (i.e., enforcement)? | B | B/A | A |
| Has the right to grant further licenses to X? | B | B/A | A |
| May maintain rights in X? | B | B/A | A |
Thus, as shown in Table 3.1, after A assigns IP right X to B, it is owned by B, and B has all rights to exploit, enforce, license and maintain X. Moving to the next column, if A grants B an exclusive license with respect to X, its ownership remains with A, but the right to exploit belongs to B (see Chapter 6 for a discussion of the rights and obligations of exclusive licensees). The right to enforce, further license and maintain X, however, may vary based on the terms of the exclusive licensing agreement between A and B. In some cases, B may obtain the right to enforce X (see Section 11.2), to grant sublicenses of X (see Section 6.5) and to maintain X (see Section 9.5). In contrast, under a nonexclusive license both A and B have the right to exploit X, while only A has the right to enforce, further license and maintain X.
3.2 Covenant Not to Sue
Various courts and commentators have weighed in on the legal nature of an IP license. One view, exemplified by the Federal Circuit in Spindelfabrik Suessen-Schurr Stahlecker & Grill v. Schubert & Salzer Maschinenfabrik AG, 829 F.2d 1075 (Fed. Cir. 1987), is that a license is simply a covenant by the licensor not to sue the licensee for IP infringement under certain conditions:
[A] patent license agreement is in essence nothing more than a promise by the licensor not to sue the licensee. Even if couched in terms of “[l]icensee is given the right to make, use, or sell X,” the agreement cannot convey that absolute right because not even the patentee of X is given that right. His right is merely one to exclude others from making, using or selling X. Indeed, the patentee of X and his licensee, when making, using, or selling X, can be subject to suit under other patents. In any event, patent license agreements can be written to convey different scopes of promises not to sue, e.g., a promise not to sue under a specific patent or, more broadly, a promise not to sue under any patent the licensor now has or may acquire in the future.
Professor Chris Newman, however, has challenged the characterization of a license as a “covenant not to sue.” In the below excerpt, he compares licenses in the real property context to copyright licenses:
If I sell you an admission ticket to my theater, I take on a contractual duty to allow you to enter the premises at the appointed time and place (as well as possibly to provide specified entertainment of some sort). If, when you arrive, I bar your entry, I violate my contractual duty, and you have a claim for breach of contract. But suppose I do allow (i.e., take no steps to obstruct or forbid) your entry as agreed. May I nevertheless charge you with trespass on the ground that while our contract imposed a duty on me to let you come in, it could not and did not grant you any privilege to do so? The answer is no … Selling you an admission ticket would be understood by all as manifesting the intent to grant you permission to enter, and so it would effectively exercise my power as a titleholder to grant you that privilege. The privilege would thus result from the same acts that give rise to a binding contract, but it would not flow from or depend upon contract formation as a legal matter. The privilege would be valid even if some technicality of contract law (say, failure to comply with a statute of frauds) prevented the creation of binding contractual duties. Even in the face of such a failure, if you were to show up at the time designated on the ticket and enter the premises, you would not be trespassing unless and until I revoked the privilege by asking you to leave.
If, on the other hand, we were to take seriously the notion that a license consists of nothing but a contractual obligation not to sue, then my hypothetical would have real bite. Under this reasoning, even though I am bound by contract to let you enter the theater, you are still technically a trespasser when you do so – it’s just that I have a contractual duty not to bring a claim charging you as one. If this sounds absurd, the absurdity resides in the contract theory of license. After all, if you were not still a trespasser, it would be superfluous to speak of my having a contractual “duty not to sue” you – sue you for what? The notion that a license is a “contract not to sue” thus assumes the implicit (and correct) premise that contracts do not create privileges. Once a privilege has been granted on the other hand, there is no need for a contract “not to sue,” though there may be for a contract not to revoke the privilege.
Is there any practical difference between having a privilege to use my property and having a right not to be sued for doing so? Indeed there is. Suppose my contractual duty not to sue you for trespass is part of a larger agreement in which you undertake other duties to me, some of which are due to be performed soon after your bargained-for use of my property is complete. If you then refuse or fail to perform in such a way as to constitute a material breach of the agreement, the contract may be terminated, thus relieving me prospectively of my duties under it, including the duty to refrain from suing you. This means that even though your prior uses of my property took place while the license (i.e., the contract) was still in force, if they are still within the statute of limitations for trespass I am now free to sue you over them. The contract theory of license cannot explain or justify the rule that licensed actions taken while the license remains in force are forever immune from claims of infringement.
Nor is the contract theory of license reconcilable with the phenomenon, common in copyright law, of multiple co-owners, each of whom is empowered to grant nonexclusive licenses without the others’ consent.Footnote 5 Were such a license a contract, it would bind only the grantor and not other co-owners, who would remain free to sue for infringement. One might seek to explain this by theorizing that co-owners of the same work exist in some sort of privity such that a license agreement by one contractually binds the others, but clearly copyright law does not hold this to be the case. If it were, a single co-owner should be equally capable of granting an exclusive license binding on all other co-owners and rendering void any subsequent attempts of theirs to grant conflicting licenses. Instead, the law prevents the creation of conflicting exclusive licenses by holding that where there are co-owners, the power to create exclusive rights can only be wielded by all of them acting jointly.Footnote 6
Notes and Questions
1. Covenant or not? Why does the Federal Circuit in Spindelfabrik refer to a patent license as a “covenant not to sue”? Why does Newman disagree with this characterization? How does he conceptualize an IP license? Does it matter that Spindelfabrik dealt with a patent license, whereas Newman is largely discussing copyright licenses?
2. Nonproperty rights. One of the difficulties with a property-based characterization of the IP license is that it does not fully account for permissions that are granted with respect to intangibles that are not generally considered to be property – data, know-how, unpatented inventions and the like.Footnote 7 How would you reconcile Newman’s description with such licenses? Are they property interests? Or is a “covenant not to sue” a better description?
3. Future rights. Another area in which conceptualizations of IP licenses are challenged is future IP rights. Licensing agreements often purport to grant rights with respect to IP that is created in the future (see, e.g., Stanford v. Roche, discussed in Section 2.3, Note 3, and Aronson v. Quick Point Pencil, discussed in Section 24.3). Are these future grants merely contractual commitments to grant licenses in the future, or are they present grants of future interests, analogous to future interests in estates that exist with respect to real property? For example, can an easement exist across a road that has not yet been built? Or is a contract relating to such an easement merely a contractual commitment to grant an easement once, and if, the road is built?
3.3 The Governing Law of IP Licenses
Closely related to the legal nature of IP licenses is the question of which law governs such licenses. There are several possible choices:
the state common law of contracts
the state common law of property
the Uniform Commercial Code enacted in various states
federal statutory law governing certain licensed IP rights (e.g., patents, copyrights, registered trademarks and federal trade secrets)
state statutory law governing certain licensed IP rights (e.g., state trade secrets)
state common law governing certain licensed IP rights (e.g., common law trademarks and rights of publicity)
federal common law relating to IP licenses.
Though there is no single, clear answer to this question, a significant amount of ink has been spilled wrangling over it. It is complicated, of course, by the diversity of IP types, which have their origins in federal statutes, state statutes and state common law. Below are various perspectives on this difficult question.
3.3.1 State Common Law of Contracts
The District Court in Sun Microsystems, Inc. v. Microsoft Corp., 81 F. Supp. 2d 1026 (N.D. Cal. 2000) held that “[t]he rules of contract construction embodied in California law control the interpretation of the [License Agreement] to the extent that such rules are consistent with federal copyright law and policy.” This position is a common one: because a licensing agreement is a contract, and because contracts are governed by state law, then the relevant state’s common law of contracts governs the interpretation and enforcement of the licensing agreement.
Of course, any interpretation supplied by state contract law cannot be inconsistent with federal law that defines the licensed IP rights. Thus, for example, a state court could not hold that a copyright licensing agreement with a duration of fifty years is too long, given that the duration of copyright protection often exceeds that period. Likewise, a state court could not create a new standard for evaluating the scope of patent claims to determine which products are subject to a royalty obligation. But the interpretation of contractual terms, whether or not they deal with federally created IP rights, is generally performed under state contract law.
3.3.2 State Common Law of Property
A slightly different approach is taken by Judge Pauline Newman of the Federal Circuit, who suggests that it is not the state common law of contracts, but that of property that should be understood as governing IP licenses:
The jurisprudence governing property interests is generally a matter of state law. Even when the property is the creation of federal statute, private rights are usually defined by state laws of property. This has long been recognized with respect to patent ownership and transfers.Footnote 8
Professor Christina Mulligan, writing about software end user license agreements (EULAs, discussed in greater detail in Chapter 17), offers efficiency-based rationales to distinguish between a contractual and a property-based understanding of licensing agreements:
One large difference between contract and property is that the number of people involved in contractual and property relationships changes how much negotiation over rights and duties is possible. Where two individuals sit down to hammer out a unique agreement for services from scratch, the costs each of them must shoulder, in terms of time and resources, to understand their agreement will be about the same. Moreover, to the extent that their contract covers unique circumstances, both parties may have similar interests in negotiating highly specific or idiosyncratic terms that advance their preferences for how the contract will be performed. And because the contract terms primarily affect those who are party to the contract, their idiosyncrasies won’t impose information-cost burdens on others.
On the other hand, the transfer and form of property rights affects many people besides the owner of the property. As a result, property rights tend to be more standardized, because the existence of idiosyncratic property rights raises the costs of understanding their scope for third parties who must respect others’ rights.Footnote 9
3.3.3 The Uniform Commercial Code
Article 2 of the Uniform Commercial Code relates to sales of goods. In general, IP licenses are not sales of goods, but the extension of rights in intangibles. Thus, as a formal matter, Article 2 does not apply to IP licenses. As Professor Ray Nimmer has observed, “In most licensing agreements and court decisions on licensing law issues, Article 2 is irrelevant and never even considered.”Footnote 10
Nevertheless, the familiarity that many attorneys have with Article 2 leads almost irresistibly to comparisons and analogies between contractual terms relating to sale of goods and transactions in IP. For example, UCC definitions of “good faith,” “bona fide purchaser” and different forms of warranty routinely inform discussions of licensing agreements, both among attorneys and in judicial decisions.Footnote 11 Similar comparisons were made between sale and lease transactions, which led to the adoption in 1987 of Article 2A of the UCC pertaining to leases of personal property.
A similar effort – proposed UCC Article 2B – was initiated in 1995 with respect to license agreements. Yet, due to disagreements between consumer and software industry groups and within IP academic circles, Article 2B was never adopted.Footnote 12 Instead, it was released in 1999 as a free-standing uniform law called the Uniform Computer Information Transactions Act (UCITA), which was adopted in only two states, Maryland and Virginia.Footnote 13
3.3.4 Federal Common Law
First-year law students are taught that the concept of federal common law was abolished when the Supreme Court held in Erie Railroad v. Tompkins, 304 U.S. 64 (1938) that there is no “federal general common law.” Yet pockets of federal common law survive to this day in a range of areas including admiralty, antitrust and bankruptcy law, as well as some areas of IP licensing. One area in which federal common law directly affects IP licensing agreements is the assignment of licensing agreements, which is discussed in Section 13.3.2.
Professor Shyamkrishna Balganesh points to the work of Professor Richard Epstein in describing the federal common law tradition in intellectual property:
Intellectual property law, or the law relating to the delineation and enforcement of rights and privileges in informational resources, remains a prominent example here. Patent, copyright and trademark law in the U.S. are today seen as statutory areas that Congress alone is authorized to modify. Together with the complex rules of federal preemption, they purport to dominate the landscape of American intellectual property law.
Yet, hidden away in the interstices of these statutory areas is a rather robust body of law that applies common law ideas, methods and principles to various informational resources without running afoul of preemption concerns. “Common law intellectual property,” as it is often referred to, represents a set of legal causes of action that create various rights, duties, and enforceable liabilities for otherwise non-rival and non-excludable assets. Its hallmark lies in its common law origins, having been developed and adapted by judges in individual cases through the deployment of the common law’s core concepts and principles.Footnote 14
The following case illustrates how courts wrestle with these seemingly esoteric issues in a real-world dispute.
204 F.3d 343 (2d Cir. 2000)
LEVAL, CIRCUIT JUDGE
Background
According to the allegations of the complaint: Plaintiff Debra Bassett operates a business, Bassett Productions, that produces films and television programs. Defendant Mashantucket Pequot Tribe is a federally recognized Indian tribe with a reservation located within the geographical boundaries of the State of Connecticut. Defendant Mashantucket Pequot Museum is a Connecticut corporation located on the Pequot Reservation.
In October 1994, Bassett met with representatives of the Tribe to discuss the possibility of producing a film for the Museum about the Pequot War of 1636–38. In November, Defendant Theresa Bell, acting individually and as a representative of the Tribe, signed a “confidential disclosure agreement” in which she agreed that all information received from Bassett Productions was proprietary, and was to be returned to Bassett Productions at its request. In May 1995, Defendant Jack Campisi, communicating with Bassett on behalf of the Tribe, advised her that the Tribe intended to hire her to produce the film, contingent on the negotiation of a satisfactory contract and the Tribe’s acceptance of a script for the film.
In August 1995, Bassett Productions entered into a letter agreement with the Tribe (the “Letter Agreement”) for the development and production of a film about the 1636–38 Pequot War. The Letter Agreement identified Bassett Productions as the “Producer” and the Tribe as the “Owner,” but did not define these terms. It stipulated that Bassett Productions would “hire and supervise the development and writing of a screenplay by Keith Merrill and George Burdeau,” and that the Tribe would “compensate” Bassett Productions for development costs according to an agreed schedule. It also stipulated that “at such time” that the Tribe approved the final draft of the screenplay, Bassett Productions would have exclusive rights to produce the film for exhibition at the Pequot Museum.
Some time before October 30, 1995, Bassett had delivered to the Tribe a script that she herself had written based on a “script scenario” she had developed with assistance from her associate Allan Eckert. The script was prominently marked on its first page, “Copr. 1995 Bassett Entertainment Corporation.”
On October 30, 1995, Bassett received a notice from the Tribe terminating the Letter Agreement. The notice asserted that Bassett had not “perform[ed] the contract as the parties anticipated.” Following the termination of the Letter Agreement, the Tribe continued to pursue the development and production of a film on the 1636–38 Pequot War for exhibition at the Museum. In October 1996, filming was completed on a motion picture entitled, “The Witness.” Bassett asserts the Tribe intends to screen the film at the Museum “in the near future” as part of “an interstate-driven tourist attraction.”
In September 1996, Bassett commenced this lawsuit in the United States District Court for the District of Connecticut. The complaint sought an injunction as well as other copyright remedies on the ground that the Tribe and the Museum used Bassett’s copyrighted script without her consent or license in order to produce their own film; it further alleged that they breached the Letter Agreement, and that they committed various state-law torts resulting in injury to Bassett … The district court granted Defendants’ motion to dismiss the complaint, and Bassett appealed.
Discussion
28 U.S.C. § 1338(a) states that federal district courts “shall” have exclusive, original jurisdiction “of any civil action arising under any Act of Congress relating to … copyrights.” It is well-established that not every complaint that refers to the Copyright Act “arises under” that law for purposes of Section 1338(a). See, e.g., T.B. Harms Co. v. Eliscu, 339 F.2d 823, 824 (2d Cir.1964) (Friendly, J.) (noting that this principle traces to “precedents going back for more than a century”). In particular, “the federal grant of a … copyright has not been thought to infuse with any national interest a dispute as to ownership or contractual enforcement turning on the facts or on ordinary principles of contract law.” Here, the district court, relying on our discussion in dictum in Schoenberg v. Shapolsky Publishers, Inc., 971 F.2d 926, 932–33 (2d Cir.1992), dismissed the claims based on the conclusion that Bassett’s “copyright infringement claims … do not ‘arise under’ federal copyright laws for purposes of 28 U.S.C. § 1338(a), but are merely incidental to [her] state law [contract] claims.” Bassett contends that the court erred in dismissing her claims on the basis of Schoenberg. She argues that her copyright claims neither depend on nor result from claims for breach of contract. She further maintains that, because she sought a remedy expressly granted by the Copyright Act, her copyright claims do “arise under” the Act pursuant to the rule of T.B. Harms.
Figure 3.1 The Mashantucket Pequot Museum & Research Center in Ledyard, Connecticut, commissioned a film about the Pequot War of 1636–38.
Whether a complaint asserting factually related copyright and contract claims “arises under” the federal copyright laws for the purposes of Section 1338(a) “poses among the knottiest procedural problems in copyright jurisprudence.” 3 Melville B. Nimmer & David Nimmer, Nimmer On Copyright § 12.01[A], at 12–4 (1999). Such claims characteristically arise where the defendant held a license to exploit the plaintiff’s copyright, but is alleged to have forfeited the license by breaching the terms of the licensing contract and thus to infringe in any further exploitation.
Prior to our landmark decision in T.B. Harms, several district courts in the Second Circuit resolved the issue of jurisdiction under Section 1338 for “hybrid” claims raising both copyright and contract issues by attempting to discern whether the copyright issues constituted the “essence” of the dispute, or whether instead the copyright issues were “incidental to” the contract dispute. That approach, however, left a class of plaintiffs who suffered copyright infringement bereft of copyright remedies. Plaintiffs whose federal lawsuits were dismissed for lack of subject matter jurisdiction on the ground that their copyright claims were “incidental to” their contract claims had no way either to obtain an adjudication of infringement or to obtain relief provided by the Copyright Act, because the Act confers exclusive jurisdiction over copyright claims on federal courts.
In T.B. Harms, Judge Friendly recognized the complexity of the problem of defining when a case “arises under” the Copyright Act. In synthesizing Supreme Court cases that had considered the issue of federal jurisdiction in a variety of contexts, Judge Friendly established a test for this circuit that focused on whether and how a complaint implicates the Copyright Act.
Judge Friendly began his analysis by examining Supreme Court precedent addressing the question when a federal court properly exercises jurisdiction under Section 1338, which creates jurisdiction in the federal courts in “any civil action arising under any Act of Congress relating to patents … [and] copyrights,” among others. He identified two lines of authority as particularly important. First, in American Well Works Co. v. Layne & Bowler Co., 241 U.S. 257, 260 (1916), Justice Holmes explained that a “suit arises under the law that creates the cause of action.” According to Judge Friendly, Justice Holmes’ interpretation of Section 1338 explained the exercise of federal jurisdiction “in a great many cases, notably copyright and patent infringement actions, both clearly authorized by the respective federal acts, and thus unquestionably within the scope of 28 U.S.C. § 1338.” Judge Friendly observed that “in the many infringement suits that depend only on some point of fact and require no construction of federal law, no other explanation may exist.” Second, Judge Friendly discussed Smith v. Kansas City Title & Trust Co., 255 U.S. 180 (1921), in which the Supreme Court held that a claim created by state law might still “arise under” federal law “if the complaint discloses a need for determining the meaning or application of such a law.”
Synthesizing the Supreme Court authorities, Judge Friendly concluded that a suit “arises under” the Copyright Act if:
(1) the complaint is for a remedy expressly granted by the Act, e.g., a suit for infringement or for the statutory royalties for record reproduction; or,
(2) the complaint … asserts a claim requiring construction of the Act
Figure 3.2 Chief Judge Henry T. Friendly, the author of the opinion in T.B. Harms, served on the US Court of Appeals for the Second Circuit from 1959–1986.
As the suit in T.B. Harms did not fall within any of these enumerated categories, the court found that it did not “arise under” the copyright laws for purposes of Section 1338 and that jurisdiction was therefore lacking.
The T.B. Harms test differed significantly from the essence-of-the-dispute or merely-incidental test. The analysis under T.B. Harms turns on what is alleged on the face of the complaint, while the essence-of-the-dispute or merely-incidental test looks rather at what defense will be proffered. For example, if the complaint alleges copyright infringement or seeks an injunction under the Copyright Act, under T.B. Harms the federal court has jurisdiction; under the other test, in contrast, the court must ascertain whether the defendant will defend only by reference to state law matters, such as a claim of contractual entitlement, or will raise defenses based on the Copyright Act.
The T.B. Harms test avoids problems that result from the essence-of-the-dispute test. By rejecting reliance on whether the copyright claim could be characterized as “incidental” and instead focusing the inquiry under Section 1338 on whether a plaintiff’s complaint “[was] for a remedy expressly granted by the Act,” T.B. Harms ensured that plaintiffs who sought copyright remedies that depended on a prior showing of contractual entitlement would not be left without the remedies promised by the Copyright Act. T.B. Harms also obviated the need for courts to determine at the outset of litigation whether copyright claims were incidental to contract claims – a difficult determination to make even after discovery and trial, and one that cannot be made reliably on the basis of the complaint alone.
Judge Friendly’s solution to the problem posed by Section 1338 has been widely admired by the leading copyright scholars. The T.B. Harms test has been adopted by all the circuits that have considered the question whether a suit arises under the Copyright Act for purposes of Section 1338, if the disputed issues include non-copyright matters.
Nearly thirty years after the T.B. Harms decision, a panel of this court in Schoenberg undertook in dictum to state the test for determining the existence of Section 1338 jurisdiction in cases alleging violations of the Copyright Act resulting from breach of contract. The plaintiff, an author, alleged that he had licensed the defendant, a publisher, to publish plaintiff’s work. The license obligated the defendant to publish within six months of plaintiff’s delivery of the manuscript, to promote and market the work, and to license foreign language editions. According to plaintiff’s allegations, the publisher breached numerous obligations of the license. As a result of these failures, plaintiff claimed that the license was terminated and that defendant’s further publication of the work constituted an infringement. Although the appeal related to a different issue, the opinion undertook to state “the appropriate test under the T.B. Harms paradigm, for determining whether a suit ‘arises under’ the Copyright Act when it alleges infringement stemming from a breach of contract.”
The opinion acknowledged that [i]n T.B. Harms, Judge Friendly wrote that, “an action ‘arises under’ the Copyright Act if and only if the complaint is for a remedy expressly granted by the Act,” and that “[b]ecause Schoenberg is seeking damages for the alleged infringement as well as an injunction against future infringements, his complaint on its face asserts a claim ‘arising under’ the Copyright Act.” It observed, however, that notwithstanding the T.B. Harms formulation, some district courts had “looked beyond the complaint in order to determine whether the plaintiff was really concerned with the infringement of his copyright, or, alternatively, was, in fact, more interested in” free enjoyment of his property or other non-copyright issues. Other courts, it noted, had adopted the even “broader proposition that no claim arises under the Copyright Act whenever an infringement would necessarily result from the breach of a contract that licensed or assigned a copyright.”
In undertaking to reconcile the varying approaches of those district court opinions (and perhaps concluding that the authority of T.B. Harms extended only to disputes over copyright ownership and not to hybrid copyright/contract claims), Schoenberg created a new, complex three-step test; the first step of the test was precisely that which T.B. Harms had rejected – whether the claim for copyright remedies is “merely incidental” to a determination of contract rights. The opinion declared that in hybrid copyright and contract cases Section 1338 jurisdiction should be analyzed in the following manner:
A district court must first ascertain whether the plaintiff’s infringement claim is only “incidental” to the plaintiff’s claim seeking a determination of ownership or contractual rights under the copyright … If it is determined that the claim is not merely incidental, then a district court must next determine whether the complaint alleges a breach of a condition to, or a covenant of, the contract licensing or assigning the copyright … [I]f a breach of a condition is alleged, then the district court has subject matter jurisdiction … But if the complaint merely alleges a breach of a contractual covenant in the agreement that licenses or assigns the copyright, then the court must undertake a third step and analyze whether the breach is so material as to create a right of rescission in the grantor. If the breach would create a right of rescission, then the asserted claim arises under the Copyright Act.
We believe for a number of reasons that the Schoenberg test is unworkable. At the outset, it overlooks that, because the Copyright Act gives federal courts exclusive jurisdiction to enforce its provisions, a plaintiff who is denied access to a federal forum on the theory that his copyright claims are incidental to a contract dispute is thereby absolutely denied the benefit of copyright remedies. Such a denial of copyright remedies undermines the Act’s capacity to protect copyright interests. A plaintiff with meritorious copyright claims and entitlement to the special remedies provided by the Act is deprived of these remedies merely because the first hurdle of proving entitlement is a showing of a contractual right.
A second problem with the Schoenberg test is that it is vague. Schoenberg characterizes the first part of its test in two ways: whether “the ‘essence’ of the plaintiff’s claim” is in contract or copyright, or whether the “infringement claim is only ‘incidental’ to the plaintiff’s claim seeking determination of ownership or contractual rights under the copyright.” The meaning of either of these phrases is difficult to discern. At one juncture, Schoenberg suggests that the focus of inquiry should be on the plaintiff’s motivations (“whether the plaintiff was really concerned with the infringement of his copyright or, alternatively, was, in fact, more interested in whether he would be allowed to enjoy his property free from the contract claims of the defendant”). District courts applying the “only incidental” test, in turn, have construed it in various other ways.
Furthermore … the Schoenberg test requires the court to make complex factual determinations relating to the merits at the outset of the litigation – before the court has any familiarity with the case. Ascertaining what are a plaintiff’s principal motives in bringing suit, and what issues will loom largest in the case, may well require extensive hearings and fact finding. The need for such fact finding recurs at each stage of Schoenberg’s three-step formula. Thus, if a court finds that a copyright claim is not “merely incidental to” a contract claim (step one), it must still determine whether the contractual term alleged to have been breached was in the nature of a covenant or a condition (step two). And if it finds that the alleged breach was of a covenant, the court must next determine “whether the breach is so material as to create a right of rescission,” failing which the case must be dismissed (step three). This third inquiry in particular, which entails an assessment of the importance of the particular covenant, as well as the seriousness of the breach, raises questions that are not appropriately, easily or reliably answered at the start of litigation.
“When a complaint alleges a claim or seeks a remedy provided by the Copyright Act, federal jurisdiction is properly invoked.”
For the reasons discussed above, we conclude that, for claims of infringement arising from, or in the context of, an alleged contractual breach, this circuit’s standard for determining jurisdiction under Section 1338 is furnished by T.B. Harms, and not by Schoenberg. When a complaint alleges a claim or seeks a remedy provided by the Copyright Act, federal jurisdiction is properly invoked.
Applying the T.B. Harms standard to this case leads us to conclude that Bassett’s copyright claims “arise under” the Copyright Act for purposes of Section 1338. Unlike the complaint in T.B. Harms, the complaint in this case alleges that the defendants, without authority, used plaintiff’s copyrighted script to produce a new film intended and advertised for imminent exhibition. The amended complaint alleged copyright infringement and sought “a remedy expressly granted by the Act,” specifically, an injunction against further infringement of Bassett’s copyrighted script. Because the complaint alleges the defendants violated the Copyright Act and seeks the injunctive remedy provided by the Act, under the rule of T.B. Harms, the action falls within the jurisdictional grant of Section 1338. The district court’s contrary holding was in error.
Notes and Questions
1. Contract versus property. The question of whether state contract or property law governs IP licensing agreements reflects the debate discussed in Section 3.2 over the nature of IP licenses themselves. How would conceptualizing a license as a “covenant not to sue” impact governing law?
2. The legacy of UCITA. After the early 2000s, little was said about UCITA or the effort to develop a consistent national body of law for IP licenses and licensing agreements. Should legislative efforts in this area be restarted? How has the law of licensing developed without such a uniform code?
3. Arising under. How do the tests under T.B. Harms and Schoenberg fundamentally differ? How often do you think these different tests would result in different outcomes? Could the court in Bassett have reached the same result using the Schoenberg test? The effect of applying the T.B. Harms test may be to authorize federal jurisdiction in many more cases and thus remove those cases from state courts. What is the impact of such a shift?
4. Contractual override? Section 11.3 discusses contractual provisions by which parties specify the law that they wish to govern their licensing agreements. If an agreement contains such a clause, are the issues discussed in this section relevant? How do you think these legal principles interact with contractual preferences of parties? How often do you think parties select “Federal common law” to govern their licensing agreements?
3.4 Obligation as Condition Versus Covenant
When a license is granted, the licensee obtains the right to perform particular acts under specified rights held by the licensor in designated fields of use. These define the “scope” of the license (which is discussed in greater detail in Chapter 6). If the licensee performs some activity outside the scope of the license, but which still infringes the licensor’s intellectual property rights, then the licensee is an infringer. The license does not grant the licensee any rights outside the scope of the license, and the licensor is within its rights to sue the licensee for infringement of those out-of-scope IP rights.
Likewise, if the grant or continuation of a license is conditioned on the licensee’s taking certain actions, then the licensee’s failure to comply with those obligations can render the license void. For example, language such as the following could be considered a condition: “Licensor grants Licensee a non-exclusive license so long as Licensee complies with the following conditions.” Because the license is conditioned on licensee’s compliance with the stated conditions, the licensee’s failure to comply with those conditions will void the license and the licensor can proceed against it in an infringement action.
If, on the other hand, the licensee violates an ordinary covenant or obligation in a license agreement (e.g., the obligation to pay royalties), then the licensor can sue the licensee for breach of contract and seek contractual damages and other remedies. It can also seek to terminate the license agreement if the provisions of the agreement permit termination for the alleged breach (see Section 12.2, discussing breach and termination of licenses). But so long as the license remains in effect, the licensee is operating under a license and is not infringing the licensor’s IP rights. Thus, a breach of a license covenant, unlike operating outside the scope of the license grant, only gives rise to contractual remedies, but not infringement claims, so long as the license remains in effect.
One of the most important remedies available for claims of IP infringement is the injunction – a court order prohibiting conduct that constitutes infringement. Injunctive relief is relatively rare in contractual actions, in which monetary damages are the normal remedy. Thus, it is often advantageous to the licensor to argue that a particular contractual provision that the licensee has violated is a condition of the license rather than a mere contractual covenant. The following case illustrates this point.Footnote 15
629 F.3d 928 (9th Cir. 2011)
CALLAHAN, CIRCUIT JUDGE
Blizzard Entertainment, Inc. (“Blizzard”) is the creator of World of Warcraft (“WoW”), a popular multiplayer online role-playing game in which players interact in a virtual world while advancing through the game’s 70 levels. MDY Industries, LLC and its sole member Michael Donnelly (“Donnelly”) (sometimes referred to collectively as “MDY”) developed and sold Glider, a software program that automatically plays the early levels of WoW for players.
MDY brought this action for a declaratory judgment to establish that its Glider sales do not infringe Blizzard’s copyright or other rights …
In November 2004, Blizzard created WoW, a “massively multiplayer online role-playing game” in which players interact in a virtual world. WoW has ten million subscribers, of which two and a half million are in North America.
WoW players roleplay different characters, such as humans, elves, and dwarves. A player’s central objective is to advance the character through the game’s 70 levels by participating in quests and engaging in battles with monsters. As a player advances, the character collects rewards such as in-game currency, weapons, and armor. WoW’s virtual world has its own economy, in which characters use their virtual currency to buy and sell items directly from each other, through vendors, or using auction houses.
Each WoW player must read and accept Blizzard’s End User License Agreement (“EULA”) and Terms of Use (“ToU”) on multiple occasions. Players who do not accept both the EULA and the ToU may return the game client for a refund.
Donnelly is a WoW player and software programmer. In March 2005, he developed Glider, a software “bot” (short for robot) that automates play of WoW’s early levels, for his personal use. A user need not be at the computer while Glider is running. Glider … moves the mouse around and pushes keys on the keyboard. You tell it about your character, where you want to kill things, and when you want to kill. Then it kills for you, automatically. You can do something else, like eat dinner or go to a movie, and when you return, you’ll have a lot more experience and loot.
Glider does not alter or copy WoW’s game client software, does not allow a player to avoid paying monthly subscription dues to Blizzard, and has no commercial use independent of WoW.
The parties dispute Glider’s impact on the WoW experience. Blizzard contends that Glider disrupts WoW’s environment for non-Glider players by enabling Glider users to advance quickly and unfairly through the game and to amass additional game assets. MDY contends that Glider has a minimal effect on non-Glider players, enhances the WoW experience for Glider users, and facilitates disabled players’ access to WoW by auto-playing the game for them.
In summer 2005, Donnelly began selling Glider through MDY’s website for fifteen to twenty-five dollars per license. As of September 2008, MDY had gross revenues of $3.5 million based on 120,000 Glider license sales.
Blizzard claims that from December 2004 to March 2008, it received 465,000 complaints about WoW bots, several thousand of which named Glider. Blizzard spends $940,000 annually to respond to these complaints.
As to the scope of the license [to use WoW], ToU § 4(B), “Limitations on Your Use of the Service,” provides:
You agree that you will not … (ii) create or use cheats, bots, “mods,” and/or hacks, or any other third-party software designed to modify the World of Warcraft experience; or (iii) use any third-party software that intercepts, “mines,” or otherwise collects information from or through the Program or Service.
A copyright owner who grants a nonexclusive, limited license ordinarily waives the right to sue licensees for copyright infringement, and it may sue only for breach of contract. However, if the licensee acts outside the scope of the license, the licensor may sue for copyright infringement. Enforcing a copyright license raises issues that lie at the intersection of copyright and contract law.
We refer to contractual terms that limit a license’s scope as “conditions,” the breach of which constitute copyright infringement. We refer to all other license terms as “covenants,” the breach of which is actionable only under contract law. We distinguish between conditions and covenants according to state contract law, to the extent consistent with federal copyright law and policy.
A Glider user commits copyright infringement by playing WoW while violating a ToU term that is a license condition. To establish copyright infringement, then, Blizzard must demonstrate that the violated term—ToU § 4(B)—is a condition rather than a covenant. Blizzard’s EULAs and ToUs provide that they are to be interpreted according to Delaware law. Accordingly, we first construe them under Delaware law, and then evaluate whether that construction is consistent with federal copyright law and policy.
A covenant is a contractual promise, i.e., a manifestation of intention to act or refrain from acting in a particular way, such that the promisee is justified in understanding that the promisor has made a commitment. A condition precedent is an act or event that must occur before a duty to perform a promise arises. Conditions precedent are disfavored because they tend to work forfeitures. Wherever possible, equity construes ambiguous contract provisions as covenants rather than conditions. However, if the contract is unambiguous, the court construes it according to its terms.
Applying these principles, ToU § 4(B)(ii) and (iii)’s prohibitions against bots and unauthorized third-party software are covenants rather than copyright-enforceable conditions. Although ToU § 4 is titled, “Limitations on Your Use of the Service,” nothing in that section conditions Blizzard’s grant of a limited license on players’ compliance with ToU § 4’s restrictions.
To recover for copyright infringement based on breach of a license agreement, (1) the copying must exceed the scope of the defendant’s license and (2) the copyright owner’s complaint must be grounded in an exclusive right of copyright (e.g., unlawful reproduction or distribution). Contractual rights, however, can be much broader.
“To recover for copyright infringement based on breach of a license agreement … the copyright owner’s complaint must be grounded in an exclusive right of copyright”
[C]onsider a license in which the copyright owner grants a person the right to make one and only one copy of a book with the caveat that the licensee may not read the last ten pages. Obviously, a licensee who made a hundred copies of the book would be liable for copyright infringement because the copying would violate the Copyright Act’s prohibition on reproduction and would exceed the scope of the license. Alternatively, if the licensee made a single copy of the book, but read the last ten pages, the only cause of action would be for breach of contract, because reading a book does not violate any right protected by copyright law.
Here, ToU § 4 contains certain restrictions that are grounded in Blizzard’s exclusive rights of copyright and other restrictions that are not. For instance, ToU § 4(D) forbids creation of derivative works based on WoW without Blizzard’s consent. A player who violates this prohibition would exceed the scope of her license and violate one of Blizzard’s exclusive rights under the Copyright Act. In contrast, ToU § 4(C)(ii) prohibits a player’s disruption of another player’s game experience. A player might violate this prohibition while playing the game by harassing another player with unsolicited instant messages. Although this conduct may violate the contractual covenants with Blizzard, it would not violate any of Blizzard’s exclusive rights of copyright. The antibot provisions at issue in this case, ToU § 4(B)(ii) and (iii), are similarly covenants rather than conditions. A Glider user violates the covenants with Blizzard, but does not thereby commit copyright infringement because Glider does not infringe any of Blizzard’s exclusive rights. For instance, the use does not alter or copy WoW software.
Were we to hold otherwise, Blizzard—or any software copyright holder—could designate any disfavored conduct during software use as copyright infringement, by purporting to condition the license on the player’s abstention from the disfavored conduct. The rationale would be that because the conduct occurs while the player’s computer is copying the software code into RAM in order for it to run, the violation is copyright infringement. This would allow software copyright owners far greater rights than Congress has generally conferred on copyright owners.
We conclude that for a licensee’s violation of a contract to constitute copyright infringement, there must be a nexus between the condition and the licensor’s exclusive rights of copyright. Here, WoW players do not commit copyright infringement by using Glider in violation of the ToU. MDY is thus not liable for secondary copyright infringement, which requires the existence of direct copyright infringement.
Notes and questions
1. Grounded in exclusive rights of copyright. The Ninth Circuit in MDY holds that a license condition is created when a contractual restriction is grounded in the licensor’s exclusive rights of copyright. Other types of restrictions are simply contractual covenants. What kinds of contractual restrictions are “grounded in the exclusive rights of copyright”?
2. Beyond copyright. Should the Ninth Circuit’s reasoning in MDY apply when distinguishing between license conditions and contractual covenants in licensing agreements that related to IP other than copyright? What about licenses of information not covered by any statutory form of IP, such as data and know-how? What might be an “exclusive right of patents” giving rise to a condition?
3. Drafting conditions rather than covenants. In Sun Microsystems, Inc. v. Microsoft Corp., 188 F.3d 1115 (9th Cir. 1999) (decided prior to MDY), the Ninth Circuit found that a provision to ensure compatibility between the licensor’s (Sun’s) and licensee’s (Microsoft’s) software was a covenant rather than a condition of the license grant. As a result, Microsoft’s failure to ensure compatibility was a breach of contract rather than infringement of Sun’s intellectual property rights. The relevant sections of the licensing agreement between Sun and Microsoft are set forth below:
§ 2.1(a) Sun grants to Licensee a perpetual non-exclusive development license under the Intellectual Property Rights of Sun to make, access, use, copy, distribute, view, display, modify, adapt, and create Derivative Works of the Technology and resulting Products.
§ 2.6(a)(vi) Licensee agrees that any new version of a Product that Licensee makes commercially available to the public after the most recent Compatibility Date shall only include the corresponding Compatible Implementation.
Do you think it would have been possible for Sun to draft §2.6(a)(vi) as a condition of the license grant? Could the two provisions have been combined so as to ensure that continued compatibility was a clear condition of the license grant? Why might Microsoft object to this provision as a condition, but consent to it as a covenant?
4. Breach over infringement? If a licensee uses licensed IP beyond the scope of the license, can the licensor bring a breach of contract claim in addition to, or in lieu of, an IP infringement claim? The answer may depend on the language of the license grant clause. If the grant simply allows the licensee to use the licensed IP for Purpose A, and the licensee in addition uses the IP for Purpose B, it is not clear that the licensee has breached the contractual provisions authorizing it to use the IP for Purpose A. On the other hand, if the license grant states that the licensee may use the licensed IP for the sole and exclusive purpose of pursuing Purpose A, then its use of the IP in pursuit of Purpose B might violate the terms of the agreement. Consider Eli Lilly & Co. v. Emisphere Techs., Inc., 408 F. Supp. 2d 668, 685 (S.D. Ind. 2006), in which the court reasoned as follows:
Emisphere granted Lilly an exclusive license to use Emisphere information for the “Field,” which was defined as oral delivery of PTH. Section 2.1 concludes: “Lilly shall not have any rights to use the Emisphere Technology … other than insofar as they relate directly to the Field and are expressly granted herein.” By its plain terms, this provision bars Lilly from using Emisphere Technology … outside the agreed field of PTH research.
Do you agree with the court? Does the language “Lilly shall not have any right to use the [Technology] other than insofar as they relate directly to the Field” create a contractual prohibition that Lilly breached by using the Technology outside the Field? Or does “shall not have any right” define the scope of the license, meaning that Lilly’s operation outside of the Field constitutes an infringement of Emisphere’s IP? And if, as the court holds, Lilly breached the licensing agreement, could Emisphere both terminate the contract and sue for contractual damages, as well as bring suit for IP infringement?
3.5 Effect of IP Transfer on Licenses
If an IP license is akin to a property interest, like a lease or a servitude upon land, then what happens to that license when the original licensor assigns the licensed IP to someone else? Does the license “run with the IP,” as a real property servitude often (but not always) “runs with the land”? Or is the licensee out of luck (i.e., an infringer) when its licensor divests itself of the IP rights underlying the license?
The Copyright Act expressly addresses this issue in 17 U.S.C. § 205(e), which provides that:
A nonexclusive license, whether recorded or not, prevails over a conflicting transfer of copyright ownership if the license is evidenced by a written instrument signed by the owner of the rights licensed or such owner’s duly authorized agent, and if
(1) the license was taken before execution of the transfer; or
(2) the license was taken in good faith before recordation of the transfer and without notice of it.
Thus, a nonexclusive copyright license will survive a transfer of the underlying copyright if the license was granted before the transfer. It will also survive if the license was granted after the licensor transferred the copyright but before the transfer was recorded with the Copyright Office, so long as the licensee acted in good faith and did not have notice of the transfer.Footnote 16 Interestingly, § 205(e) does not apply to exclusive copyright licenses, presumably on the theory that exclusive licenses should be recorded if the licensee wishes to guard against the loss of its rights as a result of future transfers.Footnote 17
Unlike the Copyright Act, the Patent Act does not explicitly address the issue of a transfer of underlying rights. Nevertheless, it has been long-established in the case law that “the purchaser of a patent takes subject to outstanding licenses”Footnote 18 and “a [patent] license is good against the world, whether it is recorded or not. A purchaser of a patent takes it subject to all outstanding licenses.”Footnote 19
But what about the multitude of other contractual obligations contained in a licensing agreement? Licensor obligations relating to service, maintenance, technical assistance, indemnification and confidentiality are not likely to constitute part of the core licensed property interest that travels with the patent, so what happens to them when the licensor transfers the underlying IP to a new owner without assigning the entire agreement? In Datatreasury Corp. v. Wells Fargo & Co., 522 F.3d 1368, 1372 (Fed. Cir. 2008), the Federal Circuit considered whether a contractual requirement to arbitrate disputes followed patents to their new owner. The court held that it did not, reasoning that
the legal encumbrances deemed to “run with the patent” in these cases involved the right to use the patented product, not a duty to arbitrate. The cases do not support a conclusion that procedural terms of a licensing agreement unrelated to the actual use of the patent (e.g. an arbitration clause) are binding on a subsequent owner of the patent.
So what becomes of these non-transferred contractual obligations? One theory is that the original licensor and patent owner remains obligated to perform these contractual obligations so long as they have not been assigned to (and assumed by) someone else. Thus, if the original licensor does not assign a licensing agreement to the acquirer of the underlying IP, the original licensor is still required to perform these obligations. But this requirement may offer only cold comfort to the licensee, as the original licensor may have few remaining resources with which to perform those obligations, and without the underlying IP may be unable to perform some of those obligations. Section 13.3.5 discusses contractual provisions that help to ensure that these contractual licensor obligations are transferred to the new owner of the underlying IP.Footnote 20
Notes and Questions
1. Other forms of IP. The principles discussed in this section are seldom raised in the context of trademark or trade secret licenses. Why? If a suitable case arose, do you think that a trademark or trade secret license should “run with the right”? What would be the practical effect of such a rule?
2. Running of the FRAND commitment. Chapter 20.6 discusses the effect of a transfer of a patent on the original owner’s obligation to license that patent to others on “fair, reasonable and nondiscriminatory” (FRAND) terms. As mentioned in Note 3 of that section, courts have generally not been amenable to treating such FRAND commitments as property-like encumbrances on patents. How does this reluctance square with the reasoning of courts in this section? Do FRAND commitments more resemble licenses, which do run with the rights, or arbitration commitments, which do not?
Open access