Copyright © 1999 David Nimmer, Elliot Brown, and Gary N. Frischling.
* The authors practice copyright law at Irell & Manella LLP in Los Angeles.
... Article 2B of the Uniform Commercial Code (U.C.C.) provides model rules to govern transactions in the digital domain, such as the licensing of software and electronic contracting. ... When a copyright owner distributes its software, it is free to grant a license extending only to specified uses, while excluding others. ... At least for the modes of software distribution used today, copyright law provides all the teeth a publisher needs to control use and dissemination of her work. ... Since the distribution right is an exclusive right in copyright law, distributions outside the license infringe the copyright. ... If, however, the copyright owner elected a licensing framework, given the structure of the transactions, the end user's right to "use" (e.g., copy) the software depends on the end user license. ... The contract at issue in ProCD, Inc. v. Zeidenberg differs from the foregoing examples in the one respect relevant to nonstatutory preemption: it contravenes one of the core policies of the Copyright Act by extending quasi-copyright protection to works that do not qualify as "original." ... Although the Act limits the copyright owner's rights to "public" distribution, publishers who follow the logic of ProCD, Inc. v. Zeidenberg may amplify their statutory rights simply by wrapping books in cellophane, subject to the limitation that the buyer is barred from passing the purchased copy on to a friend. ...
Article 2B of the Uniform Commercial Code (U.C.C.) provides model rules to govern transactions in the digital domain, such as the licensing of software and electronic contracting. By addressing fundamental contract issues in the burgeoning world of digital commerce, it provides a salutary update to extant provisions of the U.C.C. dealing with traditional goods sold in traditional modes of commerce. However, to the extent that Article 2B aspires to protect copyright owners from improper uses of copyrighted works, it solves a non-problem. Copyright owners already enjoy robust and adequate protections under the Copyright Act. Far more troubling than solving this non-problem, however, is the possibility that Article 2B will be used to upset copyright law's "delicate balance" between the rights of copyright owners and copyright users. This balance is disrupted when state law is permitted to enlarge the rights that copyright owners enjoy. Attempts to alter the "delicate balance" through contract should fail under the doctrine of preemption. Article 2B assumes a pose of neutrality on the extent to which copyright law preempts contractual encroachment, yet it facilitates emerging practices designed to alter the balance and place the burden of defending the proper bounds of copyright on copyright users. In this Article, the authors argue that if Article 2B is to be enacted, it must proscribe contracting practices that seek to extend copyright protection beyond its current scope.
The Death of Copyright: [FN1] A Short, Cautionary Tale The year is 2010. With the closure of the last B&N-Walden-Borders-Broadway superstore in Upton, California, no more off-line retail content stores remain in the United States (apart from the Scholar's Palazzo in Disneyland). Theaters, music venues, and movie houses have all but disappeared in the wake of in-home - not to mention implantable - content delivery. Funding for public libraries and the arts has been diverted into providing each person in America with access to the all-purpose device for accessing anything - Microsquish Audiovisual Utilization System (MAUS).
Now, anyone who wishes to read poetry, browse great works of art, enjoy a novel, watch an epic feature film (or humble television show), or experience any other work of authorship can, as a practical matter, do so only through the instrumentality of the MAUS. To access any work through the MAUS, the user must first click "OK" (or, in the case of implantable access, blink a cortical acknowledgment) on the ubiquitous authorization screen. That screen lets the user acknowledge that she agrees to be obligated to abide by all terms and provisions of In a commercial world burgeoning with transactions involving software and other electronically-delivered copyrighted works, an oft-expressed concern arises that traditional rules of commercial contract law - which evolved to address trade in goods - will prove ill-suited to address the peculiar needs of trade in digital products. The Uniform Commercial Code, after all, arose to address the paradigm of a sale of goods, a context that typically involves, at least in part, a negotiated contract between buyer and seller and where the value lies in the physical object exchanged. The typical software transaction, by contrast, does not involve a direct sale between the software proprietor and the end user; rather, it involves a non-negotiated license (otherwise known as a "shrinkwrap" contract) governing uses of the intangible asset (for example, software) embodied in a tangible thing that is sold (for example, the diskette or CD-ROM) - or even absent the nominal sale of a tangible thing (for example, delivery directly over the Internet). It is natural to suspect that the law for widgets may be inadequate for digits. [FN2] It is here that Article 2B of the Uniform Commercial Code comes to the rescue. The proposed model law addresses many of the questions on which traditional commercial contract law is silent, for instance, whether a digital signature constitutes adequate consent to a contract, [FN3] what warranties attach to digital products, [FN4] what choice-of-law rules apply in transactions over the Internet, [FN5] what rules govern the transferability of a license, and how notions of mitigation, consequential damages, releases, inspection, etc., operate in the context of digital products. [FN6] In sum, it provides some measure of certainty to electronic contracting.
Because Article 2B accommodates contracts over copyrightable subject matter, it is relevant to both the federal and state law planes of legal discourse. As discussed below, the symbiosis between federal copyright protection and state contract law is ancient, inevitable, and fully consonant with the purposes of copyright. In developing the law of contracts for the "digital era," Article 2B therefore represents a salutary update to the U.C.C. that can benefit both buyers and sellers of digital goods by providing clear rights and guidance in matters beyond the experience and imagination of the drafters of the current U.C.C. [FN7] Article 2B thus carries on the role that state contract law has traditionally occupied in shaping commerce in copyrighted works.
But harmony is not the end of the symphony. [FN8] When examined in light of its potential impact on copyright law's "delicate balance," [FN9] Article 2B presents the specter of becoming an unwelcome meddler. On the one hand, Article 2B might erroneously be imagined to solve a fundamental problem that does not need solving - protecting the rights of copyright proprietors insofar as third parties exploit the intangible expression underlying their works. On the other hand, Article 2B ignores and potentially weakens the rights of copyright users. These two phenomena are interrelated.
As discussed below, [FN10] Article 2B solves a non-problem to the extent that it aspires to protect the exclusive rights of authors granted under the Copyright Act from improper uses of digital products by end users. The rights of copyright proprietors are already fully protected by the Copyright Act without the need for bilateral contracts, and thus a fortiori without the need for any provisions under the U.C.C. validating mass market contracts.
Solving a non-problem for copyright proprietors may do no harm, but Article 2B's framework threatens to create new problems for copyright users. As further discussed below, [FN11] the copyright laws are designed to achieve a "delicate balance" between the rights of copyright proprietors and copyright users. This balance is disrupted when state law is permitted to enlarge the rights of copyright proprietors at the expense of copyright users. Although attempts at altering the delicate balance struck by copyright law should fail under the doctrine of preemption, a recent decision from the Seventh Circuit [FN12] illustrates that courts sometimes fail to appreciate the preemptive force of copyright, even when the subject contract is intended to defeat users' rights validated by on-point United States Supreme Court precedent.
Article 2B purports to remain "neutral" on questions of federal preemption based on encroachments by contract on copyright doctrine. However, by making provisions of software licenses presumptively enforceable while providing no limitations on overreaching contract terms that proprietors may unilaterally decide to impose, Article 2B facilitates known practices designed to alter the "delicate balance" and places the costs of defending the proper bounds of copyright on copyright users. This result is neither desirable nor necessary. Article 2B can help maintain rather than undo the delicate balance that lies at the core of copyright by giving some guidance as to which types of constraints are at odds with copyright and therefore preempted. But absent appropriate corrections to its current instantiation, it is likely to result in the use of contracts - backed up by the force of the U.C.C. - systematically to displace the rights of users. It is important to appreciate that such resort to contract does not represent the election of contract protection in lieu of copyright. Instead, it represents the use of contract to distort copyright, grotesquely at times. Proprietors who might take advantage of Article 2B do not opt out of copyright protection; they enjoy all of its benefits plus all of the benefits that can be accorded by contracts diminishing the rights of users. We suggest, accordingly, that if Article 2B is to be enacted, it first be amended to evince greater sensitivity towards proscribing certain contracting practices that are inconsistent with sound copyright policy.
This Article proceeds in four Parts. Part I reviews the existing relationship between federal copyright law and state contract doctrine. Part II argues that copyright law already provides adequate protection to copyright owners who distribute software, and that attempts by copyright owners to enlarge their rights by contract conflicts with copyright law's concern for the rights of users. Part III undertakes a critical discussion of the Seventh Circuit's decision in ProCD, which upheld a "shrinkwrap" license that extended contractual protection against copying to subject matter that the Supreme Court has already declared uncopyrightable. Finally, Part IV critiques the failure of Article 2B, under a guise of "neutrality," to take into account the rights of information users and the demands of federal law.
Copyright is, at heart, a creature of the Constitution and the Copyright Act. But ownership and exploitation of copyright are structured at every turn by contract. Unlike the monistic copyright system of German law, under which authors may never separate themselves from ownership of the indivisible whole, U.S. copyright law follows a regime of infinite divisibility. [FN13] The statute itself contemplates transfers in the nature of "an assignment, mortgage...or any other conveyance, alienation, or hypothecation of a copyright or of any of the exclusive rights comprised in a copyright...." [FN14] One salient feature of the terms just quoted is the failure of the Copyright Act to define any of them. Given that the United States Code nowhere contains an established common law as to what constitutes a "mortgage," resort to state law to determine the nature of that device, as well as like hypothecations of ownership, appears inevitable.
Imagine for a moment that Atalanta transfers ownership of her copyright to Busiris, who gives it to Cadmus, who in turn mortgages it to Dindyma Bank, which then forecloses and sells out to Erigone. In a worst-case scenario for Erigone, her ownership of the copyright could be subject to challenge on the grounds that Atalanta was a minor who may disaffirm the contract because it was not confirmed by the state court of her domicile; [FN15] that Busiris (who had previously been declared insane and placed under the control of a conservator appointed by the courts of the state in which he lived) was not bound because his legal guardian failed to sign the purported grant; that Cadmus neglected to perfect the mortgage in the manner required by his own state's law; and that Dindyma Bank had previously dissolved, thereby rendering its purported transfer nugatory. [FN16] In each of those particulars, the battle is waged primarily under state law. [FN17] To take the case of an individual committed to an insane asylum, for example, it is difficult to find any governing federal law, and thus to resist wholesale descent into the minutiae of the subject state's ordinance. In the balance of the other instances, federal law likewise does not directly speak to the question of who holds the capacity to enter into a contract. Erigone therefore faces the prospect of lengthy explorations of state law in order to validate her federal copyright claims. The best she can hope for is the application not of the particular law of the state in which Atalanta, Busiris, Cadmus, and the rest chanced to live, but instead a general notion of common law as applied throughout the several states. [FN18] But even that victory does not invoke the application of federal norms; instead, it looks to an abstract notion of state law. [FN19] As the above hypotheticals demonstrate, the symbiotic relationship between copyright and contract continues
throughout the life of a copyright. Moreover, it can begin even before copyright birth, the moment an original
work of authorship is fixed for the first time in a tangible medium of expression. [FN20] In this guise, it arises as a
factor in defining who the "author" is and thus in determining the identity of the initial "copyright
owner." [FN21] Ordinarily, the author is the efficient cause of parturition, that is, the human being(s) who gave
birth to the work. The Act departs from the default rule, however, when a preexistent contract applies to
certain categories of specially commissioned works "if the parties expressly agree in a written instrument
signed by them that the work shall be considered a work made for hire." [FN22] Thus, both at gestation and
throughout its life, a copyright is owned according to a complex scheme deriving in large part from state law.
But it is not solely the question of ownership over which state law governs. Copyright exploitation, too, can
often turn on distinctions that equally derive from state laws. In this respect, we depart from the intangible
essence of the copyrightable work and move to the tangible good in which it may be embodied. [FN23] Consider
that copyright owners enjoy the exclusive right "to distribute copies or phonorecords of the copyrighted work
to the public by sale or other transfer of ownership, or by rental, lease, or lending." [FN24] Moreover, one in
possession of a lawfully made copy "is entitled, without the authority of the copyright owner, to sell or
otherwise dispose of the possession of that copy...." [FN25] As was the case with respect to "mortgage" and the
rest, neither the Copyright Act itself nor other applicable features of federal law define when a "sale" or
"rental" or act of "lending" of a physical item has taken place. For these questions as well, resort to state law
appears inevitable. [FN26] In sum, federal copyright doctrine leaves to state law the vast bulk of issues concerning contracts affecting
copyright. It follows that state contract law (and cognate doctrines arising under state law) determine to a
great extent the destiny of a copyrighted work and the physical object in which it is embodied. Those state
rules play a critical role in maintaining the "delicate equilibrium" between the rights of copyright holders
to reap the rewards of their intellectual property and the rights of the public to unimpeded advancement of
knowledge and expression.
Consonant with the traditional interplay between state contract law and federal copyright law, the U.C.C. can
help define the mechanics of contract law in the context of contemporary transactions. For example, Article
2B creates rules to govern electronic contracting and provides that the fact that a contract is in electronic
form does not alter or reduce its effect, validity, or enforceability. It gives binding weight to electronic
signatures, and it sets forth rules for determining who shall be held responsible for electronic messages. [FN27] All
of these factors may arise in the context of an electronic license of copyrightable subject matter - an
electronic contract for distribution rights in a book or motion picture, for example - which parties modify by
e-mail and sign using digital signatures. Because federal copyright law, standing alone, is silent as to whether
such a contract is enforceable, the U.C.C. can usefully fill the doctrinal gap in this and like instances. Article
2B therefore can provide important support to the goal of maintaining copyright's "delicate equilibrium" in the
digital age.
Nonetheless, federal abdication in favor of determinations of contract principles under state law has its limits.
In certain particulars, the Copyright Act itself sets forth some governing parameters applicable to contracts
and other matters typically reserved to determination under state law. When those circumstances obtain,
federal law controls, notwithstanding contrary state doctrines. [FN28] One example is so pronounced as to have virtually escaped notice. Undoubtedly the most well-known aspect
of Anglo-American contract law is its requirement of a quid pro quo - the doctrine of consideration. [FN29] It is
doubtful that the law of any state in the union dispenses with that general requirement. [FN30] Were it applicable
to the copyright sphere, that doctrine would invalidate grants of copyright ownership unrequited by
the grantee. [FN31] Yet "notwithstanding that feature of state law, no consideration is necessary under federal
law to effectuate a transfer of copyright ownership that does not purport to require consideration." [FN32] (Nonetheless, one must acknowledge that few, if any, cases have tested the boundary of consideration-less
copyright grants, presumably because grantees of valuable copyrights invariably recite the delivery of "$ 10
and other good and valuable consideration" in order to escape serving as a test case. n33)
The most prominent example of federal contract requirements trumping contrary state doctrine that has
received treatment in published decisions is the Act's requirement that any transfer of copyright ownership
[FN34] "is not valid unless an instrument of conveyance...is in writing and signed by the owner of the rights
conveyed...." [FN35] Even if state law validates oral grants - attested by the grantor before the mythical bench
of fifty bishops, for example - that law must bow to the superior force of the federal enactment. [FN36] Another example of the same phenomenon - though this one has gone unlitigated - arises under copyright's
termination-of-transfers doctrine, which allows authors a "second bite of the apple" for works that they long
ago sold, gave away, or otherwise alienated. [FN37] In particular, the Act itself provides with respect to
transfers of copyright ownership that, following the lapse of a set period, [FN38] "termination of the grant may be
effected notwithstanding any agreement to the contrary, including an agreement to make a will or to make
any future grant." [FN39] Accordingly, a contract not to exercise an author's termination rights may be
fully operational under state law, yet the superior force of federal law nonetheless bars its enforcement,
effectively rendering it a nullity. [FN40] Much confusion arises in attempting to reconcile these strands. [FN41] As an example, consider the ruling that
when Congress used the term "children" in the context of termination of pre-1978 transfers [FN42] it intended to
adopt antecedent state family law definitions as to who qualifies for that label, [FN43] whereas when Congress
used the same word in the same Act to apply to termination of post-1978 transfers, it intended to adopt a
federal definition incorporated into the Copyright Act itself. [FN44] The court reached that result by attempting
to follow Congress's will in enacting a given provision of the Copyright Act. [FN45] As we shall see, that
desideratum furnishes the touchstone for proper analysis in the journey that follows.
Though Article 2B, as noted above, can usefully serve a complementary role to copyright, there is one
significant function for which Article 2B is not needed: to protect the copyright interests of copyright
proprietors, especially in the context of mass market distribution of software, one of the paradigmatic
transactions under Article 2B. Contrary to the claim that Article 2B is needed to protect copyright interests in
that context, existing copyright law adequately protects those owners when they distribute
copyrighted mass market software, even in a world in which shrinkwrap agreements are not deemed
enforceable contracts. [FN46] That conclusion follows because the exclusive rights granted under the copyright
laws effectively preclude use of computer software - to the extent the Constitution and Congress accord a
monopolytherein - without the express or implied permission of the copyright owner.
When a copyright owner distributes its software, it is free to grant a license extending only to specified uses,
while excluding others. Moreover, any such license does not require a bilateral contract. A simple, unilateral
statement by the copyright owner of the scope of its license suffices. [FN47] In most cases, use beyond the
scope of that license constitutes actionable copyright infringement under existing copyright law. [FN48] To the
extent that any such use beyond the scope of the unilateral license is not copyright infringement - for
instance, because it constitutes fair use under section 107 - state contract law cannot produce a different
result.
Several hypotheticals illustrate the ability of publishers to protect their intellectual property rights when
engaging in mass distribution of software. First, Procne picks up the latest copy of SuperSmart321, a nifty
spreadsheet program, at CompUSA. She purchases the program without opening the box and takes it home.
Unlike most software products, SuperSmart321 contains no license terms of any kind.
In this hypothetical, existing copyright law permits Procne to do exactly what a typical publisher and typical
buyer would contemplate: use the software on a single computer and make a backup copy. Because Procne
purchased a copy of the software, she clearly falls within the ambit of the statutory section securing rights to
those owners. [FN49] She thus is entitled to copy the software onto her computer's hard drive in order to run it,
[FN50] as well as to make a tangible backup copy. [FN51] Any further copying of the software - and thus,
effectively, use on any other computer system - constitutes copyright infringement unless excused,
for instance as a fair use. [FN52] Thus, copyright law alone affords the publisher of SuperSmart321 ample power
to prevent Procne from making or distributing improper copies of the software, or even duplicating the
software on multiple machines in her home or office. No bilateral contract is necessary to protect the software
publisher's rights.
Second, Pandion picks up the latest copy of SuperSmart321 by purchasing it via the Internet. He pays for it
with a credit card and downloads it to the hard drive of his computer. Once again, the results will be
effectively the same as above, even without a written license agreement. Pandion still owns the copy of the
computer program on his hard drive and is authorized to use it (but not reproduce it except for backup
purposes) pursuant to the statutory sections invoked above. [FN53] Third, Itys purchases a "10 User Pack" of Virulator, a software package that locates all computer viruses on a
user's hard drive, removes them and e-mails them to Iraq. He installs the software on the network server in his
office. The envelope containing the CD-ROM, as well as the install screen, inform Itys that: "This software
product is licensed for installation on a network server, to be accessed by no more than 10users
simultaneously. All other rights are reserved."
Under existing copyright law, the manufacturer's 10-user limitation is enforceable regardless of whether state
law treats the above notice as part of a binding bilateral contract. [FN54] A copyright owner may grant a
non-exclusive license by any words or conduct tending to show such a license. [FN55] Thus, by virtue of the
above language, the publisher of Virulator has expanded Itys's right to use his copy on a single machine [FN56] to include making one copy on a server and up to nine other copies in the random access memory (RAM) of client computers. If Itys were to allow 20 users to access the software, the copies existing in the RAM of machines 11-20 would be unlicensed and hence infringing. [FN57] The publisher thus has the lever it needs to preclude unlicensed use, without obtaining Itys's enforceable promise via the U.C.C. not to use the software on more machines than authorized.
Fourth, after her company's IPO, Philomela decides that she has grown bored with life in the Silicon Valley,
exercises her stock options and moves for a year to Tahiti (after thrashing Thrace). Before she moves, she
posts an ad on the Internet offering to rent her copy of MegaCAD 3D, an elaborate $ 10,000 software
package. Tereus has an eight-month project for which he desperately needs MegaCAD. He spots Philomela's
ad and jumps at her offer. Tereus pays $ 1,000 for a year's use of the disks. Can the publisher of MegaCAD
prevent this transaction?
Under existing copyright law, Philomela may not rent or lease her copy of the software. [FN58] Similarly, Tereus is
not entitled to use the software, as to do so would involve copying the software into the computer's RAM,
which constitutes infringement in the absence of a license. [FN59] Thus, the publisher has a viable copyright
infringement claim against Philomela and Tereus. An enforceable promise by Philomela not to rent her software
is unnecessary here, too.
Fifth, Bacchus needs software to help him manage a fleet of trucks for his party-hearty business. Like all good
small businessmen, Bacchus is cost-conscious. He finds a shareware package on the Internet, OINOS, which
looks like it will do the trick. Bacchus downloads the OINOS software, reads the license, which provides that
he may use the software for his own business purposes, but cannot modify or redistribute it. He pays the $
19.95 registration fee and uses the software. Two years later, Bacchus realizes he needs an integrated
system to link his truck management software with the rest of his business. He hires an independent software
consultant to build such a system. Because he loves OINOS so much, Bacchus specifically asks the consultant
to make his new system work just the same way. The consultant, who is convinced he is underpaid, takes a
shortcut and includes a modified version of some of the OINOS code in the system he has been
fermenting. Ultimately, the new system is so successful that Bacchus markets it to other trucking businesses.
Can the author of OINOS complain? Of course. It does not matter whether the license agreement that
accompanied the OINOS software is deemed an enforceable contract. Under existing copyright law, Bacchus
had no right to have the OINOS software modified or incorporated into another system, thereby creating
unauthorized derivative works. [FN60] Nor did he have the right to reproduce it for distribution to others, whether
as part of a new system or separately. Article 2B adds nothing of substance to the rights of OINOS's owner.
One could spin out numerous other hypotheticals involving typical modes of software distribution and reach
the same result. At least for the modes of software distribution used today, copyright law provides all the
teeth a publisher needs to control use and dissemination of her work. No ersatz shark via contractual promise
is necessary to enforce these rights.
The conclusion that contract is not needed to protect copyright interests further pertains when we consider
other forms of distribution in the digital realm. Consider, for example, the Divx (Digital Video Express)
technology for distribution of movies, which may be upon us in the very near future. Divx operates like a DVD
disk containing a movie, except that the encryption software included on the Divx disk limits the user to
playing the movie for a set number of days following the first time she plays the disk. [FN61] Once the initial
viewing period expires, the owner of a Divx disk can obtain additional play time, or in some cases convert the
disk to unlimited play, for a fee. [FN62] What if a studio selling movies on Divx disks feels it needs an enforceable
promise by the buyer not to attempt to circumvent the lockout technology built into the software?
Although one can appreciate the desire of the studio to seek any and all legal protections it can, the copyright
laws of today (and certainly those of tomorrow) should prove more than adequate to protect the studio's
interest, even absent the proposed contract. [FN63] Modifying the Divx software to defeat the lockout
(assuming, for the sake of argument, it were technically possible) likely would involve either unauthorized
reproduction of at least a portion of the copyrighted work, or the creation of an unauthorized derivative work.
Either way, copyright infringement liability would result. Moreover, any doubt about the impropriety of
defeating anti-copying technology will likely be laid to rest by pending federal legislation. [FN64] Thus, even on
the so-called bleeding edge of technology, we find it difficult to see a need for state law protection of
copyright rights in connection with the mass-market distribution of copyrighted works.
An alternative methodology uses actual copies of the software. Here, for example, Quicken may license a
distributor to distribute its accounting software in packages provided to the distributor by Quicken. A license is
used in the software industry here, although some other industries may sell copies to the distributor for resale.
In the license, the distributor may be allowed to distribute copies to retailers, provided that certain conditions
are met, such as terms of payment, retention of the original packaging, and making the eventual end user
distribution occur subject to an end user license. Since the distribution right is an exclusive right in copyright
law, distributions outside the license infringe the copyright.
In both sequences, the information product eventually reaches an end user. If it does so in an ordinary
chain of distribution complying with the distribution licenses, the end user is in rightful possession of a copy. If
the distribution involved sales of copies, nothing more is required. The end user is the owner of the copy.
Copyright law spells out limited rights that flow to the owner of the copy (e.g., to distribute it, make a
back-up if it is software, make some changes essential to use if [sic] its software). There is no direct
contractual relationship between the copyright owner and the "end user."
If, however, the copyright owner elected a licensing framework, given the structure of the transactions, the
end user's right to "use" (e.g., copy) the software depends on the end user license. Typically, this is
characterized as a license from the producer to the end user. It creates a direct contractual relationship that
would not otherwise exist and which, in light of concepts of privity, might not be implied as between these
parties. The contract, then, at this point, jumps past the chain of distribution and creates a direct link to the
producer by the end user. It is also, in this sequence, the only contract that enables the end user to make
copies of the software in its own machine. [FN66] Given that, as the hypotheticals set forth in the previous subsection of this Article reveal, [FN67] copyright law
itself regulates the activities of remote purchasers of software, why does Article 2B attempt to create "a
direct link to the producer by the end user"? [FN68] The answer stems from the language italicized above.
Through these various references, the draft posits a framework that falls short of a sale, constituting merely a
license.
The first two paragraphs quoted above contrast "access to a single master copy" with the "alternative
methodology [that] uses actual copies of the software." It is indeed possible to imagine access without making
actual copies. For instance, instead of purchasing diskettes containing Microsoft Word 97 and loading it onto
your hard drive, you could pay a monthly fee to log onto the Microsoft web site and create and edit
documents there. At the end of each session, you would download your text and save it, but you would never
obtain any copy of the computer program itself. That scenario indeed involves access without the alternative
of obtaining actual copies. By contrast, if you do download Word 97 onto your hard drive, then you have
already moved to the realm in which copyright law's first-sale doctrine applies. [FN69] This realm is
entered, moreover, regardless of whether the label "license" applies to Microsoft's granting of rights in the
copyright to the program.
For these purposes, it is vital to differentiate between tangible and intangible property. [FN70] When a software
publisher distributes its product, it certainly does not part with copyright ownership. [FN71] Instead, the only
matter under examination is whether it has parted with ownership of the physical media incorporating that
software. Article 2B evidently takes the view that a status short of sale exists - which it calls "licensing" -
whereby end users nonetheless acquire full dominion over the tangible property that comes into their
safekeeping.
One line of cases cited by the drafters of Article 2B apparently vindicates the existence of this type of
"licensing" framework. It is exemplified by Microsoft Corp. v. Harmony Computers & Electronics, Inc., [FN72] which
arose over the distribution of Microsoft's MS-DOS and Windows software. The facts in the case are unclear. It
may be that some of the software at issue in the case was pirated. To that extent, the subject diskettes
were clearly unauthorized, outside the safe harbor of the first-sale doctrine, [FN73] and thus rightfully subject to
seizure and suppression. [FN74] On the other hand, the opinion can also be read to arise from a factual posture in
which Microsoft produced copies of Windows software, which it then distributed to Original Equipment
Manufacturers (OEMs). The OEMs then disposed of the copies in their possession, as they were entitled to do
under the first-sale doctrine. To the extent that the case implicated the first scenario, it is uninteresting -
counterfeit products are clearly not entitled to further distribution. But the second scenario is evidently the
basis for the Article 2B drafters' citation to the case. [FN75] It is accordingly necessary to evaluate it from that
standpoint.
Microsoft v. Harmony rejected the defendants' first-sale defense on the basis that "Microsoft only licenses and
does not sell its Products." [FN76] What does that holding mean? To appreciate its import, the buzzwords
"licenses" and "Products" must be unscrambled.
License of tangible medium. If the underlying facts were that Microsoft leased diskettes or CD-ROMs to
end users and sent its agent to collect those physical media at a time specified in the lease, then no sale of
those physical products occurred. Under those assumed facts, Microsoft had not parted with possession of a
physical copy. On that basis, no sale occurred and copyright's first-sale doctrine does not come into play. [FN77] That is one possible sense in which Microsoft may have "licensed" its "Product."
License of copyrighted work embodied in tangible medium. By contrast, if the underlying facts were that
Microsoft issued only a license to its MS-DOS and Windows products, but did so through the sale or other
permanent disposition of tangible items embodying those products, then the first-sale implications are wholly
different. Imagine, for example, that Microsoft gave OEMs diskettes with the intent that they would be
distributed to end users who could discard the diskettes in the trash or erase and reuse them [FN78] - so long as
the users did not reproduce the subject software. [FN79] Under this latter scenario, a sale of the physical
medium has occurred, and its purchaser is clothed under the Copyright Act with the rights belonging to rightful
owners of physical property, subject to all of the obligations under the Act that are reserved to Microsoft of
exercising copyright dominion. This is the second sense in which Microsoft may be said to "license" its
"Product."
These two paradigms are illustrated by the evolution of motion-picture exploitation. Film owners have the
option not to sell their works, but only to "license" them. For decades, the motion-picture studios followed
exactly that course, jealously guarding ownership of the celluloid prints and only renting them to theaters for
exhibition, retrieving the physical stock at the end of the theatrical run. [FN80] That course of action matches
the first set of facts hypothesized above. Alternatively, motion-picture studios also have the option of
distributing tangible copies of their films - as they have done since the advent of the videotape era. [FN81] That
conduct from the 1980s to the present matches the second scenario.
Software publishers likewise enjoy the same rights. They can engage solely in rental or lease of physical
media, mandating return of the subject tangible items at the end of the term. Or they can incorporate
their software onto physical items that they then release from their control, setting them into the stream of
commerce. In the latter instance, the end user, while admittedly a licensee of the copyright, is not a licensee
of the diskette or CD-ROM in her possession. Instead, she is the owner of those physical media containing
licensed works. As such, she is clothed with full rights under the first-sale doctrine. No innovation in software
distribution so far has forced reevaluation of the traditional paradigm.
Which circumstance actually obtained in Microsoft v. Harmony, the exemplar of the "licensing" paradigm?
Sadly, the opinion fails to clarify the matter, and that inability to distinguish between differing paradigms is
only too typical. [FN82] Nonetheless, lack of clarity does not create a new "licensing" paradigm. Instead, if there
were a bona fide lease of the physical goods, then one legal regime pertained; if Microsoft actually sold or
otherwise permanently disposed of those physical goods, retaining full copyright ownership in itself, then
another legal regime governed. Current copyright law does not recognize any regime of "licensing" [FN83] that
stands intermediate between those two possibilities. [FN84] Nonetheless, an innovation might be said to be occurring through Internet distribution. Are the old
barriers breaking down such that U.C.C. Article 2B must come to the rescue of a tottering copyright system?
We answer that question, too, in the negative. As the foregoing example of Pandion reveals, [FN85] purchases
over the Internet do not suffer from a fatal lack of privity between software owner and end user which
requires legal redress. Instead, copyright law itself governs the usages to which Pandion may put the product
that he purchased over the Internet.
Indeed, it is possible to go further here. Someone like Pandion who purchases SuperSmart321 as embodied in
CD-ROM form has unambiguously acquired the right to pass the CD-ROM on to a friend. Does a parity of
reasoning indicate that Pandion himself likewise has the right to sell his computer when he wishes to upgrade
his whole system, even if the hard drive thereby transferred contains a copy of SuperSmart321 that he
purchased over the Internet?
Now the pedal comes down to the metal: Can Pandion keep his computer, deferring upgrades to a later day,
but nonetheless pass along solely SuperSmart321 (by transferring the files to a friend and deleting them from
his own hard drive, let us say)? There are two possible answers to this question:
Yes. Under this rationale, Pandion can take advantage of a "digital first-sale defense." [FN86] Pandion, by this
logic, is as much an owner of the "copy" purchased over the Internet as he would be of the CD-ROM. Granted,
the copy in this instance cannot be as conveniently hefted and hoisted aloft, but the mere fact of its
dispersion over manifold sectors of the hard drive does not detract from its status as a material object in
which the subject software is fixed and from which it can be retrieved. [FN87] No. An alternative view is that Internet distribution is different. No tangible copy has been released, because
there is no discrete "material object" of the type invoked above that contains SuperSmart321. The
ever-shifting sectors and buffers where the work flits and dances fail to qualify as either "material" or an
"object" under the statutory language.
The latter view - that Internet distribution is different because it does not result in a "copy" - must be
rejected as implausible. For were it correct that the recipient of an Internet instantiation does not obtain a
"copy," Pandion would be able to distribute that instantiation freely over the Internet to thousands of
remote recipients without infringing the copyright in SuperSmart 321. [FN88] Such a construction would be
disastrous for copyright owners, and should not be viewed as implementing Congress's intent.
It is not necessary, however, to resolve whether Internet distribution results in a "copy" to test the
adaptability of extant copyright law to new technologies. Under the "yes" view, Pandion may dispose of his
volatile "copy" under the first-sale doctrine; under the "no" view, he cannot. Under the former view, Internet
sales of software are assimilated to current methods of film exploitation via videocassettes, laser discs, DVD,
and the rest (that is, a first sale arises), whereas under the latter view, such sales are assimilated to
motion-picture distribution in the pre-videotape era (that is, there was no first sale). In either event, no new
paradigm is required. [FN89] In sum, the new paradigm of "licensing" instead of sales, on the one hand, and lending, on the other hand,
collapses. For that reason, no new conceptual breakthrough via U.C.C. Article 2B is required.
Copyright law and contract law not only clash overtly but may clash covertly to the extent that they pursue
different implicit purposes and objectives. When conflicts occur, preemption principles force state contract law
to yield. The source of copyright's preemptive power is the United States Constitution. To the extent that any
state law "stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress," [FN90] the Supremacy Clause mandates that the law of Congress reign supreme. [FN91] Some cases describe preemption as the upshot of a clash between state and federal law in which state law is
vanquished. [FN92] Other cases have enunciated even stricter principles according to which federal law does not
tolerate parallel state regimes. [FN93] As the Supreme Court put it in its most recent pronouncement on the
subject of parallel regimes of federal and state intellectual property protection, "The offer of federal protection
from competitive exploitation of intellectual property would be rendered meaningless in a world where
substantially similar state law protections were readily available." [FN94] On the other hand, in one case predating the adoption of the current Act, the Supreme Court allowed greater
tolerance for state schemes covering the same subject matter as copyright. In particular, Goldstein v.
California [FN95] held that the states retain concurrent power to afford copyright protection to the works of
authors as long as such protection does not conflict with federal law. Nonetheless, state laws enacted
pursuant to such concurrent power can, of course, be subject to preemption by federal statute.
When enacting section 301 of the current Act, Congress took precisely that action of preempting concurrent
state law in the copyright domain. Unlike the parallel federal and state tracks that previously applied to the
copyright realm, section 301 federalizes much of the domain of protection for copyrightable expression. [FN96] By
reason of that explicit federal preemption, states' concurrent copyright powers lack almost all practical
significance. [FN97] In ProCD, Inc. v. Zeidenberg, [FN98] plaintiff spent millions of dollars to produce a massive "telephone book" of
nationwide scope. Because the "book" contained almost one hundred million listings, plaintiff released it on
CD-ROM with copyrighted search software designed to navigate through the mass of information. The CD-ROM
was placed in a box that stated that the software came with restrictions listed in an enclosed license. The
license was encoded on the CD-ROM discs, printed in the manual, and appeared on a user's screen every time
the software ran. It expressly provided that the end user "will not make the [search] Software or the
[telephone] Listings in whole or in part available to any other user in any networked or time-shared
environment, or transfer the Listings in whole or in part to any computer other than the computer used to
access the Listings." [FN99] When the matter under discussion combines listing from the white pages of the telephone book and allegations
of copyright infringement, the Supreme Court's landmark holding in Feist Publications, Inc. v. Rural Telephone
Service Co. springs immediately to mind. [FN100] If that unanimous decision made anything clear, it is that not
even a massive expenditure of funds to compile phone listings can render them copyrightable. Instead, they
repose in the public domain - both as a matter of statutory construction and of constitutional necessity.
Relying on Feist, defendants in ProCD copied all the listings off the plaintiff's CD-ROMs, composed their own
software to access the names and addresses, and went into business in competition with plaintiff by making all
the listings available for search on an Internet web page.
The plaintiff responded by bringing suit. Barred by Feist from bringing a copyright claim against the copying of
telephone listings, plaintiffs sought redress against defendants' constitutionally privileged copying by alleging
breach of contract and misappropriation.
2. Issues State and Federal
On the latter cause of action, the district court concluded that "because plaintiff's misappropriation claim is
not qualitatively different from a copyright infringement claim, the underlying rights plaintiff seeks to vindicate
are equivalent to federal rights and are preempted by the Copyright Act." [FN101] In reference to the contract
issue, the district court noted that most commentators disfavor rights asserted under shrinkwrap
licenses, given users' inability to bargain over precise terms. More fundamentally, those licenses "pose
important questions about the extent to which individual contract provisions can supplement or expand federal
copyright protection." [FN102] The district court accordingly concluded that section 301 preempted the contract
claim. As Chief Judge Crabb held, any other ruling would "alter the "delicate balance' of copyright law" and,
more particularly, constitute an impermissible end-run around Feist. [FN103] This solicitude for "delicate balance" is not an aberration. As the Supreme Court itself has noted,
It is Congress that has been assigned the task of defining the scope of the limited monopoly that should be
granted to authors or to inventors.... This task involves a difficult balance between the interests of authors
and inventors in the control and exploitation of their writings and discoveries on the one hand, and society's
competing interest in the free flow of ideas, information, and commerce on the other hand.... [FN104] This ventilation of the contract issue in the context of copyright poses two analytically separate issues. [FN105] The first question is whether, as a matter of contract law, the shrinkwrap license unilaterally imposed by the
manufacturer constitutes a binding agreement. That first question implicates construction of state law,
namely, the Uniform Commercial Code as implemented into Wisconsin law. Assuming an affirmative answer, the
second question is whether that contract can govern in the copyright context. This second question arises
under federal law, as a matter of preemption via the statute or the Constitution.
The district court resolved both matters adversely to ProCD, Inc. In reversing, the Seventh Circuit reached
the contrary conclusion on both scores. [FN106] To isolate the preemption issue posed by this case, [FN107] we
assume for current purposes that the subject contract is either enforceable under the U.C.C. as
currently drafted [FN108] or that it would be enforceable under Article 2B. The question thereby framed is
whether ProCD, Inc.'s contract restrictions on copying can stand consistent with federal norms. This
preemption inquiry itself occupies two levels, statutory and constitutional.
3. A Step Back into History
Before proceeding to a full-scale exploration of the issue posed in ProCD of whether copyright law preempts
application of the U.C.C. in this arena, it is worthwhile to ponder precursor scenarios. ProCD, Inc. was far from
the first copyright owner to attempt to magnify its rights via contract. Indeed, it is almost commonplace in the
history of copyright jurisprudence that when new technology establishes products or media considered
incapable of being protected, copyright owners seek self-help through the unilateral declaration of expanded
rights via purported contractual limitations.
Consider the diminution of sheet music sales and the advent of sound recordings played over the radio. [FN109] Given the lack of a public performance right in those sound recordings as a matter of law, [FN110] RCA decided to
cure the law's lacunae by affixing the following language to its products:
Only For Non-Commercial Use on Phonographs in Homes. Mfr.& Original Purchaser Have Agreed This Record
Shall Not Be Resold Or Used For Any Other Purpose. See Detailed Notice on Envelope. [FN111] When the purchaser [FN112] of the records ignored that "shrinkwrap" contract - thus anticipating Zeidenberg's
conduct by a half-century - by broadcasting them for profit over the air, the manufacturer responded
by filing suit. No less an authority than Learned Hand summarily rebuffed the claim. [FN113] One could reach back even further and likewise find the same rejection of copyright proprietors' attempts to
tilt the "delicate balance" in their favor. In 1908 the Supreme Court enunciated the "first sale" [FN114] doctrine
when it refused to enforce a book publisher's proto-shrinkwrap license barring any retail sale of the books
there at issue for a price less than $ 1.00. [FN115] Even earlier than that, a lower court had likewise refused to
enforce use restrictions placed on the inside cover of a book. [FN116] With that history in mind, ProCD, Inc. v. Zeidenberg is not novel. Earlier actors throughout the twentieth
century had similarly attempted to magnify their rights through use of contract. The novelty of ProCD thus lies
not in the use of that device, but instead in the fact that the Seventh Circuit was the first court that did not
dismiss as ineffectual a copyright owner's purported recalibration of the "delicate balance."
1. section 301
As already noted, the district court concluded that a ruling in favor of ProCD, Inc., would subvert Feist;
indeed, there can be little doubt that plaintiff crafted its shrinkwrap with that precise goal in mind. The court's
invocation of a subversive assault on a recent on-point Supreme Court precedent should have brought to mind
general conflict pre-emption under the Supremacy Clause. [FN118] Yet although its reference to
copyright's "delicate balance" adverts to general preemption under the Supremacy Clause, the district court
ultimately failed to contemplate preemption on that non-statutory basis. Instead, Chief Judge Crabb grounded
her analysis entirely on the narrower statutory grounds for preemption under section 301.
section 301 preempts state-created legal or equitable rights, whether based upon common law or statute,
under the following conditions: (1) the state law creates "legal or equitable rights that are equivalent to any
of the exclusive rights within the general scope of copyright as specified by section 106" and, (2) such rights
under such state law may be claimed in "works of authorship that are fixed in a tangible medium of expression
and come within the subject matter of copyright... whether created before or after [January1, 1978] and
whether published or unpublished...." [FN119] Thus, two elements must coalesce in order to effectuate section
301 preemption, the first relating to the nature of the rights granted under state law, the second to the
nature of the work in which such rights may be claimed.
The first element was the one at issue in this case. [FN120] It is triggered if the state-created act is infringed
merely by engaging in one of the exclusive rights of authors under the Copyright Act - such as the right to
reproduce. If, however, in addition to, or instead of, amounting to one of the exclusive rights of authors under
the Copyright Act, a qualitatively different element is required, then the right does not lie "within the general
scope of copyright," and there is no preemption. This is known as the "extra element" test. [FN121] Reported cases confronting alleged copyright preemption under section 301 of contract causes of action have
almost uniformly rebuffed the preemption claim. That doctrinal result is as it should be: the vast majority of
copyright contracts easily withstand muster under section 301 because the breach alleged amounts to more
than reproduction, distribution, etc., of a copyrighted work. In other words, the typical contract case satisfies
the "extra element" test. The rub arises in ProCD in that plaintiff there drafted a contract that was exceptional
when contrasted with that vast majority.
The district court in ProCD concluded that efforts to enforce the contract there at issue were precisely
congruent with a copyright claim (and therefore, implicitly, failed the "extra element" test). But in an
inexplicable move that then undermined the basis for its own ruling, the court viewed itself as disagreeing with
a trinity of appellate decisions that had found particular contract claims not preempted: the Eighth Circuit's
ruling in National Car Rental System, Inc. v. Computer Associates International, Inc., [FN122] the Fifth Circuit's
ruling in Taquino v. Teledyne Monarch Rubber, [FN123] and the Fourth Circuit's ruling in Acorn Structures, Inc. v.
Swantz. [FN124] The district court in ProCD stated, "To the extent that National Car Rental, Taquino,...and Acorn
support the proposition that a copyright infringement claim is not equivalent to a contract claim merely
because the contract claim requires a plaintiff to show the additional element of breach, I disagree
respectfully with their conclusions." [FN125] Couched in those terms, an affirmance would seem to proclaim "circuit conflict," thus inviting certiorari. In
fact, however, the holding of preemption under these circumstances was not as unprecedented as Chief Judge
Crabb allowed. To appreciate why, we must broaden our sights from causes of action alleging pure breach of
contract to those that arise in a contracts-plus situation.
When the current Act was being deliberated, the House Committee Report stated that the tort of "interference
with contract relations" is "merely the equivalent of copyright protection, [and therefore] would be
preempted." [FN126] On that basis, courts have held that particular contract-based tort preempted under section
301 of the Copyright Act. [FN127] This result would appear doctrinally correct. [FN128] Moreover, it has been
extended as well to closely allied torts, such as interference with prospective economic advantage for
foiling consummation of a copyright contract under negotiation. [FN129] With this larger perspective in mind, the rectitude of Chief Judge Crabb's conclusion can be appreciated. When
two parties to a negotiated contract - even one for exploitation of copyrighted goods - dispute its terms and
application, the allegation of breach is typically not preempted. But just as the tort of interference with
contractual relations attempts to regulate the same subject matter as does copyright law and is therefore
preempted, when a breach of contract cause of action - particularly one that does not result from the
bargained-for agreement of both parties to its putative execution - is used as a subterfuge to control nothing
other than the reproduction, adaptation, public distribution, etc., of works within the subject matter of
copyright, then it too should be deemed preempted.
The facts of the last scenario underlay those in ProCD. The plaintiff was seeking to control the exclusive rights
granted by copyright law every bit as much as in a preempted cause of action denominated "interference with
contract relations." It sought, in short, to bar further public distribution of materials, a right that belongs to
copyright owners under the statute. [FN130] But it sought to do so with respect to a subject matter - listings in
the white pages of a telephone book - that a unanimous Supreme Court had ruled to be in the public domain.
Because the district court did not cite those considerations, and instead proclaimed itself at odds with
previous circuit precedent on the subject, it set itself up for reversal. The Seventh Circuit, as it were,
accepted the invitation: "But are rights created by contract "equivalent to any of the exclusive rights within
the general scope of copyright'? Three courts of appeals have answered "no.' The district court disagreed with
these decisions, but we think them sound." [FN131] The first comment about the Seventh Circuit's adherence to the district court's supposed disagreement with
the three previous cases is that each examined contracts in a context apart from shrinkwrap licenses. By
contrast, the previous pronouncement by a circuit court on that latter subject upheld copyright preemption.
[FN132] Accordingly, strict adherence to precedent, without even examining the issues afresh, supports,
rather than undermines, the district court's preemption ruling.
Moreover, examination of each of the three cited circuit court decisions reveals that it is, in fact, unnecessary
to disagree with them in order to preserve the scope of section 301 and concur in the district court's
preemption holding. Rather than viewing those cases as being governed by a wholly different rule than the
district court followed, each of their holdings can best be understood as turning on the specific facts
presented. None supports the Seventh Circuit's broad conclusion that "whether a particular license is generous
or restrictive, a simple two-party contract is not "equivalent to any of the exclusive rights within the general
scope of copyright' and therefore may be enforced." [FN133] First, Judge Easterbrook's reversal of the district court's holding in ProCD cites National Car Rental for the
sweeping proposition that "rights created by contract" are not "'equivalent to any of the exclusive rights
within the general scope of copyright.'" [FN134] Although that latter case did hold that the specific contract
there at issue was not preempted, it did not extend its holding to contracts in general. In fact, explicitly
relying on the "extra element rule," National Car Rental held that "the contractual restriction" that prohibited
"processing of data for third parties" was qualitatively different from a contract that could be "breached "by
the mere act of reproduction, performance, distribution or display."' [FN135] Because "none of the exclusive
copyright rights grant [the copyright owner] that right of their own force," and because "absent the parties'
agreement, this restriction would not exist," the court concluded that the "contractual restriction on use of
the programs constitutes an extra element that makes this cause of action qualitatively different from one for
copyright." [FN136] Invoking the extra element rule, National Car Rental approvingly cited the Ninth Circuit's
Kalitta ruling, "Copyright preemption is both explicit and broad: [It] prohibits state-law protection for any right
equivalent to those in the Copyright Act." [FN137] Turning to the copying of the software at issue in ProCD, it
was (in the words of Kalitta) ""in and of itself,' [an act which] "would infringe one of the exclusive rights listed
in 106.'" [FN138] Under the Kalitta standard approved by National Car Rental, the license at issue in ProCD should
therefore be held preempted, thus bolstering rather than undermining the district court's conclusion.
In Taquino, the contract at issue "forbade [the defendant] from representing a competing company
prior to termination," a restriction not equivalent to copyright:
The right to claim this breach of contract is not preempted by the copyright laws. 17 U.S.C. 301 only
preempts rights equivalent to the exclusive rights within the general scope of copyright law. A right is
equivalent if the mere act of reproduction, distribution, or display infringes it. This action for breach of
contract involves an element in addition to mere reproduction, distribution, or display.... [FN139] Similarly, the contractual restriction in Acorn did not implicate the rights granted by the Copyright Act:
Implicit in the contract between Acorn and Swantz was an agreement that while Swantz did not have to use
Acorn's plans, if he did use Acorn's plans then he was obligated either to purchase the plans from Acorn or to
purchase his building materials from Acorn....This implicit provision of the contract...does not arise out of the
subject matter of copyright and is therefore a separate and distinct cause of action. [FN140] The fact-specific holdings of these cases - that contracts that did not merely forbid reproduction, distribution,
or display are not preempted - follow the same rule as National Car Rental. Accordingly, parallel logic dictates
that Taquino and Acorn in no way undermine the district court's ruling.
Having run through the triad of cases that underlay both the district and circuit courts' analysis in ProCD, it
appears that the rule safeguarding contract causes of action against copyright preemption is less than
categorical. Although the vast majority of contract claims will presumably survive scrutiny - as did each of the
contract claims confronted in that trio - preemption should continue to strike down claims that, though
denominated "contract," nonetheless complain directly about the reproduction right. It is precisely into that
paradigm that the facts of ProCD fall.
2. Giving Supremacy to the Supremes
We have already seen that rights premised under state law cannot stand as "an obstacle to the full purposes
and objectives of Congress." [FN141] As construed by a unanimous Supreme Court in Feist, Congress did not
(and, indeed, constitutionally could not) extend copyright protection to alphabetical telephone listings. The
shrinkwrap license at issue in ProCD undid the right of the public that Feist conferred - the ability to
copy telephone listings without liability. [FN142] As such, there would seem to be a direct conflict between that
which federal law permits and that which state law forbids. [FN143] It is remarkable, then, that the Seventh Circuit reversed. Judge Easterbrook's opinion, after determining
shrinkwrap licenses to be binding under the Uniform Commercial Code, [FN144] concluded that the district court
erred in its construction of section 301. [FN145] The Seventh Circuit did not consider constitutional preemption
apart from section 301, notwithstanding that, as discussed above, non-301 preemption occupies its own
capacious niche in copyright jurisprudence. [FN146] Combined with its facile reading of precedent under section
301, [FN147] its failure even to consider the broader constitutional issues dooms the Seventh Circuit's preemption
analysis.
The court attempted to highlight the error of the district court's approach through a reductio ad absurdum
A customer visits a video store and rents a copy of Night of the Lepus. The customer's contract with the
store limits use of the tape to home viewing and requires its return in two days. May the customer keep the
tape, on the ground that 301(a) makes the promise unenforceable? [FN148] The answer to that rhetorical question is as patent as it is incomplete. In this particular instance, the
contract relating to copyrightable goods should be deemed enforceable, as it passes the "extra element" test
and contravenes no federal policy. That solitary hypothetical at best demonstrates what is uncontroversial:
not every contract relating to copyrightable goods is preempted. [FN149] It does not demonstrate the far greater
proposition at which it hints: No contract relating to copyright-able goods ever is preempted under section
301(a). Moreover, even when a contract passes muster under section 301, it is still necessary to evaluate
whether it runs afoul of the more general preemption prerequisites mandated by the Constitution.
The court's hypothetical video store rental contract, albeit incomplete, still serves as a useful tool to further
the inquiry. The contract at issue requiring the return of the physical copy at the end of the rental period
meets the "extra element" test. Further, it contravenes no policy of the Copyright Act. To the contrary, it
fosters the purposes underlying copyright law by encouraging the dissemination of copyrightable works in an
orderly fashion. [FN150] Likewise, the vast majority of contracts that one may posit with respect to
copyrightable goods should prove easy to reconcile with the purposes underlying copyright. Whether the
contract consists of imposing time limits on when films can be shown, specifying geographic limitations on
where magazines can be distributed, or mandating quality controls on how sculptures can be displayed, each
disserves no readily apparent doctrine of copyright law. Accordingly, these contracts are presumptively
enforceable.
The contract at issue in ProCD, Inc. v. Zeidenberg differs from the foregoing examples in the one respect
relevant to nonstatutory preemption: [FN151] it contravenes one of the core policies of the Copyright Act by
extending quasi-copyright protection to works that do not qualify as "original." [FN152] It further fails the test of
encouraging the dissemination of copyrightable works in an orderly fashion in that it seeks to bar the
dissemination of uncopyrightable materials. It is, in short, nothing other than an attempt in effect to
overrule by contract binding Supreme Court precedent. [FN153] The Seventh Circuit conceded in ProCD, Inc. v. Zeidenberg that the structure of copyright law "prevents
states from substituting their own regulatory systems for those of the national government." [FN154] On that
basis, the Seventh Circuit refrained from "adopting a rule that anything with the label "contract' is necessarily
outside the preemption clause: the variations and possibilities are too numerous to foresee." [FN155] One
wonders what variation the court had in mind as embodying the substitution of a contractual scheme "for
those of the national government" if not a factual scenario on all fours with a recent, unanimous Supreme
Court opinion, but reaching the contrary result. [FN156] Yet the opinion did not even mention Feist in its
preemption analysis. [FN157] To appreciate what is at issue here, it is useful to conjure up similar examples of contracts designed to evade
the strictures of copyright law. Those variants shine a brighter light on the preemption inquiry than the
Seventh Circuit's pedestrian example of a contractual obligation to return a single copy of a rented videotape.
In each of the following examples, the postulated conduct disturbs the "delicate balance" that the district
court in ProCD was attempting to maintain. Adverting to the entire spectrum of preemption concerns (rather
than simply to section 301) illustrates that each of these contracts must bow before the superior force of the
federal enactment of copyright law.
Consider [FN158] first a state law that validates all oral contracts solemnly adjured before a panel of three
clergymen. [FN159] To the extent that that law were applied to a contract transferring copyright ownership, the
unanimous view of the cases is that it would be preempted under the Supremacy Clause, given the
federal command that transfers of copyright ownership be executed in writing. [FN160] That single hypothetical
by itself proves that absolute freedom of contract under state law relating to copyrightable works is
insupportable. [FN161] Consider next a video store that does not simply limit customers to home viewing (as imagined by the Seventh
Circuit's hypothetical), but goes further and stamps each videotape in its collection with the legend: MAY BE
VIEWED BY NO MORE THAN THREE PEOPLE AT ONE SITTING. [FN162] Even better, a studio that invariably
stamped its product with that proviso in a shrinkwrap encasing all videotapes that it manufactured would
closely mirror the actual facts of ProCD, Inc. v. Zeidenberg. For just as the plaintiff in that actual case sought
to use a shrinkwrap to deny the public the right to use phone listings guaranteed by Feist, so the fictive
studio in this hypothetical case would use its own shrinkwrap to avoid Congress' limitation of the performance
rights in videotapes to public performances, defined as occurring only when "a substantial number of persons"
is gathered. [FN163] The subject proviso, even if deemed operative as a matter of state contract law, should be
deemed preempted based on its conflict with the federal scheme. [FN164] Consider next an on-screen announcement prior to the airing of a television program that its broadcast to
viewers' homes is conditional on their agreement not to engage in private home taping. Assuming the threshold
determination that the viewers' decision to watch it notwithstanding the initial admonition constitutes a
contractual bargain just as much as receiving software by wire, [FN165] then the viewer's activation of her
VCR constitutes breach of contract. Now, not only has the Supreme Court's Feist decision been
effectively nullified - so has its ruling in Sony. [FN166]
In Sony, the Supreme Court stated that "[copyright] protection has never accorded the copyright owner
complete control over all possible uses of his work." [FN167] Yet in a world governed by Judge Easterbrook's
radical freedom to impose terms by shrinkwrap "contract," there is no reason that such a conclusion should
pertain. Instead, the imagination of shrinkwrap drafters can come close indeed to achieving the type of
complete control that Sony expressly denied them.
Consider the following example of that complete control. Although the Act limits the copyright owner's rights
to "public" distribution, [FN168] publishers who follow the logic of ProCD, Inc. v. Zeidenberg may amplify their
statutory rights simply by wrapping books [FN169] in cellophane, subject to the limitation that the buyer is barred
from passing the purchased copy on to a friend. [FN170] Nor is there any reason that the publisher should stop
there. It could likewise require the reader not to skip chapters, not to read any paragraph more than three
times, [FN171] not to reveal the surprise plot twists to family or acquaintances, and certainly not to quote in a
book review the few short excerpts that the fair use doctrine would otherwise permit. [FN172] The foregoing examples represent simply the tip of ProCD's iceberg. For not only could Sony and Feist be
nullified under its approach, but so could virtually every other court decision ever to rule in a defendant's
favor.
The Supreme Court gave wide berth to the parody defense in Campbell v. Acuff-Rose Music, Inc. [FN173] That defense essentially disappears once Roy Orbison and his fellow composers wrap their tapes and CDs in
jewel boxes armed with the appropriate shrinkwrap language forbidding parodic exploitation. In another
unanimous ruling during the same term, the Court held in Fogerty v. Fantasy, Inc. that attorney's fees need to
be awarded evenhandedly to plaintiffs and defendants. [FN174] But plaintiffs savvy to the options that ProCD
affords can avoid that nettlesome limitation by adopting the expedient of wrapping their goods in packages
that guarantee by contract treble awards of attorney's fees incurred for copyright infringement. [FN175] Circuit courts have validated reverse engineering of software when undertaken for proper purposes. [FN176] Merely by prohibiting that conduct under a shrinkwrap license, the nominally "fair use" is constricted out of
existence. [FN177] Before ending, one more example of recent vintage springs to mind. A unanimous Supreme Court has just
rejected copyright proprietors' rights to bar certain parallel importations under the Copyright Act. [FN178] That
impediment need not deter those shampoo manufacturers and others who simply include with their products
shrinkwrap licenses forever barring entry of the subject goods into the United States. Another Supreme Court
precedent likewise bites the dust under ProCD's view of the law. The list could be multiplied endlessly.
However, each example in the foregoing parade of horribles differs in one critical respect from the facts of
ProCD in that each involves interposing contract into the realm in which Congress had the ability to legislate.
For example, Congress could have validated oral transfers of copyright. It could have defined "public
performance" to mean performance before no more than three unrelated adults. It could have excluded parody
from "fair use" and instead imposed a mandatory license scheme for all would-be parodists. [FN179] It could have
limited the first sale doctrine effectively to bar all parallel imports. In other words, Congress could have
made these choices, but it did not do so when it set the bounds of the "delicate balance" between that which
enjoys copyright protection, and that which is free for all to use.
By contrast, ProCD deals with matters outside the ambit of Congress's legislative power as delimited by the
Copyright Clause. For example, the text of the Copyright Clause makes it clear that Congress cannot grant
copyright protection of indefinite duration. [FN180] Feist makes equally clear that Congress is constitutionally
barred under the Copyright Clause [FN181] from giving copyright protection to alphabetical telephone listings,
which by their nature fall below the constitutional threshold for copyrightability. [FN182] It may be argued that
the shrinkwrap license in ProCD therefore falls outside of the realm of preemption altogether because by
definition it cannot "stand[ ] as an obstacle to the accomplishment of the full purposes and objectives of
Congress." [FN183] This facile argument overlooks the fact that a state law need not fall within Congress's authority under the
Copyright Clause to interfere with the objectives of Congress. Consider, for example, a state law that provided
as follows:
Upon expiration of the limited term of copyright granted by Congress in a particular work, the owner of the
copyright shall, from the moment of copyright expiration forward into perpetuity, enjoy under the laws of the
State of California the exclusive right to do and to authorize any of the following: (1)to reproduce the formerly
copyrighted work; (2) to prepare derivative works based on the formerly copyrighted work; and (3) to
distribute copies of the formerly copyrighted work to the public by sale or other transfer of ownership, or by
rental, lease, or lending.
Strictly speaking, the foregoing law is outside the lawmaking authority of Congress under the Copyright
Clause for, as noted above, Congress is constitutionally constrained to provide limited terms of copyright
protection. But the foregoing provision would render meaningless the term limitations imposed by the Copyright
Act and should therefore be viewed as "standing as an obstacle to the accomplishment of the full purposes
and objectives of Congress" - copyright protection of limited duration. Similarly, the Feist-defeating provision
of ProCD should be viewed as "standing as an obstacle to the accomplishment of the full purposes and
objective of Congress" - extending copyright protection only to original works of authorship. Were that not the
case, state law could effectively defeat all of the constitutionally generated limitations of the Copyright Act.
Such a result would offend not only the Congressional scheme, but would run contrary to the design of the
Copyright Clause itself. [FN184] * * *
The Seventh Circuit's analysis of Supreme Court preemption cases further reveals the infirmities of its ruling.
For instance, ProCD cites Kewanee Oil Co. v. Bicron Corp., [FN185] a case in which the Supreme Court validated
trade secret law. [FN186] The point is well taken that the states are free to protect under that rubric materials
that are otherwise uncopyrightable - an alphabetical customer list, for example. [FN187] Trade secret laws
survive the "extra element" test and contravene no federal policy pertaining to copyright. For that reason,
legions of cases rule that copyright law does not preempt trade secret causes of action. [FN188] On the other hand, the same line of cases recognizes that it is only because those state laws limit protection
to works held confidentially that they survive preemption; otherwise, the distinction between trade secret law
and copyright would collapse, and trade secret law would be preempted under section 301. Thus, to the
extent that a state, as a matter of its own internal law, labels a published book of alphabetical telephone
listings a "trade secret," the law would be invalid under the Supremacy Clause.
That last result exactly parallels ProCD's circumstances. Although labeled "contract under the Uniform
Commercial Code" instead of "trade secret," the state rights under examination were precisely
congruent to those litigated (adversely to their claimants) in the copyright sphere. Thus, the Seventh Circuit's
reference to trade secret law, far from proving its approach correct, in fact reveals the error of its way. [FN189] Besides Kewanee Oil, the only other Supreme Court cases in the intellectual property arena that ProCD cites to
explicate its preemption analysis are Aronson v. Quick Point Pencil Co. [FN190] and Bonito Boats, Inc. v. Thunder
Craft Boats, Inc. [FN191] The former illustrates the general principle that contracts for non-commercialized goods
deserve enforcement; it does not illuminate the commercialized goods at issue in ProCD. [FN192]
Moreover, the Aronson Court's efforts to distinguish its ruling from Brulotte v. Thys Co. [FN193] cast further
doubt on the Seventh Circuit's reasoning. Brulotte held that parties may not extend by contract the maximum
federal term for patents. It therefore proves that freedom to contract intellectual property must bow before
the federal policies implicated in that law - the precise point that the district court in ProCD attempted to
vindicate.
Bonito Boats is even more powerful proof of how Judge Easterbrook's approach cannot stand. In Bonito Boats,
a unanimous Supreme Court ruled that states lack the power to pass laws barring the plug-molding of boat
hulls. [FN194] But under the logic of ProCD, relief for the losing plaintiff should be forthcoming under the state law
of contract, for the boat manufacturer simply needs to outfit its products with a shrinkwrap license
forbidding plug-molding. [FN195] A month after Bonito Boats was decided, Judge Easterbrook went on record condemning it as "a step in the
wrong direction." [FN196] There is nothing illegitimate about that point of view; it may one day turn the whole
Court around. But it is quite another matter for the Seventh Circuit to write Bonito Boats - along with Feist,
Sony, Campbell, Fogerty, and a host of other decisions - out of existence before that day has dawned. [FN197] * * *
The threat from potential elimination of user rights through contract, moreover, is not limited to the world of
shrinkwrap contracts. Greater freedom of contract should typically attach to negotiated contracts than to
non-negotiated, or "mass market" contracts. This distinction is reasonable if, in the negotiated context, the
potential licensee (or more generally, the potential recipient of copyright rights) can walk away and pursue
other options. The distinction between negotiated and non-negotiated agreements breaks down, however,
when all or significantly all access to copyrighted works is mediated through contracts purporting to control
uses of the copyrighted work. [FN198] A given copyrighted work (for instance, a particular operating system
software for personal computers) may be de facto necessary to take advantage of hardware and software
offered by other copyright owners. No matter how vigorously a potential licensee engages in an arms-length
transaction about how she may use that operating system software, she ultimately lacks any real option of
seeking better terms from a different source. One can imagine, as in the cautionary tale spun at the start of
this Article, a market in which copyrighted works as a whole can only be accessed through contract.
In these take-it-or-leave-it contexts, the lack of true choice means that the copyright owners' contract terms
operate in effect as "private legislation" that serves to alter en masse the public's rights granted under the
Copyright Act. [FN199] One concern about undermining the holding of ProCD is that to do so exposes valuable databases of
uncopyrightable materials to parasitic copying and undermines efforts to recoup investment costs through
price discrimination. [FN200] One must concede that if neither copyright law nor contract protects
uncopyrightable but valuable databases from copying, the incentive to create such databases and the
attendant social benefits of having such databases created may diminish. [FN201] Accordingly, one may argue
that it is socially desirable to offer protection for uncopyrightable databases [FN202] and to foster price
discrimination as an incentive to investment. [FN203] But a unanimous Supreme Court has recently accorded no
weight whatsoever to price discrimination as a basis on which to resolve dilemmas in copyright doctrine. [FN204]
Moreover, as the Feist court's abolition of the "sweat of the brow" doctrine should make clear, general
considerations of social desiderata plainly cannot justify ignoring preemption problems of constitutional
dimension. Result-oriented jurisprudence must be resisted, even if one believes the result correct. The
solution to parasitic copying of historical [FN205] databases, to the extent such protection is needed, [FN206] is
to be found in creating proper federal protection for uncopyrighted materials, not in encouraging a contract
regime to perform an end-run around the limits of copyright.
To conclude, if copyright law is to maintain an autonomous existence, instead of becoming an adjunct
to whatever lawyers can draft into shrinkwrap "contracts," then its delicate balance must be respected. For
that reason, the Seventh Circuit's holding in ProCD, Inc. v. Zeidenberg, it is submitted, is in error.
From the foregoing discussion, it should be clear that contracts can not only coexist within the overarching
copyright framework, but are essential for its proper implementation. Because U.S. copyrights are infinitely
divisible, contract is the only sensible means for dividing up spheres of exploitation. On the other hand, those
convinced by the foregoing discussion can harbor no doubt that not every device unilaterally imposed under
the rubric "contract" can pass constitutional muster. ProCD v. Zeidenberg clearly illustrates that phenomenon.
The question therefore remains where the dividing line lies between the permissible and the forbidden.
Abstracting from the above criticism of the Seventh Circuit's decision in ProCD, the appropriate dividing point
emerges organically from the copyright monopoly that Congress, acting within the framework of the
Constitution, bestows upon authors. If a copyright owner contracts to exploit a work up to the limits of his
constitutionally and congressionally conferred monopoly, he is acting legitimately; conversely, if an author
uses contract law to enlarge that monopoly to apply to exploitations beyond its congressionally sanctioned
orbit, she is behaving illegitimately. That latter conclusion follows whether the expansion derives directly from
state law [FN207] or as a matter of contract law, which ultimately derives its enforceability from the same body
of state law.
* * *
To illustrate, let us imagine several scenarios, each applicable to the new entity that hypothetically has
purchased all right, title, and interest in and to the MGM classic, Gone with the Wind.
Series (i):
. The owner grants A a license to screen the film theatrically, limited to theaters located on the south side of
Pico Boulevard in Los Angeles.
. B receives a license to engage in theatrical distribution anywhere in the United States, but limited to
alternate Tuesdays during leap years.
. C bargains for a license to engage in worldwide [FN208] distribution by videotape, for a period of three weeks
only, at a 6% royalty.
. D purchases rights of television broadcast in the Cambodian-dubbed version. [FN209] . E sews up all rights to Gone with the Wind in DVD format.
. F is the lucky winner of the same rights in CD++ format.
Series (ii):
. G receives the right, after C's three weeks expire, to engage in distribution of videotapes at the same 6%
royalty, which G agrees to pay for a term of 100years.
. H receives the right to engage in theatrical screenings on the north side of Pico Boulevard and throughout
the rest of Los Angeles. However, as a condition for granting that license, the proprietor insists that G also
pay a like amount to license Night of the Lepus.
. MYRRHA stands as guardian to the portals of all copyrightable compositions. [FN210] In order to see Gone with
the Wind, viewers must pay the current freight, set at $ 5 for each 37-minute segment. A comparable charge
is imposed for Night of the Lepus and every other motion picture. Novels and poems are also metered through
MYRRHA, albeit subject to a different tariff table.
Series (iii):
. The proprietor of Gone with the Wind has waited past the expiration of the film's term. Nonetheless, taking
advantage of MYRRHA access gates, it charges the same rates at present as were previously imposed for
anyone to make home copies of the work.
. In addition, the proprietor takes out a license in the menu tree of Lotus 1-2-3, invoking MYRRHA's technology
to impose on software developers a surcharge for incorporating that time-tested model into their latest
products.
Series (iv):
. I receives a license to display a still frame from the film in connection with news reports surrounding Clark
Gable's obituary. [FN211] . J receives a license to screen ten seconds of the film in conjunction with her UCLA seminar on great
films from the heyday of
Hollywood.
. K buys a shipment of 5000 videotapes of the film, each wrapped in a shrinkwrap license forbidding any and all
uses in Nevada, even if such qualify as "fair uses" under applicable law.
In analyzing the permissibility of all of the contracts hypothesized above, our fundamental axiom furnishes the
answer, at least to the first three series: contracts are legitimate up to the full extent of the copyright
owner's monopoly as sanctioned by Congress, and are illegitimate to the extent that they exceed the bounds
of that monopoly.
1. Permissible Limitations
Series(i) represents the uncontroversial cases. The Copyright Act grants proprietors the exclusive right to do
and to authorize each of the affected activities. Accordingly, those six contracts should each find
enforcement under the copyright regime.
2. Misuse and Related Notions
Moving to series(ii), the grant to G binds that licensee to pay royalties for a century into the future, long after
the film's copyright will have expired. That contractual term is illegitimate, as ruled by the Fourth Circuit in a
landmark case curtailing the legitimate scope of contracts over copyrightable compositions - even those
bargained for by both contracting parties. [FN212] A point of terminology is in order here. Although the landmark case just cited labeled the contract infirm under
the doctrine of "copyright misuse," that term is redolent of antitrust law. It is true that instances of antitrust
violations represent one particularly egregious instance of illegitimate behavior, but the doctrine nonetheless
should not be limited to that sphere. [FN213] The fatal flaw against which the Fourth Circuit reacted is that the
copyright owner there was seeking judicial enforcement of a contract that explicitly contravened the user
limitations incorporated by Congress into the Copyright Act, not that it was acting in violation of the Clayton
Act or the Sherman Act. Based on the elaborate edifice constructed above, it is submitted that the rubric of
"preemption" better describes the contract's infirmity than does "misuse."
Nonetheless, to the extent that an antitrust violation is present, the same result should inure a fortiori and the
subject contract must be denied enforcement. We thus reach H, in which the copyright proprietor has
engaged in what appears to be an illegal tying agreement (assuming, for the sake of discussion, that the
requisite requirements for market power in the market for the tying good are met and that buyers would not
otherwise license Night of the Lepus). [FN214] That improper behavior results in the invalidation of the subject
license.
Moving now to MYRRHA, one can envision a variety of other potential antitrust violations, from tying to
monopolization to price fixing. Insofar as the facts support a finding of an antitrust violation, the MYRRHA
licenses must be denied enforcement by virtue of the illicit goals they attempt to achieve.
3. Within the Subject Matter of Copyright Even as to Unprotected Works
Under the Constitution, as we observed earlier, federal copyrights endure for terms of limited duration, after
which public policy demands that the subject expression repose in the public domain. We have already
reviewed why state law cannot attempt to accord protection anew on such lapsed works. [FN215] Could
proprietors nonetheless unilaterally achieve the same result by wrapping their expired works in self-styled
proclamations of resurrection or in the other devices hypothesized in series (iii)? Because such devices would
render the "limited times" provision of the Constitution a nullity, they cannot stand.
The same consideration governs the other usage hypothesized in series (iii) concerning the Lotus spreadsheet.
That work likewise lies outside the limits of Congress' ambit to protect. [FN216] To allow the U.C.C. or other
devices enforced under guise of state law to achieve the contrary result again vitiates the federal scheme
and, as such, cannot stand.
4. Difficulty of Drawing the Lines of Fair Use
As always, issues at the boundaries can bedevil application of any straightforward standard. In this particular
application, copyright's infinitely elastic doctrine of fair use can pose its share of mischief. We thus
reach series (iv).
In the abstract, it would seem perfectly straightforward that use of a single frame of Gone with the Wind from
the hundreds of thousands of frames that comprise the picture - and for the purpose of illustrating the
newsworthy event of reporting the death of its leading man - qualifies as fair use. By the same token, showing
a few seconds to a film school class would similarly seem to qualify. Solely from that perspective, the licenses
to I and J must be stricken as illegitimate.
To spell out the foregoing perspective, the analysis runs as follows: First, a copyright owner is only entitled to
exploit her monopoly, not to expand it. Given that the statute itself carves fair use out of the scope of
monopoly granted the copyright owner, [FN217] the copyright owner cannot require a user to contract out of fair
use. Because the licenses to I and J deal with a use that ex hypothesi qualifies for the label of fair, these
contracts are overreaching.
The fly in the ointment here is that fair use is notoriously difficult to ascertain in the abstract. Indeed, one
usually cannot know whether or not a use is fair until litigation is resolved at the level of the U.S. Supreme
Court. [FN218] For that reason, it may be wise in almost any circumstance to license a work rather than roll the
dice in a lengthy judicial process. Particularly when one reflects that cases have held against fair use in the
selling of coursepacks destined for educational purposes, [FN219] and even in the context of using short clips for
seemingly highly newsworthy events, [FN220] the vector towards settling rather than fighting takes on
tremendous magnitude. Accordingly, the conclusion follows that the licenses to I and J should not be deemed
illegitimate. Instead, because a court might in fact confound expectations by ruling the subject utilizations
outside the fair-use doctrine, it is reasonable for two consenting parties to contract now to pay an
ounce of royalties in order to avoid many pounds of attorneys' fees later.
On the other hand, the fair-use doctrine cannot be ignored in the calculus of permissible contracts. The
license to K makes no bones about explicitly contravening it. In this particular instance, fine judgments as to
future predictions need not take place, for the contract itself eliminates rights guaranteed to users by statute
under any view of its appropriate construction. [FN221] The license, therefore, is impermissible.
Turning to the more vexing issue of reverse engineering, we would argue that the same considerations apply
as in license K considered above. In other words, to the extent that a shrinkwrap contract for a
commercialized good required blanket waiver of any and all reverse engineering rights, it should be deemed
preempted. Precedent up to the circuit court level has upheld that activity, when appropriately undertaken, as
falling within the fair-use doctrine. [FN222] Although the argument remains that the U.S. Supreme Court might
ultimately reject that construction, until it does so those circuit-court pronouncements should be deemed
declarative of the law. On that basis, a blanket diminution of user rights as defined by governing construction
of the statute runs afoul of the primary axiom against illegitimate expansion of the copyright owner's
monopoly. [FN223] At this juncture, it is legitimate to inquire what standards should inform a sensible drafting of U.C.C. rules. It
will be recalled that those rules govern state contract formation, that such state law contracts form an
essential ingredient in the scheme of federal copyright exploitation, but that certain state law contracts go
too far. The question posed is the extent to which Article 2B successfully mediates the potential clash
between copyright and contract.
The drafters of Article 2B are well aware that transactions in "information" are rife with the potential for clash
between copyright and contract. Theirs is an awareness born of abundant recent experience. Distribution of
software, the quintessential "information" product, does not typically transpire as a classic sale of a
copyrighted article, such as occurs when an end user purchases a copyrighted novel. In that classic sale
context, the end user is free to make use of the purchased product, to sell it, lend it, annotate it, all subject
only to the constraints of the Copyright Act.
Copyright owners of software, by contrast, increasingly do not simply sell a tangible embodiment of their
copyrighted work - a diskette or CD-ROM - and do not rely on the Copyright Act alone to constrain the end
user. Rather, by selling "goods" intangibly, such as via "clickwraps" executed over the Internet, they seek to
interpose a licensing relationship with the end user as an absolute prerequisite to access and use. Even
without that instrumentality, they still sell tangible goods via a purported contract designed to create a
licensor/licensee relationship. The familiar name for that device, as we have seen at length, is the "shrinkwrap
license." [FN224] The preemption issue arises in an acute form in the shrinkwrap arena - and thus for Article 2B - because of
the attempted use of shrinkwrap licenses to create an intellectual property regime based on contract law that
enlarges the copyright holder's rights by denying users their rights under the Copyright Act. In this parallel
universe of contract-created "copyright plus," the copyright owner enjoys all of the benefits conferred by the
Copyright Act, while users are required to forfeit some or all of the rights secured to them under the Copyright
Act, such as the right to make fair use of the work without the copyright owner's permission. [FN225] The theoretical brake against imperialistic aspirations here is preemption, which (as discussed above) forbids
state law from undermining the intellectual property scheme created by Congress. But the current draft of
U.C.C. Article 2B largely avoids any attempt to limit copyright contracts so as to respect the preemptive
boundaries of copyright. Instead, the drafters have adopted the position that "Article 2B should not deal with
federal preemption but should be neutral" and that that position should be stated in the comments. [FN226] Implementing this principle of "neutrality," Article 2B directly addresses preemption only by providing,
"[a] provision of this article which is preempted by federal law is unenforceable to the extent of such
preemption." [FN227] But the professed goal of neutrality is not one that the drafters of Article 2B consistently attempt to achieve.
For instance, the March Draft for Article 2B consciously departs from the model for sales of goods under U.C.C.
Article 2, in order to bring the former into harmony with "clear rules of federal law." [FN228] Implementing that
approach, the draft attempts to accommodate the Copyright Act's requirement of a writing in order to transfer
ownership of a copyright. [FN229] Similarly, Article 2B departs from the good-faith purchaser rules of state law,
because copyright law offers no such defense to claims of infringement. [FN230] Furthermore, the draft follows
the copyright cases that have held that the licensee's interest is not transferable without the licensor's
consent despite the fact that this rule contradicts some state law assumptions about transferability. [FN231] Instead, the U.C.C. drafters purport to reserve their strategy of "neutrality" - that is, doing nothing affirmative
- only for "controversial or context-determined rules whose application cannot be predicted and must of
necessity await determinations by individual courts in particular cases or by Congress as a general federal
policy question." [FN232] In other words, the draft is "neutral" only where it chooses to be. [FN233] One criticism of this purportedly "neutral approach" is that it is otiose. By simply restating in state law
the constitutionally mandated effect of preemption as contained in the U.S. Constitution, it accomplishes
nothing more than the exact same result that would inure even in its absence. [FN234] In particular, it does
nothing to elucidate what is preempted or to forestall contracting behavior that is impermissible in light of the
supremacy of copyright.
Further, as a practical matter, this selective "neutrality" suffers from precisely the same shortcoming of
political neutrality in response to real-world conflict: [FN235] it de facto favors those with concentrated interests
and large financial resources and thus tacitly invites abuses. By taking no position on preemption other than
that "it preempts," the draft ratifies the status quo and makes every imaginable shrinkwrap encroachment on
users' rights presumptively enforceable. It thus forces into the arena of litigation any determination of whether
a given license or portion thereof overreaches the supreme bounds of copyright.
The displacement into litigation of allegedly controversial issues favors vested interests over a diffused public
whose members lack both the resources and incentives to litigate. In addition, the draft favors the contraction
of user rights over the public's right to make fair use of copyrighted materials and to use freely what lies in the
public domain. It is in this sense that the metamorphosis from "contract" to "expand" becomes most apparent.
The draft's justification for not taking an affirmative stand on allegedly "controversial" issues is that
they "must be resolved by courts and Congress, rather than through state legislation." [FN236] That proposition
itself is suspect, as the two hypothetical bills discussed below demonstrate.
One must concede, of course, that Article 2B cannot anticipate every contract or context in which issues of
preemption will arise and provide an a priori fix. But limiting the focus for the moment to copyright preemption,
why could Article 2B not at least address the means by which licenses have been used to alter the "delicate
equilibrium" - that is, the known evils? Let us assume, for example, that the next draft were to take an
affirmative stand on copyright preemption by incorporating the following language, which has been proposed
as an amendment to the Copyright Act:
Boucher Bill: [FN237] Respecting copyright's delicate balance - When a work is distributed to the public subject to non-negotiable
license terms, such terms shall not be enforceable... to the extent that they: (1) limit the reproduction,
adaptation, distribution, performance, or display, by means of transmission or otherwise, of material that is
uncopyrightable under section 102(b) [of the Copyright Act] or otherwise; or (2) abrogate or restrict the
limitations on exclusive rights specified in sections107 through114 and sections117 and118 of [the Copyright
Act].
Although the foregoing language is not presented as the "silver bullet" to eliminate preemption problems, it
would at a minimum specifically address at least one preemption concern identified by the current draft as too
controversial: reverse engineering. Given that case law upholds certain species of such reverse engineering as
defensible under the fair use doctrine codified in section 107 of the Copyright Act, the Boucher Bill would
clarify that a reverse engineering prohibition in a shrinkwrap license is unenforceable to the extent that it
abrogates or limits the statutory fair use privilege. [FN238] More fundamentally, an Article 2B incorporating the Boucher Bill would generally safeguard users' rights by
rendering unenforceable provisions of licenses that seek to alter the "delicate equilibrium" that the
Copyright Act establishes between the rights of copyright owners and the rights of the public. Because it is
salutary, this hypothetical feature of a newly revised U.C.C. debunks the notion that debates in this sphere
"must be resolved by courts and Congress, rather than through state legislation." [FN239] The foregoing thought experiment [FN240] reveals no reason why Article 2B could not adopt the foregoing
proposed language, thereby taking a non-neutral stand on how preemption limits freedom of contract. The
bare fact that the U.C.C. drafters would thereby be "taking sides," by itself, is unobjectionable: choices are
inevitable and neutrality usually a chimera. Indeed, Article 2B self-consciously makes choices in numerous
domains to confer differential benefits on different constituencies. [FN241] The fact that the language proposed above overtly addresses preemption concerns is also unobjectionable.
For as previously noted, Article 2B consciously carves out exceptions to its scope in deference to what it
deems to be "clear rules of federal law." Moreover, the draft expressly states that if anyone desires to
challenge its provisions as an impermissible, or indeed preempted, state encroachment on the federal scheme,
he or she is free to seek redress in the "courts and Congress."
The legitimacy of the thought experiment contained in the Boucher Bill, which is premised on respect for
copyright law's delicate balance, emerges further when it is contrasted with a hypothetical addition to Article
2B designed for the opposite purpose, that is, one that subverts that balance. As we have noted above
repeatedly, the problem with the current draft of Article 2B is that it presumptively validates provisions which
impermissibly tilt in favor of proprietors. To appreciate why that one-sidedness is illegitimate, let us entertain
conversely a hypothetical provision crafted to enlarge users' rights, that is, for the opposite purpose from the
current draft:
Charcutier Bill:
Solicitude for the downtrodden - When a work is distributed to the public subject to non-negotiable license
terms, such terms shall not be enforceable unless the licensor: (1) grants the licensee the right to reproduce
an unlimited number of copies for persons falling below the poverty line as defined in the pertinent provision of
the Code of Federal Regulations; and (2) agrees that he or she will not seek more than $ 300 for any copyright
infringement committed by an orphan, widow or widower or by an individual adjudged "insane" pursuant to the
laws of the state in which the alleged infringement occurred.
It should be patently obvious that the language just proposed is preempted and hence unenforceable, for it
baldly encroaches on the exclusive rights granted to authors under the Copyright Act. [FN242] One hopes that
the drafters of Article 2B would peremptorily dismiss the Charcutier Bill as slop, even were massive lobbying
brought to bear for its adoption.
Parallel considerations dictate that license provisions should also be deemed preempted when they effect a
blatant encroachment on the rights of users; for example their right to reproduce, adapt, distribute, perform,
or publicly display material that was once copyrighted and the term for which has lapsed, or the right to make
fair use of a copyrighted work. It is here that the current draft of Article 2B needs fixing. By failing to protect
even those obvious user rights and by blessing as presumptively valid provisions that would rob users of their
rights, Article 2B's "neutrality" in effect tilts towards a regime that could be condemned as no less absurd than
the Charcutier Bill hypothesized above, when both are viewed from the perspective of rights that Congress has
denied to copyright owners and thus reserved to users. Better than either the pro-proprietor tilt of the current
draft or the pro-user tilt of the Charcutier Bill, what is needed is respect for copyright law's delicate balance.
Though the Boucher Bill itself is concededly not a panacea, it at least points in the right direction for a
balanced reformulation of Article 2B's stance on preemption.
Some proponents of the draft's current "neutrality" have suggested that it would be a mistake to build specific
preemption-driven restrictions into Article 2B, because this will force state courts into the quagmire of
preemption analysis, a task that they view as properly left to the federal courts as the usual arbiters of
interpreting the Copyright Act and its interrelationship with other laws. This leave-it-to-the-feds argument is built on the fallacy that copyright preemption issues can be neatly cabined within federal venues. In reality, as a necessary adjunct of resolving contract disputes, state courts cannot avoid making determinations regarding whether copyright law preempts particular contract provisions. [FN243] Under an Article 2B regime, state courts still will not be able to avoid determining whether given contracts (or contract terms) are preempted if the defense is raised. Accordingly, Article 2B can be silent on the issue, and offer no guidance (the faux "neutral" position), or it can offer some guidance by ruling out as unenforceable contract terms whose very purpose is to recalibrate the "delicate balance." The faux-neutral strategy does not eliminate adjudication of preemption issues in state court. Rather, it guarantees that state courts will face
preemption questions more frequently and on a broader range of issues.
A second leave-it-to-the-feds argument adduced by the proponents of faux neutrality is that questions of
what should be preempted, to the extent addressed in legislation, are best left to Congress. As noted above,
what is really being advocated is selective "neutrality" about the relationship between Article 2B and federal
law. For in numerous particulars recounted above, Article 2B is consciously crafted to be in harmony with the
supremacy of federal law. Given that the drafters of Article 2B clearly have the ability to take into account
issues of federal law and that they do so when they wish to, their antagonism to Article 2B forbidding certain
overreaching contracting practices on preemption grounds smacks of tendentious selectivity. At a bare
minimum, they should clarify why they have singled out this particular issue as the federal-law question to be
ducked, particularly when it is reasonably certain that precisely the evils contemplated herein (for example,
attempting to contract away fair use) will take place if not curbed. The worst that can be said about
implementing limits on contracting practices designed to restrict known evils is that if such contracts are really
not preempted, copyright proprietors will not be able to enjoy fully the freedom to contract - or at least will
have to fight to do so. [FN244] Copyright law itself will not suffer. By contrast, a failure in the other direction
threatens to impinge directly on the federal copyright scheme.
To sum up, whether Article 2B should choose "neutrality" over proscribing illegitimate contracting
practices is not an either/or choice between locating a debate in a federal forum as opposed to a state forum.
The real questions are: (1) How will the preemption analysis unfold in state courts and under what
presumptive standards (not whether it will take place)? (2) Who will bear the costs of attempting to diminish
users' rights to increase protections for copyright owners (not whether such costs exist)? and (3) Who should
be forced to make a federal case out of it before Congress, those who propose to tilt the delicate balance, or
those who would seek to maintain it?
In deciding each of these questions, we come down firmly in favor of copyright, and suggest that the
attempts to rework, alter, or eviscerate aspects of copyright through the vehicle of state contract law are
illegitimate. Accordingly, in our view, it is those who alter the delicate balance who should bear the cost of
this enterprise, and should be forced to rework copyright law where it has traditionally been reworked - in
Congress.
It is emphatically not necessary to view copyright law as a "law of users' rights" (as have some n245) to adopt
the logic that undergirds this Article - that copyright has solicitude for the rights of both authors and of users.
Instead, it is necessary simply to view copyright law as the carefully constructed compromise between
society's disparate goals, reflecting the delicate equilibrium invoked so often throughout the discussion above.
As the Supreme Court has taught, "the economic philosophy behind the clause empowering Congress to grant
patents and copyrights is the conviction that encouragement of individual effort by personal gain is the best
way to advance public welfare through the talents of authors and inventors in "Science and useful Arts.'" [FN246]
By contrast, [FN247] some may view copyright not as embodying any balance at all, but instead solely as a
device to maximize the financial interest of proprietors. To them, unilateral expansion of the copyright
monopoly beyond its congressionally sanctioned orbit is to be celebrated. Obviously, those Overview
I The Inevitable Coexistence of Copyright and Contract
A. Transfer
B. Works for Hire
C. Sale, Rental, Lease
D. Contract Formation
E. Limits of Contract Rights
II Contract Need Play No Role in Protecting Copyright Rights
A. Enforceability of Unilateral License Terms
B. The Conflation of Licensing with Distribution
The considerations set forth above assimilate software contracts to traditional means of copyright
exploitation. The Article 2B posits "two distinct frameworks" in this regard. [FN65] It is worth reviewing the
drafters' view of the matter at some length:
The first [framework] involves use of a master copy and is common in the movie industry and in software
contracts. Under this framework, a "distributor" receives access to a single master copy of the information
work and a license to make and distribute additional copies or to make and publicly perform a copy. For
example, Correl Software may license a distributor to allow its software to be loaded into the distributor's
computers or video games. The contract will contain a number of terms. Correl may limit the distributor to no
more than 1,000 to be distributed only in the computers and only if subject to an end user license. Since both
the making copies of and the distribution of copies are within the scope of the owner's copyright, acts that go
outside the contractual limitations are infringements as well as contractual breaches.
III The Limits on Contract via the Preemptive Force of Copyright
Our discussion in Part I demonstrates that state contract law acts as a necessary complement to the
Copyright Act by delineating the basic mechanics of contract formation, performance, and interpretation.
Those salutary goals can find further expression in Article 2B. The analysis in Part II demonstrates that state
contract law is not needed to protect the copyright interests of copyright proprietors. Those interests are
safeguarded by the Copyright Act itself. What remains to be discussed are the ways in which state contract
law can improperly meddle with the copyright laws by altering the copyright laws in favor of proprietors at the
expense of users. The phenomenon of attempted contractual displacement of copyright, and the limits of
contract in that regard, arises in the doctrinal context of preemption.
A. General Preemption Notions
B. A Case Study in the Contract/Copyright Clash: ProCD v. Zeidenberg
1. A Twist on Feist
C. Preemption Analysis of ProCD
[FN117] D. Framework of Preemption Principles: To the Limits of Monopoly
IV Some Lessons for the U.C.C.
A. The Neutrality Myth
B. How the U.C.C. Might Help Maintain the "Delicate Balance"
Conclusion