The Second Circuit decided The Cartoon Network v. CSC Holdings & Cablevision already two and a half weeks ago. This means light years in terms of blogging, but due to its importance I've decided to take the liberty and offer a late (and somewhat elaborate) entry discussing this ruling and some of its implications. The Cartoon Network is a very significant decision and I will not be surprised to see it surfacing in many future decisions and law review articles to come. Its central import is in pinching a big hole in the balloon often referred to as the “MAI v. Peak and its progeny.” I believe that the decision will have far-reaching ramifications on the development of temporary reproduction law in the future.
The basics first: Infringement of the reproduction right is the unauthorized reproduction of the work in copies or phonorecords (in short, the making of unauthorized “copies”). Copies are fixations of the work in a tangible medium of expression, and fixation of a copyrighted work occurs “when its embodiment in a copy is sufficiently permanent or stable to permit it to be perceived, reproduced, or otherwise communicated for a period of more than transitory duration.” (So says Sec. 101 of the Copyright Act). It is possible to break the fixation requisite into two components: The first is the “perceived, reproduced or otherwise communicated” test (or in short, the PRC test), and the second is a limitation clause, which excludes representations that last for a very short period of time (i.e., those that do not last “for a period of more than transitory duration”).
The legislative history of the 1976 Copyright Act explained the motivation behind the language of the limitation clause: “[T]he definition of 'fixation' would exclude from the concept purely evanescent or transient reproductions such as those projected briefly on a screen, shown electronically on a television or other cathode ray tube, or captured momentarily in the 'memory' of a computer.” (House Report 94-1476 (1976) at 53). A PRC fixation test without any limitation clause whatsoever would be meaningless, since any perceptible physical phenomenon can pass that test, even a reflection in the mirror (an analogy Pam Samuelson has suggested long ago). So Congress supplemented the bare PRC fixation test with a limitation clause that can be described as a quantitative-temporal restriction on the concept of fixation: It appears that the limitation on the concept of fixation is quantitative in nature, by measuring the duration of signals’ representation in time units. Notably, Congress did not stipulate the exact minimum duration that would draw the line between a “fixation” and a representation that is too short to be considered as such.
About two years after the enactment of the Copyright Act the CONTU report was published. The report muddied the water by explicitly challenging the meaning and operation of the limitation clause. It stated that “[t]he text of the [Copyright Act] makes it clear that the placement of a copyrighted work into a computer… is the preparation of a copy… Because works in computer storage may be repeatedly reproduced, they are fixed and, therefore, are copies.” That statement was accompanied by a footnote which specifically referred the aforementioned explanation of HR 94-1476. Footnote 111 of the CONTU report stated: “Insofar as a contrary conclusion is suggested in one report accompanying the new law, this should be regarded as incorrect and should not be followed.” (emphasis added). This text indicates that the CONTU drafters sough to reduce the limitation clause from an independent condition necessary for determining fixation to a subsidiary question assumed away under the bare PRC test. In other words, if the digital temporary representation at issue could be copied from the physical medium, for example, then the limitation clause should be automatically satisfied.
We fast forward to the early 1990s. The Ninth Circuit in MAI (1993) concerned RAM copying in the course of software servicing and repair by an unauthorized party. MAI cited the CONTU report (among other authorities) before concluding that “since we find that the copy created in the RAM can be ‘perceived, reproduced, or otherwise communicated,’ we hold that the loading of software into the RAM creates a copy under the Copyright Act. 17 U.S.C §101.” This is the essence of the RAM doctrine that has been virtually predestining the outcome in numerous subsequent court decisions (and not necessarily only those coming from the Ninth Circuit’s jurisdiction). One later decision was Triad Systems v. Southeastern Express, a 1994 decision handed down by the Northern District of California, which discussed electronic representation of digital signals on a memory component that lasted for no more than a millisecond. Committed to MAI as a controlling precedent, the Triad court upheld fixation under the circumstances with the following reasoning:
[T]he copyright law is not so much concerned with the temporal “duration” of a copy as it is with what that copy does, and what it is capable of doing, while it exists. “Transitory duration” is a relative term that must be interpreted and applied in context. This concept is particularly important in cases involving computer technology where the speed and complexity of machines and software is rapidly advancing, and where the diversity of computer architecture and software design is expanding at an ever-increasing rate.
That decision was reversed in part on appeal for unrelated reasons, yet it nicely represents the mainstream trend followed by post MAI courts. MAI has left an unequivocal, irremovable imprint on the case law. No court decision I am aware of seems to have challenged the RAM doctrine for about a decade and a half. This has been a crucial period in the growth, development and spread of digital technology and network transmissions. (One half exception is Costar v. Loopnet, a 2004 decision by the Fourth Circuit, which indirectly disagreed with MAI by adopting a qualitative interpretation to the statutory definition of "fixation"). Now we have The Cartoon Network. In brief, the Second Circuit has found a 1.2 seconds' representation not sufficiently long to qualify as fixation in the context of infringement of the reproduction right. The Second Circuit held:
We believe that this language [the statutory fixation definition] plainly imposes two distinct but related requirements: the work must be embodied in a medium, i.e., placed in a medium such that it can be perceived, reproduced, etc., from that medium (the “embodiment requirement”), and it must remain thus embodied “for a period of more than transitory duration” (the “duration requirement”)…
Unless both requirements are met, the work is not “fixed” in the buffer, and, as a result, the buffer data is not a “copy” of the original work whose data is buffered … we construe MAI Systems and its progeny as holding that loading a program into a computer’s RAM can result in copying that program. We do not read MAI Systems as holding that, as a matter of law, loading a program into a form of RAM always results in copying…
(emphasis in original)
Halleluiah! Finally a decision that takes the words of the statute seriously. For some reason, the Second Circuit preferred to distinguish MAI, thereby avoiding a direct confrontation and the potential for a clear circuit split. The Cartoon Network reasoned that distinction by holding that MAI decided only on the PRC prong of the fixation test, not on the limitation clause (i.e., MAI did not touch the question of duration). A careful reading of MAI suggests, I believe, that the Ninth Circuit intentionally disregarded the limitation clause in order to avoid having to answer the question of how many second (or milliseconds) constituted the modicum of fixation. Unfortunately, the Second Circuit now fell precisely to this very trap by stating that 1.2 seconds was too brief. What is then the minimum duration rendering a digital representation subject to the exclusive reproduction right (or a candidate for an exception or the fair use defense)? Well, it is safe to say that no one knows the answer to this question, and the Second Circuit in The Cartoon Network did not provide concrete indications in this regard.
Jimi Hendrix once sang about castles made of sand that fall in the sea, eventually. Well, I think that the RAM doctrine has been such castle for more that fifteen years. But The Cartoon Network runs directly into the quandary nobody wants to deal with, of thinking seriously about setting a clear temporal bar to fixation. As I see it, there are two ways to address this puzzle. One alternative is to read into the limitation clause a qualitative benchmark, somewhat in line with Triad and Costar. We then stop counting seconds and ask what the reproduction can do and what the normative arguments for considering it as implicating copyright are. The second alternative is more radical, advising us to forget about the reproduction right altogether and envision a legal reform that would be geared at creating a copyright regime that better accommodated questions of digital information in general. According to this approach, “copies” and “fixation” are terms belonging to the analog reality, and the choice we have is either to continue running around in circles, or to break the circle by revising the structure of the exclusive rights. Perhaps the sand castle we see falling to the sea is copyright law as we know it.