Claire Corlett

Fish Food, Fish Tanks, and More
William Fisher, CopyrightX: Lecture 7.1, The Rights to Reproduce and Modify: Reproduction

William Fisher, CopyrightX: Lecture 7.1, The Rights to Reproduce and Modify: Reproduction

Hello, I’m Terry Fisher. This is the seventh of
12 lectures on copyright. We’re just past the halfway
point of this lecture series. By now, you should be familiar
with the rules governing what types of creative works are
subject to copyright protection, who owns the copyrights in those things,
and how one secures and transfers a copyright. In addition, you should
feel comfortable with three of the four major
theories of copyright– specifically, the fairness theory, the
personhood theory, and the utilitarian or welfare theory. We arrive, finally, at the
heart of copyright law. In the next three lectures, we’ll
examine the principal rights that a copyright owner enjoys
and the corresponding limitations upon those rights, designed to
accommodate the interests of the public at large. Here’s how my presentation
will be organized. In this lecture and
the next one, I’ll be examining the economic rights
enjoyed by a copyright owner. As you’ll see, there are four
clusters of such rights– first, the copyright owner’s exclusive right
to reproduce the copyrighted work; second, the exclusive right to make
modifications of the copyrighted work, commonly known as the
right to make derivative works; third, the right to control
the distribution, exportation, and importation of copies of
the work; and fourth, the right to control public performances
or public displays of the work. Today, I’ll be examining the first
two of those rights– the right of reproduction and the
right of modification. Next week, I’ll take up the
third and fourth of the rights. And the week after that, I’ll discuss
the major exceptions and limitations that copyright law imposes
upon these exclusive rights. During this tour, I will focus, as I
did during the first half of the course, primarily on the law
in the United States. But I will also identify some respects
in which the law in other countries parallels or differs from US law. As usual, I’ll be
supplementing my presentation with various audiovisual materials,
one of which is in front of you– a map of copyright law, the
current version of which is available on my
homepage, So let’s begin. Section 106 of the US Copyright
Statute confers on copyright owners a broad and seemingly
unqualified set of rights. Here they are. Section 107 through 122 then
carve out of those rights a long list of exceptions. Some of those exceptions are
very specific and narrow. Others are broader. The broadest of all is the fair use
doctrine, embodied in Section 107, which we’ll discuss in two weeks. As you can see, the introductory
clause of Section 106 makes the rights it creates,
quote, “subject to” end quote– in other words, qualified
by– Sections 107 through 122. The result of this structure
is that, to get an accurate sense of the scope of a copyright
owner’s exclusive rights, you must oscillate between the generous
grants contained in Section 106 and the myriad exceptions contained
in the following provisions. And that’s what we’ll be doing. The first and simplest of the rights
conferred on a copyright owner by Section 106 is the right
to reproduce– in other words, to copy– the copyrighted work
in copies or phonorecords. This entitlement is commonly
considered the core of copyright. The name of this entire
field suggests as much. The heart of copyright is
the exclusive right to copy. As it’s used in the law,
the term “reproduction” has three independent dimensions. To prevail, a plaintiff
must satisfy all three. The first pertains to the way
in which the defendant generated the allegedly infringing things. Only if he did so by copying
is the defendant liable. The second pertains to the character
of the allegedly infringing things. Only if they constitute,
quote, “copies,” close quote, is the defendant liable. And the third pertains to the nature
and amount of the material the defendant took from the plaintiff. Only if it rises to the level of, quote,
“improper appropriation,” close quote, is the defendant liable. We’ll consider them in that order. In patent law, as some of you know,
independent invention is no excuse. A defendant cannot escape liability by
showing that he dreamed up, on his own, an invention that happens to
match the plaintiff’s patent. Not so in copyright law. There, independent
creation is an excuse. Suppose, for example, I write a poem. Later, you, never
having seen mine, write and publish exactly the same poem. I have no claim against you. You infringe my copyright in
my poem only if you copy mine. So what exactly does that mean? The clearest type of copying
is mechanical reproduction. If you make a photocopy of the piece
of paper upon which my poem is written, you have copied it. If you rip a CD containing a sound
recording, which, in turn, embodies a musical composition, you have copied,
specifically in a phonorecord, both the composition and the sound recording. If you replicate a CD-ROM
containing a software program, you infringe the
copyright in the program. None of this should be surprising. But as used in the
law, the term “copying” is not limited to
mechanical reproduction. It also includes having the
copyright work in mind when making a substantially
similar embodiment. So if you read and memorize my poem and
then write it down, you have copied it. If you listen to a performance
of my musical composition and then record your own rendition of
my composition, you have copied it. If you watch a movie and then
write a short story containing an identical plot, you have copied the
audiovisual work embodied in the film, and so forth. The most intriguing cases
in this area involve situations in which the defendant
is exposed to the plaintiff’s work and then, sometime later,
creates a very similar work without being aware that it
mimics the plaintiff’s work. In other words, the defendant believes
he is being original, but he’s not. The law governing such
situations is harsh. In the 1924 case of Fred
Fisher versus Dillingham, Judge Learned Hand summarized
the relevant rule as follows. Quote, “Everything registers
somewhere in our memories, and no one can tell what may evoke it. Once it appears that another
has, in fact, used the copyright as the source of this production,
he has invaded the author’s right. It is no excuse that, in so doing,
his memory has played him a trick.” Close quote. The most famous case of this general
sort involved the following facts. In 1963, The Chiffons, a popular
quartet in the United States, released a recording of
“He’s So Fine,” a song that had been composed by Ronald Mack. Here’s a brief excerpt. [MUSIC – THE CHIFFONS, “HE’S SO FINE”] The Chiffons’ recording was a hit in
both the United States and in England. George Harrison heard
the song, but forgot it. Six years later, he composed
and performed “My Sweet Lord.” Here’s an excerpt. [MUSIC GEORGE HARRISON, “MY SWEET LORD”] Bright Tunes Music, the
assignee of Mack’s copyright, brought an infringement suit
against Harrison and prevailed. Both the trial court
and the Court of Appeals credited Harrison’s testimony that he
was unaware of The Chiffons’ recording when he made his own song. However, they ruled, Harrison
could not escape on this basis. Here’s the key passage in the
appellate court’s opinion. Quote, “It’s not new law
in this circuit that, when a defendant’s work is copied from
the plaintiff’s, but the defendant, in good faith, has forgotten
that the plaintiff’s work was the source of his own, such
innocent copying can nevertheless constitute an infringement. It’s settled that intention
to infringe is not essential under the Copyright Act.” The court then went on to
suggest a possible policy basis for this seemingly
unforgiving rule. Quote, “As a practical
matter, the problems of proof inherent in a
rule that would permit innocent intent as a defense
to copyright infringement could substantially undermine
the protections Congress intended to afford copyright owners. We therefore see no reason to retreat
from this Circuit’s prior position that copyright infringement
can be subconscious.” The rule announced and applied by the
court in this case bears emphasis. Innocent copying is still infringement. In this special sense,
copyright infringement is a strict liability offense. The final, perhaps non-obvious,
aspect of the scope of copying is that it encompasses translation
of the plaintiff’s work into a different medium. So for example, an unauthorized
motion picture version of a novel, a doll that mimics a
cartoon character, a still photograph of a ballet are all covered. So to summarize, a
plaintiff, to establish that a defendant has violated
the plaintiff’s exclusive right of reproduction, first must
show that the defendant, quote, “copied” the plaintiff’s
work, which means either that the defendant
mechanically reproduced it or had the plaintiff’s work
in mind, even subconsciously, when making his own product. How, then, does a plaintiff
prove copying in this sense? There are four techniques listed
in the map in front of you. First, the plaintiff can introduce
direct evidence of copying. Surprisingly often, evidence survives of
a defendant’s conscious intentional use of the plaintiff’s work during the
course of his own creative processes, and that evidence turns up in discovery. Second, the plaintiff can show that
the defendant had access to his work and that the defendant’s
work and the plaintiff’s work are sufficiently similar to
raise an inference of copying. For example, the plaintiff
may be able to show that the defendant had the
plaintiff’s book on his bookshelf or received the plaintiff’s
screenplay in the mail. More commonly, the plaintiff may be
able to show that his work– say, a song or a movie– was
widely performed in areas and at times that would
make it highly likely that the defendant heard or watched it. So for example, few people, at
least in the Western hemisphere, can plausibly contend that
they’ve never heard a Beatles song or watched a James Bond movie. When the plaintiff can
demonstrate access of this sort, all he then needs to demonstrate
is sufficient similarity between the defendant’s
and the plaintiff’s work to make plagiarism more
likely than independent creation. The character and amount of similarity
that’s necessary for this purpose is sometimes called
probative similarity. It’s different from
the kind of similarity that we will discuss in a minute,
under the heading of “Improper Appropriation,” because
its function is different. Its role, to repeat, is to support an
inference that the way the defendant created his work was wrong, not that
the amount he took was too much. The courts, unfortunately, sometimes
conflate these two types of similarity, but you should not. What if the plaintiff has no evidence
that the defendant had access to his work, but can show
that the defendant’s work is extremely similar to his? Is that enough to show copying? US courts use to disagree on this point. Some Courts of Appeals
said yes; others said no. All of the Courts of Appeals now seem
to agree that sufficiently, quote, “striking,” close quote, similarity
between unusual aspects of the two works can suffice. However, the inference of copying
that arises in such circumstances can be overcome if the
defendant demonstrates that he did not have access
to the plaintiff’s work or that both the plaintiff’s
work and the defendant’s work could have been copied from
something in the public domain. Finally, an especially
powerful way of showing copying is to demonstrate
that the same errors can be found in both the plaintiff’s
and the defendant’s works. Sometimes, such errors are deliberate. The writers of travel
guides, for example, often insert fictitious entries in
their lists of hotels and restaurants precisely in order to enable them
to detect copying of their material by the publishers of other guides
to the same cities or countries. This is a very effective way
of ferreting out copying, although it can be inconvenient
for the readers who visit the non-existent sites. At other times, the
errors are inadvertent. Paul Goldstein, for
example, points to one case in which a legal publisher,
as part of an infringement suit against another publisher, admitted
that, in 50 pages of the plaintiff’s treatise, there were 138
inaccurate citations, and the same mis-citations
appeared in the defendant’s text– embarrassing, but convincing,
evidence of copying. Assuming that the plaintiff
can surmount this first hurdle, the second of the three
things the plaintiff must show is that the thing the
defendant made through copying is sufficiently concrete and
stable to be called a copy. In the overwhelming majority
of cases, this is easy. But occasionally, a
plaintiff falters here. The statutory definition the plaintiff
must satisfy is shown on your screen. From this language, the courts
have extracted the requirements that the thing produced by
the defendant be tangible– in other words, a material
object– fixed, and intelligible. The most slippery of these
requirements is fixation. You’ll recall, I hope, that fixation is
a precondition of copyright protection under US law, although not
under the law of most countries. Here, we encounter a second,
closely related manifestation of the laws concerned with fixation. Not only must the plaintiff’s
work be fixed in order to enjoy copyright protection;
the defendant’s work must also be fixed in order to violate the
plaintiff’s copyright, at least under Section 106(1). Recently, the Second
Circuit Court of Appeals decided a complicated and
important case in which this fixation requirement
proved critical. Because the case implicated
several other legal rules, as well, I’m going to describe
the facts in some detail. Then, I’ll discuss one of the
issues raised by those facts today and come back to other
dimensions of the case next week. Cablevision is a small cable TV
system based in New York City. Like all cable companies, it
aggregates audiovisual programming from several sources. For example, Warner Brothers may
produce a movie, license it to NBC, and then Cablevision
picks up NBC’s signal. Cartoon Network and HBO provide
programming directly to Cablevision, and so forth. As you might expect,
Cablevision pays for the right to broadcast this
material either according to the terms of voluntary licenses
or through a compulsory license. Those licenses are
important and complex. We’ll touch on them briefly at
the end of the next lecture. For present purposes,
all you need to know is that these licenses do not
give Cablevision the right to make copies of the TV
shows or movies in question, only to publicly perform them. So to return to the
narrative, Cablevision combines these various
streams of programs into a single, composite,
encoded signal, which it sends via cable,
of course, to its customers. Each customer receives the composite
signal through a set-top box, which decodes it, relays it to a
television, which, in turn, displays it to the subscriber sitting on his couch. If, as is likely, the
subscriber doesn’t want to watch all shows at the
times they are broadcast, he’s likely to employ a
device to record them. Once upon a time, the device
would have been a VCR. Today, it’s likely to be a digital
video recorder, commonly known as a DVR. Then, when the subscriber
wants to watch a given show, he plays the copy stored on the DVR. So this pattern should be
entirely familiar to most of you. But if you’ve been following this
course of letters from the beginning, it should make you uneasy. There’s no question that the
DVR recording, represented by the blue dot on this slide, is a copy
of the television shows in question. The subscriber who makes
those copies does not have a license from Warner Brothers
or Cartoon Networks to do so. Is the subscriber then violating
section 106(1) of the copyright statute? The answer is yes, but that
violation, as we’ll see, is excused under the fair use doctrine. When and how that crucial
principle was established we’ll discuss in a few weeks. For now, it’s enough to
recognize that the subscriber can slip the noose by pleading fair use. Now, back to our story. In 2006, Cablevision had an idea. What if, instead of relying
on individual subscribers to make copies of shows
on their home DVRs, it offered to make copies for them
using a cloud-based technology? This would be more efficient and
convenient than the home devices, sufficiently so, that the subscribers
would be willing to pay a premium for it. So Cablevision created,
though it did not immediately implement, what it called a remote
storage DVR, or RS-DVR system. Its main features, in the language
of the court eventually called upon to review the system, looked like this. Quote, “Under the new RS-DVR,
this single stream of data is split into two streams. The first is routed immediately
to customers, as before. The second stream flows into a device,
called the broadband media router– BMR– which buffers the data
stream, reformats it, and sends it to the Arroyo server, which
consists, in relevant part, of two data buffers and a number
of high-capacity hard disks. The entire stream of data
moves to the first buffer, the primary ingest
buffer, at which point the server automatically inquires
as to whether any customers might want to record any of that programming. If a customer has requested
a particular program, the data for that program move from the
primary buffer into a secondary buffer and then into a portion of one of the
hard disks allocated to that customer. As new data flow into
the primary buffer, they overwrite a corresponding
quantity of data already on the buffer. The primary ingest buffer
holds no more than 0.1 seconds of each channel’s
programming at any moment. Thus, every tenth of a second,
the data residing on this buffer are automatically erased and replaced. The data buffer in the BMR holds no
more than 1.2 seconds of programming at any time. While buffering occurs at other
points in the operation of the RS-DVR, only the BMR buffer and
the primary ingest buffers are utilized absent any request
from the individual subscriber. To the customer, however, the processes
of recording and playback on the RS-DVR are similar to that of
a standard set-top DVR. Using a remote control, the
customer can record programming by selecting a program in
advance or by pressing the Record button while playing a given program. To begin playback, the
customer selects the show from an on-screen menu of
previously recorded programs. The principal difference in operation
is that, instead of sending signals from the remote to an on-set box, the
viewer sends signals from the remote through the cable to the Arroyo server
at Cablevision’s central facility, which then delivers the program
to the subscriber’s home. In this respect, RS-DVR more closely
resembles a video-on-demand service, whereby a cable subscriber
uses his remote and cable box to request transmission of
content, such as a movie, stored on computers at the
cable company’s facility. But unlike a video-on-demand
service, RS-DVR users can only play content that they
previously requested to be recorded.” So those are the facts. You can guess what happened next. Cartoon Network and the owners of
the copyrights in the other shows that would be delivered to
subscribers through the system initiated a lawsuit
against Cablevision, not because they thought
this was a bad idea, but because they wanted Cablevision
to pay them additional license fees to deploy this new feature,
which Cablevision was unwilling to do. The plaintiffs contended that the RS-DVR
system would violate their copyrights in not just one, but
three, separate ways. First, they argued that
buffering an audiovisual work, specifically for 1.2 seconds
in the BMR and for a tenth of a second in the primary ingest
buffer of the Arroyo server, creates a series of replicated
pieces of the work, which, although each one is
stored only briefly, gives rise in the aggregate
to a copy of that work and thus violates Section
106(1) of the Copyright Act. Second, they argued that making the
much more durable copies of their shows on the Arroyo server
hard drives gives rise to a second violation of Section 106(1). Third, they argued that the delivery
of the programming from the Arroyo server to the subscriber
at the subscriber’s request would constitute an unauthorized, quote,
“public performance,” close quote, in violation of Section 106(4). We’ll discuss the second and third
of those claims in future lectures. For present purposes, we’re
concerned only with the first one. The district court agreed
with the plaintiffs that the buffer triggered a
violation of section 106(1). But the Court of Appeals reversed. Here’s the crucial language in
the appellate court’s opinion. “Copies, as defined
in the Copyright Act, are material objects in which
a work is fixed by any method and from which the
work can be reproduced. The act also provides that a work
is fixed in a tangible medium and expression when its embodiment
is sufficiently permanent or stable to permit it to be reproduced
for a period of more than transitory duration. We believe that this language 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, et cetera, from that medium;
and it must remain thus embodied,” quote, “for a period of more
than transitory duration. 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.” Because the data being channeled
toward the Arroyo server reside in no buffer for
more than 1.2 seconds before being automatically
overwritten, the court concluded that the duration
requirement had not been satisfied and thus that the defendant’s
technology did not result in the creation of any copies. For reasons we’ll explore
later, the Court of Appeals also rejected the second and
third of the plaintiff’s claims. As we’ll see, the bases of those
rulings were also highly technical, and thus arguably a
bit counterintuitive. The net result was that Cablevision was
permitted to deploy its new service. This decision, which the Supreme
Court declined to review, has facilitated the emergence of many
analogous cloud-based services, which, in turn, have had huge
economic implications. As Josh Lerner discovered, quote, “VC
investment in cloud computing firms increased significantly in the
United States relative to the EU after the Cablevision decision.” The decision, it appears, led to
additional incremental investment in US cloud computing firms
that ranged from $728 million to approximately $1.3 billion
over the two and a half years after the decision. When paired with the findings of
enhanced effects of VC investment relative to corporate
investment, this may be the equivalent of $2 billion to $5
billion in traditional R&D investment. An essential step in the ruling
that catalyzed this investment was, as we’ve seen, the court’s judgment
that the fixation requirement had not been satisfied by the buffer copies. So let’s take a break at this point. When we resume, we’ll consider
the third and arguably most important component of the
cause of action under Section 106(1), the requirement of, quote,
“improper appropriation.”

Leave a Reply

Your email address will not be published. Required fields are marked *