Today, the House Judiciary Committee Subcommittee on Courts, Intellectual Property, and the Internet is continuing its comprehensive review of US copyright law with a hearing on moral rights, termination rights, resale royalty, and copyright term.

Perhaps no other feature of copyright law is so widely criticized as its term — to the point where the declaration that “copyright lasts too long!” seemingly needs no further support.

But perhaps no other copyright doctrine is also subject to as much misinformation as the term of copyright. It’s not difficult to find those who are absolutely convinced copyright term will be extended within the next several years, because this is part of an indisputable pattern of extending copyright term every 20 years.1 These and similar assertions evaporate under the slightest scrutiny. Therefore, before we can have a useful discussion on the length of copyright protection, it’s necessary to look at how the current term of copyright protection evolved in the US and how that compares to copyright protection in other countries.

The genesis of copyright law

The journey begins in England with the Statute of Anne. Passed in 1710, the Act is considered the first modern copyright law, vesting exclusive rights to reprint books to any and all authors. The Act also limited the exclusive rights to a term of fourteen years, with an option to renew protection for an additional fourteen years. At the end of this period, the exclusive rights expired, allowing anybody besides the author or publisher to print and reprint the work. This length was borrowed from the term for protection of letters patent that the earlier Statute of Monopolies established in 1624.2 And that term was, according to one scholar, “based on the idea that 2 sets of apprentices should, in 7 years each, be trained in the new techniques” that were protected under such letters patent.3

The United States was the second country to adopt a general copyright law. In 1790, Congress passed a Copyright Act that borrowed extensively from the Statute of Anne, including the “14 + 14″ year term.4 The law was passed with only the most cursory of discussion in the legislature, so the importation of Britain’s act almost word for word was likely driven by expedience more than anything else.

The third modern copyright law came out of France, itself also emerging from a revolution, and it diverged from Britain and the US in terms of duration. In 1793, only three years after the first US Copyright Act was passed, the French National Convention passed a decree providing for copyright protection lasting the life of the author plus ten years.5 (Thomas Paine, who “counted copyright agitation among his many other revolutionary interests”, was a member of the legislature that passed the French act).6

The shift to a life-based term

England would follow France and adopt a term based on the life of the author shortly after the 19th century rolled around. In 1814, the term of protection was extended to the life of the author or 28 years, whichever was longer.7 In 1842, England extended the copyright term to the life of the author plus seven years, or forty-two years, whichever was longer.8

The US did not follow suit to a life of the author based term, though it did bump up the initial term of protection to 28 years in 1831 (the renewal period remained at 14 years).9 And thus, despite being the second nation to provide for general copyright protection, it would fall behind international norms for the majority of its history. During the 19th century, many countries with copyright laws moved toward a “life plus” duration. One 19th century treatise provides this catalog:

Sweden, formerly perpetual, now for life and fifty years, no registration; Denmark, for life and thirty years, no registration or deposit; Switzerland, for life or thirty years; Russia, for life and fifty years, registration but not deposit being required, with complicated provisions as to new editions; Turkey, for forty years, or twenty for translations; Greece, for fifteen years, subject to royal extension; Mexico, which has perpetual literary copyright, registration and deposit being obligatory; Venezuela, for life and fourteen years, or deposit and registration; Chili, for life and five years; Brazil, for life and ten years ; Japan, for thirty years, with extension to forty-five.10

France would be one of the first countries to settle on life of the author plus 50 years, adopting that term in 1866.11 Several other European countries would follow suit toward the latter half of the 19th century.

In 1887, the following countries adopted the Berne Convention, the world’s first multilateral international agreement on copyright: Australia, Belgium, Canada, France, Germany, Haiti, India, Italy, Liberia, Monaco, New Zealand, Spain, Sweden, Switzerland, Tunisia and the United Kingdom. The agreement set minimum standards for copyright protection that agreeing parties must extend to works of authors from member states. The original text did not address duration of protection, but that would soon change.12

The US effort to join the international community

At the turn of the 20th century, US lawmakers looked to revise the copyright law. At that time, R.R. Bowker of the American Copyright League pointed out:

A copyright term extending through and beyond the life of the author has been adopted by thirty-seven countries, or more than half of those which have copyright laws, while five others give perpetual copyright; the term of life and fifty years is adopted by France and fourteen other nations, and although life and thirty years is proposed in Lord Herschel’s British draft, life and fifty years has also been proposed as an alternative in England.

An early version of a US revision bill in 1906 proposed a term of life of the author plus fifty years.13 The proposal was ultimately dropped in favor of the existing convention of a set term with an optional renewal term, as this was seen as more favorable to authors. The renewal term, however, was extended to match the initial term of 28 years.14

The Berne Convention, which was gradually adding new member parties, had set a voluntary minimum term of life of the author plus 50 years in 1908. England adopted this term in 1911, which extended to its Commonwealth Countries — Australia, Canada, India, New Zealand, etc.15 Other countries would slowly but surely follow suit, and Berne made the term mandatory in 1948. The decades following the US enactment of the 1909 Act saw a number of unsuccessful efforts to join Berne, hampered in part by two World Wars and a Great Depression. Bills introduced in 1924, 1925, 1926, 1930, and 1940 called for a term of life of the author plus 50 years; advocates for this term also voiced their views at hearings on other bills during this timeframe.16

But it would not be until 1955 that the US began in earnest the process that would lead to the 1976 Copyright Act, the current law. In that year, Congress appropriated funds to the US Copyright Office to engage in a series of studies on issues relating to copyright revision. One study covered the duration of copyright and noted that by this point over half of all countries with copyright laws protected a work for life of the author plus 50 years.

In this group are Australia, Austria, Belgium, Canada, Chile, Costa Rica, Czechoslovakia, Denmark, Ecuador, Finland, France, Germany, Great Britain, Greece, Guatemala, Hungary, Iceland, Eire, Italy, Lebanon, Luxembourg, Monaco, Netherlands, New Zealand, Norway, Paraguay, Switzerland, Syria, Turkey, Union of South Africa, and Vatican.

Some countries had longer post-life terms and some had shorter; however, by the 1950s, the US stood alone with the Phillippine Islands in diverging from a “life plus” copyright term.

By 1965, a general revision bill included a term of life of the author plus 50 years. But revision efforts stalled, primarily due to stark disagreements over a number of unresolved issues such as library photocopying and community antenna television providers.17 Congress was concerned about the inequities that would fall upon authors because of a bogged-down legislative process, so during this time period, it passed a series of nine interim extensions to copyright duration.18 (Lawrence Lessig counts each of these nine interim extensions when he makes his famous claim that Congress extended copyright term 11 times since 1962 and also defines this series of interim extensions preceding the general revision as “a practice that has defined copyright law since).19 The revision finally passed in 1976, going into effect in 1978, over a century since France adopted a life plus 50 term and decades after most every other country had followed suit.20

But the US continued to lag behind. In 1965, for example, Germany adopted a term of life of the author plus seventy years.

The shift to “life plus 70″

A 1971 guide from WIPO on the Berne Convention, which then consisted of over 60 member parties, explained the justification of a life plus fifty term:

By computing the term of protection from the date of the author’s death, the Convention binds the work to its creator. Honest men can differ on how long this should be: some feel it should be for ever since the nature of the work of the mind remains, throughout the ages, a reflection of the character of its creator. Like a fine piece of furniture, it gives pleasure to generation upon generation. But the particular nature of intellectual property, resulting in a need, in the interests of the public at large, for it to be made known without let or hindrance for the enrichment of culture, suggest some limit on the duration of the monopoly enjoyed by authors and their heirs in the exploitation of their works.

It is not merely by chance that fifty years was chosen. Most countries have felt it fair and right that the average lifetime of an author and his direct descendants should be covered, i.e., three generations.

In 1993, the Council of European Communities, finding that “the minimum term of protection laid down by the Berne Convention, namely the life of the author and 50 years after his death, was intended to provide protection for the author and the first two generations of his descendants; whereas the average lifespan in the Community has grown longer, to the point where this term is no longer sufficient to cover two generations”, directed its Member States which hadn’t already granted the term to adopt a “life plus seventy” copyright duration.21 The U.S., narrowing the lag behind international norms, adopted a term of life plus seventy in 1997,22 citing a need to harmonize protection with its global trading partners.23

Today, 167 countries are Member Parties of the Berne Convention and provide for a minimum of life of the author plus fifty years. Over 65 of these countries provide for life of the author plus seventy years, while an additional seven provide greater length.24

How long should copyright last?

One can argue over the “optimal” length of copyright protection—a discussion beyond the scope of this already lengthy article—but one can’t assert after surveying this history that, for example, Congress extends copyright every time Disney’s Steamboat Willie is about to enter the public domain.25

Copyright duration is a hot topic. But it is also a bit of a red herring. Creators’ main concern is lack of effective protection now, so the length of protection is somewhat a moot point. And issues raised by user communities are only tangentially related to duration, meaning any changes to term would only result in marginal differences to the real issues.


  1. See, for example, Timothy B. Lee, “15 years ago, Congress kept Mickey Mouse out of the public domain. Will they do it again?“, The Switch (Oct. 25, 2013); Jeff John Roberts, “Will copyright be extended 20 more years? An old debate returns“, GigaOM (Aug. 20, 2013); Eugene Volokh, “Copyright Extension Prediction?“, Volokh Conspiracy (Aug. 19, 2013). []
  2. Ronan Deazley, ‘Commentary on the Statute of Monopolies 1624‘, in Primary Sources on Copyright (1450-1900) (eds L. Bently & M. Kretschmer, 2008). []
  3. Fritz Machlup, “An Economic Review of the Patent System”, pg 5, Study of the Subcommittee on Patents, Trademarks, and Copyrights of the Committee on the Judiciary, United States Senate, United States Printing Office, Washington: 1958. []
  4. See Oren Bracha, The Adventures of the Statute of Anne in the Land of Unlimited Possibilities: The Life of a Legal Transplant, 25 Berkeley Technology Law Journal 1427 (2010). []
  5. French Literary and Artistic Property Act, Paris (1793), Primary Sources on Copyright (1450-1900), eds L. Bently & M. Kretschmer, []
  6. William Patry, Copyright Law and Practice, Chapter 1 (BNA 2000). []
  7. Copyright Act, London (1814), Primary Sources on Copyright (1450-1900), eds L. Bently & M. Kretschmer. []
  8. Copyright Law Amendment Act, 1842, 5 & 6 Vict., c.45. []
  9. Copyright Act of 1831, 4 Stat. 436. []
  10. R.R. Bowker, Copyright, Its Law and Its Literature, pg. 24 (1886). []
  11. Walter Arthur Copinger, The Law of Copyright in Works of Literature and Art, pg. 238 (1870). []
  12. It did, however, set a minimum term of protection for translations. Around this time, it became common to provide different terms of protection depending on subject matter (such as one term for literary works and a different term for sculptures or phonographs) as well as different terms depending on authorship (for example, different terms for anonymous authors or corporate authors. For simplicity purposes, I will subsequently refer only to the term of a natural author of a literary work when I discuss copyright duration. []
  13. S. 6330 and H.R. 19853, Section 18(c), 59th Congress, 1st Session. []
  14. Copyright Act of 1909. []
  15. Copyright Act of 1911. []
  16. James J. Guinan, Jr., Duration of Copyright, Copyright Law Revision Study No. 30, Copyright Office (1957). []
  17. See William Patry, Copyright Law and Practice, Chapter 1 (BNA 2000). []
  18. Scott M. Martin, The Mythology of the Public Domain: Exploring the Myths Behind Attacks on the Duration of Copyright Protection, 36 Loyola L.A. Law Review 253, 260-61 (2002). []
  19. Free Culture, pg. 134. []
  20. Copyright Act of 1976, 90 Stat. 2541. []
  21. Council Directive 93/98, 1993 O.J. (L 290/9). []
  22. Copyright Term Extension Act, 112 Stat. 2827. []
  23. Senate Report No. 104-315, 104th Cong., 2d Sess (1996). []
  24. Wikipedia, List of countries’ copyright lengths, visited July 14, 2014. []
  25. In his recent book, Intellectual Privilege, Tom Bell spends considerable effort—including illustrations—to imply a correlation between US copyright term length and the entry of Steamboat Willie into the public domain before dismissing it by saying, “No one can, of course, say with certainty whether or to what degree lobbying by the Walt Disney Company drove those copyright extensions.” []

This morning, the House Judiciary Committee Subcommittee on Courts, Intellectual Property, and the Internet will continue its ongoing review of the Copyright Act, focusing this time on Section 512 of Title 17, the DMCA’s notice and takedown provisions.

No doubt, copyright’s skeptics will trot out their usual line of DMCA abuse, so I thought it would be appropriate to offer my own. Here, in no particular order, are my top examples of DMCA notice and takedown abuse: sites and services which had claimed the protection of the law while receiving millions of notices of infringing content and profiting handsomely, to the detriment not just of the large copyright holders who could afford to send takedowns (and litigate if need be) but especially to the detriment of individual creators and small businesses, who effectively have no recourse:


Filelocker Hotfile was founded in 2009. By the time it was sued for copyright infringement by the major motion picture studios in 2011, it had become one of the most popular such sites. In part, this was because the site did such a good job facilitating infringement among its users—evidence showed over 90% of files downloaded through the service were infringing, and the court noted that these efforts “worked a significant financial benefit to Hotfile and its founders.”

Hotfile, of course, argued strenuously that it was operating consistent with the DMCA.

The facts showed a different story. One of the requirements for the DMCA safe harbor is that a site adopts and reasonably implements a “repeat infringer” policy, one that terminates accounts in appropriate circumstances. The court found that Hotfile was doing anything but. When the litigation began, Hotfile had received over 10 million takedown notices for infringing content on its system.  But records show that it had only ever terminated a grand total of 43 users—though 33 of those came as a result of a court order from prior litigation. Nearly twenty-five thousand Hotfile users had accumulated over three infringement notices. Over sixty of those users were sitting on at least three hundred notices.

And as proof that Hotfile could have adopted a reasonable repeat infringement policy but didn’t, and that it would make a difference, the service revamped its policy after the litigation began and almost immediately 444 of its 500 highest paid affiliates were terminated for repeated infringement. Hotfile ended up shutting its doors and settling with the motion picture studios for $80 million the day before it was set to face a jury.

It’s difficult enough for creators to detect and deter online infringement, but some bad actors use the DMCA as legal cover, window dressing that places a façade of legitimacy over plain old for-profit exploitation. The law, however, is more than just a notice-and-takedown mechanism, and services have a number of requirements they must meet to enjoy its protection.


In 2012, the US government brought criminal charges against filelocker Megaupload and seven of its officers. Like Hotfile, Megaupload argued that it was merely an innocent cloud storage provider, shielded by the DMCA, but publicly released evidence shows the site used the law as a sword, helping it rake in $150 million and causing around $500 million in damages to copyright holders.

While the government alleges Megaupload failed to meet most of the DMCA’s requirements, one policy in particular stands out as DMCA abuse. According to the Department of Justice:

If there are multiple links to a file, then any attempt by the copyright holder to terminate access to the file using the Abuse Tool or other DMCA takedown request will fail unless all of the URL links to the infringing file are known and submitted, because the file will continue to be available through any undisclosed URL links. The infringing copy of the copyrighted work, therefore, remains on the Mega Conspiracy’s systems as long as a single link remains unknown to the copyright holder. The Mega Conspiracy maintains a record of links that have been generated by the system, but duplicative links to infringing materials are neither disclosed to copyright holders, nor are they automatically disabled or deleted when a copyright holder either uses the Abuse Tool or makes a standard DMCA copyright infringement takedown request. During the course of the conspiracy, the Mega Conspiracy has received many millions of requests to remove infringing copies of copyrighted works, and yet the Mega Conspiracy has, at best, only deleted the particular URL of which the copyright holder complained, and purposefully left the actual infringing copy of the copyrighted work on the Mega Conspiracy-controlled server and any other access links completely intact.

This is a deliberate end-run around the law’s purpose of cooperation between service providers and copyright owners to detect and deter infringement. “Takedown” should mean “takedown” (ideally, it should also mean “stay down”, but that’s another story).


One of the safe harbors in the DMCA protects against liability for material stored at the direction of users. So why couldn’t you have users store music files on your servers and then provide an on-demand streaming platform with a catalog rivaling licensed music services? That’s the too-clever-by-half strategy employed by Grooveshark (along with allegations that it requires employees to upload music themselves to plug any holes). Grooveshark is currently involved in copyright infringement litigation with a number of record labels, and indie musician Erin McKeown has repeatedly called them out for their practices:

Noted copyright scholar Peter Menell has also written about the service, calling it a case study in DMCA safe harbor abuse. The “DMCA license” that services like Grooveshark believe they are entitled to under the law—where services believe it’s ok to build full-fledged content platforms without licenses so long as they “launder” the content through users—not only allows them to divert money out of the hands of musicians, but it makes it more difficult for legitimate, sustainable online platforms to grow.


At one point, IsoHunt was one of the world’s most popular BitTorrent search engines, indexing and helping users search and find torrent files—90-96% of which were infringing. As with the services above, IsoHunt claimed DMCA protection. But last year, the Ninth Circuit rejected that claim.

The primary reason was that IsoHunt was actively inducing infringement through its service:

As noted, the record is replete with instances of Fung actively encouraging infringement, by urging his users to both upload and download particular copyrighted works, providing assistance to those seeking to watch copyrighted films, and helping his users burn copyrighted material onto DVDs. The material in question was sufficiently current and wellknown that it would have been objectively obvious to a reasonable person that the material solicited and assisted was both copyrighted and not licensed to random members of the public, and that the induced use was therefore infringing.

It would seem obvious that service provider actively engaged in encouraging infringement (and, as the record showed, profiting from it) could find refuge in a statue designed to protect service providers from incidental and innocent infringement that occurs on their facilities. But that is unfortunately not always the case. The DMCA should not be so broad that it protects against deliberate infringing behavior.

Though these sites have been or are currently the subject of legal action, there are sadly far too many like them continuing to operate, hiding behind the DMCA to profit off the work of others without their permission. This abuse penalizes good faith service providers and stymies the growth of legitimate innovative services. It exploits creators—and worse, chill their ability to create and disseminate new works for the public to enjoy.

As Congress looks at the DMCA, it should look for ways to reduce the ability of bad faith actors to abuse the important protections the law provides. That would better foster the type of sustainable and innovative online creative ecosystem that the public deserves.

Too often, copyright critics speak dismissively of permission and belittle the mere individual right involved. Such rhetoric surrounded last week’s IP Subcommittee hearing on the scope of fair use, and while the panelists largely agreed that no legislative changes were needed regarding fair use, there were sharp divisions over whether the doctrine currently goes too far or not far enough.

One does not win popularity contests critiquing fair use. How can you ignore the public benefit of Google Books? How can you be against fan fiction? But the question is not whether or not there is a public benefit or creativity involved, the question is how to mediate between public and private interests. And, as fair use is on an expansive course—and increasingly becoming part of the business model of large corporations—it is worth asking how easily public interests should trump private rights.

That’s not to say there are not, as with any form of property or right, competing equities that must be considered, and balanced against a copyright owner’s interests when they come into conflict. Fair use has traditionally played this role. But lately, it has seen a steady and troubling expansion, one that more and more is comfortable with letting the common good override individual rights.1

For those who fall on the “fair use has gone too far” side, who are worried about large companies getting a “volume discount” on fair use to excuse the wholesale copying of millions of works,2 or who share the concerns of Supreme Court Justice Kennedy—who said in his concurrence in Campbell v. Acuff-Rose Music, “If we allow any weak transformation to qualify as parody, however, we weaken the protection of copyright. And under-protection of copyright disserves the goals of copyright just as much as overprotection, by reducing the financial incentive to create”—it is worth reaffirming some of the principles that fair use is established on.

A Marketplace for Ideas

Fair use is necessary, but it necessarily must be limited—especially if the primary goal of the doctrine is to complement the goal of copyright rather than subvert it. “[T]he author’s consent to a reasonable use of his copyrighted works ha[d] always been implied by the courts as a necessary incident of the constitutional policy of promoting the progress of science and the useful arts, since a prohibition of such use would inhibit subsequent writers from attempting to improve upon prior works and thus . . . frustrate the very ends sought to be attained.”3 (Note here the reference to consent.)

One of the reasons fair use analysis includes an examination of the effect of the use upon the potential market for or value of the copyrighted work is because the creation of a marketplace is a core function of copyright law.

By understanding the predicate concepts of the Copyright Clause (limited times, authors, exclusive rights, and writings) the central purpose of the Copyright Clause can be seen: it achieves a public benefit in the progress of knowledge by securing such rights. The structure of the Copyright Clause is not limited to a proclamation that this is beneficial. Rather, it assumes that promoting the progress of knowledge is advantageous and directs Congress to achieve this benefit by securing exclusive rights in intellectual property. It mandates the creation of a marketplace, wherein this unique form of property, the copyright, may be traded and protected.4

Thus, as an economic asset, copyright provides the certainty necessary for the creation of a market, one where transactions can occur peacefully and systematically, where private ordering can emerge, and where economic efficiency can be promoted. Considering the size of this cultural, scientific, and artistic market in the U.S. reached over one trillion dollars in contributions to the GDP in 2013, it seems safe to say this approach is working generally well.

But as with any market, there will be disagreements over price, or refusal to work together. These disagreements, by themselves, should not invite fair use. There are plenty of situations where licensing might be too difficult or expensive, where some modicum of public benefit might be achieved absent the barriers of prices or permission—but this is true of all markets and forms of property. Fair use should be reserved, as with any limitation on individual rights and property, for those situations where a compelling public interest exists, one that demands recognition. As famed treatise Nimmer on Copyright explains, “if the ‘progress of science and useful arts’ is promoted by granting copyright protection to authors, such progress may well be impeded if copyright protection is virtually obliterated in the name of fair use.”5

Equally important is a defense of the notion of consent, or permission. Despite consent being a foundation of civil government—the Declaration of Independence expressly notes governments only derive “their just powers from the consent of the governed”—and a moral and legal concept that pervades our institutions, some copyright skeptics reject the idea.6 What they see as the foundation instead—divine right, perhaps?—is unclear.

Nevertheless, my guess is that most of us would choose consent over coercion.7 And this provides an additional reason to make sure fair use is properly cabined.

The “right to exclude” is perhaps overstated as an essential element of property. Property in general, and copyright specifically, create a freedom to choose how to use economic assets. Legal theorist Felix Cohen puts it this way: “Private property is a relationship among human beings such that the so-called owner can exclude others from certain activities or permit others to engage in those activities and in either case secure the assistance of the law in carrying out his decision.”8 The key here is that the choice itself (so long as it does not interfere on someone else’s interests) is irrelevant.9 I’d like to point out two general reasons why this sort of structure is important to maintain in copyright.

One, it respects the dignity and autonomy of individuals:

An autonomous person is one who directs her own life, and not someone whose life is controlled by others. Of course, we can’t all do whatever we would like without running the risk of interfering with each other. Hence, from the idea of autonomy, we might derive the idea of moral rights and duties that create for each individual a sphere of autonomous action, in which each individual can direct her own life without interfering with the like freedom of others to do the same. The moral force of consent comes naturally if one accepts autonomy as a central moral value. Consent allows others to enter one’s sphere of autonomy. So long a consent is freely given, consented-to rights violations seem perfectly consistent with the idea that rights protect a sphere of individual choice.10

This respect is especially important  regarding expressive works, which can be seen as an extension or reflection of a creator’s personality.11

Two, it provides for a plurality of valid choices. Some creators may choose to partner with larger businesses, some may choose to go their own way and stick to specific formats, some may choose to disseminate freely and as widely as possible. None of these is an inherently “right” choice. All creators place different priorities on getting compensated, seeking exposure, or maintaining creative control, priorities that can easily shift from one project to the next and evolve throughout a creator’s career. None of these is an inherently “correct” choice. One of the principles of the free market is that vesting this choice at the granular level will result in the greatest advance of liberty and efficiency. A system that narrows the range of plurality is inferior to one that preserves a wide range of reasonable choices.

Exceptions to this sort of regime, including fair use, should thus be applied judiciously and conservatively.

The Inherent Public Interest in Copyright

For all the talk about fair use’s role in facilitating public interests, it is easy to forget that copyright—indeed, all property rights—inherently serve the public interest.12 So it bears repeating. The creation of a marketplace for expressive works and the protection of autonomy, both discussed above, undoubtedly advance public interests.13

The results of copyright as being in the public interest are palpable. But for the labor of an author to write about how to cultivate a certain skill, the public would not have that particular expression of knowledge. But for the labor of a filmmaker, members of the public would not see a film that may change the way they view the world or provide them a greater understanding of themselves.

The interests of the public and the author coincide, something recognized by the drafter of the Copyright Clause, James Madison, in the early days of the republic, and reiterated to this day. 14 In Eldred v. Ashcroft, the Supreme Court explicitly rejected as missing the mark the assertion that copyright law “must serve public, not private, ends.” Instead, it said, “The two ends are not mutually exclusive; copyright law serves public ends by providing individuals with an incentive to pursue private ones.”

So, yes, fair use has a long pedigree in U.S. copyright law and an important role to play. But at the same time, courts and policy makers should be mindful that the doctrine does not end up subverting the very goals it is meant to reinforce.


  1. See, for example, Princeton University Press v. Michigan Document Services, 99 F.3d 1381, 1393 (6th Cir. 1996) (J. Martin dissent), saying fair use is intended “to protect society’s vested interest in the sharing of ideas and information against pursuits of illegitimate or excessive private proprietary claims” (Emphasis added); Monge v. Maya Magazines, 688 F.3d 1164, 1184 (9th Cir. 2012) (J. Smith dissent), “The majority’s fair use analysis in this case … thwarts the public interests of copyright by allowing newsworthy public figures to control their images in the press.” []
  2. Statement of June Besek, Hearing on Fair Use (Jan. 28, 2014), “It’s as though courts are according some kind of ‘volume discount’ for fair use, where a massive taking justifies a lower level of scrutiny in a fair use determination. It becomes increasingly difficult to explain to authors and public alike a copyright regime that rigorously examines the extent of a single scholar’s partial copying, while essentially according a free pass to a for-profit enterprise’s massive takings.” []
  3. Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 US 539, 549 (1985), quoting H. Ball, Law of Copyright and Literary Property 260 (1944). []
  4. David A. Householder, The Progress of Knowledge: A Reexamination of the Fundamental Principles of American Copyright Law, 14 Loy. L.A. Ent. L. Rev. 1, 35 (1993). See also Adam Mossoff, How Copyright Drives Innovation in Scholarly Publishing, George Mason Law & Economics Research Paper No. 13-25 (2013), “This commercialization policy—that intermediaries like publishers should be rewarded for their labors in creating the legal and market mechanisms necessary to disseminate works—is essential to the American copyright system.” []
  5. 4 Melville B. Nimmer and David Nimmer, Nimmer on Copyright § 13.05[E][1] (2013). []
  6. See, for example, here, here, and here. []
  7. See Cass R. Sunstein, It’s for Your Own Good! NY Review of Books, March 7, 2013. []
  8. Felix Cohen, Dialogue on Private Property, 9 Rutgers Law Review 357, 373 (1954). It is worth noting that Cohen considered copyrights as property. []
  9. See, for example, John Stuart Mill, On Liberty (1869), “the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right.” []
  10. Consent“, Legal Theory Lexicon. []
  11. Accord Justin Hughes, The Philosophy of Intellectual Property, 77 Geo. L. J. 287 (1988). []
  12. See Thomas W. Merrill and Henry E. Smith, The Morality of Property, 48 Wm. & Mary L. Rev. 1849 (2007). []
  13. See, for example, Daniel Solove, ‘I’ve Got Nothing to Hide’ and other Misunderstandings of Privacy, 44 San Diego Law Review 745, 762 (2007), agreeing with proposition that “Society makes space for the individual because of the social benefits this space provides.” []
  14. Writing in the Federalist Papers, Madison said of the Clause, “The utility of this power will scarcely be questioned. The copyright of authors has been solemnly adjudged in Great Britain, to be a right of common law. … The public good fully coincides…with the claims of individuals.” []

The Supreme Court a few weeks ago agreed to review the Second Circuit’s decision in ABC v. Aereo, setting the stage for a major copyright decision that could come as early as this summer.

The legal issues—and the policy issues—can be quite complex.  The case law thus far has not been terribly enlightening. The legal commentary on the issues has tended to add more layers of complexity rather than getting us closer to rules that provide guidance and clarity to courts and the public.

So it makes sense to look at the statute one step at a time. Today, I want to focus solely on what the Copyright Act means by the phrase “to the public.”

The Copyright Act provides that one of a copyright owner’s exclusive rights is the right “to perform the copyrighted work publicly.”1 The Act explains, “To ‘perform’ a work means to recite, render, play, dance, or act it, either directly or by means of any device or process or, in the case of a motion picture or other audiovisual work, to show its images in any sequence or to make the sounds accompanying it audible.” One can perform a work either privately or publicly, and only the latter is within a copyright owner’s exclusive control. the Act goes on to say:

To perform … a work “publicly” means—

(1) to perform … it at a place open to the public or at any place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered; or

(2) to transmit2 or otherwise communicate a performance … of the work to a place specified by clause (1) or to the public, by means of any device or process, whether the members of the public capable of receiving the performance … receive it in the same place or in separate places and at the same time or at different times.3

It is the last action in clause (2)—transmission of a work to the public—that is at issue in Aereo.

But what does “to the public” mean? The Copyright Act does not define it. And, perhaps surprisingly, Cablevision (which the Second Circuit relied on to reach its conclusion in Aereo) never answers this—it even admits that it reaches its conclusion “without analyzing the contours of that phrase in great detail.”4

So we have some work to do. In order to analyze the contours of the phrase, I want to first look at the history and purpose of the public performance right, and then the statutory text itself in order to provide greater clarity to “to the public.”

History and purpose of public performance right

Goal of copyright law

One of the many important ends of copyright law is to encourage the creation and release of expressive works to the public. As the Supreme Court has said

Nothing in the text of the Copyright Clause confines the “Progress of Science” exclusively to “incentives for creation.” Evidence from the founding, moreover, suggests that inducing dissemination—as opposed to creation—was viewed as an appropriate means to promote science. Until 1976, in fact, Congress made “federal copyright contingent on publication[,] [thereby] providing incentives not primarily for creation,” but for dissemination. Our decisions correspondingly recognize that “copyright supplies the economic incentive to create and disseminate ideas.”5

It does this by securing to authors the exclusive right to engage in acts that multiply the audience of works. Initially, this meant the multiplication of printed copies. For example, the first US copyright law, the 1790 Copyright Act, secured to authors the “sole right and liberty of printing, reprinting, publishing and vending” copies of works.6 Today, US copyright law gives authors the exclusive right not only to make copies, but also to distribute to the public and perform or display works publicly.

These rights are often commercially valuable; by vesting them in the author, the possibility exists to recoup investments in creation and publication. We know bringing works to new audiences is commercially valuable by sheer fact that companies like Aereo expend capital doing just that.

An interpretation of the public performance provisions that is consistent with this principle is ideal. Acts that expose works to new audiences—that multiply the audience of works—should preferably be within the ambit of copyright.

History of public performance right

As noted above, copyright originally protected only the right to reproduction and distribution. An exclusive right to perform dramatic compositions to the public for profit was added to copyright law in 1856.7 The right to public performance was extended to musical compositions in 1897.8 In 1909 Congress passed a major revision to the Copyright Act. The revision provided that copyright owners have the exclusive right “To perform or represent the copyrighted work publicly if it be a drama” and “To perform the copyrighted work publicly for profit if it be a musical composition and for the purpose of public performance for profit.”

The decades following the 1909 Act saw tremendous social and technological changes, and Congress soon felt the need to update the copyright law once again. A general revision process began in earnest in 1955, when Congress appropriated funds for the Copyright Office to lay the groundwork for revision through a comprehensive program of research and study, and the resulting reports from the Register of Copyrights document well the purpose of the public performance provisions and the issues they attempt to address.

Most cases dealing with the public performance right under the 1909 Copyright Act, said the Copyright Office in a 1958 study, revolved around the “for profit” limitation; very few involved the question of what constitutes a “public” performance.9 At the same time, concerns about “semi-public” performances, in places such as “social clubs, lodges, camps, schools, and factories” were raised during revision efforts.10

Ultimately, the 1976 Copyright Act did away with the “for profit” limitation, instead providing a broad public performance right with specific exceptions for certain nonprofit and noncommercial uses. As the Register of Copyrights pointed out in a 1965 Supplemental Report:

[I]t is becoming increasingly apparent that the transmission of works by nonprofit broadcasting, linked computers, and other new media of communication, may soon be among the most important means of disseminating them, and will be capable of reaching vast audiences. Even when these new media are not operated for profit, they may be expected to displace the demand for authors’ works by other users from whom copyright owners derive compensation.

A 1975 report on the copyright law revision from the Senate Judiciary Committee also noted:

The line between commercial and “nonprofit” organizations is increasingly difficult to draw. Many “nonprofit” organiza­tions are highly subsidized and capable of paying royalties and the widespread public exploitation of copyrighted works by’ educational broadcasters and other noncommercial organizations is likely to grow. In addition to these trends, it is worth noting that performances and displays are continuing to supplant markets for printed copies and that in the future a broad “not for profit” exemption could not only hurt authors but could dry up their incentive to write.

The new Copyright Act also had the goal of addressing the issue of public performance for broadcasting. This was a very dynamic issue; as the 1965 Supplemental Report of the Register pointed out, “Anyone looking for an example of how fast things move in the field of author’s rights can find no more striking illustration than the problem of community antennas and their liability under the copyright law. This question, which was not even referred to in the 1961 Report, now promises to be one of the most hotly debated issues in the entire revision program.”

Earlier in the Report, the effect of rapid technological change on author’s rights is discussed. Though the context of the discussion there regards the exhibition of static images or text, it is just as relevant to the performance of audio and video.

Since the Report was issued in 1961 we have become increasingly aware of the enormous potential importance of showing, rather than distributing, copies as a means of disseminating an author’s work. In addition to improved projection equipment, the use of closed-and open-circuit television for presenting images of graphic and textual material to large audiences of spectators could, in the near future, have drastic effects upon copyright owners’ rights. Equally if not more significant for the future are the implications of information storage and retrieval devices; when linked together by communications satellites or other means, these could eventually provide libraries and individuals throughout the world with access to a single copy of a work by transmission of electronic images. It is of inconceivable that, in certain areas at least, ”exhibition” may take over from ”reproduction” of ”copies” as the means of presenting authors’ works to the public, and we are now convinced that a basic right of public exhibition should be expressly recognized in the statute.

The Report says that what eventually would become the 1976 Copyright Act should provide a broad public performance right that would include any and all means of transmission, including “direct amplification (as over a loud speaker system); transmission over wires or other connections; wireless transmission by the originating transmitter and by any other transmitter who picks up his signals and passes them on; and further transmission, over wires, of a wireless transmission.” In addition, the Register foresaw the potential of statutory language being used to erode author’s rights:

The 1964 bill contained language exempting transmissions by someone acting, ”as a common carrier,” the thought being that a corporation merely leasing wires or equipment for the intermediate transmission of signals to other transmitters, rather than to the public, should not be subjected to liability to the copyright owner. It was pointed out that the concept of ”common carrier” might be extended unjustifiably to some commercial transmitters to the public, and we have therefore dropped this exception as ill-advised.

So the Register called for a broad, “future proof” approach, one that would continue to provide meaningful rights to authors even as technological advances changed the way works reach their audiences. Said the report:

Obviously no one can foresee accurately and in detail the evolving patterns in the ways author’s works will reach the public 10, 20, or 50 years from now. Lacking that kind of foresight, the bill should, we believe, adopt a general approach aimed at providing compensation to the author for future as well as present uses of his work that materially affect the value of his copyright. … A real danger to be guarded against is that of confining the scope of an author’s rights on the basis of the present technology so that, as the years go by, his copyright loses much of its value because of unforeseen technical advances. For these reasons, we believe that the author’s rights should be stated in the statute in broad terms, and that the specific limitations on them should not go any further than is shown to be necessary in the public interest.

What is “to the public”

Given that background, we can take a closer look at the statutory language and resolve any ambiguities in light of the intent and purpose behind the provisions. I’ll state upfront what I believe the definition of “to the public” entails:

Any transmission from one person to another person who is not a family member or a close social acquaintance is a public performance.

How do we get here?

Though the Copyright Act defines “publicly”, it does not define “to the public”—or even “public”. But we can deduce the definition of “to the public” from the definition of “publicly,” specifically, the Transmit Clause. The Transmit Clause states that performing a work “publicly” means “to transmit or otherwise communicate a performance … of the work to a place specified by clause (1) or to the public, by means of any device or process, whether the members of the public capable of receiving the performance … receive it in the same place or in separate places and at the same time or at different times.”

If we replace the phrase “place specified by clause (1)” with the actual language of clause (1), we see that performing a work “publicly” means “to transmit or otherwise communicate a performance … of the work to a [place open to the public[, to] any place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered] or to the public, by means of any device or process, whether the members of the public capable of receiving the performance … receive it in the same place or in separate places and at the same time or at different times.”

These three are mutually exclusive. That is, we can read them this way:

Communication of a performance to:

  1. A place open to the public
  2. A place not open to the public but where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered
  3. The public, not including a substantial number of persons outside of a normal circle of a family and its social acquaintances, and at a place not open to the public

It turns out that interpretation of each of these are based on factors that are also mutually exclusive. Courts determine the first, whether a specific place is open to the public, by looking at the nature of the place.11 The second is determined not by the nature of the place, but by the nature and quantity of the audience. How would the third be determined? Since these are exclusive, we are not looking at the nature of the place or the nature and quantity of the audience. What is left to look at?

I posit that a court should look at the relationship between performer/transmitter and audience/recipient. First, the definition contemplates, through the “same place or in separate places and at the same time or at different times”, that one individual is all that is required for a public performance. But we presume, at the very minimum, that an individual transmitting a performance to only herself is performing privately. A consideration of the relationship between transmitter and recipient will account for this in all circumstances. Second, the second transmit phrase expressly states that a performance at a private place where the normal circle of a family and its social acquaintances (and perhaps an insubstantial number of persons outside that) are gathered is a private performance. It should follow that “the public” in phrase 3 also exempts family and social acquaintances—but we can only determine this exemption if we’re looking at the relationship between transmitter and recipient. (Almost) everyone is a member of some family. Finally, this reading is most consonant with the ordinary meaning of the words, the language of the statute and the purpose of the public performance right. There is no need here to import nonstatutory requirements like “unique copies,” as the Second Circuit did in Cablevision and Aereo.

I would add that the legislative history provides an additional point of support against reading in a “unique copy” requirement. The Transmit Clause speaks of a performance of a work, not a particular copy of a work. The U.S. Copyright Office noted in the 1965 Supplemental Report, “A consistent effort has been made in this section and throughout the bill to distinguish between the ‘original work’ which is the product of the author’s creative intellect and which is the real subject of copyright protection and ‘copies’ or ‘phonorecords’ (which are the material objects embodying the work). Failure to draw this distinction under the present law has resulted in a great deal of unnecessary confusion, and has led to results in individual cases that were unpredictable or unfair.”12 This is plain evidence that if Congress intended courts to look at the audience of a performance from a copy rather than a work, it would have said so in the statute.

The Public/Private Distinction Elsewhere

This reading of “to the public” is consistent with case law on other copyright doctrines.

The first is found by looking at the distinction between “limited” and “general” publication. Prior to the 1976 Copyright Act, publication of a work rather than creation served as the genesis of federal copyright protection.13 As with the public performance right, neither “publish” nor “publication” was ever defined by statute. But in its most general sense, “a work is published when it is communicated to the public.” (Emphasis added.)14

Prior to publication, authors enjoyed absolute protection under common law copyright. Publication ended common law protection and made a work eligible for protection under federal copyright statutes. However, before the 1976 Copyright Act, federal copyright protection was also premised on formalities such as notice and registration. This led on occasion to harsh results for authors who inadvertently lost copyright protection by disseminating works without complying with formalities. Because of this, courts tended to interpret “publication” narrowly, and a distinction between “limited” publication and “general” publication emerged, with only the second acting as a publication for purposes of copyright law.15 Roughly speaking, then, the distinction between a limited publication and a general publication is a distinction between a private dissemination and a dissemination to the public.

This distinction was, as many legal distinctions tend to be, somewhat fuzzy.16 But, for our purposes, one general proposition consistently arises from case law: the distinction between the two is one of kind, not degree. That is, the number of copies that have been disseminated is not relevant, it is the nature of the dissemination that matters. And that means that there can be cases where the dissemination of a single copy of a work can be a general publication.17

“In cases where general publication has been found, the creator had made his work available in a manner that suggested that any interested person could have a copy.”18 Essentially, an offer of “come one, come all” had been made—even if no one shows up. On the other hand, “Mere limited publication has been found where the range and purpose of distribution did not suggest that the general public was free to obtain and use the work”—even if the work was circulated widely.19 That is, the audience has been limited ex ante.

Distribution to the Public

The lines drawn above carry over into other copyright doctrines.

Following the general copyright revision in 1976, publication ceased to be the dividing line between common law and statutory protection (although when a work is published remains relevant for a number of other copyright doctrines).20 The 1976 Copyright Act brought both unpublished and published works under the ambit of federal statutory protection.

Much of the language of the Act departed from prior copyright statutes. Thus, for the first time, the Act included, among a copyright owner’s exclusive rights, an 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.”21 The House Report on the 1976 Copyright Act indicates that distribution is largely synonymous with publication: “Clause (3) of section 106 establishes the exclusive right of publication.” Section 101 of the Act also defines “Publication” as “the distribution of copies or phonorecords of a work to the public by sale or other transfer of ownership, or by rental, lease, or lending,” lending support to the interchangeability of the two terms. So here we have a bridge between the pre-1976 concept of publication and the 1976 Copyright Act right of distribution, allowing the doctrines of the former to inform the latter.

The Third Circuit made this observation in Ford Motor Co. v. Summit Motor Products:

The term “public” is not defined in the Copyright Act. Thus, we need to look elsewhere for a definition. Our search commences with the observation that the right protected by section 106(3) is generally referred to as that of publication. Indeed, the statutory definition of “publication” is “the distribution of copies … of a work to the public by sale or other transfer of ownership,” a definition which tracks the language of section 106(3). “Publication” and the exclusive right protected by section 106(3), then, are for all practical purposes, synonymous. Therefore, any clarification of what is meant by “publication” would also clarify what is meant by section 106(3), and in particular, the term “public.”22

Armed with this rule, the Circuit Court turned to prior decisions to see what was meant by “publication” and found that

Courts have held that when deciding if a common law “publication” occurred, the “number of persons receiving copies is not determinative; a general publication may be found when only one copy of the work reaches a member of the general public….” Hence, because “publication” and the right protected by section 106(3) are the same, and because a “publication” can occur when only one member of the public receives a copyrighted work, it follows that a violation of section 106(3) can also occur when illicit copies of a copyrighted work are only distributed to one person.

Cablevision explicitly rejected applying the reasoning in Ford to public performance, saying

Commentators have criticized the Ford court for divesting the phrase “to the public” of “all meaning whatsoever,” and the decision does appear to have that result. Whether this result was justified in the context of the distribution right is not for us to decide in this case. We merely note that we find no compelling reason, in the context of the transmit clause and the public performance right, to interpret the phrase “to the public” out of existence.

With all due respect, the Second Circuit rushes through its dismissal of Ford to get to the result it wants. For starters, “Commentators” is perhaps overstating things, as the court cites to only one source: Nimmer on Copyright. In addition, the court, in its haste, misunderstands what Nimmer is saying.

The section cited is on “publication and public distribution,” and in it, Nimmer notes that the exclusive right to distribution is limited to distributions made “to the public”—so, like the performance right, the Copyright Act contemplates the existence of non-infringing “private” distributions. Nimmer refers to the legislative history that demonstrates a rough correspondence between the pre-1976 publication right and the 1976 distribution right; he concludes that the distinction between a “general” publication and a “limited” publication should survive as a distinction between an infringing distribution to the public and a non-infringing “private” distribution if the two were to be synonymous.

But, says Nimmer, one could argue to the contrary that the distribution right not only incorporated the pre-1976 right to publish but also the pre-1976 right “to vend” a work—and it is apparently the case that the vending right made no distinction between limited and general dissemination. Both private sales and sales to the public at large fell within the scope of the vending right. Nimmer says, “The problem with that construction is that it limits the statutory language concerning distribution ‘to the public’ so that the phrase loses almost all meaning.” Nimmer goes on to say that Ford takes this construction one step further by holding that a gift rather than a sale to a designated individual constitutes a distribution. That is, Nimmer attacked Ford first for its holding that distribution encompassed the right not only to publish but to vend and second for its conclusion that a transaction that lacked “pecuniary remuneration” implicates the right to vend. He did not attack the case for the proposition that a distribution can be to the public when only one copy is disseminated, consistent with well-settled jurisprudence regarding limited and general publication.23 In fact, Nimmer’s treatise reinforces the parallels between publication and distribution, and thus the idea that it is the nature of the act rather than the quantity involved that is important. Nimmer begins this section by explaining,

Note that it is not any distribution of copies or phonorecords that falls within the right, but only such distributions as are made “to the public.” In essence, it is a right to control the work’s publication. The term “distribution” rather than “publication” was used merely “for the sake of clarity.” Accordingly, a limited publication, i.e., a distribution made to a limited group for a limited purpose and not made to the public at large, should not infringe this right.

The Copyright Act speaks of distributions “to the public”, which is roughly synonymous with general publication. The Copyright Act also speaks of transmitting a performance “to the public.” Interpretation of this language should thus be informed by case law on publication. And that suggests we are not looking at quantity of performances—and we’re certainly not looking at whether the origin of the performance stems from a “unique copy.” Instead, we are looking at the nature of the performance. Is it, as in general publications, made available to the public at large? Or is it ex ante limited, available, at a minimum, only to an individual, her family, and her close social acquaintances?

And to answer that, we need to ask, what is the relationship between performer and audience?

This is how a California court in 1991 resolved a copyright dispute involving the Transmit Clause:

Hotel guests watching a video movie in their room through On Command’s system are not watching it in a “public place” but they are nonetheless members of “the public.” This is because the relationship between the transmitter of the performance, On Command, and the audience, hotel guests, is a commercial, “public” one regardless of where the viewing takes place.24

This is, I’d note, also the conclusion reached by Judge Chin in his dissent in Aereo. “Giving the undefined term ‘the public’ its ordinary meaning, a transmission to anyone other than oneself or an intimate relation is a communication to a ‘member[] of the public,’ because it is not in any sense ‘private.’”25

Next steps

Defining “to the public” is only the first step in determining if copyright infringement has occurred. What remains are perhaps the more difficult questions: what is a performance? And who is a performer? This last question is especially germane in the cloud computing context: when is an online service performing a work, and when is it acting merely as a “device or process” for a user to perform a work?


  1. 17 U.S.C. § 106(4). []
  2. The Act later adds, “To ‘transmit’ a performance … is to communicate it by any device or process whereby images or sounds are received beyond the place from which they are sent.” []
  3. 17 U.S.C. § 101. []
  4. Cartoon Network LP, LLLP v. CSC Holdings, Inc., 536 F. 3d 121, 138 (2nd Cir. 2008). The court merely said that the relevant audience of an individual transmission is limited to who is capable of receiving that individual transmission and then presumed that that would always be an individual subscriber. Maybe that presumption is fine at the summary judgment stage, but clearly it could be the case that a subscriber might be an establishment open to the public, such as a bar or restaurant, which would make Cablevision’s transmission a public performance. []
  5. Golan v. Holder, 132 S. Ct. 873, 888-889 (2012). []
  6. Copyright Act of 1790. []
  7. Act of August 18, 1856, 34th Cong., 1st Sess., 11 Stat. 138. Oren Bracha notes bills to this effect had been introduced as early as 1841 in the U.S., and the UK added public performance of dramatic works in 1833. See Oren Bracha, Commentary on the U.S. Copyright Act Amendment 1856, in Primary Sources on Copyright (1450-1900), eds L. Bently & M. Kretschmer (2008), []
  8. Act of January 6, 1897, 44th Cong., 2d Sess., 29 Stat. 481. []
  9. Borge Varmer, Copyright Office Revision Study 16: Limitations on Performing Rights, U.S. Copyright Office (1958). []
  10. Report of the Register of Copyrights on the General Revision of the U.S. Copyright Law at 29 (1961). []
  11. Columbia Pictures Industries, Inc. v. Professional Real Estate Inv,, Inc., 866 F. 2d 278,281 (9th Cir. 1989); Columbia Pictures Industries, Inc. v. Aveco, Inc., 800 F.2d 59, 63 (3d Cir.1986). []
  12. The House Report on the 1976 Copyright Act echoes this point, stating, ” The definitions of [‘copy’ and ‘phonorecord’] in section 101, together with their usage in section 102 and throughout the bill, reflect a fundamental distinction between the ‘original work’ which is the product of ‘authorship’ and the multitude of material objects in which it can be embodied. Thus, in the sense of the bill, a ‘book’ is not a work of authorship, but is a particular kind of ‘copy.’ Instead, the author may write a ‘literary work,’ which in turn can be embodied in a wide range of ‘copies’ and ‘phonorecords,’ including books, periodicals, computer punch cards, microfilm, taped recordings, and so forth. It is possible to have an ‘original work of authorship’ without having a ‘copy’ or ‘phonorecord’ embodying it, and it is also possible to have a ‘copy’ or ‘phonorecord’ embodying something that does not qualify as an ‘original work of authorship.’” []
  13. Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 US 539, 552 (1985). []
  14. Drone, The Law of Property in Intellectual Productions (1879). []
  15. See, e.g., American Tobacco Company v. Werckmeister, 207 US 284 (1907); Burke v. National Broadcasting Co., Inc., 598 F. 2d 688, 691-92 (1st Cir. 1979). []
  16. Deborah R. Gerhardt, Copyright Publication: An Empirical Study, 87 Notre Dame Law Review 135, 138 (2011). []
  17. Brown v. Tabb, 714 F. 2d 1088, 1091-92 (11th Cir. 1983); Burke v. National Broadcasting Co., Inc. 598 F.2d 688, 691-92. (1st. Cir. 1979); Bobbs-Merrill Co. v. Strauss, 147 F. 15, 19 (2d Cir. 1906), aff’d, 210 U.S. 339 (1908). []
  18. Burke at 692, citing Nimmer on Copyright § 4.04, at 4-20-21. []
  19. Id. []
  20. Gerhardt at 145-48. []
  21. 17 U.S.C. § 106(3). []
  22. 930 F.2d 277, 299 (3rd Cir. 1991). []
  23. I would add that Nimmer’s criticism of Ford is appropriate, but for reasons other than those offered by Nimmer. It is not the fact that the transaction involved took the form of a gift rather than a sale. Ford explains, “That the bags were a gift is of no import. The term ‘other transfer of ownership’ is broad enough to encompass gifts. In fact, the House Report on the 1976 Amendment to the Copyright Act states during the course of a discussion on section 106(3) that the copyright owner has the right to control public distribution, whether by sale, gift, loan, or some rental or lease arrangement.'” Rather, the fatal flaw of Ford is that it skips over the distinction between a limited and general publication, taking the proposition that “a publication can occur when only one member of the public receives a copyrighted work” (emphasis added) to mean that all transfers involving one individual receiving a copyrighted work constitute a general publication. This is a logical fallacy. []
  24. On Command Video Corp. v. Columbia Pictures, 777 F. Supp. 787, 790 (ND Cali, 1991). []
  25. WNET, Thirteen v. Aereo, Inc., 712 F.3d 676, 698 (2nd Cir. 2013) (J. Chin, dissent). []

This past March, the Supreme Court considered in Kirtsaeng v. John Wiley & Sons “whether the words ‘lawfully made under this title’ restrict the scope of §109(a)’s ‘first sale’ doctrine geographically.” It ultimately held that they do not — copyright’s first sale doctrine has no geographic limitation. But since then, many have looked beyond Kirtsaeng’s narrow and fact-specific holding to find what they believe to be endorsement of broad, and potentially troubling to creators, expansions to the first sale doctrine.

Digital First Sale

Almost immediately, some speculated that Kirtsaeng would have an impact on “digital first sale”, especially given the fact that a decision in Capitol Records v. ReDigi dealing with that very issue was pending in the Southern District Court of New York (ReDigi operated a service allowing users to buy and sell “used” mp3s).1 For example, Joe Wikert, of TOC, wrote the same day that “the Kirtsaeng ruling can only help ReDigi’s case and that’s a good thing for anyone who wants the ability to resell their digital goods,” though he was far from the only one to share this view.2 But the court in ReDigi didn’t see anything in Kirtsaeng‘s general remarks about first sale that rendered the provisions at issue ambiguous, and it would hold that ReDigi did not have a valid first sale defense — the first sale doctrine only applies to the distribution of particular, lawfully made, copies of a work, while ReDigi was distributing new, unlawfully made, reproductions of copies.

Sales vs. Licensing

More recently, a software reseller has sought solace in Kirtsaeng‘s first sale language for a different issue in Adobe v. Kornrumpf, currently on appeal to the Ninth Circuit.

Adobe was in the practice of licensing OEM versions of its software — such as Photoshop Elements 8 — to be sold exclusively in conjunction with hardware equipment. According to court records, Kornrumpf acquired copies of these OEM products and resold them separately. Adobe sued, alleging infringement of its exclusive distribution rights since Kornrumpf had not lawfully acquired title to the software.

The court ruled in favor of Adobe, finding that the software was licensed, not sold. The decision was a fairly straightforward application of the test developed by the Ninth Circuit in Vernor v. Autodesk: “[A] software user is a licensee rather than an owner of a copy where the copyright owner (1) specifies that the user is granted a license; (2) significantly restricts the user’s ability to transfer the software; and (3) imposes notable use restrictions.” Since the evidence demonstrated that Adobe clearly met each of these elements, the first sale doctrine did not apply, and Kornrumpf was liable for infringing Adobe’s exclusive distribution right.

But Kornrumpf appealed, with a novel argument: the Supreme Court’s decision in Kirtsaeng limits the scope of Vernor. Specifically, he argues that the Court held that “market division is not a valid aim of copyright”, that it held “courts should not be burdened with enforcing restrictions on the alienability of chattels”, and that it reaffirmed “the venerable common-law aversion to restraints on alienability in the copyright context.”

As in ReDigi, these are essentially policy arguments, as they are not supported by the legal conclusions of Kirtsaeng.

Reaffirming Traditional First Sale Doctrine

In fact, if anything, Kirtsaeng reaffirmed the point that the first sale doctrine is inapplicable when copies have been licensed rather than sold. In responding to the argument that its interpretation of the first sale doctrine would render copyright’s importation provisions superfluous, for example, the Court says that, to the contrary, the provisions “would still forbid importing … copies lawfully made abroad … where … any … licensee, consignee, or bailee sought to send them to the United States.” [Emphasis added.]

What’s more, if there is anything that is venerable, it is copyright law’s tradition of recognizing a distinction between transferring title of a copy of a work through a sale and licensing a work. Going back to what is considered the origin of the first sale doctrine in U.S. copyright law, the 1908 Supreme Court case Bobbs-Merrill v. Straus, one can see this distinction already established. As the Supreme Court said there:

The precise question, therefore, in this case is, does the sole right to vend (named in § 4952) secure to the owner of the copyright the right, after a sale of the book to a purchaser, to restrict future sales of the book at retail, to the right to sell it at a certain price per copy, because of a notice in the book that a sale at a different price will be treated as an infringement, which notice has been brought home to one undertaking to sell for less than the named sum? We do not think the statute can be given such a construction, and it is to be remembered that this is purely a question of statutory construction. There is no claim in this case of contract limitation, nor license agreement controlling the subsequent sales of the book. [Emphasis added].

So it would seem that Kornrumpf, like ReDigi, will have trouble convincing courts to accept its first sale arguments. Nevertheless, these two cases exemplify a recent push, intensified by the Kirtsaeng decision, to inflate the first sale doctrine beyond its due boundaries. No doubt its proponents will raise the issue as Congress reviews the Copyright Act. Why is this troubling?

The Benefits of Traditional First Sale

For starters, a meaningful ability to license works benefits both consumers and creators. Many have observed that there is a general shift from ownership models to access models, especially with music and films.3 Licensing facilitates these new models, by allowing a more granular grant of rights than sales can provide. This gives consumers a diversity of options to read, watch, or listen, at a range of price points.

Put another way, think of it in terms of ordinary goods. For example, you can buy a car, enter into a long-term lease, rent a car for a short period, or take a cab. No one would argue that a framework that only allows outright sale of cars is preferable to this system. The same is true with real property: you can buy a house, lease an apartment, or book a hotel room. It would be an incredibly inefficient system if only the first of these was possible.

As far as digital first sale goes, there are issues with treating digital works like physical goods. There really isn’t such a thing as a “used” digital file — the file is absolutely identical to a “new” file and will remain identical indefinitely. This means “used” digital files are exact substitutes for “new” digital files and would compete directly with them.4 Obviously, there are limited issues regarding transferability and alienability that may arise, such as providing mechanisms for passing along digital files to heirs. But forcing digital works to act more like physical objects is clearly not an ideal solution.

The “First Sale License”

Which seems to be the thinking underlying much of the push for expanding first sale in the digital world: we should ignore the beneficial characteristics of the internet and arbitrarily force it to operate like a brick and mortar store. In many ways, first sale is the square peg in the future’s round hole.

Either of these — preference of sales over licensing or digital first sale — would be detrimental to creators and consumers. But the negative effects are multiplied when taken together. Most of today’s popular and innovative services — Spotify, Netflix, Audible, to name just a few — would be unlikely to exist under a system where any transfer of digital works are presumptively sales, and once transmitted they can be resold. Even copyleft systems like Creative Commons and open source software would be in trouble since they are built on copyright licensing.

In there place would be services that would operate under a “first sale license”,5 setting up convoluted systems that would allow them to provide access to songs, movies, ebooks, and other digital works without permission from or payment to creators. Imagine a streaming service that provides on-demand access to the universe of recorded music. Identical to, say, Spotify, but this service is premised on the conceit that a user “buys” a “used” song when it begins playing and then “sells” it back automatically when the song is finished. In this fashion, the service can claim it does not need permission from creators nor does it need to pay any applicable performance royalties. I don’t think such a service is outside the realm of possibilities — and if you don’t think some company will go through the effort to build a Rube Goldberg-like contrivance to take advantage of perceived loopholes in copyright law, well, I’ve got a thousand tiny antennas in Brooklyn to sell you.


  1. Kilpatrick, Townsend & Stockton LLP, “No first-sale doctrine for digital music files“, Lexology (April 8, 2013): “ReDigi had nothing to do with foreign manufacture/sale, and importation of copyrighted goods, but because the Supreme Court’s decision changed the meaning of the first-sale doctrine in the eyes of many, there was speculation that the decision might cause the Southern District to apply a comparably broad reading of the first-sale doctrine”. []
  2. See also Maria Scheid, “The First Sale Doctrine and the Sale of Digital Goods in Light of Kirtsaeng and ReDigi“, Copyright Corner (April 23, 2013): “the Supreme Court’s decision in Kirtsaeng is particularly important because it reaffirms the basic notion that one has physical ownership of the things that you buy. A court may hold this broad rationale to be equally applicable to digital goods, meaning consumers should be able to resell their digital goods under the protection of the first sale doctrine”; Daniel Gervais, “Digital Kirtsaeng: The first-sale doctrine and online content“, Blouin Beat (April 3, 2013), “ReDigi and others, relying on Kirtsaeng, will say that there is no valid reason to distinguish physical copies from online copies. When someone is done with a copy they should have the right to transfer it to someone else. They have copyright policy (as explained by a majority of the Supreme Court in Kirtsaeng) on their side.” []
  3. See, for example, Steve Jobs Was Wrong — Consumers Want To Rent Their Music, Not Own ItFuture of Music Biz: Ownership vs. Access with Ted Cohen, Founder of TAG StrategicSpotify CEO: Music access, not ownership, is the future. []
  4. Accord Register of Copyrights, DMCA Section 104 Report, U.S. Copyright Office, pp. 82-83 (August 2001):

    Physical copies of works degrade with time and use, making used copies less desirable than new ones. Digital information does not degrade, and can be reproduced perfectly on a recipient’s computer. The “used” copy is just as desirable as (in fact, is indistinguishable from) a new copy of the same work. Time, space, effort and cost no longer act as barriers to the movement of copies, since digital copies can be transmitted nearly instantaneously anywhere in the world with minimal effort and negligible cost. The need to transport physical copies of works, which acts as a natural brake on the effect of resales on the copyright owner’s market, no longer exists in the realm of digital transmissions. The ability of such “used” copies to compete for market share with new copies is thus far greater in the digital world.


  5. Analogous to the “DMCA license” that some online service providers claim allows them to build business models around copyrighted works without permission, so long as the works are uploaded by users and the service provider responds to DMCA takedown notices. []

Last Thursday, the House Judiciary Committee Subcommittee on Courts, Intellectual Property, and the Internet held a hearing called “A Case Study for Consensus Building: The Copyright Principles Project.” The hearing is the first in an expected series of hearings that will take place in the upcoming months aimed at a comprehensive review of U.S. copyright law.

Rep. Goodlatte, the House Judiciary Committee Chairman, chose the topic of the first hearing not to endorse the substance of recommendations of the Copyright Principles Project, but to set the tone for a process that is hoped to be characterized by civil discourse between those who hold diverse views rather than rhetoric and “sharp elbows.” In his opening statement, Chairman Goodlatte elaborated on his motivation for the hearings:

We should not be in a rush to focus on specific issues without first recognizing the fundamentals of copyright and the social and economic benefits that copyright brings to our economy. It is my intention to conduct this broad overview by hearing from everyone interested in copyright law as we begin by holding hearings on important fundamentals before we begin to look at more specific issues.

There are numerous questions that will be raised by interested parties during this review. I have several myself including:

How do we measure the success of copyright and what metrics are used?

How do we ensure that everyone’s voice is heard?

How is copyright working for individual artists?

How is copyright working for our nation’s economy?

These are only a few of the issues we will be looking into. This review of copyright law will not be a quick process simply because the issues are so numerous. However, we must undertake this review to ensure that copyright law continues to incentivize creativity and innovation in the digital age.

It’s an ambitious agenda even if it doesn’t make it to a point where legislation is contemplated. But fortunately, Congress does not have to reinvent the wheel when it comes to copyright review.

There already exists a large body of literature that pokes and prods at copyright law, asking many of the same questions that Rep. Goodlatte posed. Some of it sheds light on how current copyright law came to be shaped as it is, some of it identifies what issues recur most often or have proven most challenging. Together, this literature can prove valuable to Congress as it embarks on its own copyright review.

I want to highlight some of the literature I’m talking about. One can step way back and look at the very broad strokes of copyright law. Here you might focus on its philosophical underpinnings or justifications. You can also zoom in and examine individual doctrines and issues very closely. There are many books, articles, and papers that take both these perspectives. But one can also take a mid-level view of the copyright law as a cohesive whole and how it functions as a statute. This is, it would appear, the approach taken so far by the Subcommittee. And so this is the type of existing works I want to look at.

First, obviously, is the Copyright Principles Project itself, released in 2010. The Project describes a set of broad principles that copyright law should adhere to and than analyzes current law to see how aligned it is with those principles. It then offers a series of recommendations to revise copyright law. Thursday’s hearings did not focus so much on the substance of the Project but on its process itself, and the remarks from Subcommittee members suggested that here it fell short. Concerns over the lack of creators’ input during the project were raised several times during the hearing. Along with that, witness and Project participant Jon Baumgarten pointed out that while there was civil discourse during the process, very little actual consensus was reached in the end.

During the hearing, several references were made to the European Copyright Code, a publication of the Wittem Project. Sponsored in part by the Dutch Government, the European Copyright Code was drafted by a group of copyright scholars with the aim of providing a “model law” of sorts to facilitate future harmonization or unification in European copyright law. The Code was repeatedly referred to for its accomplishment of fitting a copyright law in less than 20 pages. While it did so by leaving out major components of a copyright law — most notably, it does not address infringement or remedies — it is still much shorter than the current U.S. Copyright Act. And although, like the Copyright Principles Project, the European Copyright Code was drafted by academics, its aim was descriptive rather than normative, so it may serve as a useful starting point when looking at how other countries approach copyright law.

Domestic Resources

For U.S. law specifically, one natural starting point is the U.S. Copyright Office, which, as the primary policy advisor for Congress, regularly engages in studying the U.S. Copyright Act. It was, after all, Register of Copyrights Maria Pallante who inspired Goodlatte to conduct the review with her earlier testimony on the Next Great Copyright Act. The Copyright Office has a number of reports that it has completed since the 1976 Copyright Act. Indeed, Pallante references a few in her Next Great Copyright Act article under the heading “Preparatory Work”. These include, to name a few, reports on digital first sale, orphan works, and statutory license for cable and satellite retransmission, as well as pending reports on resale royalties for visual artists, and solutions for enforcing small copyright claims.

The 1976 Copyright Act itself has produced volumes of legislative history (Subcommittee co-chair Rep. Marino could be seen paging through one such volume during Thursday’s hearings). But in terms of systematic and comprehensive discussion of copyright law and doctrines, nothing really compares to the series of copyright law revision studies prepared by the U.S. Copyright Office during this process. Although each tackles a specific area, the 34 studies together cover a range of issues that, for the most part, remain relevant today. Each not only provides a thorough historical background, but delves into many of the challenges and sticking points, surveying the diversity of opinions, highlighting the critical questions, and analyzing the different approaches. Drafts of each reports were submitted to an advisory panel of experts appointed by the Librarian of Congress; the comments and responses of panel members are also included in the reports, providing even more insights.

Although there hasn’t been a general revision since the 1976 Copyright Act was enacted, the 1997 Digital Millennium Copyright Act brought fairly substantial changes to the law, including many of the more contentious provisions today. The DMCA could be traced back to work done by the Working Group on Intellectual Property Rights in the National Information Infrastructure (NII). The NII grew out of the High Performance Computing Act of 1991 and laid a lot of the groundwork for the modern internet. In 1995, the Working Group released a report on Intellectual Property and the National Information Infrastructure. The report examined thoroughly current copyright law and doctrine, with an eye toward how that law would operate online, concluding with a set of recommendations for modernizing the law where necessary. The usefulness here is seeing the thought process behind the DMCA — where, for example, is it outdated because predictions it was based on failed to pan out and where has it fallen short despite accurate foresight because it hasn’t been implemented or interpreted properly.

Foreign and International Resources

A number of countries outside the U.S. have (or currently are) engaged in similar copyright reviews. Many of these reviews have resulted in comprehensive reports that could provide guidance and illumination on many copyright issues.

In the UK, for example, an independent review of copyright law was commissioned by Prime Minister David Cameron. The review, by Cardiff University professor Ian Hargreaves, was released in May 2011, and contains extensive findings, as well as ten recommendations for modernizing that country’s copyright law.

Australia has similarly engaged in a review of its copyright law. The Australian Law Review Commission, an independent statutory body that was established to conduct reviews of Australian law and offer recommendations, is currently in the midst of a review of copyright law. Last August, it released an issues paper on Copyright and the Digital Economy, identifying a number of questions it wished to examine and soliciting comments on the questions. Its final report is scheduled to be released November 2013.

In the international realm, nothing compares to the Berne Convention for the Protection of Literary and Artistic Works both in breadth (it is currently joined by 165 parties) and in vintage (it was first implemented in 1886). WIPO, which administers the treaty, has previously published a guide to the Berne Convention. Although the guide may be outdated in places (it was written over three decades ago), it remains essential reading for anyone engaged in copyright review. The guide explains each of the treaty’s provisions, along with its “nature, aims, and scope.” If one is looking for international norms in copyright law, this is a good place to start.

This is far from an exhaustive list of literature relating to copyright review, but it does provide an adequate foundation before one begins jumping into specific issues.