Sunday, December 31, 2006

Of the case for fair use: digital distribution of course materials: creative destruction

5. Creative Destruction

There is a theory that departs from the market failure framework entirely. Raymond Shih Ray Ku suggests that “creative destruction” could supply an economic rationale for a broader interpretation of fair use.[1]

Creative destruction describes the process by which industries change from within when radical new ways of doing business undermine and eventually replace old ways. We can see this process at work in the evolution of distribution systems for copyrighted materials. Digital networks, home computers and Internet connectivity are poised to replace the physical distribution of analog media. Ku argues that law should not take sides in this process by artificially supporting the old ways of doing business through, in this case, that grant of a strong monopoly right, at the expense of the newer alternatives. Law should be neutral and let markets and evolving business models sort things out. He suggests that applying the concept of creative destruction to the analysis of the fourth fair use factor would facilitate this neutral approach (appropriately weakening the monopoly) by yielding different results from those obtained with market failure analysis.

As Ku describes it, a court can find that the fourth factor favors fair use, in other words that there has been no harm to the market for or value of a copyrighted work 1) where the copying at issue has been carried out by the person who will use the copy[2] and 2) the copying does not undermine the copyright owner’s incentive because creation of the work does not depend on the sale of copies.[3] Clearly this theory incorporates the idea that a copyright owner is not entitled to all possible sources of revenue for his creations, just those required to fund creation. It draws the line in a different place from Sony. Sony considered harm to current sources of funding and took external societal benefits into consideration as a counterbalance to harm to the copyright owner. A court relying on creative destruction can consider whether creation depends on the funding.[4] Copyright owners would not be entitled to any revenues that do not directly fund creation, even if they had received them traditionally and the proposed use eliminated or greatly reduced them.



[1] Shih Ray Ku, R., Consumers and creative destruction: Fair use beyond market failure. 18 Berkeley Technology & Law Journal 539 (2003).

[2] Indeed, the person making a copy has bought and paid for the entire distribution system himself: he buys a computer, a printer, a CD burner, and he subscribes to Internet service. The total cost of the system to make and distribute copies is no longer an expense borne by the copyright owner. Shih Ray Ku, Consumers and Creative Destruction, pp. 544 – 545.

[3] Shih Ray Ku, Consumers and creative destruction, 565 - 566.

[4] This argument would probably suffer from similar definitional problems to those described above for market dysfunction: Which revenues are necessary to fund creation and which are not?

Tuesday, December 12, 2006

Of the case for fair use: digital distribution of course materials: market dysfunction

4. Market Dysfuntion

Although one might characterize engaging copyright owners with just such an aggressive argument [the circular reasoning -- it's all fair argument, see earlier post] as fighting fire with fire, it seems to me an extreme case to make, and as indicated above, an easy case to get around because of the ease of manipulating the four factors. After all, publishers are likely to be able to show real harm to established markets from the hypothetical Electronic Distribution uses I have described. Any court employing a market failure analysis would be hard-pressed to find the use fair: there simply is no market failure, and there are probably demonstrable lost revenues. A court that believes it is important to preserve existing revenue streams in iterative contexts can easily characterize the first three factors to favor getting permission. VoilĂ : End of circularity problem.

One could take a different tack, however. One could try to convince the court that even though there is a functional market for paying permission (in other words, that there is no absolute market failure), that should not be the end of the inquiry.

In fact, total market failure is not the only market failure. One can argue that market dysfunction also is market failure – and if it can be demonstrated, it should tip the fourth factor towards fair use without having to insist that lost revenues are always logically irrelevant in a nonprofit context. Although this is theoretically appealing and would appeal to the nonprofit educational and library communities, there is scant evidence that it would prove persuasive in this context. Nonetheless, it is worth our consideration.

Market dysfunction occurs when particular kinds of transactions other than those frustrated by high transaction costs do not occur as often as they should and therefore justify a finding of fair use. Normally markets give us good information about what people want, and vendors can use that information to decide where to invest their resources. But if a market is in effect misrepresenting what people want, we cannot count on it for this information. Vendors will not fill a need if they do not know it exists, or if they do not recognize the size of the potential market. Market dysfunction can happen when people forego a purchase because the price is too high. One example is where society benefits from the use an individual makes but the parties to the transaction cannot factor this benefit into their bargain. The vendor tries to charge a price that reflects the overall societal benefit, but the buyer will not, perhaps can not, pay that much. So some clearly socially beneficial transactions will not take place if an individual is expected to pay a price that reflects value he personally does not receive, value that benefits society overall.

The use of articles and book chapters by teachers in classrooms and by academic scientists in university labs illustrates just such a dysfunctional market. Educators, their students, academic researchers as well as their academic employers often cannot afford costly permission fees or database licenses, yet these uses would yield very high social benefits, far exceeding the benefit the individuals themselves derive. Conversely, the social detriment resulting from a decision not to use materials because they cost too much is significantly higher than the detriment to a particular individual who foregoes a use. If this social value weighs in favor of fair use as a counterbalance to the copyright owner’s database and license revenue losses, it is at least conceivable that a court could conclude that such uses should be termed fair even following a microeconomics theory. Thus society would receive the diffuse benefits of those educational uses regardless of the institutional inability to pay by permitting them as fair uses.

So, characterizing the educational and research markets as dysfunctional may be an appropriate and promising argument, but winning that argument will be challenging. As Wendy Gordon cautioned, courts that follow market failure reasoning will be urged to err on the side of letting markets work, to whatever extent they can, rather than to “tax copyright owners to subsidize impecunious but meritorious users in the guise of maximizing value.”[1] This description of fair use as a “tax” on copyright owners is repugnant to fair use supporters, but it is easy to understand from a market failure perspective: if your initial premise is that copyright owners are entitled to all they can conceivably get, fair use quite directly harms them by taking money out of their pockets and must be as narrowly construed as possible. In fact, our own university presses can amply demonstrate how directly economically threatened they are by the loss of these revenues, so again, it would be a hard case to make that the diffuse social benefit outweighs such demonstrable harm.

Another challenge in making a cogent market dysfunction argument is definitional: When is a university, a library or a research institution unable to pay? When is a price too high? Which institutions would qualify to exercise fair use of which materials? Who would set the “fair” prices, or would there simply be no price at which a poor school might be required to pay?

These challenges suggest that within the market failure framework, market dysfunction provides at most a theoretical relaxing of a narrow scope for fair use. Courts have not been willing to embrace this theory so far. Winning this argument would require an advocate who can envision a creative application, one that appropriately tailors the relief to need (does Harvard need this exception? does the University of Ohio? how about Florida State?), without undermining the incentive for publishers, many of whom desperately depend on permission revenues. The exception would also have to take into consideration the relative importance of the educational market to the publishers affected by its exercise. In its nuance, it sounds more appropriate for a legislative “financial hardship exception,” than a court-crafted financial hardship application of the fair use test. But, legislation that favors anyone other than copyright owners is unlikely to survive the legislative process, as we all must recognize. Failing these efforts, perhaps the argument could be successfully pressed in negotiations with publishers or their representatives regarding pricing and business models.[2]


[1] Gordon, Fair Use as Market Failure, p. 1632. Her point is that courts should not interfere in a working market, even if it is not working perfectly, and even if the interference is designed to subsidize a worthy cause – in this case, financially strapped educators, libraries, and researchers.

[2] As CCC is in the early stages of developing an educational blanket license, educators and libraries might urge that pricing should be based on many factors, among them what the institution has already licensed, and institutional ability to pay.

if:books creates the perfect "future of the book" course project

On the if:book blog today, director Ben Vershbow posts this provacative question: how would you design the iraq study group report for the web?

How would you design an unauthorized web edition of the ISG Report? Would you keep to the sober, no-nonsense aesthetic of the iconic print editions of past government documents like the 9/11 Commission Report or the Warren Commission Report? Or would you shake things up? What functionality would you add? What kind of discussion capabilities would you like to build into it? Who would you most like to see annotate or publicly comment on the document?

The electronic edition that has been making the rounds is an austere PDF made available by the United States Institute of Peace. A far more useful resource for close reading of the text was put out by Vivismo as a demonstration of its new Velocity Search Engine. They crawled the PDF and broke it into individual paragraphs, adding powerful clustered search tools.

The US Government Printing Office has a slew of public documents available on its website, mostly as PDFs or bare-bones HTML pages. How should texts of "national import" be reconceived for the network?

Yes, indeed. What a great question. Into the course it goes...

Saturday, December 09, 2006

Semester is over, and I am free to muse now, for 6 weeks

Even though I am officially on vacation now, for 6 weeks, I am just so curious about this idea of planning for a course on the future of the book: my idea is to spend 1 semester creating a networked book/paper/project of some sort. The second semester would document that effort and flesh out the design of an iSchool course on the subject.
There are several ebook editors/readers to use and experiment with: dotReader and fckeditor (open source), but I can't get fckeditor 2.3.2 to install properly. Every time I try, with either browser, I get a big zipped file with no executable in it, that I can tell. I had this problem with vkb also, and had to get Benn to install it for me. I didn't figure out what he did to make it work, however, so if I get him to help me again, I need to find out what he does to get it to install and run. There is a pay editor/reader, tk3, that seems to have a good audience, but I'd rather go with the open source dotReader/fckeditor.
I'm finding a lot more written on the future of the book via the topic of ebooks. This isn't really what I envision, but maybe it's where we are right now. An ebook seems separated from the rest of the web, which doesn't mean it's not networked, and interactive, and social, but it's not the same in some significant ways. I need to articulate what's wrong with that, understand the trade offs in being 100% in the Web (ie, readable through a browser), and decide where the best bet lies for me, and for other authors whose values might be different from mine. So I'm sort of thinking of doing an assessment of the state of the art in edistribution of literary and a/v works, and classifying the different products according to what their strengths and weaknesses or benefits and costs are, sort of how they would stack up depending on what you want to achieve as an author/illustrator/artist, etc.
I'll no doubt have moved on to something else by then, but it also occurs to me that this could make an interesting dissertation topic -- the future of the book: state of the art 200x. But where would you go with that? What's needed? Implications for publishing and libraries? Is there really a groundswell of interest in publishing in new ways or is my perspective skewed? And to what extent is the interest pecuniary versus people just wanting to get what they have to say out there in front of an audience? And how effective is private epublishing not-for-profit? What are the implications of all the different authoring and reading tools? I already sense, just in the tiny amount of time I've spent thinking about where to start for my first networked publication, that the choice of a reader is major. You are seemingly stuck with it, whatever it is, and it's like the Betamax/VCR thing. You don't want to choose the format that gets sidelined in a couple of years. And only those people who have your reader can read your book. That just doesn't seem to make sense. Not for the long-term viability of the creative community. Who would want to create for a tiny audience when you could create for a huge one? I know the standards people are working to make this better (OEPSS -- Open ebook publication structure specification), but they don't seem to be in any particular hurry...
But, I'm supposedly into experimentation here, not getting married. So I need to publish small things whose lack of circulation won't keep me up at night nor will I lose sleep over whether they are in obsolete formats in a few years. I can keep the source files and move them if I want to.
So the first thing to do is to choose a first publication. I thought of maybe doing the digital distribution article, because I wrote it specifically to start a conversation, but I only get occasional visitors to the fair use article. I'd hardly call it a conversation. I just don't know what the options are for reaching the audience I want to reach. Geez, I could just send the .doc file to 100 people, but that doesn't start a conversation. And most of my copyright friends are way too busy to blog and comment on blogs and engage over sections and paragraphs. I just don't know how to get started. Maybe a more interesting piece, the story of the smoke in Guatemala, with some illustrations. But it's not a conversation piece. Well, I have to start somewhere...

Monday, December 04, 2006

Of the case for fair use: digital distribution of course materials: the circular reasoning argument

The Strict Logic, or “Circular Reasoning” Argument

The fundamental point I am making here is that it is a mistake to think that one can win a fair use case merely on the strength of a logical, four factor analysis. That, simply put, is not how it works. The reality of how courts apply fair use renders another very old friend of nonprofit educational uses, the circularity argument, sadly ineffective. It is just too easy to get around it given the ease with which the factors can be manipulated.

Theoretically, even with a less than stellar first three factors, so long as at least two weigh in favor of fair use (in other words, the first three factors weigh at least marginally in favor of fair use), the circularity argument suggests that courts should ignore evidence of “lost” revenues, allowing only the social benefit of the use in question to tilt the fourth factor towards fair use, following the Williams & Wilkins Company v. United States case.[1]

It is wrong to measure the detriment to plaintiff by loss of presumed royalty income -- a standard which necessarily assumes that plaintiff had a right to issue licenses. That would be true, of course, only if it were first decided that the defendant's practices did not constitute “fair use.” In determining whether the company has been sufficiently hurt to cause these practices to become “unfair,” one cannot assume at the start the merit of the plaintiff's position, i.e., that plaintiff had the right to license. That conclusion results only if it is first determined that the photocopying is “unfair.”

The argument is always made in contexts like Electronic Distribution (iterative uses), but thus far, without further success.

Wendy Gordon dealt summarily with the circularity argument in her 1984 Betamax article. She called it a “formalistic question” that should not be the focus of the inquiry.[2] She began with the premise that a copyright owner is ordinarily entitled to revenue from all substantial uses of his work within the statutorily protected categories.[3] She went on to say that, “both fairness to the copyright owner and economic efficiency demand that the assessment of his injury include the loss of revenues he would receive in the market were his entitlement to be enforced.”[4] For her, the central question about whether copyright law could be adapted to new technologies was whether markets were likely to form around new uses.[5] Fair use, thus, was only for cases of present, and likely continuing, substantial market failure or market dysfunction resulting in unacceptably high social costs.[6]

The concept of market failure is part of microeconomics, the branch of economics that analyzes the market behavior of individual consumers and firms. The interaction of these individual decision-makers creates patterns of supply and demand that fix the prices of goods and factors of production and determine how resources will be allocated among competing uses.[7]

The premise underlying this kind of economic analysis facilitates sidestepping the demands of strict logic: if a copyright owner is entitled to all he can possibly get, there is little need for line drawing and no place for a “formalistic question.” If, on the other hand, he is entitled to only what is needed to provide an incentive to create, where to draw the line is central, and avoiding circular reasoning could provide a basis for drawing it. A loss of some revenues not critical to creation might be acceptable.

The fair use test in use today for iterative uses functions much as Gordon described. Because of the way the test has been simplified, the conclusion is guaranteed when it is the same as the underlying premise. “A copyright owner is entitled to any revenues that he conceivably might be able to get. If there is a functioning market for the use at issue, in other words, no market failure, the copyright owner is entitled to payment for the use. If there is a market failure, but the copyright owner can show that he can remedy it (that is, find a way to charge for the use), he is entitled to these theoretical revenues and the court cannot find that the use is fair.” This is simple, elegant and seductive. In fact, one need hardly bother with lawsuits if the law is this clear and easy to understand. Judicial economy has much to recommend it.[8]

Should universities and libraries argue something just as simple, elegant and seductive on the other side based on the circularity argument? “If a use is nonprofit and educational, even if it uses the entirety of a creative work, courts must exclude all evidence of harm to markets. Thus, all nonprofit educational uses are fair uses all the time.” These two diametrically opposed, I would say extreme, arguments nicely illustrate the “values conflict” between the commercial for-profit sector and nonprofit higher education. [9]


[1] Williams & Wilkins Company v. United States, 487 F.2d 1345, 1357 fn19 (U.S. Ct. of Claims 1973), affirmed, 420 U.S. 376 (1975).

[2] Gordon, Fair Use as Market Failure, p. 1651 – 1652.

[3] Id.

[4] Id.

[5] Id.

[6] Gordon, Fair Use as Market Failure, pp. 1620 – 1621.

[7] The Politics of International Economic Relations by Joan Spero and Jeffrey Hart http://www.indiana.edu/~ipe/glossry.html.

[8] Bridgeport Music Inc. v. Dimension Films, 383 F.3d 390 6th Cir. (Tenn.), Sep 07, 2004. The court in Bridgeport, in part, explicitly based its finding that 3 seconds sampled from a recording required licensing on the resulting judicial economy that the holding would foster. In other words, the court felt that finding fair use would encourage lawsuits; finding no fair use would discourage them.

[9] Laura N. Gasaway, Values Conflict in the Digital Environment:
Librarians Versus Copyright Holders http://www.unc.edu/~unclng/Columbia-article3.htm.

Friday, December 01, 2006

Interesting idea for a course on future of the book

I had an interesting idea today for a course on the future of the book. Wouldn't it be neat to write a book that had layers -- an upfront layer that was sort of like a normal book; a second layer that was heavily linked, providing paths in and out of the text; a third layer that included other people's comments and links in and out of the text, including commentary and response by the author. I can see it as a sort of wiki, but not quite like the currently available wiki software. One could choose to be in, read in, any of the three versions. Has someone already done this? If so, where is it? I would like to do this and maybe add it to the alternatives that are available to author who wish to make their works available in new and more social ways, more interactive ways. I guess I could start with an article or something short, but I need some help with the layers idea. Anyone know what I'm talking about better than I do?

Monday, November 27, 2006

Of the case for fair use: digital distribution of course materials -- Applying Sony

Applying Sony to Electronic Distribution of Course Material

Twenty-two years have elapsed since the Court decided Sony, and time has not been kind to its fair use holding. Its presumption in favor of non-commercial uses, its shifting of the burden of proof to the plaintiffs in non-commercial cases and its rejection of the idea that fair use favored only transformational fair uses have all been subsequently abandoned.[1] Moreover, the dissent’s approach, which adopted market failure concepts, has over time become the approach of the majority opinions in iterative use cases.[2] Nevertheless, for purposes of argument, we shall explore whether faithful adherence to Sony might support a fair use claim for repeatedly posting articles within course management systems or in electronic reserves. We should also consider whether, given the availability of instant permission through the Blackboard/CCC collaboration, Sony could support even first time use as fair. Put another way, is there still a fair use argument to be made in favor of spontaneous fair use if one can easily obtain permission to use the work through the CCC?

If a court acknowledged the benefit to the public and balanced it against the harm to the copyright owners as Justice Stevens did, and limited the sources of revenue it would consider when evaluating that harm to those revenues already in existence at the time the suit was filed and the record developed for trial, could the court conclude that Electronic Distribution of educational materials is fair use?

The statute’s four factors include:

1. the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;

2. the nature of the copyrighted work;

3. the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and

4. the effect of the use upon the potential market for or value of the copyrighted work.[3]

Home videotaping and noncommercial Electronic Distribution have similarities under the first factor. Both are widespread although not-for-profit. This finding in turn heavily influenced Justice Stevens’ analysis of the second and third factors. He concluded that in a nonprofit context, it was not that important that entire (third factor), creative (second factor) works were copied.[4] He emphasized that the copying at issue promoted a public good, given its obvious benefits to society.[5]

Copying for Electronic Distribution is equally, if not more, beneficial to society and enjoys the additional argument that many, although not all, of the works copied will tend to be factual. Thus, the second factor will more strongly favor fair use in many cases.

On the other hand, there are significant differences between Electronic Distribution and home videotaping. Electronic Distribution involves distribution of copies that the copyright owner offers for sale or license right now. Home videotaping involved copying by an individual for the individual’s own use broadcast programming that was offered to the public for free. These differences could affect the analysis of each of the first three factors.

So, following Sony would not necessarily result in the first three factors weighing for fair use. Sony can certainly be criticized for its reasoning on these points, and the dissent roundly does so. And, as I will suggest later, the ease with which one can manipulate the four factors is a key to understanding why we cannot rely on urging a particular view of those factors.[6] But for the moment, we will assume the best possible outcome for the first three factors in order to focus on the fourth, and to show ultimately how fallible any factor-based analysis is.

On the critical fourth factor issue of the loss of revenues, Sony placed the burden of proof on the publishers because the use at issue was nonprofit.[7] Further, the Sony assumption, that not all conceivable revenues should be considered, allowed the court to exclude revenues from any source that was not in existence at the time the case was filed and the record developed for trial. Even within this framework, copyright owners today would doubtless submit evidence to show harm to their database and permission licenses, if not to their subscriptions and sales of books and journals.

Even if some harm were shown, however, the benefit to the public from universities’ freely using articles and book chapters for Electronic Distribution would be weighed against any losses the copyright owners were able to demonstrate to their database and permission revenues. Ultimately, where the court places the burden of proving harm to revenues could affect the outcome, but even if the burden were on the plaintiffs, they might prevail on the fourth factor. That is, they might be able to show that the harm Electronic Distribution causes to their database and permissions revenues, if not their subscriptions and sales revenues, outweighs the public benefit of allowing the uses. Nevertheless, a result along these lines could still be favorable: possibly three to one in favor of fair use for Electronic Distribution. But with this result in hand, we now can see how fair use’s flexibility proves to be its liability.

We fool ourselves if we indulge the belief that Sony provides us a fair use pass for Electronic Distribution, for the truth is that all four factors are so easily manipulated that they can be made to conform to any desired outcome without difficulty.[8] For example, if the court believes that it is socially desirable to make nonprofit educational uses free of permissions and royalties, it will logically conclude that the social benefit outweighs harm to markets and minimize the effect of the use under factors one, two and three. On the other hand, if the court believes that uses that can be paid for through an existing market should be paid for, it will find that the harm to markets outweighs the social benefit of allowing uncompensated uses. The first three factors can be made to come out however they need to come out to support the conclusion the court has otherwise reached. Thus, even though by the numbers, following Sony, one can make the fair use case for Electronic Distribution (three out of four should logically result in a finding of fair use), if a court wants to protect revenues in established markets, to say nothing of desiring to protect possible revenues that might form around markets in the future, it can easily emphasize the facts that would support a determination that at least one and possibly two of the first three factors favor getting permission. The plaintiffs will have supplied the court with a plethora of such facts. The amount and nature of the work used, factors two and three, are quite flexible. Even the first factor can be made to reflect negatively on a proposed use. Words like “systematic,” “repeated,” “multiple,” or “blatant” can be applied even to nonprofit educational activity to sully the character of the use under the first factor. A use that loses support along the way to the fourth factor is doomed if the copyright owner can demonstrate harm to revenues.


[1] Campbell v. Acuff-Rose Music.

[2] Texaco; MDS; Kinko’s; Kelly; Field.

[3] 17 U.S.C. 107.

[4] Sony, pp. 448 – 450.

[5] Sony, pp. 443 – 447.

[6] David Nimmer, Fairest of Them All and Other Fairy Tales of Fair Use, 66 Law & Contemp. Probs. 263 (Winter/Spring 2003).

[7] Sony, p. 451.

[8] David Nimmer, “Fairest of Them All.”

Monday, November 20, 2006

Of the case for fair use: digital distribution of course materials -- The Sony Case

The Sony Case

Most of the time. Sony was an important exception.[1] In 1976, several major motion picture studios sued Sony, the maker of the Betamax video tape recorder (VTR), alleging that Sony was vicariously responsible for the way television audiences used the VTR (much as we use the videocassette recorder today). The movie studios alleged that audiences’ recording infringed the copyrights in their programs and that Sony was responsible for the infringement because it manufactured and sold the machine that made the copying possible. The majority found that Sony could be vicariously liable for television viewers’ copying only if the Betamax had no substantial non-infringing uses. It examined the way people used the VTR and found that recording a program to watch it at a later time (timeshifting) was a substantial non-infringing use because it was a fair use.

Justice Stevens delivered the majority’s opinion. He did not follow a narrow market failure rationale. In fact, a narrow market failure rationale would almost certainly have led to the opposite result, as Justice Blackmun argued in dissent. After all, if the Court had ruled that the use of the Betamax were not fair, the makers of VTRs could have gotten together with the easily located copyright owners and agreed to pay them a royalty for the privilege of manufacturing and selling a device that enabled massive copying of protected works. A ruling that such copying was an infringement and not excused by fair use would probably have brought about that result in short order (thus “curing” the market failure). But Sony found that massive copying of broadcast programming for convenience was a fair use. Glynn Lunney describes the Court’s fair use calculus:

In defining the balance between these competing public interests, Sony begins with a presumption in favor of fair use and a broad conception of the public interest that fair use protects. Merely increasing access to a work, even unauthorized access, represents a sufficient public interest to invoke the fair use doctrine. A transformative or “productive” use is not required. Once the fair use doctrine is invoked, Sony places the burden squarely on the copyright owner to justify recognition of her private ownership rights. Only where the copyright owner has demonstrated by the preponderance of the evidence that the net benefit to society will be greater if a use is prohibited, should a court conclude that a use is unfair.[2]

In evaluating the critical fourth factor, the effect of the use on the value of or market for the work, the majority seemed to assume that the copyright owners needed a certain amount of income to ensure creation of their works, but that additional sources of income not currently in existence need not be created and funneled to them in order to achieve copyright’s goals. In other words, not all the income that might conceivably flow to the copyright owners had to flow to them.[3] This is a classic statement of the limited monopoly rationale of copyright law. Courts following market failure reasoning based on high transaction costs never make this kind of assumption.[4]



[1] Sony v. Universal City Studios, 464 U.S. 417 (1984). More recent cases also support fair use of iterative copies, in Internet contexts where there would be devastating market failure if a license were required. See, for example, Kelly v. Ariba Soft, 336 F.3d 811 (9th Cir. 2003) and Field v. Google, 412 F.Supp.2d 1106 (D.Nev. 2006). Both involve iterative copying intrinsically a part of the operation of Internet search engines where requiring permission for each copy made before making it, or indeed imposing any requirement for a license would undermine the viability of search technology on the Internet. Further, the courts in both cases characterized the iterative coping as transformative, and concluded that it did not affect the copyright owner’s market for the original. Indeed, there was no market for the original in the Field case. These cases thus support the basic assertion of this paper insofar as they indicate the overriding importance of characterizing a use as transformative, and of market harm and market failure, to the results of the fair use analysis in iterative cases.

[2] Glynn S. Lunney, Jr., Fair Use and Market Failure: Sony Revisited, 82 B. U. L. Rev. 975 (2002).

[3] Sony, pp. 450 - 451.

[4] See, for example the Texaco, Kinko’s, and MDS decisions, and the Sony dissent. The Sony majority simply did not cast the case in market failure terms. Had it done so, the use could not have been fair under the description set forth earlier in this paper. On the other hand, the use might have been fair if the Court considered the market failure existing at the time as a permanent or uncurable failure.

Monday, November 13, 2006

Of the case for fair use: digital distribution of course materials -- Market Failure

The Economics of Copyright Law and the Concept of Market Failure

Wendy Gordon can be credited with suggesting in the early 1980s that courts seemed to be using a sort of rough economic calculus to make decisions in fair use cases.[1]

She hypothesized that market failure theory could account for their results, and in fact, the presence or absence of a market failure could predict when courts would likely find a use fair.[2] As she described it, a court would first establish that there was a market failure then balance injuries to the copyright owner against benefits to the user if the right to make the use were determined to be fair. A fair use would imply that the court had concluded that overall, permitting the use without permission would be socially desirable. But even with a market failure and a socially desirable use, the court should still go on to ask one more question: will a finding of fair use substantially injure the copyright owner’s incentives by depriving him of some source of revenue he might otherwise have obtained? In other words, if the court ruled against fair use, and in the wake of the court decision the parties would get together and find a way to “cure” the market failure, generating substantial revenues for the copyright owner, it would be wrong to say that the use should be fair. So, the court would conclude in such a case that the use should not be fair.[3]

This may sound complicated, but as a practical matter, the economic concept of market failure now provides a simple way to judge fair uses. Under this theory, only if the market completely fails to facilitate a socially beneficial transaction (the “use” of another’s work in this case) between what would otherwise have been a willing buyer and seller of rights, should courts recognize the use as fair and thereby limit the scope of the copyright monopoly. In concrete terms, a court following market failure theory would find that a nonprofit educational use were fair only if it were impossible or very difficult for a user to make the use if she had to get permission from the copyright owner. For example, a court would find a use fair where the user cannot find the copyright owner, or if the work involves multiple rights-holders who cannot be identified, or if the copyright owner does not respond to requests to use her work, all situations where a requirement to get permission will result in the user foregoing the use, with no benefit to the owner.[4] We might call this “narrow market failure theory” because it recognizes only a narrow, one might say one-dimensional, definition of fair uses: those justified by market failure due to current and likely continuing, unacceptably high transaction costs.[5]

Market failure could be less strictly interpreted but so far courts have not given it a more generous scope. Markets do not fail only when there are high transaction costs. Generally, however, courts do not find that mere market dysfunction, that is, facilitating some but not the optimal number of transactions, justifies limiting the scope of the copyright monopoly.[6]

Courts using the narrow market failure approach generally conclude that if a market can or even could facilitate the theoretical transaction, it should. For example, if it is possible to license a use, it should be licensed (that is, not characterized as fair). Any revenue a copyright owner can get, she should get. It is easy to see how courts tend to expand rights rather than limit them when they start with the assumption that the copyright owner is ordinarily entitled to all revenue for all substantial uses of his works.[7] They believe that expansive rights will achieve the goals of copyright in a manner consistent with our market-based economy. Markets should be allowed and encouraged to work, and it just seems right that an author (or more often the company that owns her copyright) should receive whatever revenues the initial effort may ultimately be capable of generating.[8] The changes in copyright law in the twentieth century reveal a dramatic expansion of copyright owners’ rights, premised on these assumptions.[9] Looked at in this way, it is not that hard to understand why those arguing for limits on the monopoly, or put another way, those arguing for generous interpretations of fair use in iterative use contexts, lose most of the time.[10]


[1] Gordon, W. J. (1982). Fair use as market failure: A structural analysis of the Betamax case and its predecessors. 82 Columbia Law Review 1600.

[2] A market failure here means that there is a malfunction in the normal way markets bring buyers and sellers of products, services and rights together. Usually, if there is demand for a good or service, in a free market, someone will see the opportunity and provide the good or service. If a person tries to offer a good or service that no one wants, the effort will not succeed. The “failure” occurs when there is demand, but no one steps up to provide the good or service.

[3] Gordon, Fair Use as Market Failure, pp. 1614 - 1622.

[4] This total market failure scenario also characterizes the circumstances currently applicable to “orphan works,” and suggests that if pending legislation to address the social need to utilize orphan works fails to pass, users who take reasonable steps to determine that a work is orphaned can probably rely on fair use to make many uses of it. But, fair use is not so certain a defense as the limitations on liability that Congress is considering in the orphan works bill. [cite] Nonetheless, there may be institutions and individuals willing to move forward with some uses of orphan works relying on fair use.

[5] Id.

[6] Gordon, Fair Use as Market Failure, p. 1618 - 1619, 1630 - 1632; Loren, L. P., Redefining the market failure approach to fair use in an era of copyright permission systems. 5 J. Intel. Prop. L1. Market dysfunction is discussed more fully in the next section of the paper.

[7] Gordon, Fair Use as Market Failure, p. 1651; Lemley, Free Riding at p. 1041 - 1042.

[8] Goldstein, Copyright’s Highway.

[9] Vaidhyanathan, Copyrights and Copywrongs; Litman, Digital Copyright; Lessig, The Future of Ideas & Free Culture; Goldstein, Copyright’s Highway; Lemely, id.

[10] Texaco; Kinko’s; Michigan Document Services.

Thursday, November 09, 2006

Four weeks to go

I still feel like I'm living in a dream. I keep thinking about "Waking Life." It just keeps getting better and better. Four weeks from today the whole semester is over for me. I have turned in everything deliverable I owe except for two term papers. One is about 95% finished. The other is a group project that is perking along at a good pace. The efficiency and organizational challenge did indeed turn out to be all that I needed to get the whole effort down to manageable proportions. I've had free weekend time, relaxed evenings, though I still tend to work until late on Mondays through Thursdays, and even on Fridays on the weeks that Dennis is in Houston. But I still absolutely love everything I am doing. I just don't have to do it 80 hours a week to get it all done. So I have had plenty of time to reflect, to chat with people, to blog, to write my papers. And the connections are now so numerous that I don't actually see any separation between my classes. Notes for one class are written all over the readings for another, and for my papers I'm working on for the independent study with Phil. It's all one big happy exploration of ideas.

It occurred to me today that I could modify my copyright class (assuming I'll teach it again after I graduate) to include a future of the libraries component, a policy component, not just copyright. I should talk to Phil about this because our Copyright Law and Policy class might be better suited to this adaptation. I will certainly have enough material to maybe teach a whole class on it at some point, but it might be hard to get it approved (for me to teach it) since it would be outside my established credentials (copyright lawyer). Well, it's something to think about.

I placed out of 16 hours of french and am registered for a 3rd year class in the spring. I'm not so worried about french now. I feel a little more playful about it, not so serious. I'll do fine. I do need to get started making my arrangements though. May will be here before I know it...

Of the case for fair use: digital distribution of course materials -- Assumptions Underlying Copyright Policy

The Assumptions that Underlie Copyright Policy

The Supreme Court generally defers to Congress in copyright matters, letting Congress determine what set of rights will promote progress in accordance with the Copyright Clause in the Constitution.[1] But none of us, including the members of Congress, has sufficient empirical evidence to determine what level of protection will achieve those Constitutional goals, and this lack of evidence means that decisions about the level of protection are made on the basis of something other than such facts. Thus, what we assume and what we believe in the absence of hard facts seems to determine what levels of protection we will implement.[2]

Those who favor public access and use, including but not limited to a generous scope for fair use, believe that the rights Congress grants to authors are a statutory monopoly, and must be limited to comply with the Copyright Clause: Congress is empowered to grant only so extensive a monopoly as is needed to overcome the potential underproduction problem created by the special nature of creative works. This problem, roughly stated, is that a creator’s ability to recover his or her investment in creating, producing and distributing a work may be threatened by the ease with which works can be copied and distributed. In some cases, potential creators will not make an investment at all. Granting the copyright owner a limited monopoly alleviates this problem, providing a period of time over which the copyright owner is free from normal competition. From this perspective, copyright is a temporary and limited burden on the public’s otherwise free use of creative works in order to assure that there will be an adequate supply of such works to enjoy.

Those who, on the other hand, favor strong author, producer and distributor rights assume that the grant of rights need not be limited, indeed should not be limited except in very narrowly constrained ways. They believe that we should seize every opportunity to increase the revenues to authors, producers and distributors because every additional penny may potentially increase the pool of creative works.[3] This is sometimes referred to as the “property theory” because its proponents resist any limits on copyright owners’ rights and argue that intellectual property is “the same as” other kinds of property, such as real estate, that do not have the many statutory limits imposed by copyright law.

But, as indicated above, because of inadequate empirical evidence, neither side can prove that its theory will produce a better result. Nevertheless, the view that copyright owners should have relatively unlimited protection has enjoyed phenomenal success, of late. Proponents of this view can point to some rather straightforward economic theories to support the case for circumscribing fair use.



[1] U.S. Constitution, Article I, Section 8, Clause 8; Eldred v. Ashcroft, 537 U.S. 186 (2003).

[2] McGowan, D. (2004). Copyright Nonconsequentialism.

[3] Whether it actually does increase the pool of creative works, it certainly has the effect of increasing copyright owners’ wealth.

Monday, November 06, 2006

Of the case for fair use: digital distribution of course materials -- Electronic distribution

Electronic Distribution and Fair Use

I taught copyright law at my university’s information science school for several years and found that nearly everything I wanted to use to teach my classes was either available online for free or licensed by my library for my students’ use. But if materials I wanted to use were not readily available, could I have placed copies on electronic reserve through my library or posted them within my password-protected course Web site without the copyright owner’s permission? In particular, could I have assigned the same articles and book chapters over successive semesters without obtaining permission and paying fees? This is the central question of this paper: is this practice fair use? In the U.S., only nine people know the answer, but until they make their pronouncement, we do our best to answer the question ourselves by trying to predict what a court (perhaps even the Supreme Court) would say. We use the same tools a court would use to make its judgment: the fair use statute and the fair use guidelines. So, the central question becomes: Would a court likely agree that this practice is fair use?

If I were to consult the various fair use guidelines that likely bear on this subject, I would find that repeated use is not characterized as fair because it violates the concept of spontaneity. The idea expressed in spontaneous fair use is that if I do not have time to get permission, the use is fair.[1] Conversely, if I do have time to get permission, the use is not fair. The collaboration between Blackboard and the Copyright Clearance Center (CCC) to facilitate instant permissions for the entire CCC catalog throws the spontaneous “first-time use” element of the guidelines into question. Are there no fair uses of any materials that CCC licenses now? Is that how it works?

While these are interesting questions, whatever the guidelines ultimately suggest about the scope of fair use, they are not the law; I am entitled to rely on the fair use statute.[2] I believe I can articulate a rational description of my use under the four fair use factors that would support the conclusion that the use is fair. In other words, I can describe the actions I want to take in such a way that the balance among the four factors will tip in favor of fair use. I am just not very confident that in today’s climate I could get a court to agree with me. Courts have rejected arguments like those I would make. In fact, my arguments tend to show up, if at all, in the dissents. Publishers have racked up an impressive string of victories in their effort to trim fair use’s sails in this second, iterative category of copying.[3]

I might try to shore up my argument by pointing out the traditional acceptance of educational uses or predict dire consequences for education if the courts further strengthen copyright protection by limiting the scope of fair uses in this context. But, there is nothing sacred about tradition, and predictions of gloom and doom do not actually prove anything, so they are not likely to persuade, especially in support of fair use rights. The winning and losing arguments in recent court cases, as well as legislation, show a pattern, however: our legislators and judges generally seem to embrace a vision of copyright that inexorably increases protection and shrinks limitations over time based on a premise that copyright owners are entitled to reap every benefit that might possibly result from use of their rights and therefore that more protection is always better.[4] Obtaining the opposite result would probably require a different premise, one that recognized that the relationship between the level of protection and the achievement of copyright’s goals is not linear, but the “more protection is better” premise seems to be the only one that succeeds today when iterative uses are evaluated.

The threat to fair use is thus symptomatic of a much larger trend in copyright law. The term of protection is growing longer; the scope of other exceptions such as first sale is growing smaller;[5] copyright owners are turning to technological protections buttressed by anti-circumvention law[6] and contracts to further strengthen their rights. There are many explanations for the upward trend in the scope of copyright protection. Paul Goldstein suggests that it is nothing more than an age-old struggle between those who believe the creator of a work is entitled to ownership of it as a natural right and those who believe the rights of the creator are only those fixed by statute.[7] Mark Lemley illustrates how the analogy to physical property has fostered a belief that users should pay for any benefit they obtain.[8] Others point to the need to strengthen rights because of the increasing difficulty of enforcing them in a digital environment.[9] Jessica Litman chronicles the capture of Congress by the copyright industries and argues that the form the statutes are taking is a logical result of Congress’s having turned over the legislative process to the strongest in the industry.[10] Others point to the publishing and entertainment industries’ efforts to preserve antiquated business models and revenue streams in the face of the uncertainty of the digital future.[11] Regardless of why the upward trend is happening, it is clear that it is happening.

As a result, defending limitations, or putting the brakes on the upward trend, will require some kind of strategic change. Courts must be persuaded to assume different things if they are to come to different conclusions about iterative uses.


[1] Agreement on Guidelines for Classroom Copying in Not-For-Profit Educational Institutions with Respect to Books and Periodicals, Published in House Report 94-1476; Electronic Reserves Guidelines developed during the Conference on Fair Use, though without consensus support. Neither of these guidelines describes the actual law of fair use; they are tools of convenience only. Nevertheless, both incorporate a “spontaneity” element suggesting that uses that are repeated need permission after the first time.

[2] 17 U.S.C. 107.

[3] See citations above to Kinko's, Texaco, and Michigan Document Services. The reader is again reminded to recall that this line of thought applies to iterative copies, as these are earlier defined in this paper, not to creative fair uses. This distinction is very important throughout the paper.

[4] Mark Lemley. Property, Intellectual Property and Free Riding. 83 Tex. L. Rev. 1031.

[5] Jessica Litman, Digital Copyright [need exact page references where she discusses the mods to first sale over 20th century].

[6] 17 U.S.C. 1201, et. seq.

[7] Goldstein, P. (2003). Copyright's highway: from Gutenberg to the celestial jukebox. New York: Hill and Wang. The natural rights framework for conceptualizing copyright does not always seem to take into account that in the U.S., most commercially valuable copyrights are owned by companies, not authors.

[8] Lemley, Free Riding at 1031 – 1032.

[9] Mark A. Lemley & R. Anthony Reese, Reducing Digital Copyright Infringement Without Restricting Innovation, U.C. Berkeley Public Law Research Paper No. 525662; U. of Texas Law, Public Law Research Paper No. 63; U. of Texas Law and Econ. Research Paper No. 025, Stanford Law Review, Vol. 56, June 2004.

[10] Jessica Litman, Digital Copyright, Prometheus Books 2001.

[11] Lawrence Lessig, The Future of Ideas, Random House 2001; Lawrence Lessig, Free Culture, How Big Media Uses Technology and the Law to Lock Down Culture and Control Creativity, The Penguin Press 2004.

Saturday, November 04, 2006

Of the case for fair use: digital distribution of course materials -- Introduction

I've been working on, struggling with, a paper on the fair use argument for electronic reserves and digital distribution of reading materials (and other materials) in course management systems like Blackboard. I want to experiment with collaborative authoring in a scholarly context, so I'm going to work with the text here, inviting comments and suggestions. I've already solicited comments privately, as is the norm in academe, but I'm still concerned about the argument, the reasoning, the factual predicates, and most of all, the conclusion I reach. So if it's not working, what better opportunity to try a new approach. Further, like other authors today, I am excited about the possibilities in collaborative authoring, and networked texts. I want to see how it works. What better way?

Here's the opening from the paper. I'll add other segments over the next couple of weeks as I see how this goes.

Many perceive fair use as less “useful” today than it was in the past. Publishers deem reliance on it as too risky. All it stands for is a right to litigate the question of whether the use is fair. To be safe, one must ask for permission even if the use might well be fair. Ultimately, the “permission culture” actually diminishes the value of fair use, though no change has taken place in the wording of the statute.

Some creators are resisting this trend, urging within their industries a return to reasonable reliance on fair use.[1] And educators and librarians who encourage access to and creative use of the works of others in general understand the importance of fair use and want to maintain its scope. Of course they can and do defend fair use in many fora, but in the courts, fair use seems to be losing ground in certain contexts. What’s happening to fair use, and what can be done to turn things around?

First, it is necessary to distinguish different functions for fair use, because all fair uses are not uniformly in trouble. Creative fair uses, those that use another author’s expression in a critique, a parody, artistic expression or news reporting and similar uses that build upon existing works in new ways, can be threatened by the general trend towards expanded owners’ rights,[2] but these types of uses still receive solid support from the courts today.[3] As the references cited in footnote two indicate, many scholars vigorously defend this transformational aspect of fair use. And as cited in footnote one, at least one entire industry is advancing the notion that permission for every creative use actually works against the creation of new works. Fair use for creative, transformational uses is not faring so badly then.

On the other hand, uses that reproduce an author’s work in order to make an exact copy, perhaps to conveniently use that copy at a later time or in another format or to give the copy to others (I will refer to these copies as “iterative uses”) are under relentless attack. Few seem eager to defend these uses in court anymore after a series of high-profile courtroom defeats.[4] This article focuses on these iterative uses, not on creative, transformational uses. In particular, it focuses on the digital distribution of educational materials to students through course management systems and library electronic reserves (“Electronic Distribution”). Only at its end does the article consider whether threats to iterative fair uses may affect creative expression that relies on core fair use. The reader is urged to keep the distinction between iterative uses and core creative, transformational uses in mind throughout the article.


[1] Cite to Fair Use Principles for Documentary Filmmaking and Duke University’s Bound by Law.

[2] See Illegal Art, www.illegal-art.org; Siva Vaidhyanathan, Copyrights and Copywrongs, The Rise of Intellectual Property and How It Threatens Creativity, New York University Press 2001; Lawrence Lessig, Free Culture, How Big Media Uses Technology and the Law to Lock Down Culture and Control Creativity, The Penguin Press 2004; Jessica Litman, Digital Copyright, Prometheus Books 2001; James Boyle, Software, Shamans and Spleens, Law and the Construction of the Information Society, Harvard University Press 1996; Eldred v. Ashcroft, 537 U.S. 186 (2003) affirming Congress’s power to lengthen copyright terms by 20 years retrospectively.

[3] Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569 (1994); Sun Trust Bank v. Houghton Mifflin Co., 252 F. 3d 1165 (11th Cir. 2001); Lee v. Art Co., 125 F. 3d 580 CA 7 (Ill.) (1997); Kelly v. Arriba Soft; Field v. Google; Bill Graham Archives. These cases show that parody, derivative works, copying to create a searchable image index on the Web, caching copies to aid Web search, and using small versions of larger poster images to illustrate a timeline can be fair use.

[4] Basic Books, Inc. v. Kinko's Graphics Corporation, 758 F. Supp. 1522 (S.D. N.Y. 1991) (commercial coursepacks); American Geophysical Union v. Texaco, Inc., 60 F.3d (2d Cir. 1995) (Texaco’s research copies); Princeton University Press v. Michigan Document Services, Inc., 99 F.3d 1381 (6th Cir. 1996) (en banc) (commercial coursepacks).