A Paper presented at the Annual Meeting of the College Art Association
February 21, 2003, New York City

First a disclaimer: The opinions presented below are not necessarily those of the College Art Association or of the Committee on Intellectual Property.

The Fair Use Provisions are codified as Section 107 of Chapter 1 of US Copyright code. Fair Use represents an exception to the exclusive privileges that the US gives to creators. Its purpose is to prevent these exclusive rights from unduly hindering the free expression of ideas that is guaranteed to us in the First Amendment of the Constitution. In this way Fair Use is meant to encourage learning, to guarantee the right to comment and criticize, and to promote new creativity in those situations where the exercise of copyright's exclusive privileges may unfairly obstruct creators. Its main purpose, therefore, is to serve the public interest by counter-balancing the exclusively private interests of copyright. In this sense, Fair Use echoes the longstanding theory called the "copyright bargain," whereby the US Constitution made copyright extend for limited times under the assumption that the public would eventually benefit when the term of copyright expired. This theory was repudiated in the recent Supreme Court decision upholding the constitutionality of the Copyright Term Extension Act. Though, it may have been discredited by a Supreme Court that appears to hold no stock in balancing the public and private good, its root fairness has not and will not be forgotten.

Section 107 says, in part:

the fair use of a copyrighted work, ... for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright.

Fair use is not a free ride past the copyright barricade. The way it works now requires an analysis of the facts pertaining to each use. We have just heard Jeff Cunard summarize the four factors that typically must be considered when the courts are asked to decide whether a use is fair. [n001] These, of course are the factors that users will also assess when they decide to claim that their individual uses will qualify as fair ones.

Briefly, again, among the factors they will consider will be the following:

1. How a work is used -- with greater weight given to nonprofit educational uses. A scholarly journal or book published by a non-profit organization may be more highly favored than one published by a for-profit enterprise. Even so, the fair use status of a nonprofit work may be subject to the number of copies produced. An image in a short run scholarly journal, some say might stand a greater chance of being accepted as a Fair Use, than one placed in a journal that circulates to many thousands of subscribers. [n002] There are always exceptions. The intent to make a profit is not an automatic black mark. There is one case in which Larry Flint Enterprises won a fair use claim for having published an artist's works in one of their magazines. Flint prevailed because the images were published with commentary and did not substitute for the originals. [n003]
    
2. The nature of the copyrighted work itself. This criterion is generally understood as an assessment of the degree of originality of the original work. Greater originality in the copy deserves greater deference. Lesser originality in the underlying work deserves lesser deference. This factor therefore may consider the relation of the copy to the original. [n004] For instance, a full-sized color reproduction on faux-paint-textured board suitable for framing and hanging would not be as likely to be seen as fairly reproduced as might a small black and white image of the same work. There is a benefit for creating what is called "transformative" works -- works that are entirely new creations, despite any obvious dependency. There are always surprises. Sensitive observers might say that the famous (or infamous) Jeff Koons String of Puppies was wholly transformative in its recontextualization of someone else’s image. Charged with infringement, Koons lost anyway. [n005]
    
3. The third factor considers the amount of the original work used, and is typically understood as mitigating against the reproduction of an entire work. Its ostensible purpose is to prevent entire works of literature and other written works from being reproduced to the detriment of the copyright owner. This factor justifies using short quotations in reviews and commentaries. Its purpose is to prevent users from taking the "heart" -- the quintessential creative element of the work; for, to do so might be tantamount to taking the work as a whole and, as such, could serve as a market substitution for the original work. One might argue that small reproductions of whole works of art are vastly different from the original and are not commercial substitutes. Thus one recent case decided that thumbnail images posted on the internet do not automatically infringe the original work, especially when they have no commercial value, and no financial impact on the value of the original, which leads us to factor four. [n006]
    
4. This last factor asks if the reproduction will affect the value or the "potential" market for the original work. It is easy enough to understand how a reproduction should be considered unfair if it significantly cuts into the economic benefit of the copyright holder. But, to me, it is harder to understand why a "potential" (presumably unknown) market might have to be considered. As I see it, it is like asking the user to contribute new ideas to the creator's business plan.

Two rarely mentioned elements in the Fair Use code should be cited here:

1. While the four factors are to be judged together, they need not be considered exclusively. The court may judge other factors it deems relevant.
   
2. Unpublished works are not immune to a fair use analysis.

From the above analyses, it is easy to conclude that none of the listed criteria operate independently when deciding a fair use. Each factor tends to implicate the others, and any decision must take into account the whole picture -- the balance of separate elements and separate interests.

Keeping the four-pronged test in mind, it is one thing to decide that a particular use you you are considering is fair; but it is something altogether different for your assessment to be held up in court. Today, proving Fair Use requires what is called an "affirmative defense," meaning that the defendant has the burden of proof, and must supply the facts to support his case. This process seems to imply that all unlicensed uses of copyrighted works are, on their face, infringements until proven otherwise. Under this rule, were infringers felons, they'd be considered guilty until proven innocent. Luckily infringement is not a criminal matter. From the point of view of the copyright owner, a user who reproduces a copyrighted work under the assumption of fair use, and has not been asked to support his claim (that is, has not been charged with infringement) has just slipped by under the surveillance of radar. Owners claim that, technically, Fair Use is not a right; since it is always comes as the result of a court judgment. From this scenario rise many problems that Fair Use poses to users, hitting artists and scholars most deeply. The copyright community categorizes undetected uses, infringements and putative fair uses together, as "market failures." In this way, they seem to acknowledge neither the moral nor the Constitutional imperative that mandates Fair Use.

Making Fair Use the consequence of an affirmative defense places prospective fair users at a great disadvantage. It turns the application of Fair Use into a chess game weighted against the user. Users must first conduct a risk assessment to determine whether their particular unlicensed application is fair; but more important (unfortunately), they then must assess the risk of being sued and then assess their own determination to stand up to a suit. In the real world there are few uses that are so important to which poor scholars and artists will be able or willing to stand up and proclaim their use fair. Generally only publishers might take that risk; but they don't. Goodbye, Fair Use.

Obviously, there is something wrong here. The theory of Fair Use is wrapped around a paradox. A Fair Use should be a fair use, whether validated in court or not. Fair use, uncontested, from the user’s viewpoint is a legal use. It is only the perspective of the owner that stigmatizes untested Fair Uses as infringements. Accepting the owner’s paradigm unwittingly turns the user into supplicant and transforms an entitlement into something that must be repeatedly won.

During the Conference on Fair Use (CONFU) that took place during President Clinton's administration, copyright holders were telling educational users of copyrighted properties that their collections of images could be digitized if and only if the original images were "legally obtained." "Legally obtained" was industry code for works that had been acquired through license, since to them unproven Fair Use was synonymous to infringement. Industry held that Fair Use did not cover making slides for teaching. Following such advice would entangle teaching departments in a vast administration of rights acquisition, slow the development of course offerings, and strangle departments in unending financial obligations. The result would be to cripple the teaching of art and art history in the US. This is the kind of scenario that the Fair Use laws were intended to avoid.

Wouldn’t the entire system be fairer, and serve the creative goals of copyright better if the kinds of uses cited in the preamble to the fair use provisions were presumed to be fair – perhaps until a complaining copyright owner was able to prove that in a specific case the four factors that define fair use showed the use to be unfair -- thus placing the burden of proof on the owner? The presumption should be that "criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research" and creative uses are automatically fair until proven otherwise.

If so, the balance probably won’t change much, copyright owners will still bother suing only if there is some likelihood of obtaining satisfaction (rightly or wrongly) in the form of monetary compensation or in the form of stopping what they hold to be an egregious misuse or pattern of misuses. In the current system, Fair Use frequently resides in a state of stalemate as it would under presumptive Fair Use.

For instance:

What happens when a copyright owner decides to pursue a potential infringement? If a use can be stopped, an owner will frequently send out a cease-and-desist letter ordering the user to withdraw the use. If such a work is on the internet, for instance, or if a work is in a store (a postcard, for instance), they may request that the work be removed. Not infrequently, some arrangement can be crafted that will satisfy both parties. For instance, I've heard of works, clearly not Fair Uses, being allowed to continue circulating merely by agreeing to credit the artist properly and to provide suitable contact information. Even fair users may wish to accommodate copyright holders -- making friends out of potential foes. Sometimes an explanation of Fair Use and how your use qualifies as fair is enough to do the trick.

The above are the informal remedies; but, as Jeff has already related, the formal consequences for infringement can be severe, beginning with a request for injunctive relief. Relief may subsequently include seizure of infringing materials, their eventual destruction, restitution of lost profits and, under certain circumstances, an award of substantial and debilitating statutory damages. [n008]

Accordingly, many user risk analyses are skewed against deciding that Fair Use is a serviceable option -- even when it should be. In this way, the law in its laudable attempt to stop the traffic in infringed goods predisposes itself against the interests of potential fair users, since no organization, including publishers and non-profits will willingly risk an unfavorable infringement decision when the cost of rights would appear to be equivalent to potentially cheap insurance -- especially true when publishers can pass the task and the costs of securing image rights down to scholars, writers and creators. That’s why scholars are asked to obtain rights for every bit of intellectual property they use – Fair Use or not. They even must clear short quotes.

How do scholars and artists handle this mess? One way, often suggested, is, whenever possible, not to ask for permission, presuming that your uses are clearly fair – indeed, so transparently fair that there is no need even to ask. For instance, Stephen Weil, Emeritus Senior Scholar at the Smithsonian Center for Museum Studies, notes that if all uses were arranged in a spectrum from most to least fair, "illustrations used within a scholarly text" would be viewed as most fair. [n009] Writers handbooks, such as the Chicago Manual of Style, as Christine Sundt will tell you in the Town Meeting following this session, sometimes will advise not to request permission to use works when uses are likely fair, so as not to reveal that there may have been any doubt on the part of the user -- a strategy that might work in the user's favor in court -- should the user find herself there.  [n010]

A writer (if he can persuade the publisher to cooperate) may elect to assume all liability for potential infringement -- in an attempt to deflect an owner's desire for compensation from a target with attachable assets to one without. Fearing that all their teaching slides could be infringements, I know of one art history department that transferred legal title of their entire slide collection to one of the faculty members. Such a subterfuge may not work, but the fact it was done indicates the level of fear that the ineffectiveness of Fair Use, as it now exists, has engendered.

An author may effectively choose to self-publish, thereby abandoning the benefit of appearing under a respectable imprint. Indeed, there are efforts underway to bypass the customary publishing paradigm, without abandoning the benefits of peer review. (I direct listeners, in this regard, to stay for the Town Meeting on Electronic Publishing, that directly follows this session.)

Fair Use is most effectively used when the author or artist is "flying under the radar," as they say. Flying solo is better. Charles Lindbergh, here, makes for more supportive metaphor than Amelia Earhart. Private uses are most likely to pass unchallenged and undetected -- such as collecting materials for private research, or circulating items among colleagues and friends. But, coming soon to your nearest computer may be methods to control even private access to copyrighted materials and to track their distribution and dissemination. It is a sad irony of our information age that privacy is increasingly threatened, you might be much better off if you don’t let your technology proceed beyond the age of paper and xerography.

In her majority opinion in the matter of the challenge to the Sonny Bono Copyright Term Extension Act, Justice Ruth Bader Ginsburg rejected the notion that continuously extending the duration of copyright violates the First Amendment right of free speech; but in doing so, she noted that Fair Use and the "idea/expression dichotomy" (as it is called) are sanctioned escapes from copyright’s restrictions. [n011]

Coupled with Fair Use, in this regard, is the policy that ideas cannot be copyrighted. Copyright’s protection only extends to the expression of ideas. Expression is protected; ideas are not. This means that there is no restriction to communicate an idea imbedded within a copyrighted work, that is, so long as one does not take the specific expression used by the work's author. This so-called "idea/expression dichotomy" permits scholars to describe copyrighted works, to relate plots, to review books, to convey facts and concepts, and so on.

But how does the idea/expression escape-hatch work when an author must convey something visual? How do you isolate the idea from the expression? Unlike the majority of written works (poetry is an exception), in visual creations, the idea is wrapped up in the expression. In visual works, as in poetry, idea and expression are almost always inseparable. [n012] Without reproducing the work, it is impossible fairly to convey its meaning. How can you write a book about, say, Picasso, without including illustrations? The only way to do that (assuming that you are writing about his art and not his lust) is to presume that readers retain an image of his works in their mind's eye -- one of the last vestiges of royalty free copying. (How long will that last?) Indeed, to omit pictures seems ludicrous. Eve Sinaiko, Director of Publications for the College Art Association, recently showed me a book about Andy Warhol -- entirely unillustrated -- an anomaly certainly bound to become a collector's item. How does an author or teacher communicate facts and ideas about the development of art, styles, or significance without reproducing the aesthetic expression? In education, you can’t have one without the other.

If using the "idea/expression dichotomy" is an unreliable strategy for the visual arts, this leaves Fair Use and the Public Domain as the only ways to convey an unlicensed visual idea, and as we have seen, Fair Use, in most circumstances is so susceptible to fear management -- in publishing, especially, and also in the creative arts -- that it is, at best, a severely dysfunctional utility. While the Public Domain is rich, its future is in jeopardy, its growth now, and perhaps forever, stunted.

In short, when it comes to pictures, copyright's reach extends deep, severely truncating the rights of the public. Indeed, because Fair Use is weak, copyright holders have been encouraged to fashion an arsenal of weapons with which to exert control over the free expression of visual ideas.

What if the copyright holder decides to disallow a writer from incorporating a visual work in her text? Perhaps the copyright holder disagrees with or is offended by the writer's ideas. What if the copyright holder wants to impose unacceptable restrictions on a use -- such as requiring the author to accept an attribution or date of a work? What if a creator or her heir have placed unusual restrictions on the use of a work when it is relinquished to a repository? What if the heirs of Matisse, for instance, said that none of his works could ever be exhibited next to a Picasso or even printed on the same page?

Since stories of this sort are not at all uncommon, these crucial issues need to be discussed. In the Town Meeting that follows this session, two speakers, Christine Sundt and Siva Vaidhyanathan will address this very predicament.

I don't want to trespass onto their territory, but let me say that I've encountered authors who have been prevented from using certain images, presumably because the rights-holder disagrees with what they wish to say, and authors who willingly (or reluctantly) agree to change their texts to reflect the opinions of those who have the power to grant or withhold permission to use images. This kind of intellectual coercion nudges closely to Big Brother scholarship. Its consequences may frequently be benign, but in the extreme, it may compromise the perception of a scholar's credibility and scholarship's ultimate authority.

The above notwithstanding, there are reasons why copyright owners choose not to pursue those who claim fair use (or seem to be claiming fair use) -- and at times may even encourage unlicensed uses. For instance,

• Some copyright owners -- especially when they are young and/or relatively unknown -- look at fair use as a kind of free advertising, as a way to become known.
• Some owners believe that they have a debt to society and wish either to donate certain works to the public domain or to license them to the public with certain allowances and/or restrictions. For more, I refer you to the licensing examples provided by the Creative Commons.
• Rights-holders may acknowledge that for some purposes some uses are not worth pursuing,  including the following:
  • They embrace the notion of fair use, themselves.
  • The cost of legal action is greater than the potential return from a successful infringement suit.
  • The administrative cost of licensing a work may be greater than the return from the license.
  • Rights-holders may want to be associated with certain kinds of projects -- as when some museums don't charge reproduction fees to selected scholarly journals.
  • The rights-holders might be embarrassed pursuing uses that society generally accepts as desirable -- as when images are used in educational situations.
  • They may wish to build up good will in other ways.
  • If the creator is still alive, he may acknowledge that creation is a two-way street: As he depended upon the example of others, others should be allowed to depend upon him.
  • Some creators realize that when copyright is too long or too strong, it places a strangle-hold on the very process of creation. They may be personally invested in the health of the creative method.
  • Copyright holders may wish not to chance losing a fair use complaint, thus establishing a precedent for others to follow. For them, the uncertainty and self-policing of users is sufficient deterrent. While not in favor of fair use, the accommodate it for their own good.

We have already noted that publishers, and, indeed all those who perceive themselves in jeopardy for having claimed fair use, will probably not be anxious to claim their just right under its authority. In addition, because Fair Use, as we have described it, is wholly an American institution -- with specific significance with respect to the right of free speech and the need to maintain and encourage public discourse and public access to information, it does not travel well beyond our borders. So, if a work about foreign authors is to be published both in the US and elsewhere, for practical purposes, many consider it safer to proceed under the assumption that every use must be licensed -- or put another way, to be certain that the responsibility for error in the granting of rights is deflected onto someone else.

All this pessimism and uncertainty exists because the consequences of academic Fair Use have not been definitively established. What may be needed is for some sympathetic organization -- one with deep pockets, perhaps -- to test the resilience of the Fair Use concept by enticing a suit, or, at least be ready to defend certain well-chosen allegations of infringement. Even better would be for the law to carve out a sanctuary of Fair Use -- one in which users needn't stand up to mount an affirmative defense. Thursday, at this conference, David Lang suggested that "creative appropriation of copyrighted work be presumptively privileged in every instance." The same should apply to the use of copyrighted materials in scholarly creations. [n013]

Today there are strong tools copyright holders can use to inhibit and/or to prevent the exercise of Fair Use. Each of these, in their own way, adds impediments to the free exchange of the kinds of information upon which scholarly and artistic progress so often depend. In addition, as information resources increasingly become site-licensed – controlled by contract -- the right to access them is drawn ever further away from the general populace and becomes ever more closely invested in a group of individuals privileged by virtue of their association with specific institutions. If permitted to develop unchecked, the consequence of this trend will likely be the creation of a two-tiered information society, distinguished by a set of contractually determined rights that will either encourage or discourage access to information – determining who gets it and who doesn’t. The consequences of such a system are inimical to a free society and to the kind of access to information needed to promote and sustain democratic governments. Access by contract, in the scheme of fair uses, may be the biggest most unrecognized threat to the equitable practice of academic inquiry in modern times. The restriction on Fair Use of images is but one chapter in this evolving story. Seen another way, with increasing frequency, rights are provided for institutions and not for individuals.

Another chapter in this saga is about the shrinking Public Domain. As the term of copyright is made to increase, as it did with the Sony Bono Copyright Term Extension Act, works do not enter the public domain at the prearranged time. In this way, added pressure is placed on Fair Use to serve as a safety valve to answer the public's demand for access to their cultural heritage. But as we have seen, in many ways Fair Use is more of a promise than an actuality. Current trends do not look favorable. Thus,

Bridgeman Art Library, Ltd. v. Corel Corp. There are some favorable developments that hold promise for an expanded and more useable Fair Use. In 1998, the southern district court in New York threw out a claim by Bridgeman Art Library that the Corel Corporation had infringed on their digital images by reproducing them for sale. The fact that Corel did copy Bridgeman pictures was not contested. The court held, however, that these Bridgeman images -- the originals of which were all works in the public domain -- were merely reproductions, and as such, had no original content of their own. Without "originality" they were uncopyrightable (essentially, just like facts), and therefore, under the laws of copyright in the US and the UK (where Bridgeman does business and where "originality" is prerequisite for copyright), the Bridgeman pictures were not protectable. While the ruling does not serve as a precedent, and, at most, applies only to the district in which it was decided, nonetheless it sent a chill down the spine of image providers, who, in the wake of Bridgeman, expect to have a more difficult time protecting their reproductive photographs. The issue here concerns the lack of copyrightability. As such, Fair Use is not even called into play; if the underlying work is in the public domain, so is the copy.

Unfortunately for users, contract law trumps copyright, so acquiring an image from a museum may still oblige the applicant to honor conditions demanded by the source. Nevertheless, those who photograph public domain images out of books may not be obliged to honor any claim of copyright in the reproduction. Mind you, this decision does not seem to apply to photographs of architecture or three-dimensional works, or to any photograph to which the photographer has added his own interpretive agenda. Furthermore, just as is the case when claiming Fair Use, it will be necessary to convince publishers to accept both the legality of images so obtained and the suitability of the quality of images copied out of books. In addition, if a work is to be published outside of the US or UK, the decision doesn't hold, and you are back to square one.

Moreover, it is uncertain whether publishing on the World Wide Web, is to be construed as a publication in whatever locality it is accessed. Recently, under its own national law, Australia has allowed an Australian national to sue Dow Jones' web version of Barron's, a US-based publication, coming out of New Jersey, for libel. [n014] The Bridgeman decision is gathering momentum, of sorts. Lately arguments have been put forward to rationalize extending the rationale behind the court's finding to certain photographs of sculptural and architectural subjects. This writer is not aware of any wide-scale use of Bridgeman. However, recently, in a probable attempt to cut off the potential for Bridgeman's Art Library's holdings to be subject to Fair Use, the company has developed a licensed product to be marketed specifically for educational use. Unlicensed uses can then be interpreted as going against Fair Use factor four.

DMCRA. Recently, on another front, Representative Rick Boucher (Virginia) has re-introduced into the House, his Digital Media Consumers' Rights Act (the DMCRA), the purpose of which is to restore the historical balance copyright law gives to the public by de-criminalizing the use of encryption-breaking devices when their uses would not violate copyright, thus enabling the Fair Use of encrypted materials. [n015]

In Mattel Inc. v. MCA Records Inc., the Supreme Court refused to hear Mattel in its attempt to reverse a lower court's ruling denying its claim that a pop song that mocked the Barbie Doll as a "blond bimbo," infringed the toy company's trademark. At issue was the assertion that the song confused potential customers into thinking that the song was an advertisement for the famous toy doll and served through its parody to dilute the value and alter the meaning of its Barbie trademark. In 2002 the "California-based 9th U.S. Circuit Court of Appeals rejected the notion that consumers were misled by the song, noting that a pop art painting of a can of Campbell's Soup does not make viewers think Campbell's 'has branched into the art business.'"  [n016] This case helps answer the question concerning to what degree common commercial symbols that serve to define cultural values and populate the cultural landscape may be used in art and social commentary without permission.

Also cutting in favor of the application of Fair Use, is a decision in the 7th Circuit authored by Judge Richard Posner in the matter of Ty, Inc., v. Publications International Ltd. [n017], which found a publisher's illustrated catalogue of Beanie Babies to be a fair use. Cutting in the favor of fair use, in this case, was the accompanying descriptive text. Posner found that the text was nearly useless without the images and added information and opinions about the products in such a way that the production was within the parameters of fair use, such as typically applies to book reviews and commentary. Without images this text, he noted "would sink like a stone." He suggested that another illustrated book, one for children, that merely described the images in a way that did not add any critical or significantly descriptive element, would probably not be a fair use. Moreover, the illustrations do not serve as market substitutes for the items illustrated, but, rather, are "complementary" to the items under copyright. The images of the Beanie Babies, by themselves, while they can't substitute for the objects, nevertheless might be protectable as derivative works were it not for the effect the added text has in the creation of content, comment and meaning.

This Beanie Baby decision would seem to have considerable applicability to the fair use of images for teaching and scholarly use when they appear in the context of analysis, criticism, teaching and in catalogues raisonnés. Indeed, because these uses tend to be non-profit activities, the fair use argument might well cut even deeper in favor of scholarship and teaching than it did for the publishers of the Beanie Baby catalogue. One key difference should be noted. Ty made no market in images of Beanie Babies; their use in the catalogue had no market implications. On the other hand, images of works of art are often made available through museums and other vendors; they can be purchased. Were a scholarly Fair Use equivalent to the above case to be adjudicated, the resolution my hinge on the fairness of the prices being asked by the image vendors. Still, for educational uses and publishing uses, the most crucial problem remains. Educational Fair Use does not follow hard on the heels of case histories. Instead it often is about the resolve of publishers and educational institutions to withstand a challenge to Fair Use, even though it may be unlikely to occur. Displays of web-based "virtual museums," should they be composed of copyrighted works without commentary, following Ty, may not pass as fair use.

In Kelly v. Arribasoft, the appeal of which was filed February 6, 2002, in the United States District Court for the Central District of California, Judge Gary L. Taylor, presiding, the issue concerned the Fair Use of thumbnail images placed on the Web. The original case found that the thumbnail-scaled images produced by Aribasoft's image-indexing web-crawler had been within the domain of fair use, as well as full-sized so-called "in-line" images that one could view by executing a link from the thumbnail. In appeal, however, Judge Taylor only partly concurred, finding with the lower court that the thumbnails served a different purpose than those full-sized images photographer Kelley created, and did not substitute for the originals. They were transformative, served a public need and helped the process of creativity and were consistent with the function of the Web. Nor did the Aribasoft images serve to replace the original or have conflicting economic purposes. Therefore Aribasoft's commercial intent did not weigh in against a finding of fair use.

On the other hand, the full-sized images that Aribasoft took from Kelley's website and displayed "in-line" were held to be an infringement of the exclusive right of display without a redeemng Fair Use component. Judge Taylor ruled that the full-size images displayed on Arabasoft's site tended to attract viewers away from the site maintained by photographer Kelley, and thus reduced the likelihood that they would benefit Kelly. Nor were they accompanied by sufficient explanatory or interpretive text or warrant a more extensive fair use analysis. As such, they were found not to be fairly used.

The Aribasoft decision tends to favor the public presentation of unlicensed thumbnail images of larger works when they don't cut into the commercial viability or reproduce the scale of the original work. It does not establish guidelines for the size of thumbnails, nor does it suggest any degree of resolution reduction that would qualify as non-commercial. Rather, the decision seeks to determine whether to copy is sufficiently reduced to turn the new image into one that will be transformative and not tend to substitute for the original. Art historians and publishers might wish to extend this principle to apply to the publication of small images of larger works -- especially those images that merely serve to identify the work in question. Frequently these images are reproduced in black and white and are of small dimension -- perhaps a few inches on a side. Following Aribasoft, it seems likely that image catalogues of art collections, including surrogate collections, can be introduced onto the web without consequence. Judge Taylor, as did Judge Posner in the Beanie Baby case, maintains that Fair Use factor three, in which the amount and substantiality of the portion used might weigh against Fair Use does not apply. Both Posner and Taylor find that if a detail of a whole image had to be used, the purpose of the use would have been thwarted. In this, the judges seem to take "amount and substantiality" as literal concepts -- nominally meaning, in visual works, a detail of a whole -- analogous to short quotations from literary works. In his article "Fair Use/Museum Use" Stephen Weil, proposes that miniaturization and visual reductions for pictures serve the same purpose as using details. [n018] An 8x10" black and white photograph of a full painting could be seen as an "insubstantial" amount of a 20x25 ft. original since so much of the fine detail, texture and color has been lost. Following Aribasoft, it will be harder to mount images of copyrighted works of substantial proportional relationship to the original, unless the reproductions are qualified by the presence of significant analytical and critical content. Furthermore, the Aribasoft decision might have a chilling effect on a common on-line publishing and scholarly technique of cross-referencing by calling up segments of web content using in-line links.

In each of the cases discussed above, Fair Use can be claimed only by oneself. If the author or artist needs to depend upon an intermediary to publish, without a clear patch of uncontestable uses, creators will be forced to suffer the consequences of the fear of Fair Use. So while a few pro-fair use decisions may shine a ray of hope, it will take some time to learn whether they will have much influence on the mode of operations within academic institutions.

8. Aftermath:

In the aftermath of the defeat of the challenge to the Sonny Bono Copyright Term Extension Act, we detect a heightened effort aimed at ensuring that scholars and creators have fair access to the materials they need. If works aren’t going to be eventually obtained through the Public Domain, then they will have to be made available by Fair Use, by dedication to the public domain, through liberal contract provisions or by rewriting the copyright code. To this point commentator Siva Vaidhyanathan, one of the speakers in the following session, has written in Salon.com that the majority opinion of Ruth Bader Ginsburg, despite its evisceration of the constitutional basis of American Copyright law, still keeps faith in the efficacy and necessity of Fair Use and in the idea/expression dichotomy as means of counteracting the undesirable consequences of copyright's exclusive privileges. It makes sense that if works are not freely available because their passage into the public domain has been delayed, Fair Use will have to take over the role of providing access.

There may yet be some hope for an expanding Public Domain, where all uses are fair uses. From Lawrence Lessig, the attorney to challenged the Copyright Term Extension Act, comes a proposal to allow copyright owners to apply for copyright extensions upon payment of a small fee. He suggests that these extensions for a small fee may be renewable every fifty years. This scheme will allow those works not renewed -- works that no longer are believed to have any commercial value -- to enter the Public Domain by default. The purpose of this scheme is multifold: It builds up the public domain with works of no or diminished economic worth, works to which scholars, researchers and creators need access. It permits copyright owners to obtain the extensions they want. It will create a database of renewed and unrenewed works. A potential user can consult this database to discover 1) who has rights to registered and renewed works, 2) which works were allowed to pass into the public domain, and 3) which works were never registered. [n019] At first glance Lessig's proposal has much to commend it; but, for now it is little but vaporware.

Based on what we have already seen, while in principle Fair Use promises much, as with so many other things, the Devil is in the details. I think our community has learned a crucial lesson. It is no longer possible to lie back and presume that creators, teachers and scholars work outside of economic and political realities. Our traditional privileges, if they exist at all today, now must come to us by way of political action. It is our responsibility to take our needs and complaints to Congress, to assemble, seek partners, marshal our forces, lobby for our rights and to promote the importance of serving the public sector through multiple tools -- a vital Fair Use regime being among the most important. Indeed, the public sector, itself, must invest itself with renewed respect. It is no longer tolerable for scholars and artists to seat themselves in the back of the bus and accept the morsels handed their way. In 1998 the Copyright Term Extension Act passed both houses of Congress unanimously, with no discussion. That level of passivity will never occur again.

We have come a long way since the time when artists took for granted their right, indeed, their obligation, to create and develop by carefully studying, and perhaps appropriating elements of the works of the old and recent masters. How odd, that the customary and respected practice of imitating and incorporating the lessons of those who preceded us would have metamorphosed into paying events. Is this the expected consequence of the demise of the patronage system? Who would have thought that the world of art someday would not be laid at the feet of those who would wish to study and comprehend it? One would think that the products of criticism and scholarship along-side the creative arts would, in themselves, be valued as gifts and contributions to society; who would have forecast that to give such a gift, the author would be forced to pay for the privilege? It seems so hard to accept that no work has value except as a commodity, and that the transmission of ideas would be seen as an economic endeavor, first and foremost.

I know most of you came to this session wishing to learn about procedures to follow to ease the process of obtaining images and the rights to use them. What I want to leave you with is the idea that these procedures, whatever they may be now or in the future, are consequences of decisions that are politically motivated and driven by competing economic concerns.

Change is in the air. In the years I have spent as chair of the CAA Committee on Intellectual Property, I have heard many complaints and gripes from art historians, and a fair number from artists, too, about how they find themselves set up as fodder in a financial war being waged between those who own or manage rights, and those who bring works to the public. Academics, squeezed by academic obligation, are asked to pay the price of rights, but rarely receive just compensation. Those affected need not take this abuse lying down. Artists are asked to pay for nearly everything that influences them – as if being asked to pay for the air they breathe – as if those upon whom they depend never owed an aesthetic debt to anyone.

This last year, under the supervision of Jeffrey Cunard (from whom you have just heard) with the approval of the Committee on Intellectual Property, the CAA filed an amicus brief supporting the challenge to the Copyright Term Extension Act -- the CTEA. As everyone knows, our cause lost. But in doing so, the issues raised have attracted public attention and are helping define an agenda for future action. Providing teeth for Fair Use and galvanizing the affected public are bound to be important tasks in the coming storm of copyright activism.