The framers of the U.S. Constitution sought to balance the rights of the creators of intellectual property and the claims of the larger community. Article 1, Section 8, grants Congress the power to give “authors and inventors the exclusive right to their respective writings and discoveries,” but it also specifies that such rights be granted only “for limited terms” and with the purpose of promoting “the progress of science and the useful arts.” Today, because of the scale of investment that is required in order to create a unified cultural record online, the participation of commercial entities is essential, and yet many people (including most of those from whom the Commission heard) believe that the balance has been upset and that the property claims of rights holders are interfering with the promotion of intellectual and educational progress.

The most notable recent U.S. Supreme Court decision on copyright—Eldred v. Ashcroft (2003)—involved someone who was seeking to disseminate works in the humanities to a broad public. Eric Eldred was the organizer of the Eldritch Press Web site, 1dedicated to providing, for free, works bynineteenth-century authors such as Nathaniel Hawthorne. Eldred had wanted to add to his Web site Robert Frost's poetry collection New Hampshire, which was slated to pass into the public domain in 1998, 2but the Sonny Bono Copyright Term Extension Act of 1998 (CTEA) halted his plans. Eldred sued to overturn CTEA on the grounds that its twenty-year extension subverted the constitutional provision of “limited” copyright terms and did nothing to promote new creativity. Eldred’s case was heard and his argument was rejected by the Supreme Court. Unrestricted access to our cultural heritage in digital form currently ends in 1923: all of Hawthorne is up on the Web, but most of F. Scott Fitzgerald is not. Copyright restrictions will limit the Library of Congress’s planned World Digital Library: because the project intends to digitize only material in the public domain, it will have to exclude the great majority of cultural works of the twentieth century.

Obtaining permission to digitize books, even if they are out of print, entails high transaction costs: it can be difficult, if not impossible, to locate the current owners of copyrighted works. In a study assessing the feasibility of obtaining permission from 209 publishers to digitize 277 titles published between 1920 and 2000, librarians at Carnegie Mellon University found that a quarter of the publishers could not be located, only half of the publishers responded after repeated efforts to contact them, and, in the end, permission was granted for only 25% of the titles. 3

It is equally frustrating that many lesser-known creative and cultural works—not just books, but also photographs, drawings, films, and other materials—from the 1920s and later years cannot be made available online simply because the rights holders are difficult or impossible to find. Because recent copyright law has eliminated the requirement that rights holders formally apply for renewal, the copyrights of these so-called orphan works are automatically extended. Although such works often lack commercial value, the expense and difficulty of locating the rights holders blocks their digitization. Most institutions want to avoid the risk of litigation should rights holders surface after the works have been made broadly accessible. In January 2006 the U.S. Copyright Office issued a report 4on orphan works; hearings were held in the House and the Senate, and, as of this writing, it seems likely that legislation will be introduced to remedy this situation.

Even more complex issues arise in providing access to unpublished works (manuscripts and letters, for example), a category of particular importance to the humanities. Many sound recordings, too, are effectively “protected” from being reproduced in the practice of scholarship until the latter half of the twenty-first century, when any scholar now engaged in research is likely to be dead. 5

Current copyright laws not only keep most twentieth-century works from becoming available in digital form but also threaten the preservation of born-digital works. Although the copyright code currently has several important provisions that enable libraries and archives to make copies for preservation, these provisions are threatened by the transition to digital distribution. Section 108 of the copyright code is one such provision. It allows libraries and archives to duplicate works under copyright (in quantities specified by case law) to preserve their intellectual content. This provision covers the right of libraries and archives to copy works from one medium to another, such as brittle paper to microfilm or nitrate film to safety stock, and permits copying to digital form for preservation purposes (not for access). Yet it is not clear that all the forms of copying needed for secure digital archiving are allowable under the law.

The provisions of Section 108, created for the world of print, need to be recast for the age of digital replication. As the 1998 Digital Millennium Copyright Act (DMCA) demonstrates, when recasting copyright law, it is important to consider unintended consequences. The DMCA lacks all of the fair use provisions outlined in Section 107 of the Copyright Act 6and criminalizes all efforts to circumvent devices that prevent duplication of digital materials, including efforts made to copy electronic materials for preservation. Without such an exception, the preservation of published electronic materials is seriously jeopardized, and the problem is bound to escalate as more and more content is distributed digitally. The DMCA has also eroded the ability of public libraries, and, indeed, of any library that is not exceptionally well funded, to serve its patrons in a digital age, while putting at risk many digital projects such as those described earlier. In other words, we could become much worse off than we have been, historically, simply because existing law thwarts a reliable and cost-effective means to preserve cultural content as a public service. 7