In 2001, a group of prominent scientists urged a boycott of scholarly journals that refused to provide free online access to research articles within six months after publication. In an open letter to colleagues they pledged to "publish in, edit or review for, and personally subscribe to only those scholarly and scientific journals" that complied with their demand. They defended their stand with the proposition that "[a]s scientists, we are particularly dependent on ready and unimpeded access to our published literature, the only permanent record of our ideas, discoveries, and research results, upon which future scientific activity and progress are based." Over 30,000 scientists from 177 countries signed the pledge to boycott. It did not work.

Librarians are also unhappy with the current state of scholarly publishing. An annual subscription to some academic journals exceeds $20,000, forcing many university libraries to cancel hundreds of titles.4 The irony of the current system is not lost on university administrators, who complain that commercial publishers obtain research papers for free from university faculty, enlist other faculty as unpaid referees and editors, and then charge exorbitant prices to sell the results back to the universities that paid for the research in the first place.

The benefits of open access to scholarly research seem largely beyond debate. Whether that access can be achieved without seriously disrupting the production and publication of scholarly research, however, is a different matter. Proposals have generally taken two forms. One advocates reliance on a new generation of "open-access" journals committed to offering free online access to users. Funding, and the reluctance of researchers to forgo the prestige of publishing in established journals, pose major challenges for this approach. Another approach centers on "self-archiving"—the deposit by authors of published articles in an accessible electronic archive, whether a personal website, institutional repository, or discipline-wide archive. Here the law of copyright presents a major obstacle. Researchers typically assign the copyright in their work to the journal that has agreed to publish it. Any subsequent uploading of the work by a self-archiving author to a publicly accessible website may well infringe the publisher's copyright. The usual rejoinder urges faculty to be better stewards of their copyrights. However, even researchers knowledgeable about copyright are a poor bargaining match for the giant commercial publishers that dominate the industry.

This Article makes a more controversial suggestion. Universities should exercise their legal right to claim ownership of copyright in the research publications produced by their faculty. Only universities can wield sufficient leverage to compel fundamental change in scholarly publishing. Although traditionally an anathema to faculty, university ownership of copyright in research can be implemented without undermining academic freedom or the economic and reputational interests of university faculty.