The Gentlemen's Agreement of 1935 was a voluntary agreement that set guidelines for the limits of acceptable reproduction of copyrighted materials on behalf of scholars. Developed in response to the challenge posed by the easy and inexpensive photographic reproduction of research materials, the Agreement allowed library, archives, museum, or similar institutions to make single photographic copies of a part of a copyrighted work in lieu of loaning the physical item. The copies were not supposed to substitute for the purchase of the original work, and they were intended solely to facilitate research. Liability for misuse was to rest with the individual requesting the copy, and not with the institution making the reproduction.

The Gentlemen's Agreement has long been recognized as one of the most important landmarks in the history of the fair use privilege. In addition, the model of consensual voluntary guidelines agreed to by copyright owners and users, first used with the Gentlemen's Agreement, has become an important technique in clarifying the limits of fair use. Yet little attention has been paid to its genesis or intended audience. Most commentators view the agreement as primarily a product of long and thoughtful negotiation with librarians, and hence a reflection of their interests. A closer examination of the history of the creation of the Gentlemen's Agreement, however, reveals both the limitations of the common assumptions about the Gentlemen's Agreement and also the limitations of mutually-agreed upon guidelines. The individuals involved with the negotiations from both the scholarly and publishing side were far from representative of their respective areas, and had no authority to negotiate on behalf of their respective spheres. The Agreement itself was largely a product of one afternoon's meeting, with limited discussion and review afterwards.

Furthermore, the Gentlemen's Agreement was intended to serve the needs of research scholars, not librarians. Through an accident of history, however, it was a librarian who conducted the primary negotiations with publishers. As a result, library interests, and not the interests of the research community, came to dominate. Furthermore, the librarian who led the negotiations was different from most of his colleagues in both his professional dependence on the good will of New York publishers and the limited scope of his own library's involvement with library reproduction. As a result, broader issues, such as the educational use of copyrighted material or the extent of acceptable copying under fair use, were consciously excluded from the discussions.

Most of all, the Agreement began the process of subjecting to legal scrutiny private behaviors that up to that point had existed outside of the copyright system. Private actions that had needed no defense in the past came to be viewed as potential infringements of copyright that needed the permission of the copyright owner. Copyright, which up to this time had been a system for controlling publication and widespread commercial distribution of material, began to be seen as a system for controlling private reproduction and use of copyrighted material.

Codifying an agreed-upon set of sanctioned behaviors was not without its dangers. In particular, behaviors that were not part of the initial discussions and hence were not officially sanctioned by the Agreement suddenly seemed suspect rather than simply unresolved. The Gentlemen's Agreement thus began to be seen by some as a defacto cap on the extent of acceptable reproduction by librarians and researchers.

In the 1976 Copyright Act, the limited vision of acceptable behavior by librarians acting on behalf of researchers became codified in law in Section 108. In very real ways, researchers, librarians, archivists, and museum specialists still live with the consequences of the process that led to the development of the Gentlemen's Agreement.