Editorial policy

Since the initiation of the project in 1968, the mandate of the AFI Catalog has been to catalog every American motion picture either produced in the United States or sponsored and financed by American companies as an aid to the preservation of the American national film heritage. In accordance with the international film archival body FIAF (La Fédération Internationale des Archives du Film), AFI Catalog includes only those films that are 40 minutes or longer in duration, or 4 reels or longer in length. As well as the feature films from 1911-70, over 17,000 short films (those less than 40 minutes or 4 reels) have also been included from the first era of filmmaking, 1893–1910.

In support of its basic mandate, it has been the goal of the Catalog staff to codify the available documentation, present it in an accurate, decade-by-decade arrangement and provide the scholarly and archival communities, as well as the general population, with a significant historical and cultural document that delineates America’s film history.

For the past twenty years, the AFI Catalog project’s first goal to fulfill its mandate has been to include only contemporary credits and plot information in the main body of the entries, and relegate ‘modern’ information (i.e. information obtained from sources not contemporary to the era being researched) to textual notes. When this practice was initiated in 1983, more than fifteen years after the project began, it was a departure from the majority of film books, which coalesced modern and contemporary credits without equivocation.

The AFI Catalog staff recognizes the value of modern film scholarship and relies upon it for interpretation and insight into film history. However, as one of the goals of the Catalog is to provide an accurate historical record for each American film, similar to national book bibliographies, for the last twenty years it has been a priority of the project to present as factual only that information which can be documented in contemporary historical records. For motion pictures, these historical records consist of the films themselves and contemporary written sources.

As each decade of filmmaking has had its own characteristics, research into each new decade has necessitated extensive research to determine which historical records are extant and which are the most reliable. During research on the 1911–20 films, for example, the project staff had a relatively small number of extant film prints from which to create historical records. Thus, while preparing the 1910s decade of the Catalog, primary source materials for the majority of entries were written sources such as copyright records, motion picture trade journals, corporate records and censorship records.

When research began on the 1930s, however, the films themselves became the primary source material for the entries: of the approximately 5,200 domestic films produced during the 1930s, the Catalog staff was able to view almost 4,200 titles in 35mm or 16mm prints or various video formats. Secondary source materials for the viewed films included written contemporary materials such as copyright records, motion picture trade journals and corporate records. Entries of films that were not viewed relied exclusively on the secondary materials, with modern, or tertiary sources again relegated to textual notes.

For the 1940s, of the 4,323 titles in the project, the staff viewed approximately 93% of all domestically produced films of the decade. For the 1950s [only 1951–56 are included in this release of AFI Catalog (as of October 2003)] of the 3,418 films in the database, 95%+ have either been viewed, obtained on videotape or identified as viewable in American archives.

To ensure that all films that meet the stated criteria for inclusion in the project are cataloged, title lists from the database are compared among contemporary and modern printed sources, including corporate records, film almanacs, narrative histories of studios, release charts, copyright and censorship records.

Publications routinely checked for each film stage include, among others, Variety, Daily Variety, Moving Picture World, Exhibitor, The Hollywood Reporter, Box Office, Filmfacts, issues of The Players’ Directory, the Los Angeles Times and The New York Times. The Catalog of Copyright Entries: Motion Pictures, and the microfilm records of the Copyright Office at the Library of Congress are also checked.

The main research process also includes extensive reliance on the resources of AFI’s Louis B. Mayer Library and the use of production files at the Margaret Herrick Library at the Academy of Motion Picture Arts and Sciences, which provide invaluable clippings of reviews and other contemporary filmographic information for virtually all films made since the 1930s. In addition to production files, film books and extensive special materials, the AMPAS Library maintains the largest number of film pressbooks in the world.

Researchers may occasionally be required to read files of the MPAA (known as the ‘Hays’ or ‘Breen’ office at various eras from the 1930s through 1950s), the organization that administers the motion picture industry’s audience advisory ratings, G, PG, R, etc. Modern sources, including critical, historical and biographical works, are also consulted at this stage.

Each AFI Catalog record documents any underlying literary or dramatic source of the film, as well as all music compositions, requiring considerable research to be performed among standard literary and music sources available both in the Catalog office and in major research libraries in Southern California. Periodically, staff members have used archival and library resources at the Library of Congress in Washington, D.C. and other archives and libraries throughout the United States. Resources of foreign libraries and archives, in particular the National Film Archive and the British Film Institute in London, have also provided invaluable resources. In recent years, AFI Catalog researchers increasingly have been able to utilize the Internet for some background information, especially in the area of historical biographies, literary and music sources.

It is a primary goal of the Catalog project to have all entries reflect the original release of the film, thus it is a priority to view each film in its most complete form. While every effort is made to locate quality viewing copies of all films, because of the number of years that have passed since their production, it is not always possible to view them exactly as released. For this reason, Catalog researchers, in consultation with the editors, must determine, based on the knowledge they have acquired and in-depth analysis of contemporary and later written documentation, whether or not they have seen a ‘complete’ film. While modern sources (i.e., sources not contemporary to the production of the film being cataloged) are consulted for each film, credits found only in modern sources are not included in the credit fields but are mentioned in the note section of the records.

When a viewing copy of the film has not been located, the researcher is responsible for collating all available written sources and compiling the entry based on contemporary documentation. The summaries for non-viewed films are based on screen cutting continuities whenever possible. When continuities are not available, summaries are based on a cumulative interpretation of plot details obtained from contemporary reviews and/or studio pressbooks. For the combined 1940s and 1950s projects, for example, Catalog entries were based solely on written documentation approximately 7% of the time. Incompletely viewed films are always indicated in the body of the Catalog record. In many cases, if a film is determined to be incomplete, but the staff has access to an original cutting continuity for the film, the note section of the entry will indicate that some scenes were missing from the viewed print but reconstructed from a cutting continuity.

Once the film has been viewed (or reconstructed) and the secondary research completed, researchers enter credits into the Catalog database. After the production credits are entered, a summary is composed. Researchers then assign subject headings, divided among two fields, Principal and Additional, primarily derived from the current edition of the Library of Congress Subject Heading, but contained in a Catalog subject thesaurus.

The body of the entries for all films from 1931 through 1956 and AFI’s ‘Top 10’ films from the years 2000–2002, as well as selected entries from the 1910s and 1920s, indicate whether or not a print has been viewed by the Catalog staff. For viewed films, which are tagged in the database by the icon , credit fields distinguish between credits obtained from the screen and those from other contemporary sources. Users of the Catalog database can easily determine which credits were listed onscreen, and which offscreen, as the offscreen credits are placed in brackets [  ].

The primary area of textual information, the summary, offers a description of each film’s plot, ranging in size from approximately 300 to 800 words for films from 1911–50 and 700 to 1,100 words for films from the 1950s. The increased average size of the Summary reflects the increased average length of films over the decades. During the late 1940s, the average length of a film increased from what was prevalent in the 1930s and early 1940s. By the 1950s, the average length increased from 75–80 to 90–100 minutes per film, with many exceeding 120 minutes. The 1950s also saw the rise of ‘epic’ films such as Quo Vadis and The Robe, which often exceeded 150 minutes in length.

The summaries, as well as the credits, are important sources of information for scholars. For archivists, summaries may provide information necessary to identify a film with missing credits and unidentified actors; for film scholars, they provide detailed descriptions of film stories; and for social historians, they provide sociological data on past eras.

It is an editorial goal of the Catalog to make all text succinct, grammatically correct and syntactically accessible to all readers. Care is taken to assure the lasting relevance and scholarly integrity of the Catalog by making all summaries and notes non-judgmental and non-colloquial. Although dialogue, characterizations and plot motifs from past eras often confront Catalog researchers with modes of behavior or social mores unacceptable by current standards, special care has been taken to assure that no Catalog record is reflective of a bias of the staff against any race, religion, country or political viewpoint.

The second area of textual information, the note, provides an analysis of disputed credits, descriptions of location shooting, budgets, literary sources, pertinent historic relevance, censorship and alternate titles. Notes also refer the user to other films of the same series, sequels and remakes. These sections of the records would include such statements as “The working title of this film was Neon Rainbow” [the working title of Las Vegas Shakedown] as well as explanations of literary background, disputed credits, production information, critical reception, remakes and/or modern research. Notes are written in accordance with the established research and grammatical practices of the AFI Catalog.

Source citations to contemporary periodicals used by the Catalog staff are input in each record to aid other serious researchers who may need to look at each review or news item pertaining to a particular film. Because of the value of these citations to researchers, efforts have been made since 1996 to add citations for reviews from trade publications and other periodicals, such as The New York Times, to 1920s and 1960s entries as they were not included when the print form of the Catalogs were published. Currently, approximately 1,500 of the 5,000 entries from the 1920s have been augmented with source citations; addition of source citations to the 1920s and 1960s entries will continue for several years.

To ensure the smallest possible number of mistakes within Catalog entries, each record is edited in three stages: editing for typographical and stylistic errors, editing for content and reviewing each record for consistency in style, content and historical perspective, with special emphasis on the notes. A final editorial pass takes place as each decade is completed, to ensure overall consistency and accuracy.