Close this search box.
Close this search box.

by Catherine J. Johnson

dancing woman

Members of the dance community-dance makers, dance educators, dance presenters, dance administrators, and dance scholars-need access to information and materials to help them make new dances, restage dances that have left the repertory, teach dance history and dance practice, and extend the body of research in the history and aesthetics of dance. But, often, dance users cannot easily find the materials they need. When librarians speak of access or making information accessible, they generally mean cataloging an item or collection, preserving it, and making it available either onsite or through photoduplication services, according to the library’s restrictions and schedule. The user has access to the material when he can find information about its existence and location in a catalog or other source and develop a plan for getting either the item itself or a copy of it. While this definition of access is clear to the librarian, users with different perspectives or less familiarity with the systematic storage and retrieval of information may understand the term to mean something altogether different.

To many within the dance community, including scholars and researchers, access connotes having complete and open entrée-in effect, a kind of ownership. Access may also mean that the material is easily available to the user, to be loaned with little or no restriction or on demand. Access may further connote that a particular dance has been documented and a historical record exists. Within the dance community, there is an appreciation for the dedication of particular dance librarians and collections, as well as an understanding that libraries may lack the resources to consistently meet users’ demands (Keens 1991, 23-24). Nevertheless, when libraries fail to meet these demands, they are sometimes seen as unfriendly gatekeepers, withholding information or denying access to it.

The origin of this tension is historical. A quick glance at the history of library cataloging and classification reveals why the subject of dance might appear neglected compared to other disciplines, such as theater, art, or music. Early books in the West focused on how to dance rather than on dance descriptions, history, theory, or analysis. The discipline was viewed more as a recreational activity than as an art form. Dance programs in college and universities were often created as a subdiscipline in the physical education department rather than under the performing arts. The Library of Congress (LC) classification system defined dance within the classification scheme for sports and recreation (the GV class) (Library of Congress 1976). The Library of Congress established most headings for dance subjects under the term dancing, which describes it as a verb or an activity as opposed to a noun or an art form. Those looking for information on dance today are more inclined to search under dance or to browse shelves in the area of the theatrical arts or anthropology, frequently missing a significant amount of literature shelved elsewhere.

To compound the situation, until the 1980s library cataloging systems concentrated on book and print materials, while information about dance is more often found in materials other than books or published records: choreographic notes, anthropological field notes, moving image materials, photographs, and theatrical programs and other ephemeral publications. A research collection that truly served the needs of a dance researcher would need to be multimedia and multidisciplinary in scope. Book-centered cataloging systems designed to provide access to printed items have proven to be unsatisfactory for the field. Resources that document dance are found in many disciplines-including music, anthropology, ethnology, physical education, history, art history, aesthetics, and education-which has made the collection of materials more difficult for researchers. While these access issues are specific to dance, they may also apply to other disciplines where a written record is less central to the activity, such as popular entertainment, ethnographic studies, and folklore. The cooperative approach described below to improve access to dance research materials might also serve as a model for other fields challenged by similar issues.

Where are the Materials that Document Dance?

Research in Dance: A Guide to Resources lists 75 library collections and archives in the United States (Bopp 1994). Of these, six are performing arts companies or organizations that maintained their own archives at the time of publication, two are significant museum collections,1 and six are private collections. An additional 40 entries describe collections of secondary resources or book collections.

Over the past 50 years, dance has become an important focus of collection development efforts at several major American libraries. Most significantly, The New York Public Library (NYPL) established its Dance Collection, now a division of its Performing Arts Library, in 1944. Over time, the NYPL Dance Collection grew to be the largest single collection in the United States (and quite possibly in the world) devoted exclusively to the discipline of dance. The Harvard Theatre Collection has collected dance materials related to the theatrical arts since it was founded in 1891; in the 1950s it also began collecting in the area of theatrical dance and music. The Library of Congress holds materials documenting dance in many of its divisions, from the American Folklife Center to Prints and Photographs. The Library of Congress’s Music Division has collected dance-related titles and has a history of supporting dance creation, having commissioned Martha Graham’s Appalachian Spring in 1944. The curators of these collections maintain close ties with the dance community in an effort to ensure that archival materials are saved and eventually find a home. Through the diligent efforts of these institutions, a great deal of historical material documenting dance has been preserved and is accessible to the public.

Other libraries and institutions have collected dance materials on a more focused or limited scale. The San Francisco Performing Arts Library and Museum (SFPALM), originally founded as a collection documenting the San Francisco Ballet, eventually expanded to include all performing arts activity in the San Francisco Bay area. Ohio State University’s Jerome Lawrence and Robert E. Lee Theatre Research Institute has grown in recent years to include more dance materials and larger dance collections, such as the Twyla Tharp Archive. The institute supports the curriculum of the university’s dance department, one of the largest graduate programs for dance in the country. The Theatre Arts Collection of the Harry Ransom Humanities Research Center at the University of Texas, Austin, has acquired significant materials, especially those related to American musical theater, film, and the Ballets Russes. Historical societies and other organizations, such as the Chicago Historical Society’s Midwest Dance Archive, have developed regionally focused collections.

Since the publication of Research in Dance, and partially in response to a national study on dance documentation and preservation conducted in 1990, additional collections have been established at the libraries of The George Washington University (Department of Special Collections), Wayne State University, and the University of California at Irvine. Two major annual dance festivals-the American Dance Festival in Durham, North Carolina, and Jacob’s Pillow Dance Festival in Beckett, Massachusetts-have created archival programs that document not only their institutional histories but also performances, master classes, and lectures held during the festival period. Moreover, individual dance companies and other dance-related organizations have begun archival projects with the intention of either maintaining an archive for the long term or donating or selling their records to an appropriate repository.2 While public collections are growing, many materials remain in private hands, where access for the general user or researcher is limited or nonexistent. Efforts have been made to encourage those holding dance materials to work with major repositories to care for these materials and eventually to make them accessible on a national scale.

Steps Toward Improving Access

In 1990, a study commissioned by The Andrew W. Mellon Foundation and the Dance Program of the National Endowment for the Arts produced a report, Images of American Dance: Documenting and Preserving a Cultural Heritage (Keens 1991). The study involved extensive interviews on a national scale with a broad range of individuals in the dance and library communities. The study’s purpose was to “ascertain the breadth of dance documentation and preservation needs in the field, to gain a sense of the range of institutions and individuals involved, and to discuss with representatives from the dance archives field how best to achieve effective coordination among them.” The report concluded that while great gaps exist in the national collections, this fact is difficult to substantiate because many significant holdings remain uncataloged. It noted that “major archival institutions have cataloging backlogs of two and three years [or more] . . . irreplaceable material is daily discarded because the general public remains unaware of its historical and artistic significance; and . . . there are no organized channels by which to retrieve information” (Keens 1991, 11).

Formation of the Dance Heritage Coalition

One of the five principal conclusions of the Keens report was that “‘access’ has become everyone’s byword”-for the artists who create the work and the records of it, for the repositories that house those records, and for scholars and others who want to use those materials (Keens 1991, 23). The report recommended the formation of a coalition of the heads of major dance repositories to address access problems and other concerns.

A Coalition Planning Group was formed and began looking for collaborative solutions. In November 1991, representatives from the Harvard Theatre Collection, the Library of Congress, the NYPL Dance Collection, and the San Francisco Performing Arts Library and Museum, and an advisor from the Research Libraries Group met to discuss access, which seemed to be the most pressing need. Two discussion papers were written describing the status of access to the collections, the use of standard MARC formats, existing cataloging rules and manuals, name authorities and subject headings for dance, and archival arrangement and description for dance materials and suggestions for a future union catalog (Flecker 1991, Johnson 1991).

After further study, the Coalition Planning Group found that the participating institutions were struggling with major cataloging arrearages. While staff shortages were noted as a major contributing factor, a number of other impediments to access were identified. The planning group reviewed these impediments and discussed possible solutions and model approaches for overcoming them. The principal conclusions were as follows:

  • Cataloging efforts should be geared toward creating a single union catalog, and that catalog should be integrated into the national bibliographic utilities, rather than residing as a separate database, so that dance materials can be located with research materials in other disciplines.
  • A manual with guidelines for cataloging performing arts materials should be created cooperatively.
  • A coordinated effort for access, including cooperative work on name authorities and coordinated subject heading development, should be launched.
  • The cooperating institutions should define their backlogs in order to plan a cooperative cataloging project, and other collections should be identified for inclusion in the project.

As a result, the four major institutions in the planning group-The Harvard Theatre Collection, the Library of Congress, the NYPL Dance Collection, and the San Francisco Performing Arts Library and Museum-formed the Dance Heritage Coalition (DHC) and developed a cooperative undertaking called the Access to Resources for the History of Dance in Seven Repositories Project. The project was funded primarily by the National Endowment for the Humanities with additional support from private foundations. It lasted six years and successfully carried out the tasks that the planning group had recommended. Members found the collaboration to be beneficial in many areas and have developed additional cooperative work under the aegis of the DHC.3

Creation of a Virtual Union Catalog

To create the foundation for a union catalog of dance materials, 204,518 machine-readable records for the holdings of the NYPL Dance Collection were updated into current MARC format and added to the Research Libraries Information Network (RLIN) database. This addition makes the cataloging work of other institutions easier as more titles, name authorities, and subjects are now available nationally. Moreover, an additional 6,500 catalog records were entered in RLIN describing 2,000 linear feet of archival collections and other materials at SFPALM, the Minnesota Dance Theatre Archives at the University of Minnesota, the Harvard Theatre Collection, the American Dance Festival, LC, and NYPL.

Coordinating Name and Subject Authorities

A difficulty frequently cited in cataloging dance materials is the inadequacy of Library of Congress Subject Headings (LCSH) for dance. Over the years, the NYPL Dance Collection adapted and broadened the LCSH, establishing more specific subject terms for in-house purposes. For example, NYPL’s headings provide subject access to the more than 50 named systems of dance notation, whereas LCSH identifies only three. Some of NYPL’s subject headings were submitted to the existing LCSH headings to expand that system and to make headings widely available for all. Most significantly, about 53,000 personal name headings and title headings for choreographic works from the authority file became part of the National Authority File in 1996. The NYPL Dance Collection and others continue to add to this system through LC’s National Cooperative Name Authorities and Subject Heading Program (NACO). The NYPL Dance Collection manages a cooperative dance “funnel” through which other libraries creating dance authority headings can submit them to NACO under the supervision of the DHC. Now, as new choreographic works are created and performed, new title headings are established and contributed. For example, every year the American Dance Festival contributes headings for all works premiering at the festival.

Addressing the Cataloging Quandary

Dance and other performing arts collections include a variety of formats that require a range of expertise to catalog. Certain cataloging decisions common to these materials previously resulted in different solutions at different repositories. Decisions regarding the treatment of particular types of materials, the formation of headings, and the application of MARC tagging rules needed to be agreed upon and documented. By compiling rule references and sample records, the DHC developed coordinated guidelines to increase the efficiency and consistency of cataloging the many formats common to this specific subject area.

The development of the cataloging guidelines raised several intellectual issues. Traditionally, library cataloging has focused on cataloging the item in hand. This necessarily divorces the item from its context. A group of photographs that document a single event or performance, if cataloged as individual items, may result in a catalog record that yields little or no information about the content documented. The cataloging may provide adequate access to the photographs as individual objects but does little to provide access to the context, which was why the photographs were taken. The event or choreographic work may not be represented in any other format (that is, there may be no script, notation, or videotape that represents the dance). This problem struck the staff of the American Folklife Center at LC as they began work on the American Memory Project in the early 1990s. They developed the concept of an event-level record that both provided information about the ethnographic event, whether a powwow, quilting bee, or religious procession, and described all of the types of materials that documented it.

This concept is particularly appealing to performing arts collections, but not always easy or practical to implement. The event-level record has tremendous potential for new archives and has been recommended as a method of organizing and describing portions of the Merce Cunningham Dance Foundation Archives. An authority record for a choreographic work can form the basis for specific event-level records that describe all materials documenting different iterations (performances) of that work. At institutions with established cataloging systems, such as the NYPL Dance Collection, catalogers have been able to apply this concept in some cases, particularly for the cataloging of photographic materials.

Sample record

Fig.5. Sample record from NYPL’s event cataloging of photographs. The companion authority record for A Mirror for Witches provides complete production information.

It is unlikely that the creation or performance of an individual choreographic work will ever result in a single item that accurately defines it. The closest record of this sort at this time is a videotape that documents a performance of the dance. The catalog record for the videotape often can serve as the base record that mentions or links to other items. Since linking records can be time consuming, catalogers’ thorough and careful assignment of added entries and subject headings becomes important to ensure access.

While describing materials that document an event collectively in an event-level record can be beneficial, for magnetic media materials such as film and videotape, where serious preservation concerns require continual monitoring and reformatting, item-level control is necessary for collection maintenance as well as access. Similarly, item-level cataloging may be recommended for a particular type of object or work of art. For example, an individual drawing by Pablo Picasso for a production of Les Tricornes may be part of a larger collection (the Howard D. Rothschild Collection on Sergei Pavlovitch Diaghilev’s Ballets Russes), but adequate curatorial control over and access to this valuable object warrant a thorough item-level description.

Another difficulty faced by the cataloger and the curator of a dance or performing arts collection is the frequent dispersal of documentation. In many cases, the materials that serve to document a single work do not reside together. When an archive from an individual or a performing company arrives at the repository, the archivist is expected to maintain the original order of the material insofar as possible and generally cannot spend time reorganizing. Providing access to materials documenting the choreographic works becomes a challenge when they are dispersed throughout various sections of a collection or archive. Additionally, materials documenting a single artist, company or group, production, or event might be scattered among collections held by different institutions. Linking and cross-referencing are possible within a single repository but become more difficult across institutions. Cooperative work can ease this problem. For example, NYPL and Ohio State University are currently working on a joint finding aid for a shared collection, the papers of lighting designer Tom Skelton. The two finding aids will be linked and indexed in the same finding aid database, making it easier for users to understand the connections between the two collections.

It is clear that neither one set of cataloging rules nor one perfect cataloging record can generally fit all circumstances and needs. Access to performing arts materials requires flexibility and creativity in applying rules and guidelines, bearing in mind the goals of providing access to the performance work or dance event itself and the need for efficiency in cataloging. Cataloging with great detail or describing the material object at the expense of being unable to describe materials representing many other works defeats the purpose of increasing access. In other words, a little access is better than none.

Access in a Changing Electronic Environment

The rapidly changing technological environment of the library and archival community presents ongoing challenges. The development of the World Wide Web, library catalogs with Web interfaces, and the implementation of search engines that comply with the International Standards Organization’s Z39.50 standard for search and retrieval software have affected the way in which users interact with bibliographic utilities. Individual library catalogs available over the Internet are no longer primarily used by individuals associated with the institution and a few scholarly subject experts. Rather, whole new populations are using the catalogs of individual libraries for bibliographic and related research, regardless of institutional affiliation.

The increased use of and access to library catalogs is only one of the effects of the World Wide Web. Catalogs now refer and link to online versions of full-text documents, images, and finding aids describing archival collections. There has been explosive growth in the amount of information presented by libraries in general and by special collections and archives in particular. On the subject of dance, some 300 representative images from the Paget-Fredericks Collection of roughly 2,000 original drawings, paintings, and photographs owned by the University of California at Berkeley, depicting Isadora Duncan, Loie Fuller, Vaslav Nijinsky, Ruth St. Denis, and other figures of early twentieth-century dance, have been digitized and may be viewed on the UC Berkeley Web site. The Library of Congress has digitized its collection of early dance manuals and created its American Ballroom Web site, which provides the texts and video clips of examples of contemporary reenactments of the dances.

Finding aids describing dance archival collections at DHC member institutions have been encoded in a standard generalized mark-up language (SGML) using the encoding standard for archival description (EAD). These finding aids, capable of full-text searching, are becoming available on the Web via a collaborative DHC site. As with online finding aids in other subject areas and institutions, they offer the user the ability to review the contents of unique holdings located at great distances from one another. An interested researcher or potential user can then request access to specific materials either through microfilming or copying (when possible) or by visiting the repository. Users can discover the depth and breadth of the collection before they request to see the materials, saving time and energy. Because finding aids for dance collections in different repositories are mounted together on one site, the user also has the ability to search across the collections and to view and compare finding aids for related collections. Although the DHC site provides some centralized linking and navigation for dance materials, there is still a need to provide more detailed and appropriate maps of the information landscape, along with navigational tools and links between the online catalogs of repositories, union databases in bibliographic networks, and collection finding aids.

In addition to access through traditional library and archival channels, the World Wide Web has provided a platform on which others can easily present information and documentation to a wide audience. Individuals with a passion for dance have created Web sites of great diversity and varying quality. When the search engine Mosaic was introduced, one of the most popular sites on the Web was a Swiss-based tango site, set up by a tango aficionado, that included numerous video clips. Other dance companies and organizations have established sites providing materials and information that were previously much more difficult to obtain. The José Limón Dance Foundation Web site, for example, includes information about the foundation’s archival holdings. Nowadays, any individual with a collection of dance photographs can easily digitize them and make them available on his or her own Web site. Increasingly, photographers are using this method to publicize their work.

Although the DHC’s work has demonstrated that library and archival descriptive standards can be used for dance materials, alternative ways of representing large volumes of data are still being sought. European theater data collection agencies have discussed developing a standard format for describing information about theatrical performances that would be available in a unified database.4 A project launched recently at Cornell University aims to develop an international standard for performing arts metadata to make sharing and interlinking data easier.5

Access in the Twenty-first Century

The beginning of an information retrieval system for dance is now in place through a variety of modes: bibliographic networks, individual library catalogs, informational Web sites and links, finding aid databases, text-retrieval systems, and digital collections. A more seamless web is needed to ease navigation.6 Just before the explosion of the World Wide Web, Lawrence Dowler, then head of Research Services for Harvard College Library, described subject pathfinders that could help users navigate the resources for a particular area of study, from published sources to original manuscripts and visual images (1997). This idea has been carried into the Web environment. A thorough, sophisticated pathfinder for dance has yet to be developed, but the building blocks are in place.

On another front, more work is needed to identify additional dance documentation that is not accounted for or readily accessible. The Images of American Dance report noted that large gaps exist in the documentary record of dance in America that are historical, geographic, cultural, and artistic in nature (Keens 1991, 20). A systematic approach to identifying gaps has yet to be developed, but pieces of the process are now within reach. For example, as a result of the DHC’s Access Project, the major collections’ holdings are recorded in RLIN or OCLC, or both. With the recent publication of the International Encyclopedia of Dance by Oxford University Press (Cohen 1998), a method could be developed to search bibliographic networks and other information sources using the Encyclopedia to determine the scope of repository holdings on all subjects. However, unless other institutions housing dance materials contribute information to national systems, the gaps will be more difficult to identify and analyze. Tools such as the new DHC cataloging guidelines and a processing procedures manual for performing arts collections should standardize this process and make it easier. In particular, the manual is currently being revised to incorporate information on EAD. Furthermore, repositories will soon be able to use a finding aid template on the DHC Web site to add descriptions of their archival holdings. Still under development, the finding aid template allows a person to enter basic information about a collection and list the collection’s contents. As outreach efforts continue, it is hoped that more materials to document the gaps in America’s dance history will be discovered and made accessible.

Assessing the impact of the DHC Access Project on the participating repositories is much easier than assessing the impact on dance research generally or on research in other disciplines. Anecdotal evidence shows that processed collections and the tools and guidelines have been used. For instance, a scholar processing a collection of materials from a native hula artist at the University of Hawaii made use of the DHC processing guidelines. The Agnes DeMille papers processed as part of the project have subsequently been used for a new biography of DeMille. The relationships established with scholarly organizations have led to a greater awareness of the materials available and of the online resources, and have helped improve scholars’ understanding of the parameters of access in the information science environment. However, the true impact on research and scholarship is more difficult to gauge. Whether or not efforts to date have really improved users’ ability to obtain materials for their work will probably not be known until a body of resulting research and writing can be evaluated.

Although the accomplishments of the DHC Access Project and other efforts have solved some of the fundamental problems that contribute to backlogs in cataloging and processing, a major quandary remains. That is, the dance field continues to create and produce materials, and curators continue to acquire these materials at a rate that often exceeds the ability of their institutions to catalog and process them. Conventional wisdom says that a curator should not accept materials that cannot be cataloged and made available within a reasonable amount of time or that cannot be appropriately cared for. However, the urgent need of the dance field to find safe homes for materials at risk raises questions about the wisdom of that approach. Does such a policy result in a loss of documentation in fields that are less financially sound, while the records of more lucrative fields of endeavor are left to represent our cultural heritage? The DHC’s work and other efforts, such as the Dance Notation Bureau’s work to computerize the cataloging of its holdings, have made strides to reduce cataloging backlogs. Other efforts to improve access include the Dance Librarians Discussion Group’s (DLDG) forthcoming publication, A Core Collection in Dance, a bibliography for libraries interested in creating a basic, representative collection of secondary resources, and the Alliance for the Arts’ project to identify the documentation of choreographic works created about or by artists lost to AIDS. The DLDG’s participation in promoting the interlibrary loan of videotape among libraries will also ease access restrictions, at least for published videotapes. But the continued decrease in funding for such organizations means that backlogs will continue or begin to build again.

The concept of access is multidimensional. Librarians, archivists, and curators, with the support of library administrators, must be creative in making the materials in their charge as accessible as possible without causing damage. Providing access to materials without a long-term preservation plan is wasteful at best and futile at worst. Access and preservation must work hand-in-hand; access decisions must reflect the preservation needs of the materials, and the accessioning and processing systems should integrate both goals. Collaboration among repositories in all aspects, from documentation and collection strategies to cooperative access projects and coordinated preservation, will be necessary to ensure that the history of American dance of the past century is known and understood into the twenty-first century. The library and archives communities have the ability and responsibility to help the dance field move out of what critic John Martin once called a “limbo of illiteracy” (Keens 1991, 10). They can do this by defragmenting the scattered documentation of the past, uniting it intellectually through the Internet, caring for it, and encouraging more comprehensive documentation and collection for the future.


1 The two significant museum collections are the Serge Lifar Collection of Ballet, Set and Costume Designs at the Wadsworth Athenaeum in Hartford, Connecticut, and the Tobin Collection at the Marian Koogler McNay Art Museum in San Antonio, Texas.

2 Well-known companies with active archives include the Paul Taylor Dance Company, the Merce Cunningham Foundation, and the Nikolais Louis Dance Company, whose archive is housed at Ohio University in Athens, Ohio.

3 The Dance Heritage Coalition originally established itself as an informal organization under the fiscal agency of the San Francisco Performing Arts Library and Museum. Its initial activity included the development of the cooperative Access Project. The benefits of working together led to an increase in membership. The coalition, now formed as a tax exempt, not-for-profit organization, currently has seven members.

4 At the 20th International Congress of the Société Internationale des Bibliothèques et de Musées des Arts du Spectacle (SIBMAS) [International Association of Libraries and Museums of the Performing Arts] entitled “Documents et Témoignages des Arts du Spectacle: Pourquoi et Comment?” [“Collecting and Recording the Performing Arts: Why and How?”] held in Antwerp, September 4-7, 1994, several papers presented in the session on “Documentation et Informatique: Miracle ou désastre?” [Electronic Documentation: Breakthrough or Breakdown?] included discussion of shared electronic data collection and distribution of theatrical information. Of particular note were papers by Jürgen Kirschner of Germany, “Coordination and Standardization: Two Assumptions for Communication;” and Margret Schild, also of Germany, “Considerations for Developing an Electronic Theatre Information System.” The presentation of these papers was followed by extensive discussions regarding international cooperation.

5 At Cornell, the digital library staff, under the direction of Peter Hirtle, have created a simple Web-accessible relational database to test what sort of data people might need in such a database. The theater faculty is putting records into that database for Noh and Kabuki theater; Ann Ferguson, of the Manuscripts and Archives Department, is adding records relating to the playwright, G.B. Shaw; and other theater partners are testing it with some documents from Russian theater archives. The Cornell theater staff and faculty are including a wide range of photographic, printed, and manuscript items in the prototype database, and would like eventually to link all items to the text or video of the performance. The database is currently in English; making it multilingual is a long-term goal.

6 Steven Hensen, of the Special Collections Library at Duke University, coined the term seamless web to describe a system in which users can navigate easily from a broad description of a group of materials to a more specific finding aid, thereby giving contextual information to a digital surrogate of the item itself. Hensen referred to this concept in several presentations to the Society of American Archivists (1998).

Skip to content