When a library adds a book to its collection, it adds a surrogate record for that book in the library's catalog. To get this record the library will either download it or create a record for the book from an international bibliographic record database. Authors have records too. These are known as name authority records. Recently the standards for creating these records changed to allow library catalogers to record more personal information about authors in authority records. This includes information about gender. There began a collective effort by a handful of catalogers to revise the new instructions so that binary gender was not encoded into the metadata of library records. This paper outlines the developments, results, and implications of this work.
Libraries use metadata schemes to describe and organize materials in order to facilitate access. This is foundational to—and inextricable from—the library project. Without classification and cataloging schemes, books would simply be in a pile; each time a user wanted to retrieve a particular book or books on a particular topic or by a particular author, they would have to sort through the pile at random. Metadata—data that describes the book in a number of different ways—makes it possible to search more precisely. When users enter a title or author into an online library catalog, they are using metadata to pull a book from the middle of the stack. These structures also enable serendipitous browsing. They describe individual titles, but also build syndetic relationships between them, collating like with like on library shelves so that readers who locate one book about the history of transgender identities will find all the others on the same shelf.
When a library adds a book to its collection, it adds a surrogate record for that book in the library's catalog. To get this record, the library will either download it or create a record for the book from an international bibliographic record database. Authors have records too. These are known as “name authority records.” An accumulation of authority records in a single database is called an “authority file.” The most widely used authority file in the United States is the Library of Congress Name Authority File (LCNAF). Library catalogers create new name authority records for all new authors and contribute those records to an international authority record database.
Metadata also ensures the maintenance of difference, preserving the boundaries between items that are like each other in some way but not identical. For example, two books might share the same title, or two authors might share the same name. “Authority control” is a term used in library science to describe the process of ensuring that every author in a library's collection is uniquely recorded in the catalog and disambiguated from authors with similar names. Authority control determines how we tell one author (e.g., Smith, Jane) from another (e.g., Smith, Jane, 1982–) in the library catalog by establishing parameters for the creation of authority records to contain a unique primary name for every author.
While these systems are necessary to enable access to materials in library collections, they are also subject to critique. Ostensibly objective and simply descriptive, all knowledge organization schemes reflect the ideologies from which they emerge. For example, the Library of Congress Classification shelves materials about bestiality near those about transgender people, and shelve both under a broad umbrella of deviant lifestyles. What might seem a set of category errors to individuals from these communities is embedded in the system and on the shelves as uncontested reality. In turn, the system reproduces those assumptions about the way the world works as patrons browse shelves and catalogers describe books. This has implications for trans people who are marked as similar to and different from on shelves and in the dominant systems that organize people and things in information environments around the world. Activist catalogers have worked to change the ways that materials by trans authors and about trans experience are represented in these settings.
An Evolution of Library Cataloging Standards
For more than 175 years, libraries have followed fairly straightforward sets of standards and instructions to create descriptions of the books and other resources held in their collections, as well as rules on how to record information about authors, subject headings, and assign classification numbers. A brief (and incomplete) history is as follows: Antonio Panizzi issued his 91 Rules for cataloging the books at the British Museum (now the British Library) in 1841. Charles Ammi Cutter published his Rules for a Dictionary Catalog in 1876. The American Library Association published their Condensed Rules for an Author and Title Catalog in 1904. The Paris Principles from the 1961 International Conference on Cataloging Principles outlined the functions and structure of a library catalog (International Federation of Library Associations 1961). The first edition of the Anglo-American Cataloging Rules was published in 1966 jointly by the American Library Association, the Canadian Library Association, and the (British) Library Association. The ISBD(M), International Standard of Bibliographic Description Monographic Publications, was issued by the International Federation of Library Associations Committee on Cataloguing in 1974. The second edition of the Anglo-American Cataloging Rules (AACR2) was published in 1978 to bring the cataloging rules in line with ISBD. The Functional Requirements for Bibliographic Records (FRBR) was published in 1998; the Functional Requirements Authority Data (FRAD) was published in 2009, and the Functional Requirements of Subject Authority Data (FRSAD) was published in 2010. This new “functional requirements” family was not necessarily rules, but recommended data models that introduced entity-relationship modeling theory for information-systems design to library metadata. AACR2 remained the primary cataloging code until 2010, when it was replaced by the Resource Description and Access (RDA). RDA follows the recommendations and entity-relationship models outlined in FRBR, FRAD, and FRSAD. In 2017, IFLA published the Library Reference Model, an approach that unifies and reconciles FRBR, FRAD, and FRSAD into a single model. RDA is expected to be updated to adhere to this new IFLA standard model in 2018.
RDA was adopted as the new primary cataloging standard by the Library of Congress in 2013, and ushered in new ways of describing library resources and the people associated with those resources for all libraries in the United States. One major change in particular was that RDA introduced many new attributes for describing people. Prior to RDA, catalogers created name authority records solely in order to disambiguate and construct a unique primary name for indexing. With RDA, catalogers are now asked to create contextualized biographical sketches in addition to constructing the unique name string for indexing. When they describe people they have the opportunity to include much more personal information:
Name of the person
Date associated with the person
Title of the person
Fuller form of name
Other designation associated with the person
Place of birth
Place of death
Country associated with the person
Place of residence, etc.
Address of person
Language of person
Field of activity of the person
Profession or occupation
This more richly detailed record anticipates future search-and-retrieval systems where the authority files themselves will be searchable. For example, researchers will be able to retrieve all records for authors who write in English, or who come from France. For the first time, catalogers are being asked to describe and classify people.
RDA is different from all earlier standards because it is based on theoretical data models—FRBR, FRAD, and FRSAD. This data model facilitates linking of library records to other data forms that are not held in the library. RDA enables library data to be linked to data elsewhere on the Internet. For example, a patron might search for information about Kate Bornstein on the Internet. With RDA, search results can include links to Wikipedia articles and documentary film clips as well as books by Bornstein held in the library. In other words, RDA facilitates search and retrieval that is both more complex and much vaster than the card catalogs envisioned by Cutter and earlier models of information organization. That theoretical model has material ramifications that are actualized both in the everyday practical task of cataloging library resources and by the creators (or contributors) of those resources through applying the instructions in RDA to library catalog records.
Addressing Binary Gender in Cataloging
Such a change in practice has particular implications for the description of the gender of authors. In its initial formation, the options available to catalogers for marking gender was binary: male or female, with a third option, “unknown.” For catalogers informed by queer and trans theory and experience, this binary gender rule was an alarming, if accurate, reflection of dominant ideologies around gender. The rule invited catalogers to assess and assign gender through the use of book jackets or information found on the open web, information that cannot be assumed to be accurate. For example, the hip-hop artist Big Freedia was misgendered as “female” in his Library of Congress name authority record1 even though she identifies as a male and uses pronouns fluidly (Welch 2011). Encoding gender also assumed that binary gender was both universal and eternal, not subject to change as ways of thinking about gender identity shift in geographical space and chronological time. Finally, the rule left out transgender and nonconforming identities altogether, as if these authors did not and will not exist at all. There was much debate in the library cataloging community on library listservs that exposed a level of ignorance and transphobia in the profession. In 2014, Amber Billey, Emily Drabinski, and K. R. Roberto addressed in depth the problem and the professional discourse of the binary gender bias in RDA Instruction 9.7 on recording gender in name authority records.
What followed was a collective effort by a handful of catalogers to revise the RDA instructions so that binary gender was not encoded into the metadata of library records. To change the specific RDA instruction 9.7 for recording gender of persons, a proposal was submitted to CC:DA in February 2015 to add the term “transgender” to the list of gender-term options defined by RDA instruction 9.7. This would list the options as “Male, Female, Transgender, and Unknown.” The proposal was accepted by CC:DA, and so it was sent to RSC for their summer 2015 meeting. This proposal was actually deferred, and the broader topic was discussed at the November 2015 meeting. A new proposal was submitted to the RSC at this meeting to continue to allow catalogers to record gender about authors but completely remove the predefined terms for gender as outlined in RDA instruction 9.7. Terms for gender would instead be decided by local cataloging communities. This change means that each individual cataloging community can define its own terminology for describing gender in name authority records. This reverting to local control acknowledges the highly contextual nature of gendered language, enabling cataloging communities to use—or not use—geographically specific terms for gender categories. The deprecation proposal was accepted and the rule was updated in February 2016. The new rule reads as follows:
9.7.1 Basic Instructions on Recording Gender
Gender is the gender with which a person identifies.
22.214.171.124 Sources of Information
Take information on gender from any source.
126.96.36.199 Recording Gender
Record the gender of the person, using an appropriate term in a language preferred by the agency creating the data. Select a term from a standard list, if available.
Record gender as a separate element. Gender is not recorded as part of an access point.
This means that a local cataloging community can determine the terms they want to use for recording gender for people, and that gender will never be recorded as part of a unique name string. So, something like Billey, Amber (female) will not be recorded as the unique name string in the LCNAF in an RDA-based authority record. Since the rule was changed, a task group within the Program of Cooperative Cataloging crafted best practices to be used by North American catalogers for recording gender in name authority records for persons. These best practices are currently under review by the cooperative membership. The task group expects these best practices to be formally adopted by the cataloging cooperative in fall 2018.
Since the authority record is independent of the bibliographic record, once an author's authority record is changed, then works associated with the author's authority record will link to author's current/correct identity. However, this change in RDA could have implications outside of library data management. The LCNAF data is openly published as linked data on id.loc.gov and is used for data reconciliation and normalization by other open datasets such as Wikidata, ORCID, the Virtual International Authority File, the International Standard Name Identifier database, and most likely many other unknown databases. With all this data reuse, it's impossible to control changes to data across all the platforms. This is why advocacy continues within the library community to encourage recording personal information such as gender only when absolutely necessary to disambiguate or contextualize information about the author.
Gendered norms are maintained and reproduced in systems and structures usually by people to whom binary gender is normal, natural, and obvious. In library systems, this was certainly the case in the initial rollout of RDA. Binary gender was an obvious way to differentiate authors from one another, and an essential component of describing who people are. For catalogers and classifiers who understand and experience gender in a different way, this rule needed to be contested. While largely invisible to users of library collections and catalogs, the change to the RDA rule adds transgender people to the project of information description and access while producing a useful story of one way that dominant systems can be resisted and changed: deliberately and slowly, by the people who administer them, one rule at a time.
Library of Congress Name Authority File, s.v. “Big Freedia,” accessed July 11, 2018, id.loc.gov/authorities/names/no2013076308.