Gender Identification and Cataloging Practice

Reconsidering Gender in Personal Name Authority Records

Identifying an individual’s public persona (or personae) can be complex. Historically, people have had a variety of reasons for masking or concealing their biological sex. For example, it was difficult—if not impossible—for women to publish their work in the past without assuming a male identity. More recently, with the acknowledgment of gender fluidity and the recognition that people have the right and empowerment to assert their gender identity, traditional views, guidelines, and standards all need to be reconsidered in light of these more inclusive social norms. 

Last month, the Program for Cooperative Cataloging (PCC) Ad Hoc Task Group on Recording Gender in Personal Name Authority Records released their revised guidance. Their initial report, in 2016, recommended revising the taxonomy for encoding gender and recording gender information that is readily and publicly available in cataloging records. The experience over the past five-plus years has indicated that this might not be the best approach. 

Their new guidance is that catalogers should no longer record the RDA gender element (MARC 375) in personal name authority records. Furthermore, catalogers should delete existing information in the 375 field when editing a record for any other reason. They also suggest that groups such as the Library of Congress, NACO, OCLC, and others, which aggregate cataloging records, consider how to programmatically apply this guidance across their collection of records at scale.

ISO/IEC 5218, Codes for the Representation of Human Sexes, first published in 1977 and updated in 2004, specifies only four options: “Not Known,” “Male,” “Female,” or “Not Applicable.” It provides numeric encoding for these values. When the standard was first published—and even when it was revised—there was a much less nuanced view of the spectrum of gender identity. Today, however, this standard is out of date and in need of revising, and systems that rely on it should be updated. Many government agencies already provide other, nonbinary options, and the movement to recognize this fluidity of representation is gaining traction. On March 31 of this year, the White House issued a statement on Advancing Equality and Visibility for Transgender Americans. This statement covered a variety of transgender-supportive activities, including, from April 2022, the option for all US citizens to select an “X” as their gender marker on their passport application. It also described how gender markers would be adjusted in other federal systems.

Regardless of the system or the use of the data, an individual’s wishes should be paramount in any decision-making on the sharing of gender information. Individuals should be able to express their wishes regarding the distribution (if any) of this information, and what the identified representations should be. While this is most pressing for living people, we shouldn’t presume that these issues are unimportant after death, as family or other interested parties may wish to continue to maintain an individual’s privacy.

Of course, there are historical and biographical reasons why gender information might be of interest, but whether this information is necessary for the purposes of discovery—one of the main goals of cataloging—is spurious. A researcher might have an analytical reason to assess the genders of a corpus of authors, but there may be other ways to discern this. Other rationales for including gender identification, such as disambiguation of records for authority control, also seem invalid upon closer examination. Some initial (unpublished) studies of the question of whether gender data improves disambiguation provide little support for its use in this context. Potentially valuable uses of gender information, such as reporting on use and circulation, e.g., if radio stations were only playing male musicians, might be tracked using authority data. However, this data does not necessarily need to be publicly distributed in the way that cataloging records are, and it can be reported in aggregate.

These sorts of challenges are compounded by how gender status can change over time. What might have been true at the time of publication may have changed since release or cataloging. When should records be changed, and by what means would people become aware that a correction is necessary? Similarly, connecting previous works to current identity might not be something the author is interested in—or welcomes.  NISO hosted a series of conversations about some of these topics last fall, with speakers including Bri Watson, who touched on these issues during their talk on minoritized vocabularies and metadata collections.  

There are certainly many ways that the standards community can support better capturing and more sensitive description of gender in the work of individual human beings.