Skip to main content

Decolonizing the Catalog

Text graphic showing possible changes to Library of Congress Subject Headings, such as "undocumented immigrant" for "illegal alien."

In summer 2020, during the national outcry that followed the murder of George Floyd at the hands of Minneapolis police, the concept of antiracism—or actively opposing racism and promoting tolerance and inclusion—gained traction in critical conversations about library work. Earlier this year the American Library Association’s Reference and User Services Association explored this theme further in a webinar titled “Decolonizing the Catalog: Antiracist Description Practices from Authority Records to Discovery Layers.” The panel brought together academic librarians who have worked to promote inclusive language in cataloging, taking advantage of opportunities to improve the Library of Congress (LC) classification system and within their own institutions, with a particular focus on issues related to African-American materials and anti-Black racism.

The panelists included Elizabeth Hobart, special collections cataloging librarian at Penn State University; Staci Ross, cataloging and metadata librarian at University of Pittsburgh; Michelle Cronquist, special collections cataloger at University of North Carolina at Chapel Hill; and Kelly Farrell, program officer for the Triangle Research Libraries Network, a consortium of academic libraries in North Carolina. Ross and Cronquist serve as cochairs of the African American Subject Funnel Project at LC.

What are some of the challenges users face when searching for titles related to racism and antiracism in the catalog? What led you to start researching this topic?

Photo of RUSA panelist Elizabeth Hobart
Elizabeth Hobart

ELIZABETH HOBART: Slate reported in June 2020 that bookstores were selling out of titles related to antiracism, as antiracism reading lists became popular. This piqued my curiosity as to how well libraries were representing these titles in their own catalogs and supporting discovery. If a user wanted to learn about racism or antiracism and wasn’t starting from one of these reading lists, how much could the catalog help? I wanted to look at those pieces of information that are not transcribed directly from sources, as catalogers’ judgment in these areas can really affect how usable and findable the material is.

I started my research by developing a core list of 21 titles related to racism and antiracism, culled from bestseller lists from Amazon and Barnes & Noble and popular antiracism reading lists by author Ibram X. Kendi and BuzzFeed. I then analyzed the catalog records for each of these titles in OCLC Connexion, particularly the summary and subject heading fields.

The summary field is a free-text field, meaning it doesn’t draw from a controlled vocabulary. This text can be provided by the cataloger or material can be quoted from another source, such as publisher information or dust jacket copy. Catalogers have a great deal of freedom in how we construct these fields.

I noticed a few things when I looked at the summaries and subject headings on all of these records. First was inconsistent usage of keywords. Of the 21 summaries, 10 included some form of the word race, racial, or racially—that’s less than half, despite being the most frequently used term on this list. The next most common term was some version of racism, racist, or racists, followed by antiracism, antiracist, or antiracists. Subject headings suffered from the same problem; there isn’t one term that brings the titles together.

Looking at the publisher-provided summary for Me and White Supremacy by Layla F. Saad, we can see it’s fairly long and detailed and contains many of those important keywords: racist, privileged, people of color, and white people. A user searching those terms would find this book pretty quickly. By contrast, the cataloger-provided summary for I Know Why the Caged Bird Sings by Maya Angelou is one sentence: “An African-American woman recalls the anguish of her childhood in Arkansas and her adolescence in northern slums in the 1930s and 1940s.” We might infer from this summary that the protagonist encountered racism but it’s never stated, so it wouldn’t come up in a search for race or racism.

The subject headings for The Bluest Eye by Toni Morrison, a seminal work of African-American literature, are limited to African Americans, girls, Ohio, and fiction. This is an extremely small number of headings, and some of the major themes of the book, including racism, discrimination, and poverty, aren’t addressed at all. This work appears on many antiracism reading lists, but it would be completely missed in a general search of the catalog. By contrast, the record for The New Jim Crow by Michelle Alexander has a high number of subject headings, but there’s an obvious one missing: Jim Crow laws, which doesn’t exist in the Library of Congress Subject Headings (LCSH).

The scope notes that govern how subject headings are used can be tricky as well. The LCSH scope note for racism, for example, essentially splits the topic into two headings: racism and race discrimination. The distinction is subtle and, on top of that, in most discovery systems scope notes are completely invisible to users, so most users would have no idea that they should be searching both headings.

(Hobart’s full analysis is available in College & Research Libraries News.)

How did the African American Subject Funnel Project come to exist? How does the group work with LC to propose additions or modifications to LCSH?

Photo of RUSA panelist Staci Ross
Staci Ross

STACI ROSS: The Subject Authority Cooperative Program (SACO) is one of four programs under LC’s Program for Cooperative Cataloging, and the African American Subject Funnel Project was initially conceived in 1999 as part of SACO. Since 2017, funnel project leadership has updated and created outward-facing documentation, established workflows, and built an active membership comprising both subject and cataloging experts.

Our work has increasingly dealt with advocating for more inclusive terminology in LCSH. For instance, members of the funnel project have drafted a letter urging the cancellation of the illegal aliens heading from LCSH. (For more on this topic, see “Conscientious Cataloging,” Sept./Oct. 2020, p. 16.) We have also asserted antiracist policies in our work and have pushed back on LC’s attempts to neutralize more challenging headings. For instance, LC initially made drastic changes to our proposal for Night Riders, a term that describes persons who committed violent acts against African Americans and other people of color. LC wanted to include a more neutral usage, which describes a group of tobacco farmers who were resisting monopolistic business practices. As a group, we were able to get LC to repeal their changes, and we continue to see more discussion between us and the LCSH editors.

Once someone proposes a new term, we collectively gather sources to support the proposal. These sources are authoritative texts as well as popular media, both of which serve as references for common usage, which is a key factor of successful proposals. After submission there’s a lengthy review period, and LC may make changes to proposals reflecting editorial deliberations.

Our recent work has focused on adding new concepts from the African-American experience not formally represented in LCSH. Terms such as Black Wall Streets and sundown towns are not new to those familiar with African-American history and culture. But these terms did not exist in LCSH until funnel project members collaborated to develop these headings from insider perspectives and scholarly references.

Over the past two years, the revamped funnel has submitted many new LC subject headings related to the African-American experience. Blackface and other related terms, for example, were developed at a time when it was being discovered that well-known politicians across North America had donned blackface and committed other caricature-based offenses, as documented in yearbooks and college photos. Before these terms were added, LCSH included only Blackface entertainers and headings related to minstrelsy. These were not suitable for describing this resurgent phenomenon.

Additionally, the funnel project has worked to make improvements to outdated and offensive terms. African-American members have expressed concerns with the outdated and increasingly offensive terminology used to describe Black and white people. We have begun the work of updating existing headings Blacks and whites to include the word people, and have compiled a spreadsheet of more than 180 related terms requiring changes, such as church work with Blacks and working-class whites. We’ve alerted LC early in the process and have been working closely with SACO specialists to tackle the many changes needed. In the same vein we are exploring changing the heading slaves to enslaved persons, following the guidance of slavery scholars. Work on these changes continues.

What challenges has the funnel project faced in this process? Do proposals ever get revised or rejected?

Photo of RUSA panelist Michelle Cronquist
Michelle Cronquist

MICHELLE CRONQUIST: There are some general challenges we face. LCSH is a very old vocabulary that dates back to the late 19th century, and the terms in LCSH are interconnected. In order to change Blacks to Black people and whites to white people, as Staci said, you need to revisit every heading that includes the word Blacks or whites. So it’s not just a matter of changing one thing but changing many things and convincing LC that it is worthwhile to make those changes.

We did a proposal two years ago for a new Jim Crow laws heading. If you were looking at a catalog record for a book about Jim Crow laws, you would see subject headings like African Americans—legal status, laws, etc.; African Americans—segregation; and segregation—law and legislation. Some members of our group felt that the Jim Crow era is a distinct period in history and that it would be useful for people to be able to search that in our catalogs. LC rejected that proposal, reasoning that there are segregationist laws from both before and after the Jim Crow period and it would be better to keep them all together.

Elizabeth made a very good point about the distinction between racism and race discrimination in LC’s scope notes. If you have a work on racism directed against a particular group, the scope note directs you to enter it under the name of that group with a subdivision, such as African Americans—social conditions instead of racism. We felt that this practice basically hides works on racism, so we proposed changes, which were ultimately rejected by LC.

Fortunately, the conversation didn’t end there: LC at least understood what we were getting at and acknowledged that the headings used for racism against specific groups are probably not intuitive to users. They gave us the option of proposing headings with the format racism against a particular group, so we have already put in a proposal for racism against Black people. I hope that soon there will be multiple headings in LCSH for racism against particular groups, and I think that that will improve access in the catalog.

We’re also working on a proposal for the concept of leasing of enslaved labor. If you look at the historical literature, it consistently refers to this concept as slave hiring. We would prefer not to use the term slave and instead use more inclusive language to describe this concept, but LCSH tends to use the language that is commonly used in the literature. That’s a challenge, because it takes a long time for the literature to catch up with current usage.

How can institutions promote more inclusive language within their own institutions and local catalogs?

Photo of RUSA panelist Kelly Farrell
Kelly Farrell

KELLY FARRELL: I work with Triangle Research Libraries Network (TRLN), a consortium of four university library systems in North Carolina. We have a shared online index called TRLN Discovery that allows users to browse the catalogs of these four institutions.

In 2019 TRLN hosted a screening of the documentary Change the Subject, about a group of Dartmouth College students who challenged anti-immigrant language and LCSH. This inspired us to put together a metadata team to review the subject terms used in TRLN’s discovery layer and then overlay problematic subject headings with alternative vocabulary terms.

An early project for the team was to draft and pilot a process for staff at TRLN institutions to submit terms for remapping, and then the team would shepherd the term through a review and decision process.

People submit headings they’d like to remap through a simple Google form. The only required field is the problematic term they think should be overlaid, but we also asked for suggestions on what term to use instead, sources for the suggested term, contact information, and comments. Submissions from the Google form populate a spreadsheet the metadata team uses to review proposals. Some terms require additional research, and usually this involves some follow-up discussion. For some terms, we reach out to the metadata interest group for their feedback; for others, we contact specific metadata experts who can help determine whether we should remap.

There’s a variety of reasons the team might not recommend a term for remapping. These include, as Michelle and Staci suggested, that the term might already be getting updated via LC through SACO, or that the team could not agree on an adequate term to remap with. But if a term is approved and the product owners are all in agreement, the original term is added to the mapping file.

It’s a pretty subtle change in terms of user experience. If the user searches the original term, such as illegal aliens, it will return results using the term; however, the subject facet and the item record subject will display the overlaid term undocumented immigrants. The remapping applies only to subject headings, so if illegal aliens displays elsewhere, such as the title field, then this information will not be overlaid with undocumented immigrants.

Our next step is to consider how to increase the transparency and visibility of the project. This might include wider distribution of the suggestion form beyond staff. We’ve also raised the idea of making some version of the form accessible via the user interface of Blacklight discovery services. There’s also opportunity for us to improve and increase access of the underlying documentation for this project.

Source of Article

Similar posts