This article originally appeared on ChrisSalvano.com on December 5, 2010

Subject cataloging has opened up a whole new aspect of librarianship to me. Namely, how does an indexer or bibliographer correctly and comprehensively determine an article’s “aboutness”? I’ve been working with the Hispanic American Periodicals Index (HAPI) since January 2009 helping update the organization’s thesaurus and index articles from Spanish language periodicals. Only since about June of this year have I started learning the more challenging process of subject cataloging.

Unlike subject cataloging, descriptive indexing is a standards-driven process that mostly eliminates the possibility of subjective interpretation. In contrast, subject cataloging requires that the bibliographer perform content analysis in order to determine what an article is about so that they can then assign meaning to an article. This process, as you can imagine, opens the door to subjective interpretation. After all, it’s a human who analyzes the content and applies subject headings based on what they think the article is about. So this begs the question, how does a bibliographer not only determine what an article is about, but how do they also assign subject headings that will help the end user correctly determine “aboutness”? What happens if my interpretation of the content differs from another person’s interpretation of the same content? Is the article then about two different things? The process of determining “aboutness” is something I’ve been thinking a lot about lately. After all, as an indexer, you have to use a limited number of terms to describe meaning, a potentially limitless concept. And as far as I know we don’t yet have a software program that performs automated content analysis, interprets meaning, and assigns perfect subject headings. This is no doubt part of what makes subject cataloging a fascinating and challenging aspect of librarianship.

At HAPI, I can apply up to four subject headings per article (including appropriate sub-headings) that describe an article’s content. Sometimes this is a straightforward process; sometimes it’s challenging. For instance, assigning subject headings to a literary critique of a work by another author is straightforward. For instance, a literary critique of, for example, Rubén Darío’s work Azul… should be indexed:

Darío, Rubén–Criticism of specific works–Azul

Of course, this is only an example and a bibliographer should also assign terms for any other themes or time periods discussed in the article to comprehensively determine “aboutness.”

However, what happens when an article addresses a variety of issues, or if the article focuses on a particular topic for which there is no thesaurus heading? For instance, I’m currently indexing a series of articles that discuss the spread of books, reading, and, in particular, the rise of popular literature in Latin America in the late 19th Century (more or less, the concept of the “best seller”). Many of these articles address specific authors, time periods, and geographic locations (all of which help determine “aboutness” and should be indexed), but many of them focus on what is known as the “folletín.” Folletines where short, serialized publications that appeared in late 19th century Latin American newspapers or magazines, many of which were eventually published as books. For the most part, these articles address the relationship between folletines and the development of a commercial book industry and the rise of popular literature in the region.

What if the term that makes the most sense to describe “aboutness” is not part of the official taxonomy structure? In this case, the term folletín is not an official HAPI subject term, so I have to use other terms to describe this content. At this point, the limitations of controlled vocabularies become evident. Here are the terms that I’ve consistently looked at to determine “aboutness” for these articles. Some work; some don’t:

• Publishers and publishing
• Booksellers and bookselling
• Books
• Novels
• Reading
• Literacy
• Literature (country-specific)
• Literature and society
• Newspapers
• Journalism
• Magazines
• Popular culture
• Mass Media

.
It would be great if there was a subject heading that was something like “commercialization of literature,” “popular literature,” “book as object,” or simply “folletín.” But a bibliographer can’t always get what a bibliographer wants.

Additionally, this topic is interrelated with some many other topics, making the determination of “aboutness” even more challenging. Folletines, for example, helped give rise to new literary genres and trends throughout Latin America (such as the novel) in the late 19th and early 20th centuries, and according to other sources, they also helped promote the development of national identities. Writing about the connection between folletines and the formation of national identity, Amy E. Wright notes in Novels, Newspapers, and Nation: The Beginnings of Serial Fiction in Nineteenth-Century Mexico (In Building Nineteenth Century Latin America: Re-Rooted Cultures, Identities, and Nations, Nashville: Vanderbilt University Press, 2009):

“[g]iven its flexibility as a container for multiple discourses and its possibilities for mass dissemination—inherent in its very conception as a hybrid between literature and journalism—this new narrative format was particularly poised to play a role in the nation-building process (p. 61).”

So, within the context folletines and the rise of popular literature, where does “aboutness” stop? Do I also include a subject heading for “nationalism”? Certainly the articles I’m indexing are tangentially related to this concept but they do not explicitly discuss nationalism or national identity and these terms are not prominent in the article’s text or paragraph headings. So, I think it’s safe to say that these articles are not about national identity or nationalism. Now, returning to the subject headings listed above, which ones help determine “aboutness” most accurately and comprehensively? One could argue that the article is about all of them. What about literacy? Certainly the spread of popular literature and new genre forms couldn’t have occurred without a literate populace. But while literacy may be an implied sub-theme, these articles do not overtly discuss the topic and the term does not show up in the title, abstract, section headings, or the article’s text. I think it’s also safe to say that the articles are not really about literacy.

In the end, the subject headings I’ve consistently used to describe these folletin articles have been “Publishers and publishing,” and depending on the particular article, “Novels,” “Reading,” “Literature (country-specific with appropriate date sub-headings), and “Newspapers.”

Subject cataloging and determining “aboutness,” as I’ve learned, is a whole new intellectual challenge than descriptive indexing. It’s an art form and a skill set that I continue to learn and improve on each day. I’m also learning a lot about the limitations of controlled vocabularies. There is no doubt that controlled vocabularies and thesauri bring a sense of rationality to how information is organized and recalled in an online database, which ultimately benefits the end user. But no matter how good the vocabulary is, all terms cannot be applied in all situations. Controlled vocabularies contain inherent limitations and cultural biases, as evidenced by the first project I worked on with HAPI, which was to update obsolete terms like, “Indians of Mexico” to “Indigenous peoples of Mexico” (an antiquated term still used by the Library of Congress).

Determining “aboutness” is an interpretive process, but in the absence of some sort of computer program that can analyze content, interpret meaning, and assign accurate subject headings with 100% certainty, bibliographers have the responsibility to perform the best content analysis possible and apply the best subject headings they have at their disposal in order to help the end user with their information needs. In addition, having a taxonomy structure that is robust and comprehensive, yet flexible enough to allow the addition of new terms and modernization of outdated terms will help both the bibliographer and the end user determine an article’s “aboutness.”

—
This posting is the opinion of the author and does not necessarily reflect the opinion of the HAPI organization.

HAPI Online contains more than 275,000 bibliographic citations to articles, book reviews, and original literary works from more than 500 social science and humanities periodicals published throughout the world. The organization, housed in UCLA’s Latin American Institute, indexes these periodicals, maintains a subject heading thesaurus, and publishes the bibliographic citations which are searchable through HAPI’s online search interface (HAPI Online).