Welcome back! This is a long one.
bad news, good news
During good-natured ribbing at our family’s Thanksgiving dinner, my cousin Matthew asked me if I had yet learned the Dewey Decimal Classification, and I said that Dewey was actually a deplorable dude, explaining that the way he constructed the DDC reflects very limited priorities that favor a small sliver of the American population: white Christian men (see the first ever Dunking on Dewey segment in Issue 3). As my sister’s eyes glazed over, Matt’s widened with fascination,1 and I decided it was time to stop putting off this next issue on naming and placing.
Last time (ish), I explained how entities are placed in knowledge schemes using thesauri and controlled vocabularies; I left off discussing the mutability of these placements, wondering how we could (and why we should) actively change them. I pointed out that nesting or linking entities is fraught, because we regularly encounter things that cross “established” boundaries (care workers, for example). Even when we do place entities in a seemingly consistent way, we might label them poorly (as “severe,” “moderate,” or “mild” disabilities, for example). In my conversation with Matt, I noted Dewey’s classification of religion leaves little intellectual space for non-Christian religions, grouping many non-Christian religions into a blanket “Other” category.
There are literally hundreds of similarly fraught examples within major classification schemes, but few make the news. A counter example to this, and an example of poor, harmful naming, is the Library of Congress Subject Heading “illegal alien.” The story of this heading and how it changed is an unusual case study that illustrates just what I mean about intervention in information and how none of this is inherent.
A note about the Library of Congress Subject Headings: they are sprawling. The catalog’s mission at times has been to organize “all of human knowledge.” As anyone who has ever tried to multitask knows, trying to do too much means something’s gotta give. In classification, this results in headings that no longer reflect common parlance and cultural consensus. One such example came to light in 2016, when undocumented students at Dartmouth kept encountering the term “illegal aliens” in LCSH. They launched a movement to change the heading to “undocumented immigrants,” running up against a few unusual barriers along the way that demonstrate some of the fraught aspects of controlling knowledge to begin with:
Usually, if you want to change an LCSH heading, you submit a proposal that argues warrant (the grounds for change), which can fall into the following categories (categories! They’re everywhere): user, organizational, cultural, scientific, or ethical.
Then, a committee reviews your proposal and either (A) approves it or (B) sends it back or (C) agrees with it but decides to implement it slightly differently. These reviews happen quarterly, so interrogating the catalog is labor-intensive but not infrequent.
The proposal argued that changing “illegal aliens” to “undocumented immigrants” fell under two primary categories: organizational (a significant number of institutions had adopted different language in lieu of “illegal aliens,” establishing authority for alternative headings) and cultural (a marked semantic shift had retired use of “illegal alien”). These were strong grounds for change.
Then, something entirely unexpected happened!
The then-majority Republican House of Representatives intervened in this process and ordered Library of Congress to continue to use the heading “illegal alien.” One representative argued that the term “illegal aliens” had been the precedent for “hundreds of years” (it hadn’t). There’s a long history of the LCSH as a nation-making process, which I will return to another time, but the short version of this is that Congress hadn’t bothered Library of Congress about changing terms in a long while, and they certainly hadn’t passed a resolution about it, and this behavior should make you pretty furious. I run a low-grade fury temp all of the time, so I’m not a great barometer, but this is bad stuff!
a short interlude about fascism
Governments deciding what is and is not good information is not a good practice. See: Texas’s “every opinion has an equal and opposite opinion” law that led one school administrator to recommend to her teachers that they teach the “opposing views” of the Holocaust. Originally, I made this paragraph a footnote but this is not footnote material, and here’s where I need to be exceedingly clear: representation and organization is not information itself.
Rep & org describe the information, they arrange the information, they convey new kinds of information about the information. Think of information as a coin, and rep & org as the sleight-of-hand performing a trick. Rep & org are not the coin itself. The coin has not performed the trick. The magician has. There are facts, and then there is the expression of those facts, and then there is who is expressing those facts, and that’s where everything goes sideways and we have to have these long explanations.
Classification has long served the knowledge organization of a narrow, dominant, powerful group. That power is richly reflected in the density of accurate, respectful naming and arranging of those groups at the tops of these hierarchies, gathered and coherent.2 The naming and arranging of everyone else are scattered throughout different classes of aberration, disorder, deviation, and loss. Right now, we are seeing that falsely assigned deviance and experiencing that loss in the strongarmed bans of materials from curricula and catalogs by labeling LGBTQ books as “pornographic” (they are not) and children’s books about Black protagonists as racist (they are not) or threatening to burn books.
As a resolution to this term debacle, by the way, Rep. Joaquin Castro introduced counter-legislation to the Republican intervention in 2019, and then President Biden proposed removing the term “alien” in all US immigration laws, so an updated resolution passed this November (2021). Now there are two headings instead: “noncitizens” and “illegal immigration.” Like I said, sometimes the LCSH reviewers take option C: they agree with you, but they decide to implement your suggestions differently.
what are we even doing here?
We understand that classifications are control, and that control has skewed in one direction, and it’s resulted in a lot of faulty classing. But these schemes can be large. Why bother trying to fix the wreckage? After all, what does the LCSH do for us (you reader, me writer)? We understand the necessity of consistent metadata for findability, yes, but do people outside of librarianship actually use subject headings? As Dorothy Berry, an archivist at Harvard, put it:
Most people who use internet-based search algorithms do not search using subject headings. The algorithms don’t even search exclusively using subject headings. When most of us search in databases or library catalogs or even Google, we use keyword searches, which draw from titles, authors, subject tags, and full text, among other data.
This changing environment of search further underscores the absurdity of classification. Not only are attempts to organize all information fraught and insufficient because our knowledge evolves, they are insufficient because our means of finding knowledge evolve as well.
I’ve turned these concepts over and over again in my head for weeks—while in class, or sitting in the Reading Room, or staring at my monitor—and I invariably hear Bo Burnham’s “A Little Bit of Everything” play on a loop behind this train of thought, the soundtrack to the ridiculousness of “universal” classification schemes.
Burnham’s song is an audio tour of the carnival of crap accessible by a flurry of fingers on our keyboards. The abyss of content is too overwhelming for any one person; it’s too much for any one attempt to organize it all. So why are we attempting that with our ways of organizing knowledge?
I return again to my conversation with Matt, who asked what the purpose of librarianship is in a world where everything is searchable online. It’s a good question. My answer is in two parts: first of all, not everything is searchable online and it never will be. But assuming it is (it won’t be), simply dumping digitized and digital-born content into the Internet doesn’t mean its findable. A purpose of librarianship is making things identifiable and findable; without attaching metadata to every document or photo or video or Tweet or bits or bobs that we stick into this fiber optic nightmare, we won’t be able to find it again.
Classification is useful, in this sense, because it structures that identification and findability. Say I’m a researcher who wants to learn more about the history between antiquarian booksellers and librarians. I’m not going to be able to find that info with a single keyword Google search. I’ll end up in a database, but no single database is dedicated exclusively to the history of bookselling & librarianship. Instead, I’ll have to sift through materials about the history of printing in colonial Latin America, and the development of ink in Germany, and reading practices in World War II Britain. All of these vaguely related articles might be gathered in a database called Book History, and they’ll be tagged and sortable by keywords that an information professional (maybe an editor or a cataloguer) assigned using field vernacular. The classification here is effective because it’s focused.
We manage best when we manage small. That’s not a failure but a feature of classification. Examples of effective small-scale classification exist: a famous one began in 1974, when Kahnawá:ke librarian Brian Deer developed the Brian Deer Classification Scheme for Indigenous communities and collections in the Pacific Northwest. He tailored each application to a community’s specific needs and desires in terms of functional requirements and organization. The bespoke schemes meant that interoperability (meaning, application) of the schemes between collections was inconsistent. He was consistent, however, in representing each community respectfully, thoughtfully, and personally in their organization schemes. His priorities were clear.
Another example, in a totally different realm, is about classifying birds: In our first week of Rep & Org, we read the first chapter of Steven Feld’s Sound and Sentiment, in which he describes one kind of Kaluli song-story (of the Kaluli people in Bosavi, Papua New Guinea). The songs include stories about community values and roles, as well as lines that mimic the calls of birds. The songs are an oral classification of the kinds of birds the Kaluli encounter daily that communicate:
priorities about intergenerational and cross-gender relationships,
expectations of care,
coping with loss or abandonment,
cultural grieving rituals, and
poetry.
In Western taxonomy, we organize birds according to Charles Darwin’s whole evolution situation, focusing on the development of functional morphological aspects and genes and such. The Kaluli taxonomy arranges birds by sound. Both classifications tell us about a society’s priorities, and both classifications communicate information about the birds (among other things), and both classifications subscribe to a specific logic. Just as with naming, placing draws on existing frameworks to relate entities to one another. Neither is right or wrong, they both simply are.
I am left wondering, after thirteen weeks discussing the deep inequities of structures that dictate finality and authority, how to change them. We’ve heard a lot of arguments from our readings: we can’t, we can but we should work within the system, we can and we should work without the system, we could but we shouldn’t because something worse might take its place. One author misappropriated an Audre Lorde quotation (“the master’s tools”). Another wrote in language so impenetrable it was almost farce. A third wrote poetry and concluded: if we try to burn this to the ground, the conflagration will last an eternity.
Let’s take a leaf from Bo’s book, or rather, a note from his songs: Fixing the whole thing is not the solution. Trying to manage the whole thing is the problem in the first place. Trying to master the whole of the bibliographic universe, trying to name and place everything, always, is absurd. This recalls to me the James Baldwin quotation:
There is reason, after all, that some people wish to colonize the moon, and others dance before it as an ancient friend.
Must we classify everything in one universal authoritative system, or can we dance with the possibilities of different knowledge? Living outside of those fraught, insufficient, unusable schemes offer us so many more things, as the ocean is deep, as there are stars in the sky. I am excited by the possibility of undefinable, unclassifiable, or classified by something else. I suggest we start small.
housekeeping and birdseeking
house
A few weeks ago, I linked to a New Yorker article about the Big Five shutting out libraries with prohibitive e-book prices, and now some Congresspeople are looking into that.
I get a daily poem to my inbox, and it’s the only email I will definitely, 100% open each day. Thanks to my cousin Hannah for the rec: Pome, from Matthew Ogle, where I first read Gregg’s “We Manage Most When We Manage Small.”
More tweets from librarians under siege:
Classification in a nutshell (click the tweet):
bird
More later.
I originally wrote “bugged out of his head,” but Matt noted that this diminished what was an enlightening conversation. So I edited it.
I imagine knowledge organization like a root system: the most powerful groups have space to grow and extend, root themselves deeply in the ground, stretch into the earth, and grow impossibly large. And in doing so, they crowd out less powerful groups, sucking up the nutrients and clouding out the sun. But we know that tree families do not act this way—they help one another grow, and so too can we if we deemphasize one single way of doing or knowing. Allow other systems to root down.
Can you please explain to me how they managed to argue this??: "children’s books about Black protagonists as racist"
“every opinion has an equal and opposite opinion” <-- I can but also simultaneously CANNOT believe this legitimately is a law that exists...