Illegal Aliens and Other Unwarranted Subject Headings


Technicalities 41(4) July/Aug 2021: 10-13

This is the fourth column in a series on the politics of metadata. And it is important to keep count, because I would think that on hearing the there was going to be such a column, betting librarians all over the world asked “what is the over/under on the number of columns until they write one on Illegal Aliens and the Library of Congress Subject Headings?” The American Library Association does not meet often in Las Vegas, which suggests that librarians do not gamble, or that librarians DO gamble and… ahem… it is a problem. ALA’s predilection for Chicago and New Orleans suggests an abiding librarian love for deep dish pizza and beignets, which are both more of a problem for me than gambling. You might remember that ALA conference in Vegas as the one with Jeff Bridges. I thought The Dude’s remarks on FRBR were pretty provocative. But if librarians are notorious gamblers, then I am imagining groans and cheers in technical services units when you find out I finally get around to Illegal Aliens in column #4. Or maybe just groans when you think “this topic, again???!!!???”

I do not want to rehash the basics too much here as most of this audience knows the issue already. Illegal aliens is the authorized term for the concept in the Library of Congress Subject Headings (LCSH). Many people and organizations, including immigration rights agencies, journalists and linguists have argued against the term, saying “illegal” should not be used to describe a person as it is dehumanizing and stigmatizing—preferred terms are “undocumented immigrant” and other variations. To be clear, the Associated Press is not so sure about “undocumented” either because of its lack of precision. Many librarians objected to the term, then some Dartmouth students jumped into the fray. Librarian opinion did not matter much until the millennials got involved. Then Congress. Suddenly a contested term in American culture wars involved subject headings. You knew this was not going to be a principled argument.

But principles matter. I learned from my mom when I was four years old. “Name-calling is wrong.” People do not like to be called names. And taunting is also wrong. So it seems pretty easy to me, though what was really at stake for me at the time was the possibility for dessert at dinner.

A slightly more sophisticated concept associated with the issue is the distinction made between endonyms and exonyms, that is, a self-designated name for a group of people (or a place, language, etc.) versus a name imposed from without. Berman used many such examples in his critique of LCSH, [1] because they are such rich ground for displaying colonizer hubris and disdain. And of course it is an opportunity to either skewer someone for being insensitive or uninformed, or for being politically correct. Names for people is a semiotic opportunity to show your cultural politics.

So, what are we supposed to do with our subject headings? Preferring endonyms seems like a good principle. We could call this endonym warrant. What could go wrong with it? The possibility that a group calls itself by a name that is generally offensive, a group has a secret name for itself, a group has more than one name for itself, a group that changes its name over time, or a group that is known to itself by one name and to another by a different name are all possible issues, probably resolvable in pratice and would generally avoid using names that the members of a group would find objectionable. Libraries are or should be inclusive places. As should all of our public institutions. The post office does not sell stamps with offensive depictions of people, or should not.

But what if a group is known to its members by more than one name? That regularly occurs because of in-group variation and the simple fact that names change over time in different contexts. We are right back where we started.

Literary Warrant and User Warrant

Another good principle to consider is, of course, our old friend “literary warrant” which most of us know is to use the name that most commonly appears in print. It does provide a relatively easy means for catalogers to determine the name of something or some group, presumably because they are working in a place surrounded by books. If writers and readers are enculturated in the same ways, then they will generally share terminology. Of course that does not always happen, which is how we got into this mess in the first place. Literary warrant is the basis for the construction of LCSH, and it leads to all sorts of problems like old language being used. And why should writers be given the authority to decide? In fact one of the major concerns about the politics of knowledge is that perhaps the world of print has not effectively represented the full range of human knowledge, and favored certain perspectives and treatments over others. IF such is the case, then print will return the same biases by which is constructed. And poses a real conundrum for libraries and others who serve up knowledge in its print-based forms.

Another principle is to use the full legal form of a name. This pertains mostly to individuals, and us old-timers remember when this was the standard approach for establishing personal names under earlier editions of the Anglo-American Cataloging Rules. I often think of this as a restricted sense of literary warrant, except that we restrict ourselves to legal documents to establish the name rather than sources written by the person. As it pertains to *Illegal aliens, Nowrasteh indicates that “illegal” is used as a preceding adjective approximately half the time in Title 8 of the US Code, for example, in the phrase “illegal alien” or “illegal immigrant”.2 The remaining portion is evenly split between “unauthorized” or “undocumented”. Nowrasteh seeks to justify the use of “illegal” and finds its use in the legal code to do so—they are called “Illegal aliens” because that is what they are called in the written law. And what supporters of literary warrant and Nowrasteh (as an implicit supporter of literary legal warrant) fail to consider is that the writers of literature are not naive uninterested promulgators of words. In fact they are quite the opposite. When Nowrasteh cites US law and court opinion, he cites politicians and judges who themselves have explicit politcal agendas who in some cases were explicitly creating both a legal basis for racial or religious discrimination and pejorative language to describe the people subject to those laws.

Indicating that “Illegal alien” is the correct legal term does not make you less political, it merely obscures the political motivations of those that created that category and terminology in the first place. It is an example of the dangers of the policy of literary warrant. What appears to be a fair policy of using the terminology that is most prevalent in source materials becomes explicitly unfair if the source language is discriminatory itself, or becomes associated with discriminatory positions in subsequent usage. (Querying on the term “illegal alien” in Google is a quick lesson on whether that term has become associated with racist political positions, and also indicates what is at stake in the Googlization of web searching). Such problems are why Julian Castro (D-TX) sponsored H.R.3776, the “Correcting Hurtful and Alienating Names in Government Expression (CHANGE) Act.” And that is an example of the way being political is not necessarily wrong. In this case it is ameliorative; to take on the act of changing our legal systems, our social sorting generally, or what we call such persons, whether in everyday language or in our controlled vocabularies. The amelioration aims to restore what was taken, or to provide what is needed, by understanding the needs of undocumented immigrants and providing tailored information services. Assuring reading programs in child detention centers would be a good start. Naturalization services in public libraries is another o=good move.

But the policy of literary warrant carries within itself a way of being political that is consistent with this column—that controlled vocabularies and the catalog are a site of social contestation. Literary warrant begins with the assumption that there is more than one way to refer to a concept, “illegal alien” or “undocumented immigrant”, etc. Unless those terms are randomly distributed across the population—and they almost never are—the premise of literary warrant is to start engaging in politics. Selecting one term or phrase above its alternatives is to engage in politics, and entails supporting one linguistic community over another. Adopting the most commonly used term is itself a political decision, whether the phrase is offensive or not. It disenfranchises linguistic minorities, which, when we do as bad of a job of integrating subject authorities into software as we have done, that disenfranchisement becomes a real burden, and is us neglecting our professional responsibilities of equal access to information. And if the terminology is offensive or discriminatory, as Sanford Berman and others have frequently indicated, we have abdicated our professional responsibilities of respect for users.

And while user warrant may be a good alternative to literary warrant, you can see that it runs the same risks of pejorative language use, the necessity of political adjudication, and the possible necessity of amelioration in order to avoid something offensive. User warrant of course is just the policy of selecting users’ search terminology to represent concepts. And all the same problems pertain there—different communities use the different words, and, if anything, there is less likelihood of moderating editorial review to tamp down offensive language. Google’s autocomplete feature is basically a tool based on user warrant, and we are all familiar with the risks when you enter “why are [name a group] so …” Basing your indexing language on user search terms seems like it would be a good idea for matching user search terms, but in the Internet Age, such a proposition is risky.
Beyond Warrant

If our various concepts of warrant are not delivering the results we want, then what other solutions do we have? Perhaps we need a reconceptualization. Why select one term amongst a set of synonyms? That is a policy developed in the context of a specific technology: the paper-based machine cabinet of catalog cards.3 There, replicating a series of bibliographic records under each of a series of synonyms would be highly inefficient. It does not get rid of offensive terms like Illegal aliens but there is no reason why we would have to choose amongst synonyms. If we replaced concept names with an arbitrary concept identifier, we could maintain a subject naming system separate from a bibliographic file whose job it is to support users in identifying concepts instead of authorized names of concepts. Once a concept or series of concepts is identified, the system would provide the concept identifier(s) and automatically search it in the bibliographic file, in a simple linked data environment. We have been begging for a similar solution for name authority data and for work identifiers since the 1990s. Siegfried and Bernstein reported on the Getty “synoname” project in 1991.[4]

Such an environment would not just separate out the concept naming component of the catalog, it would actually separate out the entire conceptual modelling apparatus, of which naming is just one component. I know there are people and scholars with a particular interest in plurals, word forms, morphological variants, alternative spellings, etc., I think it is one of the less interesting aspects of conceptual model systems, amongst which I include the Encyclopedia Britannicaand Wikipedia. What if, instead of browsing LCSH, users could browse Wikipedia to develop their search queries?

Or something like the Encyclopedia Britannica and Wikipedia. In both systems, the designated concept is Illegal immigration, so they both use the word that people find offensive. But this time in a way that is said of a phenomenon, rather than of a person, and that at least is consistent with the style guide of the Associated Press amongst others. Searching either “Illegal alien” or “Undocumented immigrant” in Wikipedia will return the entry on Illegal immigration, whose first section is dedicated to the terminological controversy.

The politics of naming is just one aspect of a general nomenclature problem involving not only concepts but also personal names. For all of Berman’s and Olson’s insights,[5] the nomenclature problem is just one aspect of politics and bias in our systems. Even if Wikipedia got it perfect, eliminating the problem of bias in naming would not solve all our problems in trying to build a just and inclusive system for accessing knowledge. I will have more to say about conceptual models in future columns.

Works Cited

1.       Sanford Berman, Prejudices and Antipathies: A Tract on the LC Subject Heads Concerning People. (Lanham, MD: Scarecrow, 1971).

2.       Alex Nowrasteh. 2019. “Illegal Alien” Is One of Many Correct Legal Terms for “Illegal Immigrant”. Cato at Liberty. Available at: https://www.cato.org/blog/illegal-alien-one-many-correct-legal-terms-illegal-immigrant(accessed September 4, 2020).

3.       Markus Krajewski. Paper Machines: About Cards & Catalogs, 1548-1929. (Cambridge, MA: MIT Press, 2011).

4.       Susan L. Siegfried and Julie Bernstein. 1991. “Synoname: The Getty’s New Approach to Pattern Matching for Personal Names.” Computers and the Humanities 25 (4): 211-226. http://www.jstor.org/stable/30204541.

5.       Berman, Prejudices and Antipathies; Hope A. Olson, The Power to Name: Locating the Limits of Subject Representation in Libraries.(Dordrecht, The Netherlands: Kluwer Academic Publishers, 2002).