Forums

XXVIII.3 May - June 2021
Page: 66
Digital Citation

How the ideology of monolingualism drives us to monolingual interaction


Authors:
Manuel Pérez-Quiñones, Consuelo Salas

back to top 

When our languages are internal, we don't know where one ends and where one begins.
    — Ofelia Garcia

back to top  Insights

ins01.gif

Most user interface designs are influenced by an ideology of monolingualism and therefore support only monolingual interactions. Monolingualism is the condition of being able to interact in only one language. The ideology of monolingualism, however, is the notion that communication happens through and by only one language at a time. A more honest examination of everyday linguistic practices reveals that even if one is familiar with one language group only—English, for example—they have fluency in multiple dialects of that language—academic, professional, and so on. Furthermore, a significant portion of the population of the U.S. and the world is bilingual or multilingual.

According to François Grosjean [1], about 60 percent of the world population uses more than one language. The number is even higher in Europe, where bilinguals account for over 50 percent of the population. According to U.S. census data, around 20 percent of the population speaks more than one language; that percentage has been increasing steadily in the past three decades.

While it is clear that linguistic diversity is the norm, not the exception, the pervasiveness of the ideology of monolingualism is such that it is present in all aspects of our daily lives—with negative consequences. Within the context of the U.S., this ideology helps prop up English-first and English-only sentiments, the results of which are damaging to the language development and intelligence assessment of multilingual school-age children as well as to those who do not have English as a dominant language (see [2,3]).

Within the design and implementation of user interfaces, the ideology of monolingualism not only neglects to account for the large portion of the population that is bilingual and multilingual; but also, by building monolingual interfaces, designers disregard the nuances of linguistic diversity and ignore the bilingual individual as a user class. Instead, designers assume that all users are monolingual and that bilingual users are comfortable communicating in one monolingual interface at a time. Through its implementation, this ideology at best glosses over the complexities of communicating in more than one language at a time, and at worst is an agent of purporting a standard language ideology. Standard language ideology is "a bias toward an abstract, idealized homogeneous language, which is imposed and maintained by dominant institutions and which has as its model the written language, but which is drawn primarily from the spoken language of the upper middle class" [2]. Standard language ideologies are harmful and damaging.

Perhaps unintentionally, monolingual interfaces fail to accommodate how multilingual users communicate with each other and interact with information in multiple languages. Bilinguals do not flip a switch and change language use; we use all of the languages available fluidly, often mixing words, phrases, and idioms. The idea that we communicate in a compartmentalized fashion, with just one language at a time, is not just wrong; it also reinforces the ideology of English first and the standardization of language. It is based on a mistaken notion that monolingualism is the norm, which has severe negative consequences for those who do not fit the monolingual mold. For example, bilingual individuals who speak with an accent may be perceived as lacking in intelligence simply for how they sound or their pronunciation of words. Additionally, bilingual students who undergo standardized testing may not do as well as others simply because of the language of the test, not because they lack knowledge of the concepts.

A quick look at the literature in human-computer interaction might explain why this monolingual interaction has been the norm. Internationalization and localization is the process by which we build interfaces that must address regional differences. It invariably creates interfaces that are either modified to be international or adapted to be more appropriate for a local (regional) context. At the center of both of these processes is the simplistic notion that region and country dictate language. The result is that design practice completely ignores bilingual users, and even users who use one language while living in a different region. For example, consider a user who can read English, Spanish, and Portuguese and lives in the U.S. If the interface used is localized for the U.S., it will invariably be in English. As a result, the ability of the user to read news stories and interact with others in more than one language is restricted. This user would have to ping-pong between keyboard, language, and system settings to be able to use multiple languages. And even after all of those settings change, the interaction will still be mostly monolingual.

ins02.gif

It is clear that linguistic and cultural diversity is already present in our communities; however, interface designs have long treated human languages as if they were an indicator of being international. In some cable systems, if we select Spanish as the language of interaction, you get offers for world packages. In most desktop systems, the multilingual setup is part of the international features of the OS (Figure 1).

Top: Message from Cable company: "Multilingual? Watch your favorite TV from around the globe with a WorldDirect package." Bottom: MacOS Installation instruction showing "International Features" with a description "New multiligual setup for macOS. Choose your languages, including those for keyboard and dictation, in Setup Assistant, and customize your language preferences from the start." Figure 1. On the top, offers from a cable service where being multilingual generates offers for the WorldDirect package. On the bottom, instructions for the macOS installation gives options for language selection under "International Features."

It is time we separate bilingualism from internationalization—the idea that people can use multiple languages is now common. Global migration patterns spread languages and cultures across the world, and businesses have negotiations in more global markets. Electronic communication makes global information exchange more feasible. Within these globally distributed communities, communication across languages is more prevalent today than ever before. Yet, user interfaces insist that we use just one language.

Figure 2 shows one of many examples of how bilinguals change language (e.g., code switching) in the middle of a message. While the change follows some well-defined patterns, it often confuses spell checkers and auto-complete services, frequently rendering these services useless for bilingual users. Bilinguals are more likely to forgo using these services than to have to negotiate language use with a monolingual interface.

Bilingual communication via text between child and parent. Message reads "mom is making arroz con salchichas with chicken patties." Arroz con salchichas (Rice with Vienna Sausages, a staple of Puerto Rican cooking) is displayed with red lines indicating spell-checking errors. Figure 2. Bilingual communication at home via text between child and parent. Arroz con salchichas means Rice with Vienna sausages, a staple of Puerto Rican cooking. Note the red lines indicating spelling errors.

Language acceptance also happens in the display of character sets that are foreign from an English first and English only point of view. Figure 3 shows what happens when characters from the Spanish language, in this case é and ñ, are displayed in the inappropriate character set.

Top: We'd love to know what you think! Note addressed to one of the authors, Professor P?rez Qui?ones but characters with diacritical marks shown garbled. Bottom: Header from US Congress Representative addressed to P?rez-Padilla but accented-e in P?rez is displayed in Unicode Hex form é Figure 3. Using the wrong character set in the interface often yields to strange characters for names with diacritical marks. Top image is from a survey sent after service to one of the authors, meant to display Pérez Quiñones. The bottom is from the header of a letter from a congressman to Pérez-Padilla.

Translation services are improving but are being used in poorly chosen ways, providing further evidence of the ideology of monolingualism. Look at the next two examples. The first is a Web search for a Mexican comedian, Chespirito (Figure 4). Note that the browser language is set to English and the search results are in English, yet the browser somehow decides to offer to translate the page. What are the triggers that make such an interaction needed?

Screenshot of Google search result after searching "Roberto G?mez Bola?os," a well known Mexican comedian. The browser offers to translate content though results shown are in English. Figure 4. Searching for an element that is typical of Spanish-language culture—in this example, Roberto Gómez Bolaños, also known as Chespirito—triggers the browser to offer a translation. It's hard to see in the image, but the first links shown are from the Wikipedia page and other sites that are in English. Interface and browser are set to English.

The second example is even more puzzling. Figure 5 shows the first page of a slide deck from a computer graphics lecture taught by one of the authors. The only thing in Spanish is the name of the author. Yet, the browser offers to "Translate this page." Note that this happens on the first page of the slide deck at the beginning of every class.

Offer to translate a page that has text in English (computer, graphics, january) and the author's name (which includes diacritical marks) Figure 5. Offer to translate a page that has text in English (computer, graphics, January) and the author's name (which includes diacritical marks).

Most news services provide the ability to select a language and region. Google News allows the user to see news in English for the U.S. or in Español for Estados Unidos (Figure 6). But for Spanish/English bilinguals living in the U.S., why not see both at the same time? Again, the ideology of monolingualism is designed directly into the interface.

Google News Spanish interface showing choice of language (Spanish vs English) for news from the United States. No option to see news in BOTH languages at the same time. Figure 6. Google News options to select language and region of interest for the display of news. The choice is between language (Spanish versus English) for the same region. There is no option to see news for the U.S. in both Spanish and English.

Over the past three decades, scholars within linguistics, literacy, rhetoric, and communication have studied the nuances of linguistic diversity, acknowledging the rapidly diversifying modes of communication with increased technological developments and globalization. In response, pedagogical approaches known as multiliteracies, which value and are more inclusive of cultural and linguistic diversity in communication, have been created and implemented.

The linguistic choices and maneuvers made by bilingual people are sometimes called code switching, code meshing, or translation. More recently, the term translanguaging has been used to define or describe these communication practices. Bilinguals use two languages in their day-to-day communications. They combine the use of the two languages in what might appear to be a random manner, but the reality is that the combination of the two languages follows well-understood linguistic patterns. To the uninformed, these might appear to be poor communications skills, but research in linguistics and education has shown that some of this fluid language switching is evidence of higher degrees of language acquisition. While monolinguals are "often oblivious to the presence of these bilingual practices … they reflect greater choices, and a wider range of expression … and convey not only linguistic knowledge, but also combined cultural knowledge that comes to bear upon language use" [4]. Translanguagings are the "multiple discursive practices in which bilinguals engage in order to make sense of their bilingual worlds" (emphasis in original) [4]. In short, translanguagings are the practices that allow bilingual individuals the ability to communicate across their "languaging continuum." In addition to the linguistic practice of bilingual people, translanguaging is also seen as a research practice that studies "the language practices of bilinguals from their own dynamic perspective, rather than from the static monoglossic one of monolinguals, and examines the ways in which those resources are deployed in teaching and learning" [5].

In her monograph Sites of Translation [6], Laura Gonzales made visible the concept of translation moments. Through her research, Gonzales identified the moments that users of multiple languages paused to consider the context, situation, and use of a term in one language to find the most appropriate translation for it in another language. Building on this work, we propose the concept of bilingual moments, analogous to translation moments, as the instances when the user is at a crossroads about engaging a particular language for interaction. When bilingual users are forced to engage in monolingual interfaces, we constantly have to make choices of what language and form to engage with, all while considering the audience of our message and the affordances and limitations of the technology through which we are communicating. For example, when responding to something that we find funny, we may respond to one group with "ha ha" and to another group as "ja ja," but in either case we have to face the option that autocorrect is possibly set on the wrong language. To avoid this, we may just stick with "ha ha" to both groups, in effect erasing our bilingual communication practices.

back to top  Call to Action

We hope to have conveyed the idea that we need to design interfaces to support bilingual interactions and bilingual users differently from just any user with two modalities of use.

While this idea of a bilingual interface might be new to some, we already have some examples to follow. Emojis and emoticons are a form of multiple literacy communication. Today it is common to allow alphabetic text to be mixed with graphical images. Spell checkers and autocorrect understand that between words that might exist in a dictionary, there are also these graphical glyphs that are part of the message. These are interfaces that facilitate translingual communication. We need to continue to expand this idea by allowing multiple languages in the messages. We would like the ñ and é to be treated as valid characters, just as any other emoji like intr2803_a.gif or intr2803_b.gif. In the technology domain, bilingual interfaces are also extremely important to help dispel the notion that monolingual interfaces are the norm. This, in our opinion, is essential to building an inclusive society.

Within the HCI community, we need to consider that an attribute that impacts user experience is the language of use. Bilingual users who are forced to use an interface in only one language are not going to use the full range of communication capabilities innate to them. We should consider the language of interaction as part of a UX evaluation. Maybe we even need to consider bilinguals as a separate user persona for evaluation purposes. Experience tells us that auto-correct, spell checkers, and translation services might work great for monolingual individuals but are often turned off by bilinguals because we find them extremely annoying, unhelpful, and disruptive in communication patterns.

From research in multilingual communication, we have learned about translation moments. We call the equivalent in user interaction bilingual moments, but what would the resulting bilingual interaction look like? We know that language switching in bilingual communication is much more fluid than "Change settings in the UI." Can we allow language change at a micro level (within each interaction) and more fluidly than a preference hiding in a settings dialog? These are research questions to explore for both basic human-computer interaction as well as in design considerations in user-centered design.

Finally, by continuing to build monolingual interfaces, we are truly missing an opportunity to be a part of the force that brings global communities together. Bilingual users are the bridges between these multiple communities. Allowing them to experience the full range of information from two somewhat disparate worlds will allow them to be the bridge or weak link between the multiple communities. And allowing them to communicate with the full range of language abilities they possess would make the interface more natural.

Bilingual users live in a liminal space, inhabiting two worlds, that technology is forcing apart. If (when?) we create multilingual interfaces, imagine the potential impact for bilinguals in valuing linguistic diversity. Bilingual interaction will go a long way toward eliminating the ideology of monolingualism that technology is imposing on the world today, and instead value and support the linguistic practices already occurring in a large portion of users, not only in the U.S. but also globally.

back to top  References

1. Grosjean, F. Bilingual: Life and Reality. Harvard Univ. Press, 2012.

2. Green, R.L. Accent, standard language ideology, and discriminatory pretexts in the courts. Language in Society 23, 2 (Jun. 1994), 163–198; https://www.jstor.org/stable/4168513

3. Garcia, O. and Otheguy, R. Plurilingualism and translanguaging: Commonalities and divergences. International Journal of Bilingual Education and Bilingualism 23, 1 (2020), 17–35. DOI:10.1080/13670050.2019.1598932

4. Garcia, O. Bilingual Education in the 21st Century: A Global Perspective. Wiley, MA, 2009.

5. Garcia, O. and Wei, Li. Translanguaging: Language, Bilingualism and Education. Palgrave Macmillan, Baskingtoke, U.K., 2014. DOI: 10.1057/9781137385765

6. Gonzales, L. Sites of Translation: What Multinguals Can Teach Us About Digital Writing and Rhetoric. Univ. of Michigan Press, 2018.

back to top  Authors

Manuel A. Pérez-Quiñones is professor of software and information systems at the University of North Carolina at Charlotte. His research interests include personal information management, human-computer interaction, CS education, and diversity issues in computing. He is originally from San Juan, Puerto Rico. perez.quinones@uncc.edu

Consuelo Carr Salas is an assistant professor of border rhetorics in the Department of Rhetoric and Writing Studies at San Diego State University. Her areas of interest include visual rhetoric, food studies, the commodification of Mexican and Mexican Americans, information literacy, and the intersections of translanguaging and monolingual technological interfaces. csalas@sdsu.edu

back to top 

©2021 ACM  1072-5520/21/05  $15.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2021 ACM, Inc.

Post Comment


No Comments Found