[Image: iStock / Getty Images Plus]
Today, the lingua franca of science is English. Having a common language among scientists of all origins eases international communication and collaboration and helps scientists whose primary language is not widely spoken get their work recognized by a wider audience in other regions. Despite these advantages, having a single universal language for science also hinders non-native English speakers, as they are required to master the language before they can present their work on an international stage.
A recently published paper addresses the cost of having one common language in science and proposes practical methods for breaking down language barriers (BioScience, doi: doi.org/10.1093/biosci/biac062). The set of approaches, suggested for individuals, academic societies and institutions, ranges from machine translation tools for written work to structural changes that could support “a multilingual academy.”
Science in English
English hasn’t always been the primary language of science. Its linguistic centrality began about 400 years ago with the expansion of the British colonial empire. Before English gained such a stronghold, scientific exchange, broadly construed, happened in a wide variety of languages in different times and places.
But today, 98% of peer-reviewed scientific publications are written in English, according to previous studies, and English is the official language of most scientific conferences. And while having a unified language helps scientists communicate across borders, the authors of the new study argue that the one-language approach could also harm scientists, science and global society at large.
Drawbacks of monolingualism
In the paper, the authors—including scientists from the University of California, Berkeley (UC Berkeley), USA, the Center for Ecological Research, Vácrátót, Hungary, Ben-Gurion University of the Negev, Israel, and the University of Ottawa, Canada—identified several consequences of communicating science using a single language. One is that scientists and others often overlook work written in other languages, creating knowledge gaps.
In addition, the team argues, English proficiency has become a barometer for who gets to participate in science at a global scale. With language ability front and center, it’s harder for scientists who are less fluent in English to publish in high-impact journals, receive grants and present their results at conferences. Having to take English lessons and pay for proofreading and translation services could also financially burden them, the authors observe.
All in all, such “scientific monolingualism” hinders the exchange of diverse perspectives, impeding the construction of robust and innovative scientific knowledge, the authors argue. Having to express diverse viewpoints in a single language could alter those very viewpoints, as language shapes, for example, how humans perceive color and remember previous events. Scientific monolingualism also interferes with sharing new information in non-English-speaking communities.
“Science doesn’t need to be based on a single language,” Tarvin said. “And there’s lots of additional benefits that come from incorporating multilingual approaches in every phase of science.”
Translating research
To begin to address the long-standing language issue in science, the Department of Integrative Biology at the UC Berkeley created a course called “Breaking Language Barriers in Evolution and Ecology” in 2021. During the semester, students who take the course translate a scientific paper into a second language or into a different format that communicates the science to a broader audience.
Teaching this class inspired Rebecca Tarvin, a professor in the department, and colleagues to evaluate machine translation tools—such as Google Translate, DeepL, Baidu Translate, Naver Papago and Yandex—and highlight use of the tools as a short-term action that people in the scientific community can take to realize multilingual science.
Going further, Tarvin and colleagues argue that, to make research more accessible not only to scientists but to others in society, translating paper abstracts into languages other than English should become a standard practice. Preparing a plain-language summary can help with this effort, as such a summary helps an English-speaking lay audience understand the work and is easier for machine translation tools—and humans—to translate than jargon-heavy abstracts.
Further steps toward multilingual science
Most journals currently only accept articles written in English, and few explicitly permit publishing articles in non-English languages. Many publishers also lack translation policies, and because of copyright restrictions, they charge fees for issuing translations of scientific work after its initial publication.
However, the authors believe that journals and academic societies have the power to normalize publishing scientific articles in multiple languages. They can set clear guidelines on translating articles, citing translated articles and searching for work written in languages other than English, among other strategies. The authors also suggest that societies could provide free translation services or develop proofreading networks.
Universities can also play a role in tearing down language barriers by recognizing that publication of scientific work in non-English and regional journals is important for disseminating knowledge. The institutions can reflect this when evaluating tenures and promotions, contract renewals and degree requirements.
“Science doesn't need to be based on a single language,” Tarvin said in a press release accompanying the paper. “And there's lots of additional benefits that come from incorporating multilingual approaches in every phase of science.” The paper is also available in Chinese, French, Hungarian, Portuguese and Spanish.