Overcoming language barriers in science

Language barriers have serious consequences in science, causing inequality for under-represented communities, making non-English-language knowledge inaccessible, and impeding the uptake of science by decision-makers.

Tatsuya Amano, from The University of Queensland, Brisbane, Queensland, Australia, Clarissa Rios Rojas from Centre for the Study of Existential Risk, University of Cambridge, Cambridge, UK, Yap Boum from Epicentre, Margarita Calvo from Physiology Department, Pontificia Universidad Católica de Chile, Santiago, Chile and Biswapriya B. Misra, Independent Researcher collaborated to provide a practical checklist as a starting point for tackling this overlooked issue.

The below 10 tips aim to help everyone in science, technology, engineering, and mathematics (STEM) to start tackling and solving this issue.

Researchers shouldn't assume that their science can be communicated via a single language. Scientists typically release their research in English, and in their own language/s, but not always in languages that are potentially relevant. This can lead to an insufficient uptake of scientific knowledge among key decision-makers and the general public. Uploading non-English-language equivalents of papers - now easier than ever to do through many journals and preprint servers - and distributing press releases and summaries for policymakers in a range of languages should assist in bridging this gap.

We live on a diverse planet, and not all important information can necessarily be sourced from a single language. Ignoring science from other languages could bias findings. Collaborating with speakers of multiple languages when searching through local journals and literature databases can help counter this bias. Great examples of this include our translatE project and the Conservation Evidence project. Our team has also developed resources for aiding multilingual literature searches. In cases where comprehensive searches for literature in relevant non-English languages is simply not practical, a valid reason - such as limited financial and human resources - should be disclosed. Authors should be encouraged by their editors and reviewers to identify and cite relevant literature in other languages, where applicable.

The poor international visibility of non-English-language science is an enormous barrier to its use. Unfortunately, many major search engines do not index important peer-reviewed journals published in non-English languages. One option to increase visibility is to prompt authors to provide an English title and abstract of their paper. Alternatively, existing literature could be translated, as has been done in social sciences in Colombia. Publishing non-English papers in machine translation-friendly (e.g. HTML) forms can also assist. English-language journals and databases also have a role to play; for example, Applied Ecology Resources collects scientific knowledge in other languages and supplies English-language summaries, increasing the visibility of otherwise hidden knowledge.

It's becoming harder to find non-English-language translations for new scientific terms, with English terminology becoming default in many languages. This creates barriers for those not fluent in English. This may seem negligible, but can disproportionately affect researchers with lower socioeconomic status, often associated with poorer availability of language education and therefore lower English proficiency. One solution would see multiple authors of original papers creating abstracts in relevant local languages, making an applied effort to translate main keywords into those languages.

The scientific community must offer genuine, meaningful support to help non-native speakers of English overcome language barriers. Many journals encourage the use of English-editing services at the author's expense, but some journals offer free English editorial services. The 2021 joint annual meeting of three societies in evolutionary biology offers free English translation and bilingual mentors for Spanish-language presenters, increasing the diversity of conference attendees. This kind of support is critical on many levels, for example, for students and staff at universities, and for interviewees at job interviews. And this not only applies to English but any other language. For example, funding agencies and academic societies - in countries where English is not common - should try to provide language support, eliminating any disadvantages faced by those who don't sufficiently understand the local language.

As scientists, we should try to improve our writing (and other) communication skills in whichever languages needed for our science to be understood. Less fluent language skills, however, don't necessarily equate to poorer quality science. Automatic linking of these skills, sadly, still happens; science written in “non-native-like” English is often rated lower than what's written in “native-like” English. Scientific communities must make a conscious effort to decouple language skills and scientific quality when working with colleagues and collaborators, reviewing papers submitted to English-language journals, and assessing grant proposals or student/job applications. For example, journals could emphasise in their instructions to authors and reviewers that their decisions are purely based on the quality of science, not the linguistic fluency, of papers. The need to improve the English writing can still be requested, but in a less disparaging way (e.g., do not demand blanket editing by “native English speakers”).

It's often a great idea to increase the diversity of contributors in any scientific activity. Higher diversity in language background, although often overlooked, is usually connected with higher cultural and ethnic diversities, which helps facilitate higher scientific productivity and more balanced decision-making. When picking plenary speakers for a conference, inviting new members to a journal editorial board, or recruiting a new employee, it's important to consider the balance in language background, as well as other types of diversities, such as gender, ethnicity, geographical location, and career stage, and make a conscious effort to involve non-native English speakers. Also, importantly, just involving a linguistically diverse group of contributors is often not sufficient, as the voices of native English speakers are more likely to be heard. Providing support to non-native speakers (e.g., offering English interpretation services to keynote speakers at conferences) would be effective for maximising the benefits they bring with their perspectives.

Actions and efforts taken to overcome language barriers are often ignored in the researcher performance evaluation. For example, peer-reviewed papers, translations of books and media coverage in non-English languages are typically not acknowledged as important outcomes. We need an institutional change to encourage and value those efforts, for example, when evaluating dissemination activities. It's also important to take into account the considerable disadvantage non-native English speakers are likely to have experienced during their career developments when assessing their scientific performance, for example, at job interviews, grant reviews, and employee appraisals.

When talking with non-native English speakers, or assessing their science, be tolerant of any linguistic mistakes, non-native-like English usage, clumsiness or slowness. As a group of non-native English speakers, we know (from experience) that a generous attitude makes a huge difference in relieving nervousness in the speaker and ultimately leads to more productive conversations. Maybe try experiencing some scientific activities from the perspective of non-native speakers. For example, if your first language is English, learn a new language relevant to your work, and try to give a presentation, or write a blog post in that language, and imagine having to do this throughout your academic career. This will not only help disseminate your work more broadly but you will also immediately understand the challenges and difficulties non-native speakers go through when surviving in academia or in industry.

Despite this issue of language barriers not yet being resolved sufficiently, this issue in academia certainly has come a long way. This is thanks to a variety of useful online resources, such as machine translations and sites providing pronunciation solutions for a number of languages. Mentorship programmes also help improve writing and communication skills and obtain general career advice via mentors who speak local languages. We have created a list of resources and opportunities for overcoming language barriers in science. We intend to update the list at least once a year and always welcome suggestions.

Although the authors point out the advantages of English as a lingua franca in science, they consider that the ten tips above showcase what we believe is an ideal environment for academia to overcome language barriers; that is: the latest research is available in multiple languages, science is widely utilised regardless of the language in which it was discovered, non-native speakers have access to sufficient support throughout their career, and everyone is considerate of those communicating in a non-mother-tongue language.

Some of the suggested practices would require extra effort, resources and cost. However, in the long run, it would be a considerable gain for the scientific community if we can indeed realise the above, as the wealth of talent and knowledge currently buried due to language barriers could, once released, boost the advancement of science and its application to solving global challenges. We hope that these tips will help raise awareness and prompt actions in academia and beyond.