The language barrier in scientific communication – 12/09/2023 – Science

A team of researchers from ten countries analyzed the publication policies of 736 scientific journals in biological sciences and identified barriers imposed on authors who do not speak English as their native language.

Some of these obstacles are well known. As English is the lingua franca of science, it is not uncommon for journal editors to suggest that authors hire the services of companies specialized in editing or translating scientific texts to ensure that the content of a manuscript is expressed clearly and follows the English standard — and this increases publishing costs, mainly harming authors from poor countries.

There are also less tangible barriers. Of the 736 journals analyzed, only two of them, Nature and Nature Plants, categorically stated in their guidelines that a paper will not be rejected just because the author does not express himself satisfactorily in English — the essential thing is to evaluate the relevance and quality of its content .

In addition to analyzing publication rules, the group interviewed editors-in-chief of 262 of these journals — and only 6% of them instructed reviewers not to preliminarily reject articles in English with problems with grammar, clarity and fluidity. About half of editors suggested authors use free online English editing services for grammar correction or referred them to online tutorials. Only 1% of journals offered assistance through free mentoring programs.

A novelty of the study, published in the EcoEvoRxiv repository and not yet peer-reviewed, was to point out examples of journals that provide support to researchers whose manuscripts require language adjustments. Journals maintained by scientific societies tend to be more inclusive. The Society for the Study of Evolution, based in the United States, has a mentoring program in English to support those who submit work to Evolution magazine. For free, editors with experience in scientific writing can suggest slight changes and discuss directly with the authors ways to make the manuscript clearer. Authors can request support before submitting the article or during the review process.

Another example is the Journal of Field Ornithology, linked to the Association of Field Ornithologists, a North American scientific organization that connects authors who do not have English as their primary language with volunteers who can help them improve their texts.

The American Society of Mammalogists has created a free partnership scheme, the ASM Buddy System, in which mammalian zoologists help improve the English level of manuscripts, while the British Ecological Society’s Journal of Ecology offers a free proofreading service by artificial intelligence, Writefull.

Support for authors is well received. Germana Barata, a researcher at the Laboratory for Advanced Journalism Studies at the State University of Campinas (Unicamp), told Nature magazine that she had a good experience publishing in English in the journal Cultures of Science. “The corrections and edits didn’t change my ideas, the essence of what I was writing, or the style at all,” she says. “That doesn’t happen in many other publications.”

The study proposes a set of actions to address language barriers, such as a public commitment by journals to fairly evaluate the content of papers, even those with writing problems, or the offer of free services to make articles with written content more readable in English. relevant content.

“Our work concluded that journals can fulfill a dual role. At the same time as they are a source of language barriers, they can also help authors face them,” said Brazilian biologist Pedro Albuquerque Sena, technical coordinator at the Center for Environmental Research do Nordeste (Cepan), a private research institution based in Recife, one of the study’s co-authors.

Sena is part of a community of ecologists connected by X, the old Twitter, which usually exposes and discusses embarrassments suffered by researchers from developing countries when submitting their papers to high-impact journals.

Complaints include possible linguistic errors, but also the feeling that there is injustice and discrimination in the rigor with which reviewers disqualify studies as poorly written in English or unoriginal.

On the social network, Sena found an invitation for researchers interested in surveying and analyzing the publication policies of ecology journals and joined this project, together with colleagues from countries such as Australia, the United Kingdom, the United States, Indonesia and the Czech Republic.

The team was assembled by the study’s lead author, Japanese biologist Tatsuya Amano, a researcher at the Center for Biodiversity and Conservation Science at the University of Queensland in Brisbane, Australia.

Amano is interested in the topic of linguistic bias because he feels affected by it. He left Japan in 2011 to work in the United Kingdom and then in Australia, where he was challenged to produce science exclusively in English. Even today he says he has difficulty writing articles, preparing lectures and, in particular, giving presentations at conferences in English.

“It takes me a lot of time and I need to dedicate a lot of effort to do everything in English”, he said in an interview with the Australian Broadcasting Corporation, an Australian public broadcasting company. “Language barriers create anxiety, discomfort, embarrassment. You have to be courageous to face them.”

In another work published in July in the journal PLOS Biology, Amano and collaborators interviewed 908 researchers in the field of environmental sciences. Those who did not have English as their first language took up to twice as long as native speakers to prepare articles or presentations in English. They also found a 2.5 times greater chance of having work rejected by journals and a 12.5 times greater chance of being asked to make revisions before publication. Due to a lack of confidence in communicating in English, a third of them said they had already given up going to international conferences — among those who participated, half reported having avoided giving oral presentations.

The damage is not limited to the mental health of researchers. The main line of research developed by Amano seeks to find data to fill gaps in biodiversity knowledge in order to improve conservation strategies. He highlights that this knowledge is produced in several languages ​​and that current publishing practices end up blocking its dissemination. “The planet needs the contribution of all scientists, regardless of their fluency in English,” he says.

Sigmar de Mello Rode, researcher at the Universidade Estadual Paulista (Unesp) and president of the Brazilian Association of Scientific Editors (Abec-Brasil), states that Brazilian researchers and the country’s journals have been challenged, in recent years, to publish more and more in English and had to create strategies to deal with linguistic difficulties.

“Whether we like it or not, English has become the language of science and disseminating scientific results in this language is mandatory. If the target audience for a research is in Brazil and Latin America, the solution has been to publish versions also in Portuguese and Spanish” , explains.

He says that scientific journals in Brazil have adopted strategies similar to those of international journals to guarantee the linguistic quality of the content. “Many titles ask authors to send their articles to specialized English proofreading services and they provide certificates that the text complies with the language,” he explains.

Magazines even recommend specific companies, such as Enago, which charges from US$90, the equivalent of R$456, to correct the grammar and style of a 1,500-word text. Publishers like Springer Nature have their own services. An article with 1,500 words costs US$243 (R$1,230) for the silver service, with review by an editor specialized in the manuscript’s theme and the issuance of a certificate, and US$484 (R$2,450) for the gold service, which provides adaptations to give the text a more professional and natural style.

Rode says that, in his experience as a researcher, he saw signs of prejudice towards science produced in developing countries in criticisms of the quality of the English in the manuscripts. Years ago, he did a test: he submitted an article he had written in English to an international magazine that had been reviewed by a professional service. He purposely did not attach the review certificate.

“The article was returned immediately with a warning that there were clarity and grammar problems and needed revision. I questioned the editor: he said that the manuscript had been revised, yes, I sent the certificate and asked him to point out where he had found problems, to that I could speak to the company and ask for the US$ 120 I had spent back. The editor apologized, said it was a mistake and that the review wasn’t really necessary”, he says.

Biologist Marcia Triunfol, who previously worked as an editor for Science magazine and is an expert in scientific writing, sees certain exaggerations in the way in which proficiency in English is demanded of Brazilian authors.

“Once, in a workshop I organized in São Paulo, a researcher of North American origin said that, until Brazilians learned the difference between ‘show’ and ‘demonstrate’, they would not be able to write good articles in English. I was shocked, because I don’t see it as important to assess whether an article has quality”, says the biologist, who is the founder of Publicase, a company that since 2007 has offered translation and article review services and organizes workshops and training courses to guide researchers and students in scientific writing techniques.

Currently living in Portugal, the researcher does not see this type of demand or concern about the problem among European authors who have English as a second language.

For her, the most complex challenge is not so much promoting proficiency in English, which has been improving in recent years, but training students and young researchers in scientific writing. “During the pandemic, we held virtual training via Zoom for postdoctoral fellows at Harvard University. The doubts regarding how to write a scientific article were the same as in the workshops with Brazilian researchers.”

According to her, the training offered by universities is generally improvised: “Ideally, there would be professors specialized in scientific communication techniques and not just researchers who share their experience with students.”

Triunfol believes that the problem will soon no longer have its current appearance. For her, translation and editing tools based on artificial intelligence could play a central role in adjusting the writing of non-English speaking researchers.

“In recent years, with the introduction of translation aid software, I have noticed that the quality of scientific writing has improved. More recently, due to ChatGPT, I noticed a drop in demand for translation and proofreading services in my company, which I consider positive.”

The publisher says that it is possible to use artificial intelligence to correct scientific writing in an ethical way. “As the objective is to correct and improve a text already written by the researcher himself, there is practically no danger that these tools will create situations of plagiarism”, she states.