A first-of-its-kind study has quantified the additional burdens shouldered by researchers whose primary language is not English. The analysis, published in the journal Plos Biology, suggests that the issue is largely overlooked in efforts to diversify the scientific workforce.

Lead author Tatsuya Amano, a conservation scientist at the University of Queensland, said the problem could be costing academia dearly. “We are potentially losing a huge contribution to science from a massive number of people, simply because their first language isn’t English,” he said.

The research team surveyed more than 900 environmental scientists on the time required for routine academic tasks such as reading, writing and rewriting journal articles. The subjects, who had all published at least one first-authored, peer-reviewed paper in English, came from eight countries with varying levels of English language proficiency and income.

The study uncovered “profound disadvantages” for non-native English speakers. Typically, respondents from nations with moderate English proficiency spent 47 per cent more time than native speakers reading scientific literature, while their low-proficiency counterparts invested 91 per cent more time – excess effort that persisted throughout their careers.

Non-native speakers reported similar time penalties in preparing conference presentations. They spent up to 51 per cent more time than native speakers in writing journal articles and sought proofreading help around 50 per cent more often.

They were almost three times as likely to have papers rejected for linguistic reasons, and more than a dozen times as likely to be asked to rewrite papers.

Many respondents said they sidestepped conferences altogether, apparently because of language issues rather than cost. Three in 10 early-career respondents from Japan and Spain said they customarily avoided English-language conferences.

As a non-native speaker in an English language environment, Dr Amano has experienced all this first-hand. “It just takes time to do everything,” he said. “That kind of disadvantage is never accounted for by institutions. It’s something we need to change.”

He said the “struggle” occurred in many contexts, from staffroom conversations to conferences, where native English speakers butted in while non-native speakers were composing questions in their heads. His team had only been able to measure the extra “effort” required from researchers like him – not the insecurity, embarrassment or anxiety: “We couldn’t quantify these impacts on mental stress.”

Counterintuitively, respondents from countries with moderate English proficiency or high-income levels reported the greatest disadvantages on some measures. This could be an indication of “survivorship bias”, with only the most resilient scientists from poorer or low-proficiency nations surviving in academia long enough to complete the survey, the researchers speculated.

Dr Amano said potential solutions included liberal approaches to the use of artificial intelligence in academic work, financial support from research grant agencies and free editing assistance from journals. Some publishers had already integrated an AI proofreading app in their submission systems, he noted, while the journal Evolution had initiated a bilingual mentoring service for authors.

Dr Amano said English fluency had long functioned as an entry ticket into academia. “Anyone in any part of the world should be able to participate in science and contribute to accumulating humanity’s knowledge,” he said.

[email protected]