Does language limit scientific expression?

Scientific papers

Fancy/Punchstock/Getty Images

This is a guest blogpost by Aya Nader.

More evidence is confirming that the choice of language used in scientific literature can influence access to it, and how visible its authors are – including in the Arab world.

Language can limit the transfer of knowledge for one, concludes a study that looked into the prevalence of scientific literature written in local languages. The study, published in PLOS Biology, confirmed some sentiments that many researchers across the Arab world already have.

Over one third of conservation-related scientific documents are written in non-English languages, and a large proportion of local researchers interviewed in the study identified languages as a barrier to accessing knowledge. “I was expecting to see these results, as that was the primary motivation to conduct this work,” says Tatsuya Amano, corresponding author.

Amano says that gaps in information are formed when local scientists either do not get exposed or turn away from publishing in their original language. What surprised the researcher was that over one third of non-English literature reviewed in the study provided neither the title nor the abstract in English, so it’s essentially “invisible to international communities”.

The study might explain why Arab scientists are not as visible, in terms of science research, to international peers, he opines.

“Perhaps only 25% of the global population has some understanding of English and we cannot limit science to just a fraction of the world,” says Steve Griffiths, vice president for research at Masdar Institute of Science and Technology. According to him, having scientific knowledge being somewhat confined to the English language can present a problem when collecting scientific data or disseminating information.

“While language is probably not the driving force behind the lag in scientific visibility of Arab scientists, it certainly can hinder progress,” Griffiths says. Different factors could be causing the lag, he says, which include that the region has only been recently making strides in establishing top-tier research universities and institutes. As well, regional equivalents of supportive bodies like the US National Science Foundation or the US National Institutes of Health are absent.

One of the barriers could be the language itself. A few argue that Arabic, because of the way it’s structured, cannot be adopted as a language of science. “I personally am fluent in English and have studied Arabic for some time and clearly see the translation challenges for technical information,” says Griffiths.

On one hand, English is the universal language of science. On the other, having science available in the local language can enlighten field practitioners and local policy makers.

“The availability of scientific information in relevant non-English languages is a key to the use of science in policy making in countries where English is not widely spoken,” comments Amano. It’s one factor contributing to the divide between science and public policy. “I imagine that extremely busy policy makers would prefer just using easily-accessible information in their own languages, instead of trying to understand English-written papers.”

Poor English skills are observed in many MENA countries and particularly within the government sectors, which limits the uptake of scientific information, Griffiths highlights.

In order to compile non-English scientific knowledge effectively and enhance publishing of new and existing knowledge that is otherwise available only in English, Griffiths suggests launching regional initiatives modeled after the MIT Global System for Sustainable Development. The networking hub, specialized in sustainable development, was created to give researchers that speak English, Arabic, Chinese and Spanish seamless access to its science content.

Another approach is to encourage individual researchers to translate their work, or provide lay summaries of their work in different languages.

There’s also hope in artificial intelligence (AI) for natural language processing (NLP). “Major industry players like Google, Microsoft, Amazon and IBM are deeply engaged in AI NLP for commercial reasons, and over time the outcomes will benefit the scientific community,” Griffiths suggests.