Language and geopolitical distance across multiple margins of international scientific collaboration
This paper analyzes English-language journal articles from 2018 to 2025 to demonstrate that while linguistic distance consistently hinders international scientific collaboration across all network margins, geopolitical distance primarily influences partner allocation, entry, and tie intensity, with the two factors showing substantial overlap except regarding collaboration exit.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Science has always been a global conversation, but the way researchers connect across borders is changing. For decades, we have known that geography matters; it is harder to work with someone on the other side of the world than with a neighbor. We also know that language acts as a bridge or a barrier, shaping who talks to whom. More recently, the political climate between nations has become a powerful force, capable of freezing cooperation even when scientists want to work together. The question for modern researchers is not just whether these factors matter, but exactly how they shape the daily reality of scientific work. Do language barriers simply stop new partnerships from forming, or do they also weaken existing ones? Does political tension only stop new projects, or does it also cause old, successful collaborations to fall apart? Understanding the specific mechanics of these breaks and bonds is crucial for anyone trying to manage or support international science.
A team of researchers set out to map these invisible forces by looking at the actual papers scientists write. They analyzed millions of English-language journal articles published between 2018 and 2025 in the fields of economics, mathematics, and medicine. Instead of just counting how many papers two countries wrote together, they broke the data down into six different ways a scientific relationship can change. They looked at whether a country was choosing to work with a specific partner more or less often, the total number of papers produced, whether a partnership was still active, whether a new partnership had started, whether an old one had ended, and how intense the work was within partnerships that were still going. By tracking these specific moments of change across 33 countries and their 173 partners, the researchers could see exactly where language and politics were pulling the strings.
The study found that language distance acts like a steady, quiet drag on almost every aspect of scientific collaboration. When two countries have very different languages, the researchers observed a consistent pattern: new partnerships were less likely to start, existing ones were more likely to end, and the volume of work within active partnerships tended to shrink. This effect was visible across all six measures of collaboration. Whether the scientists were trying to form a new team, keep an old one alive, or simply write more papers together, the difficulty of communicating in different languages made the process harder. The data showed that this was a stable, reliable pattern that held true regardless of how the researchers counted the papers or which specific countries were involved.
Political distance, however, told a different and more complex story. While language barriers seemed to affect every stage of a partnership, political distance showed its influence in more specific ways. The researchers found that political differences were strongly linked to whether countries chose to start working together in the first place and how intensely they worked once they were partnered. However, the evidence for political distance causing existing partnerships to end was much weaker and less consistent than the evidence for language. In fact, when the researchers directly compared the two forces, they found that language was significantly more likely to cause a partnership to break up than politics was, but this was the only area where the two forces were clearly distinct; in almost every other aspect of collaboration, the data suggested that language and politics were pulling in similar directions, making it difficult to say one was more important than the other.
The researchers were careful to note that their work shows a strong connection between these factors and scientific output, but it does not prove that one directly causes the other in a simple, isolated way. The data reflects real-world conditions where language, politics, and economics are all tangled together. Still, the findings offer a clear picture for those who manage scientific networks. It suggests that while political tensions might stop new collaborations from starting, the day-to-day friction of language is a more pervasive force that quietly erodes the strength and longevity of scientific teams over time. For science to thrive globally, the study implies that we must look beyond just the headlines of political conflict and pay close attention to the practical, everyday barriers of communication that shape how knowledge is shared.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.