Stance-taking through that-clauses in research article abstracts: A cross-disciplinary comparison of translated and non-translated English in applied linguistics and aerospace engineering
This corpus-based study reveals that stance expression via that-clauses in research article abstracts varies significantly between applied linguistics and aerospace engineering, with translation effects proving discipline-contingent rather than uniform, as translated texts exhibit distinct rhetorical shifts influenced by both disciplinary conventions and source-language interference.
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
In the global world of science, English has become the common language, a bridge allowing researchers from Tokyo to Nairobi to share their discoveries. For scholars whose first language is not English, writing in this academic tongue presents a unique challenge. They must not only convey complex data but also adopt a specific rhetorical style—a way of sounding confident yet cautious, authoritative yet open to debate. This style is known as "stance." It is the subtle art of how a writer positions themselves in relation to their own ideas, using specific words to show whether they are certain, merely suggesting, or arguing a point. One of the most powerful tools for this is a simple grammatical structure: a verb or noun followed by the word "that" to introduce a statement, such as "the study shows that" or "it is clear that." These small phrases act as the hinges on which scientific arguments turn, signaling to the reader how much weight to place on the information being presented.
A new study by researchers Yi Liu and Yiheng Xia from Northwestern Polytechnical University investigates how this delicate balancing act changes when scientific writing is translated from Chinese into English. The researchers were particularly interested in whether the act of translation smooths out the text to sound like native English, or if it leaves behind a distinct fingerprint of the original language and the specific field of study. To find out, they did not look at just one type of science. Instead, they compared two very different worlds: applied linguistics, a field focused on human language and communication where ideas are often debated and nuanced, and aerospace engineering, a field focused on machines and physics where facts are usually precise and absolute. By analyzing thousands of research abstracts—the short summaries that appear at the beginning of every scientific paper—they examined how writers in these fields use "that" phrases in both their original English writing and in texts translated from Chinese.
The researchers gathered a massive collection of 2,000 abstracts, split evenly between the two disciplines and between translated and non-translated texts. They meticulously counted every instance where a writer used a verb, noun, or adjective to control a "that" clause, categorizing these words by what they meant. Some words expressed certainty, like "show" or "prove"; others expressed communication, like "argue" or "suggest"; and some expressed likelihood or attitude. The goal was to see if the translation process made the texts sound more alike to native English, or if the deep-rooted habits of the specific scientific field remained the strongest force.
What they found was a story of two different worlds, where the rules of translation played out in opposite ways. In the field of applied linguistics, the translated abstracts sounded noticeably different from the native English ones. The researchers observed that when linguists translated their work from Chinese to English, they used fewer of these "that" phrases overall. More importantly, the translated texts lacked the rich variety of words used to express nuance. Native English linguists used a wide range of verbs to argue, suggest, and propose, creating a textured, conversational tone. The translated versions, however, were more sparse, relying on a narrower set of words and missing out on the subtle rhetorical moves that make the field feel like a lively debate. It was as if the translation process had smoothed away some of the essential texture of the argument, leaving a text that was less engaged with the reader.
In stark contrast, the aerospace engineering abstracts told a different story. Here, the translated texts did not lose their voice; instead, they became even more rigidly focused on certainty. In this hard science, native English writers already preferred words that signaled absolute fact, such as "show" or "demonstrate," because the goal is to present undeniable evidence. The translated abstracts followed this pattern but took it a step further. They relied heavily on a single, dominant word: "show." In the translated aerospace texts, this one word accounted for more than half of all the "that" phrases used. While the native English writers used a slightly broader mix of verbs to report their findings, the translated texts concentrated almost entirely on this one term of certainty. The translation process here did not introduce variety; it reinforced a narrow, highly factual style, stripping away even the small amount of communicative flexibility that existed in the original English writing.
The study suggests that translation is not a simple, one-size-fits-all process. It does not merely convert words from one language to another; it interacts with the deep cultural and intellectual habits of the specific field. In soft sciences like linguistics, where ideas are fluid and open to interpretation, translation seemed to reduce the richness of the argument. In hard sciences like engineering, where precision is paramount, translation seemed to amplify the demand for absolute certainty, making the text more repetitive and less varied. The researchers concluded that the "fingerprint" of a translated academic text is shaped by a complex negotiation between the original language, the translator's choices, and the strict expectations of the scientific discipline itself.
This discovery matters because it highlights a hidden barrier for non-native English speakers trying to publish their work. It suggests that simply translating a paper is not enough to make it sound like a native contribution; the writer or translator must also understand the specific rhetorical style of the field. For linguists, the challenge is to preserve the nuance and variety of their argument during translation. For engineers, the challenge is to avoid becoming so focused on certainty that the writing becomes repetitive and loses its connection with the reader. The study offers a clear path forward: training for academic writers and translators needs to be specific to the discipline. Generic advice on how to write in English is insufficient. Instead, scholars need to learn the unique "voice" of their field, understanding how to use the right words to build trust and authority, whether they are writing in their native tongue or navigating the complex landscape of global scientific communication.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.