What is Beijing thinking? Why Washington is struggling to understand its rival
Language is important in US-China diplomacy, but funding cuts and dwindling student exchanges have left a ‘major gap’ in capabilities
While the White House framed this vision as a “constructive relationship of strategic stability on the basis of fairness and reciprocity”, Beijing said only that it was a “constructive bilateral relationship of strategic stability”.
China watchers zoomed in on the phrase – especially the Chinese version – debating what it could reveal or conceal about Beijing’s long-term ambitions.
They put that down to funding cuts to language programmes in America as well as dwindling student exchanges and less interest in learning Chinese.
The result is a disjointed network of government programmes, think tanks and independent researchers seeking to translate and interpret China’s latest moves.
Analysts say that fragmentation could make it harder for Washington to understand its most critical bilateral relationship.
“A decline in government support and interest among American students leaves a major gap in Chinese language capabilities,” according to Kyle Chan, a fellow at the John L. Thornton China Centre at the Brookings Institution.
“This could lead to a wider gap in understanding China.”
Since Trump’s return to the White House, numerous federal language programmes have been disrupted.
Overseas fellowships such as the State Department’s Critical Language Scholarship have seen funding interrupted, while programmes including the National Resource Centre and Foreign Language and Area Studies – which provided millions of dollars in grants, tuition and stipends for university students nationwide – have been halted.
However, defence-centric initiatives under the Pentagon’s Defence Language and National Security Education Office have been largely preserved.
Similarly, language training through the Federal Bureau of Investigation, Office of the Director of National Intelligence and Central Intelligence Agency remains funded, though some programmes face operational bottlenecks due to tightened security clearance.
Shengyu Wang, a researcher at the Asia Society Policy Institute, noted that defence-related language programmes were important but he said a wider approach was needed.
“Language knowledge should not be built only around defence needs,” Wang said. “If civilian language programmes weaken, Washington may lose the broader cultural, historical, political and social knowledge needed to interpret China as a system.”
A bipartisan group of US lawmakers attempted to establish a civilian language centre in 2024 through the Open Translation Centre Act.
The legislation aimed to set up a separate, federally funded research centre tasked with locating, translating and analysing open-source materials. The bill noted that “a lack of foreign language speakers greatly hinders US policymakers, academics, students and others’ ability to understand China and other nations”. But the legislation stalled in Congress.
Rosie Levine, executive director of the US-China Education Trust, said fewer than 2,000 Americans were currently studying in China – a dramatic drop from around 11,000 in 2019.
Conversely, more than 265,000 Chinese students were enrolled in US universities during the 2024-25 academic year.
She said factors including China’s pandemic-era visa restrictions, geopolitical tensions and higher travel costs had contributed to the decline and warned that recent funding cuts could worsen the situation.
“If that trend continues … the US will face a critical shortage of grounded China expertise within a decade,” Levine said.
Apart from fewer student exchanges, Beijing has also imposed tighter controls on Western researchers and journalists in China as the focus on national security and geopolitics intensifies.
With their access to China limited, Washington think tanks and policy analysts have turned to translating and analysing open-source official texts to try to work out Beijing’s thinking.
In recent years, leading research organisations have launched translation-specific initiatives with support from both the federal government and private foundations.
One significant example is the Council on Foreign Relations. Its China Strategy Initiative features a dedicated database called the China Open Source Observatory – a project launched in 2024 that aims to translate key Chinese official papers and policy documents, with a collection that spans from 1990 to 2023.
The initiative asserts that the “future of Western-China relations now rests on open-source analysis” and notes that the project bridges the translation gap by “locating, translating and annotating documents of historic or strategic value that are currently only available in Chinese”.
It also makes use of artificial intelligence, involving engineers from Anthropic and Google to optimise large language models for translation. Through this partnership, it claims to have achieved “near-perfect accuracy” when translating military and diplomatic texts.
Georgetown University’s Centre for Security and Emerging Technology meanwhile maintains a “discovery tool” for Chinese-language news and commentary on technology issues, with more than 5,000 articles featuring analysis on AI, quantum technology and aerospace.
Individual Substack accounts such as Pekingnology, The East is Read and Sinification are also actively involved in translating and analysing Chinese-language material.
Sourabh Gupta, a resident senior fellow at the Institute for China-America Studies, argued that while some of these efforts were welcome, “current initiatives are scattershot and lack a certain coherence”.
“Unless presented coherently, systematically and consistently, these translation initiatives will, more or less, lack purchase and resonance within the consuming community,” he said.
However, Wang from the Asia Society Policy Institute said “some degree of fragmentation is natural”. He said the real challenge was “when translations circulate without enough explanation of method or context”.
A debate over context arose in October when the Rand Corporation – a Washington think tank mainly funded by the Pentagon – released and then quickly withdrew a 115-page report accusing prominent Washington-based experts with Chinese skills of “hawkish” distortions of Beijing’s terminology.
The study, titled “Stabilising the US-China Rivalry”, warned that several analysts – including Rush Doshi, director of the China Strategy Initiative and a former National Security Council official in the Joe Biden administration – had translated Chinese terms with “more hawkish English alternatives than the original Chinese language sources may imply”.
The report highlighted a specific example from a 2021 book by Doshi, which translated the Chinese phrase xianshouqi as “offensive moves”.
It argued that Doshi had overstated the phrase’s significance to claim Beijing’s strategy was more assertive than it was, noting it is actually a board game metaphor that literally means “first hand” and contextually translates to “taking the initiative”.
In response, Doshi said the paper had engaged in “creative conceptual reinterpretation”.
“In most cases they muster only 200 words of critique against several thousands words of my own empirical support, which is a bit superficial,” he wrote on social media.
The report was swiftly taken down and replaced with a note saying: “This document has been withdrawn for further review.”
Gupta from the Institute for China-America Studies said that if the goals of translation were narrow and “almost ulterior” it would reinforce a more skewed and “securitised beltway view of China”.
“The end purpose of translation is not to reinforce a view of China as the ‘other’; rather, it should be to obtain a more objective assessment of China,” he said.
But Wang noted that China’s official language “often creates room for several plausible interpretations”.
“A phrase can be linguistically accurate but politically incomplete … China’s political system is difficult to read by design,” he said.
“Its official language is formal, repetitive, coded and often intentionally ambiguous. That ambiguity gives the leadership flexibility and helps it speak to different audiences at once.”
Chan of the Brookings Institution – who has also compiled a public database of Chinese government documents – emphasised that the undertaking would always be an “imperfect project”.
“There will always be some degree of selection bias in terms of which voices get translated and read by American audiences. At the same time, it’s much better to translate a limited selection of important texts rather than to leave a vacuum of understanding,” he added.
For Levine from the US-China Education Trust, translation efforts were “better than nothing”, but she said there was no substitute for on-the-ground experience.