Few metaphors have shaped Western thinking about Chinese espionage as much as the “thousand grains of sand”. It describes an intelligence system built on volume rather than precision. It is also one of the most contested ideas in the field.
For decades, counterintelligence officers, journalists and policymakers have used this image to explain how China gathers information abroad. The idea is easy to grasp: many ordinary people each collect a tiny piece of information, and the state assembles the pieces into a complete picture.
This article explains what the concept means, how it is said to work in practice and what real espionage cases tell us about it. It also looks at why many specialists now consider the theory incomplete, and in some respects misleading.
The concept: intelligence by accumulation
The thousand grains of sand strategy describes intelligence collection through mass accumulation. Instead of relying on a few highly trained agents to steal secrets, the state gathers large amounts of small, often unclassified information from many sources. Analysts then combine those fragments into knowledge of real strategic value.
The best-known version of the metaphor is attributed to Paul Moore, a former senior China analyst at the FBI. In his telling, if a beach were an intelligence target, the Russians would send a submarine and frogmen at night to carry off a few buckets of sand. The Americans would study it with satellites. The Chinese would send a thousand tourists, each asked to bring back a single grain. Once home, they would shake out their towels, and end up knowing more about the beach than anyone else (War on the Rocks).
The image is older than Moore’s later writing. An FBI official used it to describe Chinese espionage to the press as early as 1988 (The News Lens). Over time it acquired other names: the “mosaic”, “human wave” or “vacuum cleaner” approach (Mattis, USCC testimony).
As summarised by its critics, the theory rests on a few core assumptions (Mattis, Studies in Intelligence). China is said to:
- rely mainly on amateur collectors rather than professional intelligence officers;
- focus on ethnic Chinese as sources;
- avoid the clandestine tradecraft used by Western and Russian services;
- prefer open or low-grade information to classified secrets.
The use: how the approach is said to work
In practice, the strategy is described as a division of labour between many collectors and a central point of analysis. It is usually broken into four elements:
- Wide collection. Students, visiting researchers, business travellers, official delegations and conference participants are all seen as potential collectors. Each gathers small items: a technical paper, a conversation at a trade fair, a product specification, a useful contact.
- Low risk per piece. Most of what is collected is legal to obtain or not classified. No single contribution looks like espionage, which makes detection and prosecution difficult.
- Central assembly. The real value is created at home, where analysts, research institutes and state companies combine the fragments into a coherent picture of a technology, an industry or a policy.
- Patience. The method works over years, not weeks. It seeks gradual advantage rather than one decisive theft.
Supporters of the model argue that it fits China’s long-standing priority of closing technology gaps in sectors such as aerospace, defence and advanced manufacturing. It also reflects a broader truth about intelligence: it is not only about stealing “classified” material, but about any information that helps decision-makers (China Digital Times).
For companies, universities and governments, the concept has a practical use as well. It explains why protecting classified files is not enough. Open conferences, research partnerships, hiring and business travel all become part of the security picture.
Examples: three US cases
Public court cases offer the clearest view of how Chinese collection works in reality. Three well-documented US cases are often cited in this debate.
Chi Mak: unclassified data, deliberate tasking
Chi Mak was an electrical engineer at Power Paragon, a contractor for the US Navy. In 2007 a jury convicted him of conspiring to export defence technology to China, acting as an unregistered foreign agent and lying to investigators (DOJ). The data concerned submarine and warship propulsion systems. It was not classified, but it was export-controlled (Washington Times).
Investigators found that contacts in China had sent him lists of specific topics, with instructions to attend technical seminars and record what he learned (DOJ, 2007). In 2008 he was sentenced to more than 24 years in prison.
Greg Chung: decades of document collection
The investigation into Chi Mak led to Dongfan “Greg” Chung, an engineer at Rockwell and later Boeing. Agents found around 300,000 pages of sensitive documents in his home, including material on the Space Shuttle and the Delta IV rocket (OC Weekly). In 2009 he became the first person convicted at trial under the US Economic Espionage Act. The judge found that he had acted as an agent of China for about 30 years, and in 2010 sentenced him to nearly 16 years in prison (DOJ).
Xu Yanjun: a professional officer at work
Xu Yanjun was not an amateur. He was a career officer of China’s Ministry of State Security (MSS), the country’s main civilian intelligence agency, and rose to deputy division director. Over several years he targeted employees of GE Aviation and other aerospace companies, inviting them to China to give presentations. According to prosecutors, MSS colleagues copied the contents of their computers in hotel rooms while the guests were taken to dinner (DOJ).
Xu was arrested in Belgium in 2018 and became the first Chinese intelligence officer extradited to the US to stand trial. In 2022 he was sentenced to 20 years in prison.
What the cases show
Together, the cases show both sides of the debate. Chi Mak’s case involved unclassified data gathered patiently over time, which fits the “grains of sand” image. Yet in all three, information moved through deliberate tasking and long-term relationships. In Xu’s case, a professional intelligence officer used classic tradecraft.
The debate: a useful image, a contested model
Since around 2011, the theory has faced strong criticism. Peter Mattis, a former CIA counterintelligence analyst, has played a leading role in challenging it (Axios). He argues that China’s intelligence services are professional organisations with a long history. Documented cases show officers running sources with familiar tradecraft, and those sources are not limited to ethnic Chinese (Mattis, Studies in Intelligence).
Critics also point to a social cost. Journalist Mara Hvistendahl and former FBI agent Michael German have linked the theory to suspicion of Chinese-American scientists based on ethnicity rather than evidence (China Digital Times). A model that treats every student or researcher as a potential collector can damage legitimate scientific exchange. It can also alienate the communities whose cooperation counterintelligence needs.
Not everyone agrees the theory is obsolete. Intelligence scholar Nicholas Eftimiades argues that the FBI’s original view was broadly right. In his reading, cases of professional tradecraft are few compared with the vast scale of Chinese collection, which mobilises state, academic and commercial actors alike (The News Lens). Cyber espionage, which the metaphor never described, adds a further layer.
The most balanced reading is that each side captures part of the truth. China does gather information at scale through many channels. But its most damaging operations are usually directed by professionals.
Conclusion
The thousand grains of sand is a useful image but a weak model. It reminds us that intelligence is built from many small pieces, and that open information gains strategic value once it is combined. That lesson remains relevant for any company, university or government that works with China.
Taken literally, however, the metaphor misleads. It understates the professionalism of China’s intelligence services and encourages suspicion of people rather than of behaviour. For anyone following China, the concept is best treated as a starting point for analysis, not as a conclusion.
Sources
- Peter Mattis, “A Guide to Chinese Intelligence Operations”, War on the Rocks, 2015
- Peter Mattis, “Beyond Spy vs. Spy”, Studies in Intelligence, Vol. 56, No. 3, 2012
- Peter Mattis, testimony to the US-China Economic and Security Review Commission, 2016
- Nicholas Eftimiades, essay on the thousand grains of sand theory, The News Lens
- “New book unveils China’s formidable spy agencies”, Axios, 2020
- Report on race and the grains of sand hypothesis, China Digital Times, 2020
- Mattis on Chinese intelligence methods, China Digital Times, 2014
- US Department of Justice, Chi Mak sentencing, 2008
- US Department of Justice, Mak family case, guilty pleas, 2007
- “Engineer guilty in plot to give data to China”, Washington Times, 2007
- US Attorney’s Office, Central District of California, Dongfan Chung sentencing, 2010
- Chung sentencing report, OC Weekly, 2010
- US Attorney’s Office, Southern District of Ohio, Yanjun Xu sentencing, 2022























