Key Takeaways
Treat universities as open infrastructure in a contest where both sides recruit
Two intelligence systems converge. Golden's central finding is that American universities serve both foreign intelligence services and U.S. agencies. Foreign services seek research, policy insight, and recruits; the FBI and CIA cultivate students and faculty, including international scholars who may return home. The scale of academic exchange makes campuses attractive: in 2014-15, U.S. universities enrolled 974,926 international students, while 313,415 Americans studied abroad.
Global reach brings exposure. More than 160 American-style universities operated in 80 countries, and many campuses welcomed foreign-funded programs and visiting scholars. After 9/11, U.S. intelligence and universities also rebuilt their relationship. The result is not that every foreign student is suspect, but that open academic networks have become part of the landscape of espionage.
The important shift is from seeing espionage as an intrusion into an otherwise separate academic world to recognizing that international research networks are themselves valuable terrain. The same mobility that spreads expertise also creates pathways for influence and information gathering. That is a structural vulnerability, not evidence against international education. The challenge is to protect people and research without treating nationality, language, or scholarly collaboration as proof of hostile intent.
Protect the work behind the paper, not just what journals eventually publish
Published does not mean harmless. At Duke, graduate student Ruopeng Liu joined Professor David Smith's group, which had developed an experimental microwave invisibility cloak. The research was not classified, yet Smith and lab colleagues later alleged that Liu shared their designs and data with collaborators in China, arranged for equipment to be copied, and used Duke work to support ventures there. Liu denied wrongdoing, and he was never charged with a crime.
The process has value. A published article usually presents successful results, not every failed test, design choice, or unpublished idea. Access to that working knowledge can save competitors time and help them publish or patent first. University collaborations often relied on informal trust, with little training in intellectual property or clear agreements about sharing. The book's point is not to close labs, but to recognize that openness can expose valuable work before it reaches print.
This case makes a useful distinction between open science and unprotected science. Collaboration depends on sharing, but trust is not a substitute for clear agreements about authorship, data, funding, and inventions. A university can preserve publication and international cooperation while clarifying who may use unpublished work and under what terms. Such safeguards should apply consistently to collaborators of every nationality. Otherwise, a security response can become either ineffective or discriminatory.
Treat student recruitment as a long game, not a single campus encounter
Recruitment can begin early. Soviet intelligence placed agents in universities, and the KGB cultivated students who later entered powerful positions. The Cambridge Five, recruited while studying at Cambridge, went on to penetrate British government and intelligence. China, too, encouraged students sent to the United States to report useful information, even when the material was not classified.
A modern case shows the personal route. Glenn Duffie Shriver, an American graduate of Grand Valley State University, returned to China after studying there. Chinese agents paid him to write essays, cultivated him, and encouraged him to seek work in the U.S. government. He took more than $70,000 and applied to the CIA, but never obtained access to secrets. He pleaded guilty to conspiring to commit espionage and was sentenced to four years. His story shows how money, ambition, and personal ties can precede official access.
Recruitment is better understood as relationship-building than as a dramatic demand for secrets. That makes ordinary social behavior, such as mentoring, networking, or offering paid work, difficult to distinguish from an intelligence approach without context. It also explains why institutions should teach students how to recognize pressure and seek advice, rather than ask them to become amateur counterintelligence officers. The Shriver case is a warning about manipulation, not a template for judging students who study abroad.
Do not let tuition revenue purchase institutional blindness
A partnership can serve different agendas. Marietta College, a small Ohio institution dependent on tuition, built a long relationship with China's University of International Relations, or UIR. U.S. diplomats described UIR as linked to China's Ministry of State Security, which recruits some graduates. Marietta gained Chinese students, while UIR gained legitimacy and opportunities for its students and faculty to experience the United States.
Money shaped the relationship. Chinese enrollment at Marietta reached 144 students in 2011 and 2012, about a tenth of the student body. Chinese students were asked to pay $50,000 up front. Marietta professors also taught American culture and academic subjects at UIR. The book does not establish that every UIR participant was an intelligence officer, or that every exchange was covert. It shows how institutional benefits can discourage leaders from examining risks they would rather not confront.
The case is less about proving a hidden plot than about exposing an incentive problem. When one partner controls a valuable stream of students, funding, or access, administrators may normalize uncertainty instead of investigating it. A sound review asks what each institution contributes, what it expects, and who oversees the arrangement. It should also distinguish a documented institutional affiliation from suspicion about individual students. Financial dependence can impair judgment, but suspicion without evidence can damage legitimate exchange.
Make intelligence partnerships visible to the people whose trust they use
Cooperation can become concealment. After 9/11, Penn State president Graham Spanier helped create the National Security Higher Education Advisory Board, bringing university leaders into contact with the FBI and CIA. He received a CIA medal for his work. The board gave administrators classified briefings and helped agencies build relationships across higher education.
The classroom raises harder questions. The book describes CIA officers attending Harvard's Kennedy School under State Department cover, with the administration informed but classmates and faculty generally not. Harvard's 1977 guidelines had rejected covert intelligence activity and deceptive recruiting on campus, arguing that these practices undermine trust. The school later accommodated undercover officers. Even when no secret is stolen, a hidden identity can change how people speak, teach, and form relationships.
Institutional consent is not the same as community consent. A president may judge intelligence cooperation useful, while students and faculty have no chance to assess how hidden identities affect classroom discussion or professional trust. At the same time, blanket exclusion of intelligence professionals would discard expertise and legitimate public service. The crucial boundary is transparency about the nature of the relationship, especially when an agent's role depends on classmates or colleagues believing a false account.
A recruited academic can be compromised, coerced, and still ineffective
Peng's case was no clean recruitment. Dajin Peng, a China-born professor at the University of South Florida, had been an FBI contact as a Princeton student but had declined earlier overtures. After USF investigated him for questionable expenses, visa letters, and other conduct, FBI agent Dianne Mercurio sought his help gathering information about China. Peng said the bureau used his university troubles as leverage. The book presents emails and phone records supporting FBI involvement, while USF denied that the bureau influenced its discipline.
The bargain had consequences. The university ultimately fined Peng $10,000 and suspended him without pay for a year, while preserving his tenure. The FBI had pressed for institutional support that could protect his access to Chinese contacts. Peng later resisted the bureau's demands and never became the effective source it sought. His story also includes serious allegations against him, so it cannot be reduced to a simple tale of an innocent professor persecuted by spies.
This episode exposes the ethical cost of using legal vulnerability as recruiting leverage. A person who agrees under pressure may provide unreliable information, resent the agency, or face risks that institutions cannot control. The case also shows why the competing explanations matter: the FBI's role is documented in correspondence, but the university disputed that it shaped the outcome. Careful analysis separates verified conduct, disputed causation, and the author's interpretation rather than flattening them into a verdict.
Treat academic conferences as public stages with private intelligence value
The room is part of the opportunity. Conferences gather specialists, policy makers, and researchers in one place, often before their work is published. Intelligence officers can learn from presentations, ask clarifying questions, and build relationships through repeated encounters. A casual first meeting, called a
People Also Read
Download PDF
Download EPUB
.epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.