When the world's at stake,
go beyond the headlines.

In early November, after meeting China’s defense minister, U.S. Secretary of Defense Pete Hegseth performed a longstanding ritual in military relations between the United States and China: He announced that the two countries would “set up military-to-military channels to deconflict and de-escalate” problems between them. Like most rituals, this one is unlikely to have any practical effect. The pursuit of effective crisis communications channels between the American and Chinese militaries to mitigate the risk of crisis escalation has been a work in progress since President Bill Clinton was in office. Progress, however, has been halting and largely unproductive. Washington and Beijing, it is true, have managed to negotiate multiple agreements on crisis communications mechanisms, but when actual crises erupt, Beijing rarely uses these links.

Nevertheless, Washington persists in pursuing these connections — and for good reason. When, in 1950, Washington did not ask Beijing if it should take seriously indirect warnings that China would intervene in the Korean conflict if U.S. forces crossed the 38th parallel, the results were calamitous. In contrast, during America’s Cold War standoff with the Soviet Union, institutionalized crisis communications channels were critical to escalation prevention. After the Cuban Missile Crisis, Washington and Moscow set up a crisis “hot line” that the two nuclear-armed powers went on to use dozens of times as they worked to ensure that their war remained a cold one. In 1983, Secretary of Defense Caspar Weinberger described the direct communications link between the United States and Soviet Union as “invaluable in major crises.” The success of this tool inspired the creation of bilateral hotlines between other nuclear powers and crisis-prone countries.

Fifteen years later, in April 1998, U.S. Secretary of State Madeleine Albright and Chinese Foreign Minister Tang Jiaxuan announced the establishment of the first link for crisis communications between Washington and Beijing in the form of “a secure presidential phone-link … [that] will help our leaders to communicate rapidly, directly and candidly whenever the need should arise.” This followed the signing of the Military Maritime Consultative Agreement in January of the same year. In the early 2000s, U.S. defense officials proposed to the Chinese military that the two sides establish a hotline for crisis management and confidence building. In February 2008, the two sides reached an agreement  to set up such a quick communication link in the form of a defense telephone link. The two countries expanded the defense telephone link in 2015 to include video teleconference communications. They also identified this link as a mechanism for conveying information under a 2014 bilateral understanding to notify each other of “major military activities.” In 2016, Beijing and Washington instituted a hotline on cybercrime and related issues.

Yet, when bilateral tensions between Washington and Beijing have spiked to risky levels, these carefully negotiated mechanisms have generally failed to function as the United States has anticipated. Experience with the U.S-Soviet hotline may have led U.S. officials to expect that messages sent by one side through a hotline would be seen as urgent and deserving of a swift reply. But Beijing evidently sees things differently.

Even at the leader-to-leader level, the crisis communications channels established between Washington and Beijing have failed to normalize rapid bilateral communications in crises. Following the accidental bombing of the Chinese embassy in Belgrade in 1999, Chinese leader Jiang Zemin refused to answer President Bill Clinton’s call for several days and suspended bilateral miliary-to-military contacts. Similarly, after a 2001 midair collision between a U.S. Navy EP-3 intelligence aircraft and a Chinese air force J-8 interceptor, China’s foreign ministry waited — for 12 hours — to answer a call from the U.S. ambassador in Beijing, Admiral Joseph W. Prueher. In 2016, when a Chinese vessel seized a U.S. unmanned underwater vehicle, the ship acknowledged receipt of U.S. radio messages requesting the return of the vehicle, but then ignored them. The defense telephone link seems not to have been used during this incident. During the controversial 2022 visit by U.S. House of Representatives Speaker Nancy Pelosi to Taiwan, both Defense Secretary Lloyd Austin and Joint Chiefs Chairman Gen. Mark Milley were unable to reach Chinese counterparts. In February 2023, Beijing refused Austin’s call via the defense telephone link after the U.S. Navy shot down a Chinese high-altitude balloon that crossed into U.S. airspace, although the U.S. understanding was that mutual concerns about this type of contingency was precisely why the two sides established the defense telephone link. These failures contrast with the successful use of the defense telephone link in the fall of 2020, when Milley was able to reassure his Chinese counterpart that President Donald Trump was not going to manufacture a pre-election crisis (an “October surprise”) with China.

Aside from a few success stories, the general pattern of Chinese behavior during crises is to suspend communications and suspend or terminate agreements for routine military-to-military contacts. Yet, in the wake of bilateral political-military crises, China has routinely engaged in negotiations on crisis communications and renewed military-to-military ties. For example, the two sides began negotiations that led to the 1998 presidential phone link after the 1996 Taiwan Strait confrontation.

These negotiations have rarely been easy. U.S. efforts to negotiate a defense telephone link, for instance, took years to bear fruit. The two sides did not reach agreement until February 2008— after a year bookended by China’s debris-generating antisatellite missile test in February 2007 and Beijing’s denial of permission for U.S. Navy warships to make a Hong Kong port call ahead of a storm in November 2007. At other times, China has been more open to negotiation. In May 2014, following a surge in bilateral tensions after Washington indicted Chinese hackers with alleged ties to the People’s Liberation Army and China suspended its participation in the U.S.-Chinese cybersecurity working group, Washington and Beijing concluded two nonbinding memoranda of understanding. These two agreements, aimed at mitigating military crises, were the Notification of Major Military Activities Agreement and the Rules of Behavior Memorandum of Understanding. In 2020, after an escalation in U.S.-Chinese tensions that included harsh words surrounding the COVID-19 virus, the expelling of U.S. journalists from China, and the closure of the Chinese consulate in Houston, the two sides established a Crisis Communications Working Group.

This pattern suggests that the United States should reset its expectations about China’s use of crisis communications mechanisms. Clearly, China sees negotiating such mechanisms and other bilateral dialogues as a useful diplomatic activity in terms of improving bilateral confidence and restoring diplomatic ties, including military-to-military relations. However, the pattern supports expert assessments that Chinese leaders are skeptical of the value of rapid bilateral communications and other guardrails in most crises. Beijing, say the experts, does not see the establishment of a code of conduct as a tool for de-escalation, but rather as an instrument that legitimizes the very U.S. activities that Beijing seeks to prevent, such as U.S. military operations near China. In other words, for Chinese leaders, hotlines and other crisis management mechanisms are effectively a form of “seatbelts for speeders” that embolden rather than constrain risky and aggressive American behavior.

Moreover, Chinese strategists see crisis communications less as a useful way to mitigate the potential for escalation and more as a potential constraint on China’s ability to maximize potential gains from crises. Published pieces by Chinese experts portray crises as opportunities to signal resolve, manipulate situations for strategic advantage, and create negotiating leverage. Chinese military strategists began refining a concept of “escalation control” for crises two and a half decades ago. Today, the Chinese military’s response to political-military crises appears to draw on a concept of “effective control” of crises aimed at shaping and controlling them, including through escalatory tactics, to support a strategic position that serves China’s domestic stability and growing international objectives.

Furthermore, Chinese and American perspectives of what constitutes a genuine crisis may differ fundamentally. This is illustrated by comparing Washington’s effective use of the defense telephone link during the 2020 “October surprise” with the failure to make a connection to Beijing following the 2023 balloon incident. In the first case, where China saw military action by the United States as potentially imminent, China appeared to welcome communication from the United States clarifying its intentions. In the second, Chinese Defense Minister Wei Fenghe refused Defense Secretary Austin’s outreach, with the Chinese Defense Ministry contending that the United States had “not created the proper atmosphere” for dialogue and exchange.

Beijing’s reluctance to use crisis communications mechanisms may also be inspired by perceived asymmetries in military power between the two countries. Peking University’s Hu Bo contends that whereas the United States and Soviet Union were militarily balanced, the United States and China are not. This power asymmetry erodes mutually credible deterrence, impeding China’s ability to act as openly and freely as the United States in the process of building and negotiating crisis management mechanisms. There is also the argument that the United States and the Soviet Union confronted reciprocal nuclear risk — what scholar Thomas Schelling described as a “reciprocal fear of surprise attack” that motivated efforts to reduce the risk of escalation. In contrast, Tsinghua University’s Wu Riqiang argues that Beijing “prioritizes building its retaliatory capability over reassurance and risk reduction and maintaining a high level of secrecy, both of which go against U.S. arms control principles.”

Another point of discrepancy is the organizational asymmetries between the two sides. On the U.S. side, more individuals down the chain of command may be empowered to engage in real-time decision-making. In contrast, the Chinese system is designed for downward rather than upward flows of information, limiting the authority and decision-making capacity of those not in the highest leadership positions. Furthermore, lower-level officials would be exposed to charges of foreign influence were they to interact directly with a foreign government. As a result, no one in the Chinese system wants to communicate with the United States until the top leadership in consultation with relevant actors within China’s party-state have decided on a response, which may take days if not longer.

Finally, Beijing sees the U.S. focus on crisis communications mechanisms as a distraction from what it believes should be the major thrust of bilateral interactions: forging stability in the relationship as a whole. China sees stability in addressing the “root cause” of the instability, which lies in the political context — namely, Washington’s failure to accept the legitimacy of China’s political system and respect China’s “core interests.”

Washington should move beyond the notion that negotiating and establishing crisis mechanisms with Beijing will provide a reliable way to mitigate the risk of escalation during actual crises. This does not mean abandoning such efforts entirely. These negotiations serve valuable functions for routine military-to-military communications and post-crisis diplomatic engagement. But policymakers should not mistake participation in negotiations for a genuine desire by Beijing for mechanisms that institutionalize rapid communications with Washington in a crisis. Looking at Pelosi’s visit to Taiwan, several U.S. government officials described witnessing the events surrounding this visit as “watching two trains on a collision path.” Although there was a sense that neither side wanted to escalate the situation to conflict, each “believed it was well within its rights to undertake the actions it did.” Thus, all that could be done was to seek to manage the situation.

This is the sobering reality facing U.S.-Chinese relations today. It may not be possible to move the trains off the tracks and completely avoid crises. If that is indeed the case, then the United States should adopt a more modest but still valuable goal: slowing the trains down as much as possible to reduce uncertainty, harm, and escalation. This can be achieved through modest efforts to introduce more certainty on the U.S. side.

Instead of expecting that once crisis communications channels are established, Washington can rely on them, U.S. policymakers should consider establishing or supporting communications channels that will not become targets of suspension or cancellation in U.S.-Chinese political-military crises. These include not only governmental but also extra-governmental communications mechanisms.

With respect to governmental channels, the United States may want to draw a lesson from the 1950s, when U.S.-Chinese tensions were acute, leading President Dwight D. Eisenhower to threaten China with nuclear annihilation eight times, by some counts. However, even amid these tensions, the Eisenhower administration agreed to regular talks with China, which were quietly sustained by the two governments despite mutual nonrecognition in international fora until U.S. Secretary of State Henry Kissinger visited China in 1971. Reportedly, the talks may have been little more than structured shouting matches. However, they are credited with denting the risk of U.S.-Chinese conflict at multiple decisive intervals. Low-profile meetings between designated American and Chinese officials might serve this function.

The U.S. government might also consider creating a nonpolitical, bureaucratic coordinator position affiliated with the National Security Council to serve as a point of contact for U.S.-Chinese relations. Building trust and interpersonal connections is a key part of managing the relationship. Washington is unable to control turnover on the Chinese side, which has recently seen dramatic shifts, but it can try to address the natural turnover that occurs within the U.S. political system. The designation by the United States of a nonpartisan coordinator who can weather administrative differences and maintain relations with the Chinese leadership offers a potential channel for real-time communication during crises to help avoid miscalculation.

In addition to these governmental mechanisms, Chinese and U.S. experts alike have pointed to the vitality of Track-2 exchanges — unofficial, non-governmental channels — that can serve as resilient mechanisms of communication during points of high tension in bilateral relations. Washington’s success in reassuring Beijing that Trump was not going to engineer an “October surprise” in 2020 is partly attributable to the role played by Track-2 actors — in particular, Track-2 channels that communicated Chinese concerns to the U.S. defense apparatus. Such nongovernmental forums as the U.S.-China Military-to-Military Initiative (formerly the Sanya Dialogue), which brings together retired four-star generals and admirals from both sides, offer potential conduits for mutual reassurance in crises. Independent institutions with close ties to the U.S. government have long had a special legitimacy in exchanges between the United States and China. The hollowing out of such congressionally funded institutions as the Wilson Center and the U.S. Institute of Peace (the latter of which employed both of us before it was shuttered) undermines these important avenues for communication.

There may be technical ways forward as well. Global risk expert Christian Ruhl makes a case for incorporating a text-based hotline. Ruhl points out that a text-only channel could be more easily protected against damage to its infrastructure, particularly during a conflict. A text-based system might also help overcome the proscription in China’s hierarchical political and military system against individual officials engaging in real-time communication with foreign interlocutors. Notably, the first hotline between the United States and Soviet Union involved telegraphs rather than phone calls to “allow time to carefully read and then respond.” It can serve a similar function here. Further, as Ruhl also observes, a text-based system could include a “mechanism for acknowledging receipt of information without having to formally issue a reply.” This enables a lower-level official to provide reassurance without commitment, potentially reducing time between communications and speeding up the crisis management process.

Some argue that only a Cuban Missile-style crisis could catalyze the kind of robust communications mechanisms that proved essential in reducing escalation risks between the United States and the Soviet Union. Yet, as Washington and Beijing increasingly face off militarily across all domains, diplomatic contact grows more constrained and contrived, and the dangerous trope that conflict is inevitable takes hold. Waiting for a crisis to force cooperation is a gamble neither side can afford to take.

Carla P. Freeman is director of the Foreign Policy Institute at the Johns Hopkins School of Advanced International Studies where she is also a senior lecturer in international affairs. She rejoined the school early in 2025 after working for several years as a senior expert on the China team at the U.S. Institute of Peace.

Alison McFarland is a Ph.D. student in politics at Princeton University where she focuses on Chinese foreign policy and conflict studies. She previously worked on the China team at the U.S. Institute of Peace.

**Please note, as a matter of house style, War on the Rocks will not use a different name for the U.S. Department of Defense until and unless the name is changed by statute by the U.S. Congress.

Image: Wikimedia Commons

Correction: A previous version of this article mistakenly left out a sentence in the fifth paragraph, which was erroneously deleted during the editing process.