Arms Control China East Asia Indo-Pacific Korea Missile Defense Nuclear Deterrence Security & Defense Security Partnerships
Issue Brief August 16, 2023

Biases blind us to the risk of Chinese military intervention in Korea

By Jonathan Corrado

More than seventy years ago, the US intelligence community, military leaders, and policymakers systematically downplayed and discounted the risk tolerance and threat perception of the CCP leadership on the eve of PRC intervention in the Korean War. This intervention resulted in the massive loss of lives, territory, and initiative, all of which could have been prevented. Far from an isolated incident, this intelligence failure is better understood as a pervasive and accumulating system of misperceptions widespread throughout the US national security community, including policymakers in Washington, the leadership at Tokyo’s Far East Command, and intelligence analysts in the military and the nascent Central Intelligence Agency (CIA). These misperceptions corroded the assessment at five critical junctures, including

  1. in the run-up to North Korea’s invasion on June 25, 1950;
  2. when the PRC mobilized tens of thousands of troops to the North Korean border in July 1950;
  3. when the PRC foreign minister relayed a strategic warning message on October 3, 1950;
  4. when PRC forays into North Korea and skirmishes began in late October 1950; and
  5. when the major mobilization occurred before the full assault by the PRC in late November 1950.

The reason for the intelligence failure was a collection of cognitive and organizational biases that went unrecognized and unaddressed. Variations of these factors remain prevalent today, impairing the United States’ ability to forecast and plan for PRC intervention on the Korean peninsula and the possibility of simultaneous conflict involving the Korean peninsula and Taiwan. In addition, the presence of organizational bias reinforces, exacerbates, and disguises the effect of cognitive biases. This compounds the problem, making it harder to recognize and address. Perhaps most troubling is that the alchemy of US cognitive and organizational biases is uniquely suited to advantage PRC strategic proclivities.  

This paper examines the historical record of cognitive biases, focusing on the US intelligence community’s failure to forecast PRC intervention in the Korean War, despite collecting information and evidence indicative of that outcome. The cognitive biases examined in this paper include mirror imaging, over-centralization of decision-making, under-centralization of decision-making, over-sensitivity to consistency, and the vividness criterion.1Richards J. Heuer, Jr., “Psychology of Intelligence Analysis,” Central intelligence Agency, Center for the Study of Analysis, 1999, https://www.ialeia.org/docs/Psychology_of_Intelligence_Analysis.pdf. The organizational bias examined is the law of the instrument bias, which describes a tendency to assess problems through the prism of the tools at one’s disposal, rather than viewing the problem on its own terms. This is caused by a division of labor that separates resources and authorities in such a way that stifles forecasts and plans that acknowledge the high degree of integration and overlap between the challenges posed by the PRC and North Korea. 

Mirror-imaging bias

Mirror imaging describes the tendency to impose one’s own background, culture, and beliefs when analyzing another’s behavior. From a historical perspective, US analysts have underestimated the PRC’s threat perception and risk tolerance, particularly Beijing’s willingness to expend blood and treasure to maintain influence over the Korean peninsula. One reason for this is Chinese propaganda, which has historically emphasized the cause of aiding a fellow socialist state in describing its motivation to intervene in Korea. However, that narrative has changed in recent years. The new version is more reflective of the deeper reason for that intervention, pointing to the direct security threat the PRC feels from contingencies on the Korean peninsula.2Jonathan Corrado, “China Put Itself in the Spotlight with Its Flurry of New Korean War Propaganda,” NK Pro, October 28, 2020, https://www.nknews.org/pro/china-put-itself-in-the-spotlight-with-its-flurry-of-new-korean-war-propaganda/. A few quotes by CCP Chairman Mao Zedong and CCP Chairman Xi Jinping can demonstrate this point.   

In the summer of 1950, Mao said the following at a politburo meeting: “If the U.S. imperialists win [in Korea], they may get so dizzy with success that they may threaten us. We therefore must come to [North] Korea’s aid and intervene.”3Guang Zhang Shu, Mao’s Military Romanticism: China and the Korean War, 1950-1953 (Lawrence: University Press of Kansas, 1995), p. 63. This contrasts with later propaganda treatments of the rationale for PRC intervention, which painted PRC intervention in less conflictual tones and emphasized the cause of fraternal socialism. However, at events marking the seventieth anniversary of the start of the Korean War, this narrative changed, with the new version more closely approximating the original justification advanced by Mao. This transition to a more confrontational approach also coincided with the CCP’s turn toward a more aggressive diplomatic posture, typified by “wolf warrior” diplomats. In particular, statements by Chairman Xi clarify that the PRC continues to view Korea as within its sphere of influence. He said, “Seventy years ago, the Imperialist invaders fired on the doorstep of a new China.”4Sarah Zheng, Laura Zhou, and William Zheng, “Xi Jinping Says China ‘Determined to Defeat Invaders’ in Korean War Anniversary Speech,” South China Morning Post, October 23, 2020, https://www.scmp.com/news/china/diplomacy/article/3106759/china-ready-fight-xi-jinping-says-korean-war-address-aimed-us. Xi then argued that the PRC’s military should draw inspiration from this history by dealing “head on” with threats to its security.

US analysts have historically underappreciated Beijing’s perception of the linkage of Korea and Taiwan as national security issues. This mistake is especially puzzling given that US policymakers, through their actions, have demonstrated the strategic value of this linkage, only to later suffer from sudden amnesia and disregard how the adversary could make that same connection. Prior to the Korean War, in late 1949 and early 1950, “American officials were prepared to let PRC forces cross the Strait and defeat [Kuomintang leader] Chiang.”5“The Taiwan Straits Crises: 1954–55 and 1958,” US State Department Office of the Historian, last visited December 11, 2022, https://history.state.gov/milestones/1953-1960/taiwan-strait-crises. Mao and the CCP were eager to complete their war of unification by eradicating the remaining Nationalist Kuomintang forces and claim Taiwan as their prize. US Secretary of State Dean Acheson recommended that the United States recognize the communist government as the official representation of the PRC.6Additional Background Information: US Enters the Korean Conflict, US National Archives, last visited December 11, 2022, https://www.archives.gov/education/lessons/korean-conflict. The CCP allied with the Soviets and prepared to attack. But before Mao could mobilize his forces, Kim Il-Sung beat him to the punch. This changed everything. 

Two days after the invasion, President Harry Truman said, “communism has passed beyond the use of subversion to conquer independent nations and will now use armed invasion and war.”7“President Truman’s Statement on the Situation in Korea,” US National Archives, June 27, 1950, https://www.archives.gov/education/lessons/korean-conflict. He, therefore, ordered the US Seventh Fleet into the Taiwan Strait to block a CCP invasion. However, despite seeing a North Korean invasion as an opening gambit for communist expansion, US intelligence analysts continued to downplay the possibility that PRC leaders would themselves be willing to enter the Korean theater of war for the furtherance of their own aims. On the other hand, President Truman’s assurances that the United States had no intention to carry the fight into the PRC was cold comfort for the communist leaders in Beijing. Having suffered the humiliation of a spoiled Taiwan invasion at the hands of the US Navy, Beijing’s leaders came to see the US defense of the ROK as the opening move of a larger ambition to roll back communism in Asia.

Another aspect of mirror imaging is US misunderstanding of PRC strategic culture, particularly a version of deterrence that is more akin to the Western concept of preemption. The PLA concept of a “self-defensive counterattack” involves using a sudden blow against an imminent threat in order to change the enemy’s willingness to take risks and make concessions.8Andrew Scobell, et al., The People’s Liberation Army and Contingency Planning in China (Washington, DC: National Defense University Press, 2015), https://ndupress.ndu.edu/Portals/68/Documents/Books/PLA-contingency/PLA-Contingency-Planning-China.pdf. Also see: Henry Kissinger, On China (New York: Penguin 2011). This leads to political negotiations and a favorable outcome for the PRC. This strategy was utilized by the PRC in Korea, India, and Vietnam. Deterrence failures are more likely in such cases of underestimation. 

Importantly, a simple calculation of present military conditions will not yield a satisfactory forecast. “Long term political, economic, and social prospects” must also be considered.9Michael E. Brown, “Deterrence Failures and Deterrence Strategies,” RAND, 1977, https://www.rand.org/pubs/papers/P5842.html. This requires not only understanding the political decision-making calculus, but also the strategic culture that filters the military balance calculation. What looks like “certain suicide” from one nation’s perspective could appear as an acceptable risk to another. Americans could not imagine how communist China, limping from a long and bloody civil war and suffering economically, would be willing to trade blows with a superpower, nuclear-armed United States. Only by taking PRC strategic culture into account is it possible to understand the rationality of Beijing’s risk tolerance. 

Mirror imaging remains a problem in the US assessment of PRC intervention in Korea and the possibility of simultaneous conflict. Ascribing to Chinese leaders the same strategic playbook that Washington uses, the United States is likely to continue to downplay the likelihood of an early PRC aggression designed to seize the initiative in a contingency. It is likely that US analysts will discount this move because it appears to invite an unacceptably high level of escalation risk. However, the “self-defensive counterattack” accords with the CCP’s military record and contemporary posture. This also means that the United States is likely to downplay the extent to which regional flare-ups are viewed by the CCP as interrelated and relevant to its own security. 

Over-centralization of decision-making bias

Over-centralization of decision-making describes a tendency to group actors together and view their behavior as centrally directed, even when that is not the case. In its analysis, the United States grouped adversaries together and applied the same logic of restraint demonstrated by the patron state when assessing the actions of the client state. US analysts failed to imagine that local actors would risk superpower escalation to pursue regional objectives. First, they failed to predict Kim Il-Sung’s invasion, and later, they failed to predict PRC intervention. In each case, US analysts and policymakers widely shared the belief that the Soviet Union would not be willing to risk aggression by its client states, because this could escalate into a direct confrontation between the Soviet Union and the United States, precipitating World War Three. As it turned out, the Soviet Union quite effectively used client-state aggression to tie down the United States without risking direct escalation. This logic set the parameters of engagement for the entirety of the Cold War. However, it is useful to remember how, before they became a fact of life, these conventions were seen as risky and unlikely at the time.  

This bias also explains why US analysts frequently misquote and misinterpret Mao’s appraisal of the PRC-North Korea relationship as “close as lips and teeth.” A better translation of the statement is, “when the lips are gone, the teeth will be cold.”10Dan Miller, “China and North Korea: A Tangled Partnership,” China News, April 18, 2013, https://chinanews.net.au/2013/04/18/china-and-north-korea-a-tangled-partnership/. This emphasizes the point that the PRC and North Korea are two independent actors behaving in self-interest, and occasionally benefiting from sharing the same adversary. Mao supported Kim’s war in principle, but was agitated by the timing, which he (accurately) predicted would greatly delay the CCP’s designs on Taiwan. The Korean War played out as it did because North Korean leader Kim Il-Sung was able to launch his invasion of unification before Mao could launch his. Although greatly annoyed by this outcome, once North Korea lost the initiative and UN forces streamed across the 38th Parallel, the CCP felt compelled to intervene. Much to Beijing’s frustration, most of the PRC’s military advice to North Korea—which could have been helpful—went unheeded by the Korean People’s Army in the fall of 1950.11Charles Kraus, “Zhou Enlai and China’s Response to the Korean War,” Wilson Center, last visited December 11, 2022, https://www.wilsoncenter.org/publication/zhou-enlai-and-chinas-response-to-the-korean-war. Reflecting the difficulty of influencing North Korea’s policy choices, PRC Foreign Minister Zhou Enlai told his Soviet interlocutor in 1956 that the Korean Workers’ Party leadership “does not listen to the CCP advice 100%, and it does not listen to you [the Soviet Union] 30%.”12Sergey Radchenko, “‘We Do Not Want to Overthrow Him:’ Beijing, Moscow, and Kim Il Sung, 1956,” Wilson Center, last visited December 11, 2022, https://www.wilsonceter.org/blog-post/we-do-not-want-to-overthrow-him-beijing-moscow-and-kim-il-sung-1956.

The over-centralization of decision-making bias continues today, leading US analysts to discount a wide array of potential scenarios. It is important to remember that the PRC-North Korea alliance is not unconditional; an editorial in the CCP-run Global Times hinted that Beijing would “remain neutral” if North Korea launches missiles against the United States. This bias tends to feed into the US illusion of control.13Simon Denyer and Amanda Erickson, “Beijing Warns Pyongyang: You’re on Your Own if You Go After the United States,” Washington Post, August 11, 2017, https://www.washingtonpost.com/world/china-warns-north-korea-youre-on-your-own-if-you-go-after-the-us/2017/08/11/a01a4396-7e68-11e7-9026-4a0a64977c92_story.html. Some policymakers and analysts assess that the PRC controls North Korea, and they can at least reason with the PRC, which can prevent an escalation spiral on the Korean peninsula. But sharp disparities in Pyongyang and Beijing’s risk tolerance, threat perception, and strategic culture could result in coordination failure.

Another present bias could be considered under-centralization of decision-making. Unlike in the Korean War, when the United States forecast that the PRC would be restrained by the Soviets because of a mistaken presumption about their hierarchical relationship, US strategists now tend to write off a host of intervention scenarios based on the premise of mutual distrust. As the study illustrates, the mutual distrust at the heart of the modern PRC-North Korea relationship could actually enable, rather than inhibit, intervention and simultaneous conflict. The following list is not exhaustive, but it underscores the kind of analytical errors that are likely to follow when analysts engage in the natural, automatic tendency to group actors together and disregard the divergence in their interests, and in the tendency to underappreciate the extent to which convergence in interests can lead to simultaneous conflict. 

  1. In the case of a North Korean nuclear accident, domestic instability, or mistaken escalation, it is possible that PRC-North Korean disagreement could lead to PRC intervention, especially if there are spillover effects on the PRC side of the border, such as migration or radiation.
  2. North Korea could deliberately escalate on the peninsula against the PRC’s wishes, presenting a fait accompli that prompts PRC intervention.
  3. North Korea could take advantage of chaos resulting from PRC action on Taiwan before or after US intervention, and with or without coordination with the PRC.
  4. This bias could also lead to an underestimation of the PRC instigating a provocation on the peninsula to drag North Korea into conflict with the United States as a distraction while the PRC takes action against Taiwan.

The law of the instrument bias

The law of the instrument bias is also referred to as “Maslow’s Hammer.” Abraham Maslow was a psychologist who wrote, “If the only tool you have is a hammer, it is tempting to treat everything as if it were a nail.”14Abraham H. Maslow, The Psychology of Science: A Reconnaissance (New York: Harper and Row, 1966). The United States is more focused on using the tools at its disposal to respond to a Korea/Taiwan contingency than on thinking from the ground up: what are the most likely contingencies and what tools are optimally suited to address them? The cause of this bias is rooted in organizational and bureaucratic division of labor that separates resources and authorities and stifles holistic forecasting and planning. Both historically and contemporarily, this type of organizational or office-culture bias exacerbates and conceals the impact of cognitive biases. One Pentagon official explained, “Cognitive biases are deeply rooted in the division of labor between the combatant commands and the resources and authorities underneath them.”15Anonymous source in a telephone interview with Jonathan Corrado, June 16, 2022.

The official described multiple demilitarized zones (DMZs) or firewalls that prevent effective coordination and strategic creativity. This includes a DMZ dividing conventional warfighters and nuclear weapons and a DMZ dividing USFK and USINDOPACOM. These DMZs are primarily caused by a structural division of resources and authorities. Consequently, both sides of the DMZ struggle to plan with a recognition of the region’s high degree of integration. As one example, the Pentagon official described an unclassified wargame involving a PRC invasion of Taiwan with a map that excluded North Korea and featured South Korea as an island. This visually represents the way that North Korea is absent in the strategic imagination of planners in Taiwan scenarios. This is quite common both in and outside of government. In a recent Center for a New American Security PRC-Taiwan wargame, North Korean and Russian potential to distract was mentioned once but discussed no further.16Stacie Pettyjohn, Becca Wasser, and Chris Dougherty, Dangerous Straits Wargaming a Future Conflict over Taiwan, Center for a New American Security, June 2022, https://s3.amazonaws.com/files.cnas.org/CNAS+Report-Dangerous+Straits-Defense-Jun+2022-FINAL-print.pdf.

Wishful-thinking bias: “The PRC will help us with Korea”

White House officials tend to make statements to the effect of “The PRC should or will help us with North Korea.” This message often hits a wall in Beijing. In some instances, the motivation for said statements is to persuade and pressure the PRC to collaborate. It is reasonable to make performative statements to push for coordination with the PRC on a Korea scenario, but this must be coupled with high-level military and diplomatic dialogues on contingency planning. Absent direct and continuous communication on this issue, including between combatant commanders to address contingencies such as a North Korean nuclear accident, it is mistaken to assume that US-PRC coordination can be executed smoothly in the heat of the moment. 

An assumption of PRC cooperation could play into the hands of one of the PRC’s go-to strategies, the “self-defensive counterattack,” which employs a sudden, decisive blow to seize the initiative or change the facts on the ground. What will the US president actually do if the PRC intervenes in a manner antithetical to US and ROK interests? The psychological sting would be pronounced if the United States expects cooperation. It is prudent to recall President Truman’s reaction to PRC intervention in November 1950: “Because of the historic friendship between the people of the United States and China, it is particularly shocking to us to think that Chinese are being forced into battle against our troops.” 

Oversensitivity to consistency bias

This cognitive bias describes the tendency to expect continuity, and to overattribute causality when consistency is observed, even in a very small data set.17Jonathan Corrado, “Rethinking Intelligence Failure: China’s Intervention in the Korean War,” International Journal of Intelligence and Counterintelligence 36, 1 (2023), p. 214, https://www.tandfonline.com/doi/abs/10.1080/08850607.2021.1938905?journalCode=ujic20. In the Korean War, analysts discounted PRC intervention because the PRC refrained from entering the conflict at earlier junctures (Incheon landing, UN forces crossing 38th Parallel, etc.). This was understood as an indication of unwillingness to intervene. In actuality, this restraint was an intentional tactic to lure UN forces further north into the mountainous regions, to separate lines of communication and supply and handicap the fighting force. This, therefore, serves as another example of a cognitive bias that helps the efficacy of the self-defensive counterattack. Moving forward, there are numerous applications to the cases of North Korea and Taiwan, all revolving around an initial failure to perceive the security threat posed by an inciting incident, and then the use of deception and obfuscation by the adversary to feed its expectations of inaction or misdirection.   

The vividness criterion 

This cognitive bias describes the tendency “to favor “vivid, concrete and personal” information over abstract and dull information, even when the vivid information is less causally relevant to the outcome under assessment.”18Id., p. 213. For instance, “on November 24, 1950, General Douglas MacArthur flouted the advice of his subordinates by boarding a propeller plane and flying the length of the Yalu, putting himself within range of enemy fire. Gazing down at the borderland below, MacArthur declared that he detected no sign of PRC presence.”19Id., p. 210. Just a few days later, hundreds of thousands of PLA troops streamed over the border in a full-frontal assault. This sort of bias does not exclusively apply to leaders. It also can apply to analysts, as vivid data points can make a particularly strong impression and distract from other relevant evidence pointing toward a contrary conclusion. 

Application of historical lessons

The alchemy of cognitive and organizational biases is uniquely suited to advantage a PRC self-defensive counterattack strategy. Politicization, obfuscation, and organizational biases exacerbate and disguise the impact of cognitive biases. Importantly, an assessment of capabilities alone will not only fail to yield an effective forecast but will also hamstring contingency planning. A better understanding of the enemy informs a broader range of policy response solutions, including deterrence, dissuasion, reassurance, and the “artful orchestration of political, military, and economic instruments of power.”20Paul K. Davis, et al., “Influencing Adversary States Quelling Perfect Storms,” RAND, 2021, https://www.rand.org/pubs/research_reports/RRA161-1.html.

To overcome the mirror-imaging bias, there is a need for deep regional and country-specific expertise, including knowledge of language, history, and strategic culture. Structured analytic tradecraft, particularly Red Team exercises, should take advantage of said country expertise. However, participants in the study noted a tendency to sideline Red Team insights from mainline authoritative assessments that serve as the foundation for decision-making. Participants also noted that there is a need to address these biases everywhere they exist—within both the analysts and the customers of their products. To address this, one participant suggested the creation of formal products to help analysts and policymakers recognize and address the deleterious effects of cognitive and organizational biases on these issues.  

These techniques should be buttressed by other methods to cope with the high degree of uncertainty, such as integrating quantitative analysis to complement and cross-check qualitative methods.21Daniel Irwin and David R. Mandel, “Improving Information Evaluation for Intelligence Production,” Intelligence and National Security 34, 4 (2019), https://www.tandfonline.com/doi/abs/10.1080/02684527.2019.1569343. The limits of intention analysis underscore the need for robust systems for monitoring capabilities and strategic-warning indicators to inform continuing reassessments based on new capabilities and situational factors. Finally, cognitive biases thrive wherever bureaucratic divisions of labor blunt more holistic views of the problems presented. In the case of PRC intervention on the Korean peninsula and/or a simultaneous conflict involving a PRC invasion of Taiwan and North Korean aggression, the impact of the law of the instrument bias is particularly severe. 

Author biography

Jonathan Corrado is Director of Policy for The Korea Society. He was previously a non-resident James A. Kelly Fellow at Pacific Forum, an Emerging Leader at the National Committee on American Foreign Policy, and a contributor to NK Pro, and he has published peer-reviewed articles and analyses in a wide range of journals and outlets. Beginning in Fall 2023, Jonathan will teach a class at SUNY Stony Brook University titled, “North Korea: State, Society, Diplomacy, and Security.” Jonathan received an MA from Georgetown University’s Asian Studies Program in the Walsh School of Foreign Service and a BA in anthropology and philosophy from the University of Maryland College Park.

This piece was written in collaboration with the Atlantic Council Indo-Pacific Security Initiative as a supplement to the main report, “The United States and its allies must be ready to deter a two-front war and nuclear attacks in East Asia.”

Image: Army Gen. Douglas MacArthur, U.N. Command commander in chief, observes the naval shelling of Incheon, South Korea, from the USS Mount McKinley, Sept. 15, 1950. Photo by: National Archives