‘Never Thought They Could Pull Off Such an Attack’: Prejudice and Pearl Harbor

December 7, 2021

“I never thought those little yellow sons-of-bitches could pull off such an attack, so far from Japan.” So confessed Adm. Husband Kimmel, former U.S. Pacific Fleet commander, to a member of the congressional Pearl Harbor investigation. Our 2021 reaction to Kimmel’s words is to focus on their racist invective. Yet, there’s an even more obvious problem with Kimmel’s statement. He was simply wrong, and he should have known better. Adm. Kimmel, Gen. Walter Short (the U.S. Army Hawaiian Department commander), and other military leaders on Oahu fundamentally underestimated and misunderstood the threat of Japanese carrier aviation to their detriment, and to the detriment of national security. Their underestimation and misunderstanding was deeply grounded in racial prejudice and went unchallenged in an environment of ethnocentric groupthink. Even in acknowledging his mistake, Kimmel’s words suggest he remained angrily defiant that the attack materialized the way it did — as if it were unfair that reality did not conform to his prejudices. Ethnocentric and racist attitudes and actions are objectionable for their harmful effects on the groups they malign, of course, but we should never forget that they are fundamentally stupid because they are factually incorrect. Prejudice literally means passing judgement prior to possessing adequate information. For national security professionals, prejudice is dangerous. Prejudice is fatal. The prejudicial unwillingness of Kimmel, Short, and others to posture adequately against a potential Japanese aerial attack was fatal 80 years ago today. Pearl Harbor is a concrete example that demonstrates how devastating ethnocentric bias brought on by largely homogenous institutions can be.



While there are myriad reasons for the failure of U.S. strategic warning regarding Japan’s attack on Pearl Harbor, it is clear today that specific warning of the attack days or weeks in advance was not possible given the state of U.S. intelligence and effective Japanese operational security. While signals intelligence, diplomatic reporting, and even open-source information showed that war in the Pacific was likely by late November 1941, there was no clear diagnostic evidence of a Japanese strike on Pearl Harbor. Effective defense against the attack would have come only via tactical warning from Hawaii itself and from alert defenders prepared to detect and engage a capable enemy. Yet, leaders responsible for the U.S. defense posture incorrectly focused on the internal threat of sabotage and prepared almost solely for a Japanese military attack much farther west in the Pacific, eliminating any meaningful preparation for a surprise direct attack on Pearl itself. Psychologist Irving Janis attributes the failure of military commanders at Pearl Harbor to groupthink: This small, like-minded group shared each other’s factually unsubstantiated beliefs without question — beliefs shaped by ethnocentric contempt for the Japanese — and fell victim to their collective lack of vigilance against an external threat. This groupthink thrived in a racially homogenous national security workforce. Today, increased diversity and inclusion in the national security workforce is the single best method for ensuring such flawed assumptions do not blind the U.S. to future threats.

Pearl Harbor is not the only warning failure ethnocentric groupthink has encouraged. Israelis were equally contemptuous of their Arab adversaries in 1973, discounting indications of an impending surprise attack because they discounted Arab military skill and organization. Similar attitudes toward Islamist terrorism played a role in Western underestimation of the al-Qaeda threat prior to 9/11. Moreover, ethnocentric groupthink went both ways in the Pacific war: Japan waged a war of conquest partially based on the country’s supposed racial and ethnic superiority over its neighbors. Mistaken Japanese hopes for long-term strategic success against America were grounded in ethnocentric contempt for the West.

To be sure, America’s warning failure at Pearl Harbor was not solely the product of ethnocentrism. Stove-piped intelligence, limited collection opportunities, and effective Japanese security shielded the strike from discovery as much as anything. Aircraft carriers striking silently across 3,000 miles was certainly a low-probability, high-impact threat in terms of risk calculus (though there was recent precedent, admittedly from shorter range). Kimmel, Short, and their colleagues should not be vilified for failing to deduce the location and timing of Japan’s carefully planned secret strike. Yet, ethnocentric blinders — which led to the focus on phantom saboteurs and an unwillingness to expend resources to guard against Japanese aircraft — forestalled effective tactical warning and initial defense against the attack. Commanders adopted none of the defensive measures that should have followed a prudent, objective assessment of aviation threats facing Oahu — measures that were adopted at other distant American targets. Baseless assumptions about “fifth columnists” and Japanese technical incompetence should have been challenged, reevaluated, and cast aside by national security professionals. The fact that they were not highlights the dangers of ethnocentric groupthink in a homogenous workforce — both in 1941 and 2021.

Bigots and Opinion-Makers Alike

Kimmel and Short focused on two threats to Hawaii. First, sabotage: a fear tied to yellow peril subversion paranoia at the time — but a threat that simply did not exist, and not for lack of investigation. Second, fatigue: of their men, planes, ships, and radars, which they expected could be called to fight the Japanese military soon, but much farther west in the Pacific theater. They and those around them focused on these threats, assuming the “little yellow bastards” could not pose a real military threat to Pearl Harbor but that ethnic Japanese in Hawaii must be a danger. Fear of saboteurs and spies was especially regrettable given the later testimony of Japan’s chief operative in Hawaii, based on his own attempts to recruit sources, that “the Japanese population of Hawaii we found essentially loyal to the United States.” As modern scholarship demonstrates, there was no discernible “fifth column” threat from Japanese-Americans before or after Pearl Harbor. Short and his commanders also found no actual evidence of such a threat. They merely assumed that Americans of Japanese descent must pose a threat. And so on Dec. 7, U.S. aircraft were parked wingtip to wingtip, stripped of fuel and munitions to stymie saboteurs. Antiaircraft shells were carefully locked away. No aerial patrols stressed aircrew or airframes. Novices operated the radars — not to provide warning but to fill limited training hours. Pilots on rotational training duty, not intelligence officers, manned alert desks. No torpedo nets protected the ships in port: Planners assumed Japanese torpedoes could not run in Pearl’s shallow waters. The contempt Kimmel, Short, and their subordinates felt for even the faintest chance of Japanese aerial attack influenced each decision. This contempt was born and fostered within a racist miasma in which Kimmel and Short felt they knew the “little yellow bastards” very well. In fact, they actually knew relatively little about their enemy but encountered no corrective pushback until the attack itself.

The U.S. military did have some men (and they were all men) who understood the enemy better. Arthur McCollum was born the son of missionaries in Nagasaki and studied Japanese language and culture throughout his naval career. On the eve of Pearl Harbor, he headed the Office of Naval Intelligence Far East Section. Edwin Layton also spent much of his career studying Japan, including during assignments in Tokyo and in China. He was a combat intelligence officer with the Pacific Fleet in December 1941. Joe Rochefort worked for Layton as a cryptanalyst and Japanese linguist and, like Layton, Rochefort was the graduate of a program that detailed dozens of officers to study in Japan. Unlike Kimmel, Short, and others in command positions, these men had direct sociocultural, political, and linguistic knowledge of Japan. Individually, each man pressed for greater awareness of the Japanese threat and an improved warning posture on Oahu in the months and weeks prior to Dec. 7. Yet, they represented the extent of Japanese expertise in the U.S. military: a handful of white men whose unique backgrounds and careers provided them direct contact with foreign language and culture. Though such men understood Japanese military modernization, operational security, and strategic ambitions, their presence alone was insufficient to break the stranglehold ethnocentric groupthink held on Oahu’s defenders. Organizations don’t reap the benefits of diversity and inclusion and effectively combat groupthink without actual representation of otherwise excluded groups. Though McCollum, Layton, Rochefort, and a few dozen others with similar backgrounds represented a small “diversity of thought” in the U.S. military’s perspective on Japan, this was not the same as the kind of service-wide diversity and inclusion that might have broken the ethnocentric groupthink plaguing Hawaii’s defenders.

According to David Kahn,

Americans looked upon the Japanese as bucktoothed, bespectacled little yellow men, forever photographing things with their omnipresent cameras so they could copy them. Such opinions were held not only by common bigots but by opinion-makers, as well.

These stereotypes gained traction among military leaders in part because pseudo-experts advanced them as facts, and they went unchallenged in a homogenous military environment. The recently retired Director of Naval Intelligence William Puleston wrote in his well-received Armed Forces of the Pacific (1941) that American pilots simply had “more natural aptitude for flying” than their Japanese counterparts. Fletcher Pratt, a wargame designer and popular author specializing in military affairs and science fiction, asserted in Sea Power and Today’s War (1939) that

every observer concurs in the opinion the Japanese are daring but incompetent aviators … the Japanese as a race have defects of the tubes of the inner ear, just as they are generally myopic. This gives them a defective sense of balance.

Pratt had no substantive background in aviation or naval affairs, nor was he (unsurprisingly) a scholar of Japanese history or culture — or basic human anatomy. Yet, Pratt’s books were popular with the public — and with the military. Sea Power and Today’s War received positive reviews in Foreign Affairs and various military journals and was regarded in national security circles as a serious work.

Steve Twomey’s Countdown to Pearl Harbor yields other examples: In March 1941, a popular aviation magazine explained that little was known about Japanese aircraft not because of operational security (the actual reason), but because the Japanese “are, by nature, imitators and lack originality … this blend of characteristics makes them conscious of their failings, and they seek to hide them from the world.” Rather than being critiqued or ignored by national security professionals for its shallow, racist perspective, this assessment was reprinted in the June 1941 issue of the U.S. Naval Institute’s Proceedings. Kimmel’s cruiser force commander, Adm. Wilson Brown, took it as an article of faith that the Japanese were inherently poor pilots. His evidence was a conversation with a Singer Sewing Machine Company executive, who told Brown it was company policy not to fly on Japanese airlines because of the notoriously poor airmanship of the Japanese.

Such “as everyone knows” false facts (see also: “a lot of people are saying”) swayed many a decision-maker without a shred of hard evidence and went unchallenged thanks to ethnocentric groupthink. Not even the actual Japanese attack overcame it. Several American newspaper and radio outlets insisted Germans must have directed or conducted the attack (as Japan was obviously incapable of such a feat). Douglas MacArthur, U.S. commander in the Philippines, was certain the attacking pilots must have been white mercenaries.

Many of the drivers of ethnocentric groupthink surrounding Pearl Harbor encourage similar prejudicial behavior today. “Experts” like Pratt and Puleston, with their eugenics-laden assertions, are forerunners of modern pseudo-experts who enjoy undue influence over large groups of readers and viewers thanks to the popularity of their platforms and despite their shallow credentials. The widespread belief in Japanese physical and mental shortcomings owed much to a proto-meme culture in which racist political cartoons (including those by the future Dr. Seuss), newsreel animations, and repeated movie tropes saturated popular culture. These cartoons and trope villains reinforced and normalized nonsensical beliefs about the Japanese. In this very same fashion, 21st-century jingoist punditry and racist memes pose a similar threat to effective analysis and decision-making. Their influence works against a clear-eyed, sober assessment of threats and opportunities. Indeed, the modern digital information environment exacerbates these effects, empowering pseudo-experts and overwhelming consumers’ ability to think critically about data sources and biases.

Ethnocentrism also warped U.S. technical intelligence. U.S. attachés gathered accurate technical data on the new Japanese Zero fighter plane in January 1941 but the Navy’s Bureau of Aeronautics discounted the data. American technicians could not understand (and would not accept) the details of a lightweight fighter that did not conform to American aircraft design philosophies. Not only were Japanese technologies seen as generally inferior, but ethnocentric consideration of their technical details drove assumptions that certain threats were essentially impossible. It did not occur to most U.S. decision-makers that Japanese weapons might be able to do things American-made ones couldn’t. Lacking analysts with Japanese weapons expertise, the Office of Naval Intelligence had U.S. engineers assess the potential threat from Japanese weapons systems. U.S. engineers assumed Pearl Harbor was safe from aerial torpedo attack because the harbor’s roughly 45-foot depth was too shallow for aerial torpedoes to drop and run. This was a limitation of American aerial torpedoes, and American engineers assumed Japanese torpedoes certainly wouldn’t perform better. The Japanese military extensively modified their torpedoes to run at a shallower depth, which they did to deadly effect on Dec. 7, slamming into the sides of American ships unprotected by torpedo nets.

It is compelling to consider the potential impact leaders with more diverse perspectives might have had for U.S. defenses on Oahu. Men like McCollum, Layton, and Rochefort held important positions (disproportionately in intelligence ranks) and benefited from unique life experiences. Yet, what if such positions of national security responsibility had been open to Americans of East Asian descent? What if women and ethnic minorities were not systemically excluded? Might a more realistic and objective evaluation of Japanese carrier aviation (and less preoccupation with sabotage and subversion) have tempered ethnocentric groupthink on Oahu?

Ironically, the Allies achieved a considerable comparative intelligence advantage over Axis foes during the war through gender diversity and inclusion. Large-scale employment of women in imagery intelligence and signals intelligence operations, for example, provided both the raw staffing these laborious tasks required (ultimately several thousand women worked at Medmenham, Bletchley Park, Arlington Hall, and elsewhere) and the pockets of genius that all newly included demographics bring. Consider critical figures like Constance Babington-Smith, whose imagery analysis was critical to Allied understanding of the German V-1 threat; Genevieve Grotjan, who made key breakthroughs in deciphering Japan’s diplomatic “Purple” encryption system; and others like Elizebeth Friedman and Ann Caracristi. This was not tokenism or political pandering: This was the marshalling of national resources in time of crisis. Yet it had its limits. These were overwhelmingly white women, disproportionately from middle- and upper-class Christian backgrounds. Imagine what advantages might have been realized via large-scale inclusion of capable people regardless of gender, religion, sexual orientation, race, or national origin.

Pearl Harbor Wasn’t an Outlier

The details of the warning failure at Pearl Harbor illustrate how toxic ethnocentrism, the byproduct of a homogenous workforce, taints analysis and decision-making in various ways. A lack of diversity fosters devastating shared blind spots, skewing the foundations upon which every process is built. Without diversity, some flawed beliefs go unchallenged. Pearl Harbor demonstrates the dangerous results of unchallenged ethnocentric assumptions. Pervasive ethnocentrism and racism lead to disastrous outcomes when they supplant real evidence or lead one to underestimate a foe. These dynamics do not merely reflect the prevailing racial attitudes of the American military of the 1940s. They illustrate how a lack of diversity and inclusion in the national security workforce could have lethal consequences today.

The private sector has no shortage of industry research that demonstrates how a lack of diversity and inclusion negatively impacts organizational performance. National security organizations are similarly vulnerable. Since national security leaders have made the argument that diversity and inclusion can strengthen their organizations, extrapolations from these industry findings should be further explored for their applicability.

While organizations lacking diversity risk prejudicial blind spots, teams comprised of people from diverse backgrounds are more likely to mitigate this bias thanks to multiple perspectives drawn from personal experiences. This is especially true in the realm of international affairs. Ethnocentrism can damage analytical tradecraft through groupthink, mirror-imaging, and the misreading of cultural norms and behaviors. Additionally, bias due to a lack of technical workforce diversity can lead to prejudicial data-conditioning for machine intelligence. Artificial intelligence models can be encoded with human prejudices and personal values when they are embedded in the data that feeds and trains algorithms. Human influence can, after all, still initiate machine learning. In the national security and intelligence realms, for example, baked-in ethnocentric biases in artificial intelligence tools can lead to ineffective and inaccurate natural-language processing and facial recognition. A 2011 National Institute of Standards and Technology study, for example, found that Western-designed algorithms “recognized Caucasian faces more accurately than East Asian faces and the East Asian algorithm recognized East Asian faces more accurately than Caucasian faces.”

Training and education in the national security community already addresses diversity and inclusion and issues of racial prejudice and ethnocentrism. The national security diversity and inclusion conversation needs to become more pragmatic, focusing on strategic competitors like China. It is critical, moreover, to note that inclusion creates a more substantial effect than diversity alone. It’s not enough to recruit more women and more people of color. While diversity focuses on workforce demographics, inclusion creates an environment with a sense of belonging that enables diversity to flourish. Research on human resources and organizational behavior shows that diversity policies risk resulting in failure if irreconcilable divisions among diverse groups are not addressed, resulting in dysfunctional relationships and poor performance.

National security decision-makers must be willing to incorporate more varied sources into their assessments to better avoid strategic surprises from peer competitors. Improved diversity and inclusion can drive better consideration of possible (if unlikely) threats by inhibiting groupthink and reducing cognitive biases, resulting in greater objectivity. Conversely, research shows that homogenous groups are less able to recognize the value of contrary information, leading to poor decision-making. Diversity and inclusion increase the integration of broader perspectives, reducing shared blind spots. Inclusion fosters new paradigms and heuristics that may have been absent within an otherwise homogeneous organization or team. Studies of racially diverse groups show that social differences generate teams more likely to anticipate differences of opinion, driving them to integrate multiple perspectives while building consensus.

The Japanese attack on Pearl Harbor offers other useful insights for today. As with Japan in 1941, Western powers have studied China as a possible adversary for decades. Over those same decades, major advances in technology and tradecraft have turned China from a possible adversary into a peer competitor. As was the case with Japan in 1941, longstanding assumptions regarding Chinese goals, limits, and capabilities may no longer be valid. In particular, while espionage and intellectual property theft is a serious problem, too many frame China as a copycat threat — perhaps, unsurprisingly, given some of the stereotypes that plagued America’s analysis of Japan in 1941. Not only would diverse national security teams be more likely to challenge such assumptions, they are also more prone to be objective while scrutinizing facts, and “facts.” Organizations not strictly tied to shared beliefs are more willing to consider alternatives that challenge conventional thinking. “2-D” diversity (moving beyond mere racial and gender diversity toward a combination of intrinsic traits and qualities gained from experience) further spurs innovation when present within organizations. This is the type of team where ideas outside the norm can be heard, supported by leadership, and allowed to develop.

Too many critics frame these topics as mere political correctness. National security leaders should appeal to pragmatism when it comes to diversity and inclusion, emphasizing the strategic advantage (and not just the obvious morality) of these traits. To avoid being dismissed as another buzzword, diversity and inclusion training programs should demonstrate pragmatic benefits by focusing on how diversity provides distinct advantages over more homogenous adversaries like China and Russia. Diversity and inclusion make it possible to build stronger, more accurate AI-enabled national security tools — and there is essentially no national security function or role that does not touch on an AI-derived technology in some manner.

Finally, it’s critical to start from the beginning. Much of the private sector’s focus on addressing diversity and inclusion is rightly aimed at recruitment. That focus should be the same for national security institutions, especially given their talent-management challenges. Retention is also an ever-present challenge for the military and government. Ensuring an inclusive environment that fosters belonging should be considered a national security imperative.

There is a concerning tendency among some to discuss the factors that lead to the ethnocentric groupthink displayed before Pearl Harbor in terms of personal opinions and hurt feelings, as if overt racism were merely impolite — and to dismiss diversity and inclusion as tokenism or wokeness run amok. National security professionals need to remember that ethnocentrism and racism aren’t just politically incorrect or morally objectionable. They’re fundamentally, inherently stupid. Ethnocentrism is dangerous. Ethnocentrism threatens national security. Diversity and inclusion fight ethnocentrism.



Maj. Caesar “Harley” Nafrada is a U.S. Air Force Reserve officer with more than 20 years of experience in intelligence collection and analysis. He is currently chief of diversity and inclusion for the Air Staff’s A2/6 Intelligence, Surveillance, Reconnaissance, and Cyber Effects Directorate at the Pentagon. Maj. Nafrada holds a bachelor’s degree from the University of Florida and a master’s degree in strategic intelligence from the National Intelligence University.

Joseph Caddell is an adjunct assistant professor with the National Intelligence University, where he teaches graduate courses on intelligence collection, geospatial intelligence, and U.S. intelligence history. His research and analysis has been published in Intelligence and National Security, the International Journal of Intelligence and CounterIntelligence, Studies in Intelligence, and War on the Rocks.

All statements of fact, analysis, or opinion are the authors’ and do not reflect the official policy or position of the National Intelligence University, the Department of Defense, or the U.S. government.