Where the stakes are high, political leaders have few incentives to take intelligence warnings seriously.
On 25 March, the Foreign Policy magazine carried an article that termed the coronavirus outbreak the worst intelligence failure in US history — more glaring than Pearl Harbor and the 9/11 attacks. The article in its entirety, however, suggested more than just intelligence failure and blamed the Trump administration for turning a blind eye towards repeated intelligence warnings throughout January and February. By the time US President Donald Trump declared himself a “wartime president” on 21 March, the total numbers affected by the virus had reached 35,000 and the casualty rates had risen to over 450. At the time of writing this article, the total casualty count of COVID-19 in the US had crossed 50,000. The cumulative human and economic costs of the pandemic was bound to generate speculations of intelligence failure among the public, whilst the intelligence community and policymakers were certain to shift the blame onto each another. Nevertheless, it is clear that the COVID-19 surprise reiterates three important adages held by the academic intelligence studies community.
Henry Kissinger once remarked that “there is no such thing as friendly intelligence agencies. There are only intelligence agencies of friendly powers.” This has been an age-old maxim in the intelligence world where alliances and partnerships do not forbid agencies from spying on each other. For instance, during the 20th century, despite shared interests, the French and Israeli intelligence services were found spying on the US. The untrustworthy nature of the intelligence business can get more treacherous in areas where one agency is disproportionately reliant on the other for information. During the 1970s, the CIA’s total dependence on the Iranian secret service, the SAVAK, for information on Iran’s internal political situation is cited as the single biggest factor in the US’s failure in predicting the 1979 Islamic Revolution. The Iranian episode shares a close similarity with the COVID-19 surprise and bears an important lesson.
On 9 January, the World Health Organisation (WHO), for the first time, issued a statement indicating a “cluster of pneumonia cases in Wuhan, China.” Two months later, on 11 March, the WHO declared COVID-19 a pandemic, although the tragic implications of the outbreak were clearly visible by then. The delay in the WHO’s provision of an alert is blamed on the organisation’s blind acceptance of Chinese misinformation that there was no “proof of human-to-human transmission.” Whether the WHO can be entirely blamed for this mishap is questionable since the International Health Regulations 2005 mandates member nations to share information on potential health threats with probable international implications, which China evidently violated. The larger question, therefore, is why did China refuse to share reliable information on the nature and scale of the coronavirus outbreak? The reason lies in the nature of intelligence sharing practiced by authoritarian regimes.
Academic studies on authoritarian regimes highlight that there is a tendency for the regimes to emphasise both real and perceived internal threats. However, what is less emphasised is that, notwithstanding the excessive internal focus, there has also been a tendency to downplay the scale of internal troubles. In order to command internal and international legitimacy, autocratic regimes have at times tended to display artificial power and control. It is for this reason that in 1979 the Shah and SAVAK pretended to have things under control while actual events on ground were spiraling out of hand.
Similarly, in the present crisis, China simultaneously downplayed the statistical data on the coronavirus while exaggerating the outcome of its responses to the outbreak. According to media reports, US officials attribute this concealment of information and exaggeration of results to “a desire by the country’s leaders to avoid embarrassment.” Chinese President Xi Jinping, staking his reputation on the promise of reestablishing China as the ‘Middle Kingdom,’ was less likely to allow information on COVID-19 to negatively impact the Chinese economy. Also, a Bloomberg report had warned that the bi-decadal Communist Party meeting scheduled for 2022 could raise questions about Xi’s continuance in power if unemployment increased. Thus, Chinese diplomats around the world were directed to “field questions about the virus and rebut what they believe to be misleading reports.” Therefore, where there is a serious risk of reputational damage, authoritarian regimes cannot be trusted to share reliable information.
The logical riposte to relying on foreign agencies for information is to improve independent intelligence acquisition mechanisms. This brings us to the second adage, neglecting human intelligence (HUMINT) can be perilous. While relying on the WHO to provide warnings on the emerging pandemic, the US’s epidemic intelligence agency, Centre for Disease Control and Prevention (CDC), also failed to provide timely warnings on the COVID-19 outbreak. The most important reason for this failure lies in the operational methodology of these agencies. Both the WHO and the CDC, despite a degree of collaboration with medical staff and epidemiologists around the world, mostly rely on technical means for gathering information on health threats. Whilst relying on Electronic Health Records, the agencies employ technologies that mine data from the internet and social media to identify and investigate outbreaks. The challenge, however, lies in the fact that the internet and social media in China are state controlled and the emerging information is, thus, unreliable.
Citizen Lab, a Canada-based internet censorship research organisation, published a report on 3 March exposing how Chinese social media had been actively censoring information related to COVID-19 between December 2019 and February 2020. According to the report, every message was monitored for blacklisted keywords at both the user’s application level and the remote server level to censor crucial information. It is probably for this reason that the WHO and CDC missed critical inputs required for analysis, while BlueDot, a Canadian private firm that uses big data and artificial intelligence, predicted the outbreak nine days before the WHO. BlueDot relied on a combination of information from human sources plus commercial flight travel details, animal and insect population data, and online news portals ning 65 languages to draw its analyses. However, the fact that the Taiwanese government raised an alarm before BlueDot highlights the vital role HUMINT can play.
During late December 2019, Taiwanese doctors learnt from their colleagues in China that the medical staff treating the patients in Wuhan were falling sick, which was a clear sign of human-to-human transmission. The Taiwanese epidemic intelligence officials ostensibly used this data to analyse additional inputs like pneumonia cases being “isolated for treatment.” In an expression of displeasure over the WHO’s reluctance to accept Taiwan’s warnings, the Taiwanese Health Minister Chen Shih-chung questioned “whether being treated in isolation was not a warning.” In fact, even the CDC learnt of the outbreak in Wuhan through a human source, not via technical platforms. On 3 January, CDC director Robert Redfield received a call from his counterpart in China who reported that a “mysterious respiratory illness was spreading in Wuhan.” It was this call that prompted the US intelligence community to include warnings on the coronavirus in all subsequent presidential daily briefs. Finally, it is these presidential daily briefs that Trump failed to pay attention to until end-March, which brings us to the third axiom — surprises are inevitable because political leaders will resist contrarian intelligence.
In the study of military surprises, intelligence scholars have observed a natural tendency among policymakers to disregard estimates to preserve resources and reduce military expenditure. Even in non-military surprises, it is arguable that policymakers resist acceptance of intelligence, more so when the political and economic stakes are high. This is once again evident in both the 1979 Revolution and COVID-19 cases with striking similarities.
In 1979, Iran was at the heart of the US’s regional geopolitics and anti-communist strategy. Likewise, Tehran was also central to Israel’s ‘peripheral alliance strategy’ and oil supply. Due to such high political and economic stakes, the CIA never mustered the courage to warn former US President Jimmy Carter of the Shah’s weakening position, while the Mossad’s warnings in 1977 were turned down by Tel Aviv. What the respective policymakers did, in order to avoid accepting the intelligence warnings, was invest a high degree of faith in the Shah’s capacity to establish order. Similarly, given the 2020 presidential election and Trump’s re-election calculus, accepting the presidential daily brief warnings were virtually impossible, which paved the way for increased faith in China’s response capacity.
The latter half of 2019 saw serious speculations among pundits that Trump’s electoral fortunes depended largely on the stock market, which was inextricably being linked to the trade war with China. The ‘Phase-One’ trade agreement signed in mid-January set in motion an unprecedented level of admiration of China and Xi. Trump’s faith in, and appreciation of, Xi’s ability to tackle the coronavirus outbreak comes as a consequence of wanting to avoid the politico-economic ramifications attached to accepting intelligence warnings on COVID-19. For instance, in the ‘Phase-One’ deal, China promised to buy $200 billion worth of American products and services over the next two years, with Beijing promising to buy $36.2 billion worth of US agricultural product in the first year. There were two noteworthy outcomes of this — one, the stock market found a fresh boost; two, American farmers, a potential voter base for Trump, found certain relief after being at the wrong end of the US-China trade war. However, with the COVID-19 outbreak, contrary to the US Agriculture Secretary Sonny Perdue’s promising predictions, independent experts have “cast further doubts on China’s ability” to sustain such drastic increase in buying of US farm goods. Considering such dire outcomes in areas that could severely impact the upcoming elections, it should come as little surprise that Trump spent January and February applauding Xi and China at least 15 times, with a hope that normalcy would be restored. Like in 1979, in the current crisis, the US president had more to gain from presumed stability than accept intelligence warnings that seemed unnecessarily alarming. Hence, where the stakes are high, political leaders have few incentives to take intelligence warnings seriously.
In sum, looking at the COVID-19 episode through the prism of intelligence-surprise dynamics, it is evident that its key characteristics are largely reiterations of age-old theories held by scholars and practitioners. The US, or any nation considering intelligence reforms to bring more attention to health security threats, will benefit from bearing these three points in mind.
The views expressed above belong to the author(s). ORF research and analyses now available on Telegram! Click here to access our curated content — blogs, longforms and interviews.
Dheeraj is doctoral candidate in Intelligence Studies at the University of Leicester U.K. My research focuses on the cultures of intelligence and strategic surprises. I ...
Read More +
PREV
