T-shirt diplomacy. Photographer: Tim Rue/Bloomberg
A little French word that used to play a big role in global politics is poised for a comeback: detente. The word was first used as a diplomatic term in the early 1900s, for example when the French ambassador in Berlin attempted — in vain as it proved — to improve his country’s strained relationship with the German Reich, or when British diplomats attempted the same thing in 1912. But detente became familiar to Americans in the late 1960s and 1970s, when it was used to describe a thawing in the Cold War between the U.S. and the Soviet Union.
I have argued since last year that the U.S. and the People’s Republic of China are already embroiled in Cold War II. President Donald Trump did not start that war. Rather, his election represented a belated American reaction to a Chinese challenge — economic, strategic and ideological — that had been growing since Xi Jinping became general secretary of the Chinese Communist Party in 2012.
Now, Joe Biden’s victory in the U.S. presidential election (the various legal challenges to which will achieve nothing other than to salve Trump’s huge but hurt ego) creates an opportunity to go from confrontation to detente much sooner than was possible in Cold War I.
The Chinese government waited until Friday before offering Biden congratulations on his victory. Put that down to caution. As during the election campaign, Xi and his advisers have been striving not to provoke Trump, for whom they have come to feel a mixture of contempt and fear. But a few unofficial voices have ventured to express what Xi doubtless thinks. Wang Huiyao, the president of the Beijing-based Center for China and Globalization, said last week that he hoped a Biden administration would “provide China and the U.S. … with more dialogue and cooperation channels concerning energy saving and emission reduction, economic and trade cooperation, epidemic prevention and control.” Speaking at the same event, the former foreign vice-minister He Yafei talked in similar terms.
Such language might be expected to a ring a bell with the Democratic Party’s throng of foreign policy experts, who have spent the past four years — from the Brookings Institution to the Aspen Strategy Group to Harvard’s Kennedy School — bemoaning Trump’s assault on their beloved liberal international order.
It seems pretty clear that Biden himself would gladly return to the days of the Barack Obama administration, when his meetings with Xi, Premier Li Keqiang and other Chinese leaders were all about the “win-win partnership” that I used to call “Chimerica.” Eight years ago, he was pictured beside Xi holding up a T-shirt with the slogan “Fostering Goodwill Between America & China.”
Asked about superpower competition at an event in Iowa in 2019, Biden replied: “China is going to eat our lunch? Come on, man. They can’t even figure out how to deal with the fact that they have this great division between the China Sea and the mountains in the East, I mean the West. They can’t figure out how they’re going to deal with the corruption in the system. I mean, you know, they’re not bad folks, folks. But guess what? They’re not competition for us.”
Luckily for Biden, he had few opportunities to say more in this nonsensical vein, as this year’s election campaign — including both presidential debates — scarcely touched on foreign policy, depriving Trump of the opportunity to point out how much more in touch he is with public sentiment, which has grown increasingly hostile to China since Biden left office four years ago.
Detente should not be confused with amity. Whatever comes of the diplomacy of the new presidency, it is unlikely to be a new era of Sino-American friendship. Detente means reducing the tensions inherent in a cold war and reducing the risk of its becoming a hot one.
“The United States and the Soviet Union are ideological rivals,” wrote Henry Kissinger, who was in many ways the architect of detente in the 1970s. “Detente cannot change that. The nuclear age compels us to coexist. Rhetorical crusades cannot change that, either.”
For Kissinger, detente was a middle way between the appeasement that he believed had led to World War II, “when the democracies failed to understand the designs of a totalitarian aggressor,” and the aggression that had led to World War I, “when Europe, despite the existence of a military balance, drifted into a war no one wanted and a catastrophe that no one could have imagined.”
Detente, Kissinger wrote in his memoir, “The White House Years” — published in 1979, 10 years before the effective end of the first Cold War — meant embracing “both deterrence and coexistence, both containment and an effort to relax tensions.”
Today, it is the U.S. and China who find themselves — as Kissinger observed in an interview with me in Beijing last year — “in the foothills of a Cold War.” As I said, this Cold War was not started by Trump. It grew out of China’s ambition, under Xi’s leadership, to achieve something like parity with the U.S. not only in economics but also in great-power politics. The only surprising thing about Cold War II is that it took Americans so long to realize that they were in it. Even more surprising, it took a maverick real estate developer turned reality TV star turned populist demagogue to waken them up to the magnitude of the Chinese challenge.
When Trump first threatened to impose tariffs on Chinese goods in the 2016 election campaign, the foreign policy establishment scoffed. They no longer scoff. Not only has American public sentiment toward China become markedly more hawkish since 2017. China is one of few subjects these days about which there is also a genuine bipartisan consensus within the country’s political elite.
Unlike the president-elect himself, the members of Biden’s incoming national security team have spent the past four years toughening up their stance on China. In Foreign Affairs this summer, Michele Flournoy, the favorite to become secretary of defense in the new administration, argued that “if the U.S. military had the capability to credibly threaten to sink all of China’s military vessels, submarines and merchant ships in the South China Sea within 72 hours, Chinese leaders might think twice before, say, launching a blockade or invasion of Taiwan.”
Flournoy wants the Pentagon to invest more in cyberwarfare, hypersonic missiles, robotics and drones — arguments indistinguishable from those put forward by Christian Brose, a top adviser to the late Senator John McCain, in his book “The Kill Chain.”
Biden’s key Asia advisors, Ely Ratner and Kurt Campbell, have also acknowledged that the Obama administration, like its predecessors, underestimated the global ambition of China’s leaders and their resolve to resist political liberalization. Last year, Campbell and Jake Sullivan (who was Vice President Biden’s national security adviser in 2013-14), made what seemed like an explicit argument for detente in Kissinger’s sense of the term. “Despite the many divides between the two countries,” they wrote, “each will need to be prepared to live with the other as a major power.”
U.S. policy toward China should combine “elements of competition and cooperation” rather than pursuing “competition for competition’s sake,” which could lead to a “dangerous cycle of confrontation.” Campbell and Sullivan may insist that Cold War analogies are inappropriate, but what they are proposing comes straight out of Kissinger’s 1970s playbook.
Yet the lesson of detente is surely that a superpower ruled by a Communist Party does not regard peaceful coexistence as an end in itself. Rather, the Soviet Union negotiated the 1972 Strategic Arms Limitation Talks agreement with the U.S. for tactical reasons, without deviating from its long-term aims of achieving nuclear superiority and supporting pro-Soviet forces opportunistically throughout the Third World.
The crucial leverage that forced Moscow to pursue detente was the U.S. opening to China in February 1972, when Nixon and Kissinger flew to Beijing to lay the foundations of what, 30 years later, had grown into Chimerica. Yet the compulsive revolutionary Mao Zedong was never entirely at ease with his own opening to America. At one point in late 1973, when the U.S. offered China the shelter of its nuclear umbrella from a possible Soviet attack, Mao became indignant and accused Premier Zhou Enlai of having forgotten “about the principle of preventing ‘rightism.’”
For all its achievements, detente came to be a pejorative term in the U.S., too. It is often forgotten how much of Ronald Reagan’s rise as the standard-bearer of the Republican right was based on his argument that detente was a “one-way street that the Soviet Union has used to pursue its aims.”
The danger of detente 2.0 is that Biden will be Jimmy Carter 2.0. Throughout his four years in the White House, Carter was torn between the “progressive” left wing of his own party and hawkish national security advisers. He ended up being humiliated when the Soviet Union tore up detente by invading Afghanistan.
Consider how China will approach the new administration. Beijing would like nothing more than an end to both the trade war and the tech war that the Trump administration has waged. In particular, Beijing wants to get rid of the measures introduced by the U.S. Commerce Department in September, which effectively cut off Huawei and other Chinese firms from the high-end semiconductors manufactured not only by U.S. companies but also by European and Asian companies that use U.S. technology or intellectual property.
In the tech war, Team Biden seems ready to make concessions. Some of the president-elect’s advisers want to offer wider exemptions for the foreign chipmakers who supply Huawei and to drop Trump’s executive actions against the Chinese internet companies TikTok and Tencent. But in return for what? Is China about to halt its dismantling of what remains of Hong Kong’s semi-autonomy? Clearly not. Is China going to suspend its policies of incarceration and re-education of Uighurs in Xinjiang? Not a chance. Will China stop exporting its surveillance technology to any authoritarian government that wants to buy it? Dream on.
If China’s quid pro quo is nothing more than Xi’s recent commitment to be “carbon neutral” by the distant year 2060, then the Biden administration would be nuts to do detente. China is currently building coal-burning power stations with a capacity of 250 gigawatts, more than this country’s total coal-fueled power capacity. The country accounts for roughly half of all the new CO2 emissions since the Paris Agreement on climate change was signed. If Biden is determined to take the U.S. back into the Paris accord, he needs to say explicitly that it will soon be a dead letter without meaningful Chinese actions.
The microchip race is a bit like the nuclear arms race in Kissinger’s time. Beijing lags behind qualitatively, as the Soviet Union did, though it can win in terms of quantity. China cannot yet match the sophistication of the chips made by Taiwan’s TSMC.
Its goal is to buy time while catching up and achieving “technological self-reliance.” The obvious U.S. response is to try to stay ahead. Biden (who is going to need all the bipartisan issues he can find if Republicans retain control of the Senate) may well work with the GOP to pass the CHIPS Act, which aims to promote domestic semiconductor production.
Such technological races can go on for years. Similar races are underway in the fields of artificial intelligence, digital currency and even Covid vaccines. But the lesson of Cold War I is that the ultimate test of any national security policy is its first crisis. In 2021 there is a significant chance that North Korea will provide the Biden administration with its earliest foreign-policy challenge in the form of new missile or nuclear tests. But there is another scenario: a Taiwan crisis.
Biden should have no illusions about Xi. The Chinese leader’s ultimate goal is to bring to an end Taiwan’s de facto autonomy and democracy and bring it fully under Beijing’s control. This is not just about asserting the principle of “One China, One System.” There is the eminently practical argument that China would no longer need play catch-up in the chip race if it directly controlled Taiwan. Earlier this year, one nationalist blogger proposed a simple solution: “Reunification of the two sides, take TSMC!”
Meanwhile, as we have seen, there has been a bipartisan upgrade of the U.S. commitment to Taiwan, which dates back to the 1979 Taiwan Relations Act. Not long after Flournoy’s pledge to increase America’s capacity to deter Beijing from invading the island, Richard Haass of the Council on Foreign Relations argued for an end to the “ambiguity” of the U.S. commitment to defend Taiwan. “Waiting for China to make a move on Taiwan before deciding whether to intervene,” he wrote in September, “is a recipe for disaster.” But another recipe for disaster would be a showdown over Taiwan before a Biden administration has even begun beefing up deterrence.
Relations between Washington and Beijing reached an impasse this year. Strategic dialogue gave way to Twitter abuse. Detente 2.0 would be an improvement, if only at the level of superpower communication.
The rationale for detente, as Kissinger often argued in the 1970s, was the world’s growing interdependence. That argument has even more force today. The pandemic has revealed the immense extent of our interdependence and the impossibility of a world order based on — to use French again — sauve qui peut (“every man for himself”) and chacun a son gout (“to each his own”).
A novel virus in Wuhan has caused a global plague and the deaths of hundreds of thousands of Americans. Similar interdependence will be revealed if global warming has the dire consequences projected by the Intergovernmental Panel on Climate Change. Economically, too, the U.S. and China remain interdependent. Trump’s tariffs did nothing whatever to reduce the bilateral trade deficit.
Yes, the U.S. and China are in the foothills of a Cold War. But there is no good reason to go through decades of brinkmanship before entering the detente phase of this cold war. Let Taiwan in 2021 not be Cuba in 1962, with semiconductors playing the role of missiles.
Nevertheless, as in Kissinger’s time, detente cannot mean that the U.S. gives China something for nothing. If the incoming Biden administration makes that mistake, the heirs of Ronald Reagan in the Republican Party will not be slow to remind them that detente — diplomatic French for “let’s not fight” — was once a dirty word in American English.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
3D print of a SARS-CoV-2 virus particle|© National Institute of Allergy and Infectious Diseases (NIAID)
In 2002 the Cambridge astrophysicist and Astronomer Royal Lord Rees predicted that, by the end of 2020, “bioterror or bioerror will lead to one million casualties in a single event”. In 2017 the Harvard psychologist Steven Pinker took the other side in a formal bet. As the terms of the wager defined casualties to include “victims requiring hospitalization”, Rees had already won long before the global death toll of Covid-19 passed the one million mark in September. Sadly for him, the stake was a meagre $400.
Nicholas Christakis has given his rapidly written yet magisterial book about the pandemic the title Apollo’s Arrow. The allusion is to the plague the god unleashes against the Achaeans for kidnapping the daughter of his priest Chryses in Book One of Homer’s Iliad. An alternative title might have been Rees’s Bet. In a report from 1998 the US Department of Defense observed that “historians in the next millennium may find that the 20th century’s greatest fallacy was the belief that infectious diseases were nearing elimination”. Pinker was only one of many scholars who subscribed to this belief in the twenty-first century. “Disease outbreaks don’t become pandemics” any more, he argued in Enlightenment Now, because “advances in biology … make it easier for the good guys (and there are many more of them) to identify pathogens, invent antibiotics that overcome antibiotic resistance, and rapidly develop vaccines”.
Pinker was not alone in his complacency. But complacency alone cannot explain the remarkably ineffective response of the US, UK and numerous other developed countries to the virus’s spread. The most Christakis can say in defence of the US response is that “vastly more Americans would have died – perhaps a million – had we failed to deploy the resources we marshaled, belatedly, in the spring of 2020 to cope with the first wave of the pandemic”. He himself was never one lightly to dismiss the potential lethality of Covid-19. And via his Twitter feed (@NAChristakis) he provided an exceptionally illuminating and learned commentary on the pandemic from its earliest stages. He was well placed to do so. Though Christakis’s first degree was in biology, he also holds an MD, a master’s degree in public health and a PhD in sociology. His recent books – Connected: The surprising power of our social networks (2009) and Blueprint: The evolutionary origins of a good society (2019) – established him as a leading authority on social networks and contagion of all kinds. If one graphed the social network of pandemic expertise, the node marked “Christakis” would have very high centrality.
But he is more than just a rare polymath in an academic world of narrowly specialized Fachidioten. Christakis is also a publicly engaged scientist with a notable aptitude for sharing his knowledge with lay audiences. He also possesses the quality – rare in academia – of courage. When, in October 2015, a crowd of Yale students surrounded and harangued him, demanding an apology for a perfectly innocuous public email his wife had sent on the subject of Halloween costumes and the transgressive origins of the festival, Christakis stood his ground, cordially listened – and somehow kept his temper.
In some respects, Apollo’s Arrow is an instant history of an event that is by no means over. This is a hard thing to pull off, but Christakis does it with aplomb. His account of the initial outbreak in Wuhan, which began on December 1, 2019, describes the uncertainty about the new virus’s origin, the local cover up, the lost weeks as the seriousness of the crisis slowly became clear to the Politburo, and the uniquely restrictive “lockdown” imposed not only on Hubei province but also right across China, beginning on January 23. He is a little too kind to the Chinese central government, in my view, and inexplicably says nothing about the shameful role played by the World Health Organization in swallowing Beijing’s early lies. But he is right that Chinese doctors acted with courage as well as competence. And much that we know about what happened in China is due to high-quality research by Chinese scientists.
Christakis’s account of Covid-19 is paradoxical. That the new virus would reach the US was inevitable. “Patient zero” returned to Washington state from Wuhan on January 15, but it was another carrier who first spread it. Numerous infected people were flying into the United States in January and February, not only from Wuhan but from other Chinese cities, from Europe and from Iran. No one knew exactly how many because the Centers for Disease Control and Prevention (CDC) failed utterly to expand testing in the way that its counterparts in South Korea and Taiwan did. This was the first of a woeful succession of failures of public health policy (which had their counterparts in the UK and Europe). Though much media commentary has placed nearly all the blame for the relatively high excess mortality in the US on President Trump, Christakis’s list of culprits is much longer: in addition to Trump, he singles out the governor of Georgia, Brian Kemp, who “explicitly banned cities and counties from enacting rules requiring masks”, and Vice President Pence, who “visit[ed] the Mayo Clinic … without wearing a mask while everyone else around him was wearing one”. Democrats also failed to act quickly, particularly in New York City and New York State. And, as Christakis makes clear, the public health bureaucracy was as much to blame as the elected politicians. “Our nation spends 17.7 percent of its GDP on health care,” he notes, “and this was our level of preparedness?”
What could the US have done differently? Clearly, it would have been better if the White House had not interfered with CDC recommendations on social distancing. Clearly, the president had no business endorsing what proved to be an ineffectual prophylactic (hydroxychloroquine). But, as Christakis gently hints, the administration of Barack Obama had not distinguished itself in 2009 when confronted by “swine flu” – which fortunately proved far less deadly than Covid-19 – and it failed utterly to contain the opioid pandemic, the death toll from which remains higher than the death toll from the new pandemic, and with many younger victims. Christakis clearly thinks that an earlier ramping up of testing and of social distancing would have saved American lives in 2020, though he is surprisingly ambivalent about digital contact tracing (contact-tracing apps, he writes, “do not offer enough epidemiological benefit to justify the privacy trade-off”). He ultimately concludes that the belated imposition of lockdowns “was commensurate to the threat posed by the virus”, even if viewed “strictly from an economic perspective”. I have my doubts about that conclusion. Research by economists as different in their approaches as Austen Goelsbee and John Cochrane suggests that Americans adapted spontaneously to the outbreak, with much social distancing preceding the official shelter-in-place orders. The debate on the costs and benefits of the lockdowns will not be over until we have a full accounting of their economic, social and health costs. The fact remains that there were alternative policy options that achieved better public health results at much lower economic costs, in countries such as Taiwan, South Korea and Sweden. Researchers at the Oxford Blavatnik School have shown the lack of correlation between the stringency of government measures and outcomes in terms of infections. The correlation between stringency and the magnitude of economic contraction is, by contrast, high.
Apollo’s Arrow has three great strengths. Firstly, Christakis clearly explains the nature of the virus and the disease it causes, and shows that social network structures and public policy largely explain the great variance in the pandemic’s impact from country to country. Because of the distinctive role of super- spreaders (the roughly 20 per cent of carriers who for various reasons conduct around 80 per cent of the spreading), controlling the proliferation of Covid-19 depends on testing, tracing and isolating the infected. As he shows, a laissez-faire approach that aimed to achieve herd immunity through uncontrolled spread almost certainly would have led to an intolerable number of premature deaths. (Though Sweden did not impose a blanket lockdown, it banned public gatherings and encouraged social distancing.) Given the high costs of containment, a vaccine will therefore be crucial, just as it was in controlling earlier infectious diseases such as polio and measles.
Secondly, Christakis does an exceptional job of comparing Covid-19 to other major pandemics in history. He chooses to go backwards through time, beginning with the two coronavirus epidemics (of SARS and MERS) that should have served as warnings to everyone and not just to the countries directly affected. Covid-19, he shows, is less deadly than these recent coronaviruses but it has “the ease of transmission of a common cold”. Our pandemic is nowhere near as deadly as the Spanish flu of 1918–19, as some (notably the Imperial College epidemiologists) feared back in March. It is much closer to the 1957–8 “Asian flu”, an episode now largely forgotten and which Christakis describes vividly, noting the crucial difference: the Asian flu killed the very young as well as the very old, along with a significant number of teenagers.
The third strength of Apollo’s Arrow is what it has to say about the future. On the social and psychological consequences of the pandemic, including the silver linings of markedly increased voluntarism and “catastrophe compassion”, Christakis fizzes with insights. Until 2022, he argues, “Americans will live in an acutely changed world – they will be wearing masks, for example, and avoiding crowded places”. This will be “the immediate pandemic period”, to be followed by an “intermediate pandemic period”, when “we either reach herd immunity or have a widely distributed vaccine”, by which point people will “still be recovering from the overall clinical, psychological, social, and economic shock” of what has happened. Not until 2024 will we reach the “post-pandemic period”, but even this will seem only a partial return to “normal”.
What will have changed? Handshaking will be a thing of the past, he conjectures, like spitting in the street. Telemedicine will be here to stay, as will a lot of working from home. Cities will likely be “duller, as many small retail firms go out of business”, and many smaller universities will close. Yet perhaps we shall eventually make it to our own Roaring Twenties, as those who survived 1918–19 did. “The increased religiosity and reflection of the immediate and intermediate pandemic periods”, Christakis suggests, “could give way to increased expressions of risk-taking, intemperance, or joie de vivre in the post-pandemic period.” Here’s hoping.
There certainly seems some chance that, after Covid-19 has been contained, the huge monetary and fiscal measures deployed to offset the shock of lockdowns could have inflationary consequences. Christakis foresees higher wages as a result of a political backlash against inequality. He also joins the many commentators who have argued that “the loss of American economic power and the lack of US leadership could create an opening for China to exert more influence”. My own view is that Western commentators are making the usual mistake of over-estimating the authoritarian competition and underestimating the American system’s capacity for self-repair. Let’s not forget where this pandemic began and why China’s one-party state took so long to admit the truth about it and take effective measures to protect the rest of the world, as opposed to the rest of China. Beijing is paying a price for this, as Pew’s recent report on China’s international reputation made clear. And let’s not exaggerate the scale of the American failure. The excess mortality rate in the first half of this year (23 per cent in the US) was roughly twice as high in Spain (56 per cent), the UK (45 per cent), Italy (44 per cent), and Belgium (40 per cent). Only in one of these four European countries was an incompetent populist in charge, suggesting that incompetent populist leadership may not be the crucial variable.
Of course, this pandemic isn’t over. As Christakis notes, the experience of past pandemics should have led us to expect more than one wave, as opposed to a single curve that had to be flattened. “We are likely to reach an attack rate of roughly 40 to 50 percent by 2022 no matter what we do”, the author concludes. If the infection fatality rate is 0.5 per cent (by no means the maximum estimate in the current literature), that implies at least 660,000 Covid-19 deaths in the US, three times the total to date. Lord Rees has won his bet, and Apollo is still shooting his arrows.
Niall Ferguson is the Milbank Family Senior Fellow at the Hoover Institution, Stanford University, and a senior faculty fellow of the Belfer Center for Science and International Affairs at Harvard. His next book, Doom: The politics of catastrophe, will be published in May 2021
Waiting for the day. Photographer: Olivier Douliery/AFP/Getty Images
When talking among themselves, Silicon Valley big shots sometimes say weird things. In an internal presentation in March 2018, Google executives were asked to imagine their company acting as a “Good Censor,” in order to limit the impact of users “behaving badly.”
In a 2016 internal video, Nick Foster, Google’s head of design, envisioned a “goal-driven ledger” of all users’ data, endowed with its own “volition or purpose,” which would nudge us to take decisions (say, about shopping or travel) that would “reflect Google’s values as an organization.”
If that doesn’t strike you as weird — like dialogue from some dystopian science-fiction novel — then you need to read more dystopian science fiction. (Start with Yevgeny Zamyatin’s astonishingly prescient “We.”)
The lowliest employees of big tech companies — the content moderators whose job it is to spot bad stuff online — offer a rather different perspective. “Remember ‘We’re the free speech wing of the free speech party’?” one of them asked Alex Feerst of OneZero last year, alluding to an early Twitter slogan. “How vain and oblivious does that sound now? Well, it’s the morning after the free speech party, and the place is trashed.”
I don’t know if, as the New York Post alleged last week, Democratic presidential nominee Joe Biden met with a Ukrainian energy executive named Vadym Pozharskyi in 2015. I don’t know if Biden’s son Hunter tried to broker such a meeting as part of his board directorship deal with Pozharskyi’s firm, Burisma Holdings. And I am pretty doubtful that the meeting, if indeed it happened, was the reason Biden demanded that the Ukrainian government fire its prosecutor general, Viktor Shokin, who was investigating Burisma. I am even open to the theory that the whole story is bunk, the emails fake, and the laptop and its hard-drive an infowars gift from Russia, with love.
What I do know is that if I read the story online and found it compelling, I should have been able to share it with friends. Instead, both Facebook and Twitter made a decision to try to kill the Post’s scoop.
Andy Stone, the former Democratic Party staffer who is now Facebook’s policy communications manager, announced that his company would be “reducing” the “distribution” of the Post story. Twitter barred its users from sharing it not only with followers but also through direct messages, locking the accounts of people — including White House press secretary Kayleigh McEnany — who retweeted it.
This is not an isolated incident. In May, Twitter attached a health warning to one of President Trump’s Tweets. There was uproar at Facebook when chief executive Mark Zuckerberg declined to follow Twitter’s lead. Days later, Facebook was pressured into taking down 88 Trump campaign ads that used an inverted red triangle (a Nazi symbol) to attack antifa, the far-left movement. In August, Facebook removed a group with nearly 200,000 members “for repeatedly posting content that violated our policies.” The group promoted the QAnon conspiracy theory, which is broadly pro-Trump. Earlier this month, the company deleted all QAnon accounts from its platforms.
Google has been doing the same sort of thing. In June, it excluded the website ZeroHedge from its ad platform because of “violations” in the comments sections of stories about Black Lives Matter.
The remarkable thing is not that Silicon Valley is playing a highly questionable role in the election of 2020. It is that the same was true in 2016 and, despite a great many fine words and some minor pieces of legislation, Americans did nothing about it.
Far from addressing the glaring problems created by the rise of the network platforms that now dominate the American (and indeed the global) public sphere, we largely decided to shut our eyes and ears to them. In the past 10 months, I’ve read as many op-ed articles and reports about this election as I can stand. I’m staggered by how few even mention the role of the internet and social media. (Kevin Roose’s work on the conservative dominance of Facebook shared content is an honorable exception.) You would think it was still the 1990s — as if this contest will be decided by debates on television, newspaper endorsements or stump speeches, and accurately predicted by opinion polls. (Actually, make that the 1960s.)
Yet the new role of social media is staring us in the face (literally). The number of U.S. Facebook users was 240 million in 2019, more than 72% of the population. Adults spend an average of 75 minutes of each day on social media. Half that time is on Facebook. Google accounts for 88% of the U.S. search-engine market, and 95% of all mobile searches. Between them, Google and Facebook captured a combined 60% of U.S. digital-ad spending in 2018.
The top U.S. tech companies are now among the biggest businesses on earth by market capitalization. But their size is not the important thing about them. Earlier this month, the House Judiciary Committee’s Antitrust Subcommittee released the findings of its 16-month long investigation into Big Tech. The conclusion? “Apple, Amazon, Google and Facebook each possess significant market power over large swaths of our economy. In recent years, each company has expanded and exploited their power of the marketplace in anticompetitive ways.”
Cue years of antitrust actions that will enrich a great many lawyers and have minimal consequences for competition, like the ultimately failed attempt 20 years ago to prevent Microsoft from dominating software.
An antitrust action against Amazon is doomed. Consumers love the company. It has measurably reduced the prices of innumerable products as well as rendering shopping in bricks-and-mortar stores an obsolescent activity. Good luck, too, with breaking up Google. Even the much less trusted Facebook (according to polls) will be hard to dismantle, without a complete transformation of the way the courts apply competition law. It’s free, for heaven’s sake. And there are network effects on the internet that can’t be wished away by judges.
Is it stupidity or venality that has convinced America’s legislators that antitrust is the answer to the problem of Big Tech? A bit of both, I suspect. Either way, it’s the wrong answer.
The core problem is not a lack of competition in Silicon Valley. It is that the network platforms are now the public sphere. Every other part of what we call the media — newspapers, magazines, even cable TV — is now subordinated to them. In 2019, the average American spent 6 hours and 35 minutes a day using digital media, more than television, radio and print put together.
Not only do the big tech companies dominate ad revenue, they drive the news cycle. In 2017, two-thirds of American adults said they got news from social media sites. A Pew study showed that, at the end of 2019, 18% of them relied primarily on social media for political news. Among those aged 30 to 49, the share was 40%; among those aged 18 to 29, it was 48%. The pathologies that flow from this new reality are numerous. Antitrust actions address none of them.
“I thought once everybody could speak freely and exchange information and ideas, the world is automatically going to be a better place,” Evan Williams, one of the founders of Twitter, told the New York Times in 2017. “I was wrong about that.” Indeed, he was.
Subject to the most minimal regulation in their country of origin — far less than the TV networks in their heyday — the network platforms tend, because of their central imperative to sell the attention of their users to advertisers, to pollute national discourse with a torrent of fake news and extreme views. The effects on the democratic process, not only in the U.S. but all over the world, have been deeply destabilizing.
Moreover, the vulnerability of the network platforms to outside manipulation has posed and continues to pose a serious threat to national security. Yet half-hearted and ill-considered attempts by the companies to regulate themselves better have led to legitimate complaints that they are restricting free speech.
How did we arrive at this state of affairs — when such important components of the public sphere could operate solely with regard to their own profitability as attention merchants? The answer lies in the history of American internet regulation — to be precise, in section 230 of the 1934 Communications Act, as amended by the 1996 Telecommunications Act, which was enacted after a New York court held the online service provider Prodigy liable for a user’s defamatory posts.
Previously, a company that managed content was classified as a publisher, and subject to civil liability — creating a perverse incentive not to manage content at all. Thus, Section 230c, “Protection for ‘Good Samaritan’ blocking and screening of offensive material,” was written to encourage nascent firms to protect users and prevent illegal activity without incurring massive content-management costs. It states:
1. No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.
2. No provider or user of an interactive computer service shall be held liable on account of any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing or otherwise objectionable.
In essence, Section 230 gave and still gives websites immunity from liability for what their users post (under-filtering), but it also protects them when they choose to remove content (over-filtering). The idea was to split the difference between publisher’s liability, which would have stunted the growth of the fledgling internet, and complete lack of curation, which would have led to a torrent of filth. The surely unintended result is that some of the biggest companies in the world today are utilities when they are acting as publishers, but publishers when acting as utilities, in a way rather reminiscent of Joseph Heller’s “Catch-22.”
Here’s how Catch-22 works. If one of the platforms hosts content that is mendacious, defamatory or in some other way harmful, and you sue, the Big Tech lawyers will cite Section 230: Hey, we’re just a tech company, it’s not our malicious content. But if you write something that falls afoul of their content-moderation rules and duly vanishes from the internet, they’ll cite Section 230 again: Hey, we’re a private company, the First Amendment doesn’t apply to us.
Remember the “good censor”? Another influential way of describing the network platforms is as the “New Governors.” That creeps me out the way Zuckerberg’s admiration of Augustus Caesar creeps me out.
For years, of course, the big technology companies have filtered out child pornography and (less successfully) terrorist propaganda. But there has been mission creep. In 2015, Twitter added a new line to its rules that barred promoting “violence against others … on the basis of race, ethnicity, national origin, religion, sexual orientation, gender, gender identity, age, or disability.” Repeatedly throughout the Trump presidency — for example, after the violence in Charlottesville, Virginia, in 2017 — there have been further modifications to the platforms’ terms of service and “community standards,” as well as to their non-public content moderation policies.
There is no need to detail all the occasions in recent years when mostly right-leaning content was censored, buried far down the search results, or “demonetized.” The key point is that, in the absence of a coherent reform of the way the network platforms are themselves governed, there has been a dysfunctional tug-of-war between the platforms’ spasmodic and not wholly sincere efforts to “fix” themselves and the demands of outside actors (ranging from the German government to groups of left-wing activists) for more censorship of whatever they deem to be “hate speech.”
At the same time, the founding generation of Silicon Valley entrepreneurs, most of whom had libertarian inclinations, have repeatedly yielded to internal pressure from their younger employees, schooled in the modern campus culture of “no-platforming” any individuals whose ideas they consider “unsafe.” In the words of Brian Amerige, whose career at Facebook ended not long after he created a “FB’ers for Political Diversity” group, the company’s employees “are quick to attack — often in mobs — anyone who presents a view that appears to be in opposition to left leaning ideology.”
The net result seems to be the worst of both worlds. On the one hand, conspiracy theories such as Plandemic flourish on Facebook and elsewhere. On the other, the network platforms arbitrarily intervene when a legitimate article triggers the hate-speech-spotting algorithms and the content-moderating grunts. (As one of them described the process, “I was like, ‘I can just block this entire domain, and they won’t be able to serve ads on it?’ And the answer was, ‘Yes.’ I was like, ‘But … I’m in my mid-twenties.’”)
At a lecture at Georgetown University in October 2019, Zuckerberg pledged “to continue to stand for free expression” and against an “ever-expanding definition of what speech is harmful.” But even Facebook has had to ramp up the censorship this year. The bottom line is that the good censors are not very good and the new governors can’t even govern themselves.
Two years ago, I wrote a lengthy paper on all this with a well-worn title, “What Is to Be Done?” Since then, almost nothing has been done, beyond some legislative tinkering at the margins. The public has been directed down a series of blind alleys: not only antitrust, but also net neutrality and an inchoate notion of tighter regulation. In reality, as I argued then, only two reforms will fix this godawful mess.
First, we need to repeal or significantly amend Section 230, making the network platforms legally liable for the content they host, and leaving the rest to the courts. Second, we need to impose the equivalent of First Amendment obligations on the network platforms, recognizing that they are too dominant a part of the public sphere to be able to regulate access to it on the basis of their own privately determined and almost certainly skewed “community standards.”
To such proposals, Big Tech lawyers respond by lamenting that they would massively increase their clients’ legal liabilities. Yes. That is the whole idea. The platforms will finally discover that there are risks to being a publisher and responsibilities that come with near-universal usage.
In recent few years, these ideas have won growing support — and not only among Republican legislators such as Senator Josh Hawley. In the words of Judge Alex Kozinski in Fair Housing Council v. Roommate.com (2008), “the Internet has outgrown its swaddling clothes and no longer needs to be so gently coddled.” He was referring to Section 230, which gives the tech giants a now-indefensible advantage over traditional publishers, while at the same time empowering them to act as censors.
While Section 230 protects internet companies from liability over removing any content that they believed to be “obscene, lewd, lascivious, filthy, excessively violent, harassing or otherwise objectionable,” successive court rulings have clearly established that the last two words weren’t intended to permit discrimination against particular political viewpoints.
Meanwhile, in Packingham v. North Carolina (2017), the Supreme Court overturned a state law that banned sex offenders from using social media. In the opinion, Justice Anthony Kennedy likened internet platforms to “the modern public square,” arguing that it was therefore unconstitutional to prevent even sex offenders from accessing, and expressing opinions, on social-network platforms. In other words, despite being private companies, the big tech companies have a public function.
If the network platforms are the modern public square, then it cannot be their responsibility to remove “hateful content” (as 19 prominent civil rights groups demanded of Facebook in October 2017) because hateful content — unless it explicitly instigates violence against a specific person — is protected by the First Amendment.
Unfortunately, this sea change has come too late for root-and-branch reform to be enacted under the Trump administration. And, contemplating the close links between Silicon Valley and Senator Kamala Harris, I see little prospect of progress — other than down the antitrust cul-de-sac — if she is elected vice president next month. Quite apart from the bountiful campaign contributions Harris and the rest of Democratic Party elite receive from Big Tech, they have no problem at all with Facebook, Twitter and company seeking to kill stories like Huntergate.
In 1931, British Prime Minister Stanley Baldwin accused the principal newspaper barons of the day, Lords Beaverbrook and Rothermere, of “aiming at … power, and power without responsibility — the prerogative of the harlot throughout the ages.” (The phrase was his cousin Rudyard Kipling’s.) As I contemplate the under-covered and overmighty role that Big Tech continues to play in the American political process, I don’t see good censors. I see big, bad harlots.
Niall Ferguson is the Milbank Family Senior Fellow at the Hoover Institution at Stanford University and a Bloomberg Opinion columnist. He was previously a professor of history at Harvard, New York University and Oxford. He is the founder and managing director of Greenmantle LLC, a New York-based advisory firm.
Woodrow Wilson and Warren Harding. Normal? Photographer: Topical Press Agency/Hulton Archive via Getty Images
“America’s present need,” the candidate declared, “is not heroics, but healing; not nostrums, but normalcy … My best judgment of America’s need is to steady down, to get squarely on our feet, to make sure of the right path. Let’s get out of the fevered delirium.”
The candidate was the Republican Warren G. Harding and the date was May 14, 1920. Six months later, Harding won a landslide victory over the Democratic nominee, James M. Cox, winning 60% of the popular vote and 404 Electoral College votes.
A return to normalcy: It’s an appealing prospect today, too, amid an ongoing pandemic, in the wake of an unprecedented economic shock, and after four years of political disruption. A century ago, to be sure, Americans had come through worse: the 1918-19 Spanish influenza, which killed around 675,000 people (the equivalent of 2.2 million today), and World War I.
A century ago, there was no incumbent to defeat, as Woodrow Wilson — having been struck down by the flu during the 1919 Paris peace negotiations and then by a severe stroke — was judged by his party to be unfit to run. (It remains to be seen if President Donald Trump’s admission to hospital for Covid-19 presages a premature exit for him.) But the parallel with today is still striking. In the so-called Red Scare of 1919-20, the country had been swept by strikes, protests and race riots. A severe recession had begun in January 1920. By November, what most Americans craved was indeed normalcy.
I have been thinking a lot about the election of 1920 in trying to predict that of 2020. Four years ago, chastened and educated by the experience of Brexit, I felt that Trump had at least an even chance of winning the presidency. Recall that in the week before the Nov. 8, 2016, the left-wing Daily Kos website put Hillary Clinton’s chances of winning the presidency at nearly 90%. According to The Upshot in the New York Times, the number was 85%. Betfair said 79%. Nate Silver said 65%.
So what do I think now, when even the ultra-cautious Silver puts Joe Biden’s chance of beating Trump at around 80%? Spoiler: I always said the half-life of populism was short.
Donald Trump is a classic populist, who offered disgruntled voters a heady cocktail of protectionism, nativism, easy money, isolationism and anti-elitism. Comparisons with European fascists between the World Wars always struck me as wide of the mark. Historically, it has generally been hard for mercurial figures such as Trump to win the highest political office, at least in the northern hemisphere. (I never bought former White House adviser Steve Bannon’s analogy between Trump and Andrew Jackson.) From Georges Boulanger to William Jennings Bryan to Huey Long, the history of populism is mostly of near misses — which was part of the reason most pundits assumed Trump would be a near miss four years ago.
When populists do get elected, they almost never deliver all they have promised to their supporters, and are often exposed as even more corrupt than the people they ran against. South America has a lot of experience in this regard, from Juan Peron in Argentina to Hugo Chavez in Venezuela. Latin American populists get re-elected not because they deliver higher living standards to their supporters (they may do so in the short run, but it always ends in some kind of financial crisis). They get re-elected by repressing their opponents and, when necessary, changing the constitution — a regional pastime.
To read the mainstream press, you would be forgiven for thinking something similar is about to happen in the U.S. According to Barton Gellman in the Atlantic, there is going to be voter suppression, voter intimidation, a declaration of emergency, the bypassing of election results in battleground states, and finally martial law. Trump’s going to steal the election somehow — and it may even be constitutional if he does, Fareed Zakaria has argued. Only a Biden landslide can save the Republic from violence and a constitutional crisis. Forget Hitler and Mussolini; now Trump is Richard III.
It must be said that Trump did everything possible to validate these narratives in last Tuesday night’s debate, short of opening with “Now is the winter of our discontent.” But, as Biden likes to say, “Come on, man.” Trump may have the instincts of a caudillo, but this isn’t Venezuela.
The debate would have mattered only if Biden had looked unmistakably senile. He didn’t. Instead, Trump came across as an insufferable bully. Even my ex-cop friends Mike and Gerry — who backed Trump in 2016 and were infallible guides to that year’s politics — felt their man had been too aggressive. And now it turns out that Trump was mocking Biden for wearing a mask, when he himself was probably already infected with Covid-19. (Not just mocking him, but yelling at him indoors from just six feet away. We won’t be sure for roughly a week that Trump didn’t infect Biden.)
Far from being in peril, I would guess, the Constitution is about to do what it was designed to do: Having successfully constrained a demagogic president throughout his term, in the usual ways — courts striking down executive orders, Supreme Court appointees acting independently, midterms handing the House to the Democrats — it is going to allow voters to eject him from the White House and install in his place dear, old Joe Normal.
For any of the “end of the Republic” scenarios to happen, this election needs to be close — close enough for the results in multiple states to be challengeable. But I struggle to see how this could come about.
If Jimmy Carter couldn’t get a second term after the small recession of January to July 1980, and if George H.W. Bush couldn’t get one after the comparably minor recession of July 1990 to March 1991, how on earth can Donald Trump get a second term after the disaster that has befallen the U.S. this year? Who gets re-elected after a pandemic that has killed more than 200,000 Americans and a recession that sent unemployment up to 14.7% in April, compared with peaks under Carter and the elder Bush of 7.8%? Trump’s latest jobs report has unemployment at 7.9%.
Even with the recovery that’s occurred since the lockdown low-point back in the spring, the U.S. economy is still on course to shrink by 3.8% this year, according to the Organization of Economic Cooperation and Development. The only example I know of a democratic leader getting re-elected under such economic conditions is Angela Merkel in 2009.
Now let’s look at the polls, where Biden leads Trump by an average of around 7 percentage points. The remarkable thing here is the consistency of Biden’s lead: Over more than a year he has never been less than four points ahead. We’ve seen nothing like this in our lifetimes — in most presidential elections since 1968, the polls have bounced around, sometimes wildly.
Moreover, Biden’s lead right now is not just bigger than Clinton’s four years ago at this stage in the race; it is also bigger than Barack Obama’s 32 days out in both 2008 and 2012. If the news on the pandemic and the economy is bad between now and election day, Trump could end up where John McCain did, seven points behind. Or, if the news improves, he could somehow claw his way back, as Mitt Romney did in 2012, and still lose. What I struggle to imagine is Trump getting close enough to rerun the George W. Bush-Al Gore standoff of Florida 2000 in multiple states.
Remember what happened in the final phase of the 2016 campaign. First, a relentless stream of negative news about Clinton throughout October ate away at her lead. Second, state polls seriously underestimated Trump’s support in Michigan, Pennsylvania and Wisconsin. Finally, third-party candidates took 6% of the vote in 2016. Maybe there’s a secret stash of toxic opposition research waiting to be unleashed against Biden this month, but I doubt it.
Instead, what we’re getting is a relentless stream of negative news flow about Trump, not the least of which is the New York Times expose of how insanely little income tax the guy has been paying. (As my friend Mike told me last week, “I heard a couple of blue-collar workers today, cops and firemen, talking about the Times story about him not paying any taxes ... it was the first time I ever heard anything negative about Trump from this base.”)
And that’s not the only bad news for Trump. He’s only one of nearly 300,000 Americans who tested positive for Covid-19 last week. The hospitals in Wisconsin are filling up with new Covid-19 cases. And Covid-19 is the main reason Trump is struggling with older voters, a key demographic for him four years ago.
Trump clearly wanted to announce a successful vaccine before the election. That seems less and less likely. Jared Kushner wanted the economy to be “rocking” by now. But the refusal of the pandemic to “go away, like a miracle” is clearly having some adverse effects on the economy, preventing mobility from returning to normal in the most affected states, and slowing the recovery of the labor market. Finally, the finances of the Trump campaign appear to be in disarray (though being outspent did not stop him winning four years ago).
Even if you allow for polling errors as bad as 2016’s in the key states, Trump is going to struggle to get above 240 Electoral College votes, 30 shy of victory. For all these reasons, I am inclined to think he is going to be a one-term president, and that the election result won’t be close enough for full-scale GOP lawfare to save him.
What am I missing? What could make me wish I’d stuck to my contrarian position of four years ago?
After all, only last December another populist, Boris Johnson, won a much bigger victory in the U.K. general election than almost anyone (including me) expected, sweeping a bunch of traditionally Labour-voting working class constituencies in the north of England. Could there be an equivalent surprise in this year’s U.S. election?
Leaving aside the potential Covid-19 impacts on the candidates’ health, I can think of nine reasons why the polls might be even more wrong than last time.
First, a striking 11.7% of Republicans say they would not report their true opinions about their preferred presidential candidate on telephone polls, while 10.5% of independents also fall into the “shy voter” category.
Second, the law-and-order issue really matters to those shy voters. (It also gave Trump his best debate moments.) Polls give us a sample of voters’ stated preferences. Revealed preferences are in many ways more reliable. According to Small Arms Analytics, gun sales in August 2020 were 58% higher than in August 2019, continuing a surge of purchases (especially of handguns) since the spring. In 2016, gun ownership was very closely correlated with voting for Trump.
Third, the resumption of the so-called culture wars this summer was a godsend for Trump. Judging by Tucker Carlson’s ratings — not only on cable but also on YouTube — there are at least five million Americans who share his skepticism about the Black Lives Matter movement, to say nothing of “critical race theory.”
Fourth, check out whose Facebook posts have been getting shared the most this year. On the day of the first presidential debate, five of the top 10 posts were by conservative firebrand Ben Shapiro, not an unusual occurrence this year.
Fifth, the third vacancy on the Supreme Court in as many years was another stroke of luck for Trump. Conservative voters care more about the makeup of the court than liberals, so Amy Coney Barrett was a near-perfect pick to boost Republican turnout.
Sixth, Hispanic voters seem unenthused about Biden and indeed about voting generally. That matters in Florida, obviously, but there are 11 other states where Hispanics are more than 10% of eligible voters, including Arizona and Texas.
Seventh, Republicans are winning the voter-registration game in key states, notably Pennsylvania and Wisconsin.
Eighth, as a significant percentage of mail-in ballots tend to be rejected because of errors, Trump should benefit from the higher proportion of Democratic voters intending to vote that way.
Finally, don’t underrate the economy. A third-quarter bounce as big as the one projected by the Atlanta Fed’s GDPNow would give Trump a second term if you simply plug the number into the wonderfully parsimonious model devised many years ago by Yale’s Ray Fair to predict U.S. elections with economic variables.
Usually, if you can think of nine reasons why a hypothesis might be wrong, it’s probably wrong. And yet, even when I add all these variables together, I still don’t think Trump can salvage the situation. There is a lot of overlap, after all: Most gun purchasers probably owned at least one firearm already, and they may be the same people watching Tucker Carlson, liking Ben Shapiro and rooting for the confirmation of Justice ACB. In terms of new votes in swing states, and therefore Electoral College votes, my nine reasons to be doubtful may sum to zilch.
The probability of a repeat of 2016, when the votes of fewer than 40,000 people got Trump over the line in Michigan, Pennsylvania and Wisconsin, is simply too low. The probability of another 2000 or another 1876 (when the results in four states were contested) is also low. The probability of a contingent election — when no presidential or vice-presidential candidate receives an absolute majority of Electoral College delegates — is even lower: We haven’t seen one involving the presidency since 1824. None of these scenarios is remotely as probable as a victory for the “normalcy” candidate who has been out in front every single month of this annus horribilis.
The irony is that if a Biden victory is accompanied by a Democratic majority in the Senate, then it could suddenly be the turn of Republicans to cry “Republic is in peril,” as projects such as packing the Supreme Court, getting rid of the filibuster in the Senate, and giving statehood to the District of Columbia and Puerto Rico (i.e., packing the Senate) will suddenly seem feasible to the progressives.
But that’s the trouble with voting for normalcy. Remember, Americans did just that — overwhelmingly — a hundred years ago. What they got was the Roaring Twenties, followed by the Great Depression, followed by World War II.