Just about anyone who’s ever lived in mainland China is familiar with a constant stream of “fake food” scandals. Widespread food-safety problems include reports of fake cooking oil, a ghastly concoction reprocessed from “gutter oil” collected from restaurant fryers and sewer drains. There’s also fake baby formula (fatally adulterated with melamine to improve its apparent protein content), not to mention fake eggs, fake cheese, fake watermelons—and the fact that some of those reports are themselves fake.
Living in Shanghai, as I did for several years, often made me exercise muscles of suspicion I’d barely even known I had. Nothing was the way it seemed; the air could well be more polluted on a bright sunny day than on a foggy one. If I was about to buy a bottle of soy sauce, I would study the label, note where it was produced, deduct points if it was bottled in China, and wonder if the label itself was fake. Chinese friends often told me that they simply never bought Chinese milk/eggs/meat because of worries about fakery and counseled me to do the same.
Just as I had to learn how to navigate my local supermarket, I also had to learn how to navigate the Chinese internet. On mainland China’s censored internet, where Wikipedia is often blocked, doppelgängers of many major U.S. tech firms exist in a closed-off ecosystem fiercely protected from outside competition by the Great Firewall. Chinese users query a search engine called Baidu, watch videos on Youku, and post messages on a social network called Sina Weibo; an app called Didi Chuxing acquired Uber’s China operations in August. (Crucially, that also means all of these users’ data are collected by massive Chinese firms like the messaging and payments application WeChat, which are then required to share the information with the government.) A New York Times piece last year suggested China is like a lagoon compared to the greater ocean of the internet, one within which mutated apps evolve separately under different conditions. That analogy works, as long as you keep in mind that it’s a giant lagoon roughly a fifth the size of the ocean itself.
Inside this sealed-off world, there’s a Chinese version of Wikipedia, a user-edited encyclopedia called Baidu Baike, baike, or “hundred subjects,” being the Chinese word for “encyclopedia.” (That’s not to be confused with the Chinese-language version of the original Wikipedia, which also exists, just as there are growing Wikipedias in Patois, Xhosa and Old Church Slavonic.) You’ve probably never heard of Baidu, but on Alexa’s list of the most popular websites it places fourth, after Google, YouTube and Facebook—and ahead of Wikipedia itself. It offers both a Google-like search function and a Wikipedia-like encyclopedia, Baidu Baike.
It’s perhaps unsurprising that Baidu Baike’s content policies are commercially compromised—after all, it belongs to a for-profit entity, as opposed to the nonprofit Wikimedia Foundation. Unlike Wikipedia, Baidu Baike unashamedly hawks opportunities for “content collaboration” to celebrities, companies and media outlets, emphasizing that Baidu Baike can incorporate a prominently placed external link or insert entire sections of content into an entry. Buried in the fine print is the stipulation that Baidu Baike content, instead of being available under a Creative Commons license as Wikipedia’s is, belongs to NASDAQ-listed Baidu.
But in many cases the misinformation on Baidu Baike cannot be attributed to commercial interests; much of it is bizarre or just plain wrong. For instance, Baidu Baike lists Barack Obama as a member of the “Barack family” and identifies his mother’s citizenship as “White American from Kansas.” It quotes Bill Clinton calling Obama “the worst president in American history.” It also says Obama was a “drug addict” as a teenager and inexplicably recounts an anecdote about a couple whose wedding plans were disrupted by Obama’s golf schedule. Despite having the same open-content, anyone-can-edit structure as Wikipedia’s, Baidu Baike is a virtual quagmire of arbitrary opinions and what one might call fake facts.
The existence of the fake fact gives the pleonasm “true fact” meaning. A fake fact doesn’t perform the most crucial function of a fact. If you absorb a fake fact, you might feel more informed (just as you might feel full, at least momentarily, after you consume fake food). But unlike a true fact, it doesn’t tell you anything about the world. A fake fact sits on a website just where an actual fact might be, but it can’t get you from a place where you know less about Barack Obama to a place where you know more.
Recently, to test out Baidu Baike’s reliability on an ordinary, innocuous subject, I looked up my address in Shanghai, a street in the city center called Nanjing West Road. The opening paragraph of its Baidu article states that “750 international brands and more than 80 percent of the world’s best known brands have flagships or brand stores on Nanjing West Road.” Is that an actual fact or a fake one? The footnote takes me to a news article—but it’s about a noise complaint on an identically named street in a small southeastern city, which has nothing to do with the statistics on the page. A quick search turns up dozens of websites that all cite this statistic using exactly the same wording, including the websites of a five-star hotel and the Shanghai Consumer Council. In some paragraphs of the Baidu Baike entry on Nanjing West Road, nearly every phrase appears word for word on hundreds of other websites.
Last year, the Chinese government finally cracked down on Baidu’s pernicious practice of allowing companies to buy higher-ranked search results. But the most troubling aspect of the Chinese internet is not the deep reach of ad content; it’s the fact that online information in general is held to such mournfully low standards of truth. Whereas a dubious factoid about a company or a political movement might be traced to PR or censorship, I can’t even figure out who would be motivated to make up fake facts about Nanjing West Road. Writing in the London Review of Books, the political scientist David Runciman suggests that “the fact that there are no authoritative versions on Wikipedia is what makes it possible to generate a sense of personal accountability for particular entries, since any entry at any given time is the responsibility of the last person to edit it.” He optimistically adds, “This seems to be enough to make most people want to get it right.”
Clearly, Runciman has not spent much time on Wikipedia’s Chinese counterpart. But if the two websites are similarly organized, then what accounts for the fact that one of them produces vastly better results? If an ocean organism (Wikipedia) is cloned and placed in the lagoon (Baidu Baike), but then develops strange characteristics, what does that tell us about the lagoon?
The phenomenon of Chinese fake facts differs from the much-discussed post-truth politics of the West, for the simple reason that post-truth assumes some past era when “truth” reigned in public discourse. The phrase suggests that the “birther” conspiracy theory that Barack Obama was not born in the United States, or the much-maligned 350 million pounds a week that were to go to the NHS after Britain left the European Union, represent new developments, not business as usual. Until recently, a campaign pledge was not often called a “mistake” by the winning side the morning after the election, as Nigel Farage referred to that NHS claim. Presidential candidates didn’t usually hint darkly about a sitting president’s birth certificate—as Donald Trump notoriously did. Unlike the U.S., China isn’t post-truth; if anything, given the warped ideologies that have governed its history over the past century, it’s pre-truth.
Of course, the condition cannot be separated from a history of government censorship, going back to the birth of the modern People’s Republic. Under the auspices of the Great Leap Forward, in the late Fifties and early Sixties, millions starved to death, in part because of wildly exaggerated harvest statistics. Officials told peasants to massacre sparrows because it would improve rice yields (not effective). And schoolchildren were taught that South Korea invaded North Korea to start the Korean War (not true). The heroic soldier Lei Feng was widely venerated (his diary was forged, and he may not even have existed).
Since the internet came to China, the government has attempted to filter out websites it considers pernicious. These include popular porn sites, but also Google, Facebook and news outlets such as Bloomberg and the New York Times. On platforms such as Weibo comments critical of the regime can simply be deleted. The government also tightly restricts access to news about labor unrest as well as sensitive historical subjects such as the June 4, 1989 massacre of student protesters in Tiananmen Square.
Chinese internet users frequently use wordplay and innuendo to get around Beijing’s restrictions. For instance, users successfully used “grass-mud horse” (caonima) to evade the government censorship of profanity online, since the word is a homonym for “motherfucker” (also caonima). To refer to the June 4th Tiananmen massacre, users referred to the date “May 35th”—until that was blocked too. In the cat-and-mouse game of deleting pesky comments by internet detractors, Beijing is constantly playing catch-up.
Perhaps that’s why, in recent years, the Chinese government has added a new strategy to its arsenal: an army of people hired to post fake social-media comments as if they were ordinary Chinese citizens’ genuine opinions. Harvard political scientist Gary King, along with fellow investigators Jennifer Pan and Margaret Roberts, estimates that the government fabricates and posts about 448 million fake comments a year, which the researchers call “reverse censorship.” With traditional censorship becoming less and less effective within the internet lagoon, “reverse censorship” means the government simply adds its own voices to the online cacophony. The paid commentators are often known as “fifty-centers” (wumaodang) after the widespread rumor that they’re paid fifty cents—about seven U.S. cents—per post.
The genius of this strategy is that the fake posts aren’t even meant to spread misinformation. Rather, they simply distract social media users by diverting attention away from, say, a genuine report of labor protests in a small town. The patriotic cheerleading doesn’t contradict or shout down the actual news report—it doesn’t even engage with it. It simply changes the subject. And all the noise makes it harder for users to do what the government fears most: leverage the internet as a platform for collective action.
As always, the internet can fight back, and does. Suspicion that any pro-government user is simply a paid fake commentator is widespread, though as the Harvard researchers demonstrate, that supposition is often unfounded. More often, pro-government users are taunted as “volunteer fifty-centers” (ziganwu)—literally, fifty-centers who pack their own lunch. The taunt suggests that “volunteer fifty-centers” are suckers for holding opinions so unpopular or unreasonable that you’d have to be paid to hold them, that they’re effectively working for free.
While the lion’s share of fake facts online (such as the spurious statistic about Nanjing West Road) can’t be directly attributed to political censorship, information anarchy probably works to the censors’ advantage: with such a glut of fake facts, political untruths can blend right in. Books censored by the Chinese authorities circulate online; news of factory strikes and worker unrest slips through on social media, and the Great Firewall can be scaled by anyone with a little patience and a VPN. But the world’s wiliest Communist Party has adapted: having acknowledged that it can’t entirely suppress verifiable facts, it now manufactures and efficiently disseminates its own fake facts and ersatz nationalistic outrage.
In his essay about knockoff (known as shanzhai) phones, Chairman Mao lookalikes, and rivaling copycat rebel factions during the Cultural Revolution, Chinese author Yu Hua tells this story: “More than twenty years ago I could say whatever came into my head when I was interviewed by a journalist, but the interview would undergo strict review and be drastically edited before publication; ten years ago I began to be more circumspect in interviews, because I discovered that newspapers would report everything I said, even my swear words; and now I am often amazed to read interviews I have never given—remarks that the reporter has simply concocted, a gushing stream of drivel attributed to me.” When Yu Hua confronted one such reporter about an interview that never took place, he was told, “That was a copycat interview.”
Best known for his novel To Live, which tells the tale of a survivor of China’s tumultuous twentieth century, Yu Hua is no stranger to fakery himself. During the Cultural Revolution, Yu found an unlikely first profession as a dentist at a clinic that locals more accurately called a “tooth shop.” On his first day, Yu remembers: “I had the patient open his mouth wide and fixed my eyes on the tooth that had to be pulled.” But when he looked down at all the different-sized forceps on the tray, he had no idea which ones he should use, and “at that moment I had an intense sensation of being locked in a lonely, daunting struggle, without allies or sympathizers. I didn’t dare look into the patient’s staring eyes, for I was more petrified than he was.” It was not until years later that he understood what he was—a fake dentist.
Academics and experts who ponder the oft-diagnosed “lack of trust” in Chinese society—the suspicion, in other words, that anything from a close friendship to a vital drug could be fake—often put it down to the trauma of the Cultural Revolution, or alternatively, to the upheaval caused by a rapid transition to a market-driven economy. There’s probably a kernel of truth in these historical hypotheses, as there is in the observation that the traditional Confucian view privileges obligation to one’s family or community over any sense of responsibility to outsiders. But it’s worth noting that other Confucian-influenced, predominantly Chinese societies such as Hong Kong and Taiwan don’t seem to exist in the same condition of mistrust.
I moved to Shanghai shortly after college, having grown up in colonial and then Chinese Hong Kong. At first, the city irked me. It was disagreeable without being novel: people spat noisily on the ground and elbowed their way off subway cars. And then there were the fakes. Like everyone else around me in Shanghai, I developed a set of coping mechanisms for sifting out real products, facts and states of being—or at least, those I thought might be real—from fake ones. I scrolled through internet reviews when I was about to try a new service, hunting for the touches that distinguish genuine user comments from fake ones (or so I thought). I reasoned that the air might be nearly unbreathable on a beautiful day, but it wouldn’t be quite as polluted right after a downpour. I bought vegetables marked “export quality” on the hunch that they would pass Western safety controls—although I could never quite stop worrying about whether the “export” sign was fake. Every time I read something in simplified Chinese characters, the character set used in mainland China but not in Hong Kong or Taiwan, I’d automatically discount its credibility—as if simplified Chinese were the Chinese version of the typeface Comic Sans.
But I also had a deeper difficulty in Shanghai: I felt like I was faking it. I had never had occasion to use Mandarin before, and I could speak it passably but not confidently. I felt conflicted about my plum expat package: as far as I could tell, I was getting paid more money to be less competent than most of my peers.
In this, it turns out, I was not alone. In July, an essay titled “Twenty Million People Are Faking Their Lives in Beijing” went viral on the messaging app WeChat. Writing under the pseudonym Zhang Wumao, its author wrote about how both the rapid pace of change in Beijing terrified both newcomers and native Beijingers. Zhang bemoaned the great influx of migrants from other parts of China into the capital, the near impossibility of affording housing, and the fact that wealthy Beijingers are mostly plotting to escape overseas. “That leaves more than twenty million people in Beijing, faking their lives,” he wrote. “Here we have only the dreams of a very few and the drudgery of everyone else.”
That was, of course, how Shanghai had made me feel. It didn’t last all that long: within months, I grew more fluent in Mandarin and stopped fumbling my job as a fledging banker, and Shanghai began to feel more like a real home and less like a fake one. The coping mechanisms I’d evolved to deal with a world full of fake facts became sources of small satisfaction, instead of frustration. As for Zhang Wumao, he eventually had to apologize for the controversy his essay caused. “I don’t want any more trouble,” and “the figure twenty million isn’t really accurate,” he said. It sounded like a fake apology.