Chinese censors’ looser social media grip ‘may help flag threats’
The central government could be letting some talk of protests and corrupt officials to flourish to keep tabs on social risks, researchers find
The Chinese government may be allowing some discussion of protests and allegations of corruption on social media to keep a closer eye on local officials and potential threats to stability.
That is the conclusion of a team of researchers from Hong Kong, Sweden and the United States who mined a data set of more than 13 billion blog posts in 2009-2013 on Sino Weibo, the mainland’s biggest microblogging platform.
The mainland has tightened its grip on online comments in recent years, including detaining some of the most popular Weibo commentators.
But “a shockingly large number of posts on highly sensitive topics were published and circulated on social media” during the five-year period, the researchers said in their paper in the Journal of Economic Perspectives published by the American Economics Association.
The study’s authors – University of Hong Kong assistant professor Qin Bei, University of Southern California assistant professor Wu Yanhui and Stockholm University professor David Stroemberg – found that 382,000 posts in the data set alluded to conflicts and as many as 2.5 million related to mass events such as protests and strikes.
A large number of posts also lobbed explicit allegations of corruption against officials.
“It’s different from the popular view that everything sensitive is censored,” Qin said.“But it’s not an accident.”
The researchers found that many of the sensitive posts were posted “before or concurrent with the events and predict the events”.
“The fact that people begin discussing events before they happen indicates that Sina Weibo may be used to organise or at least to coordinate collective action events,” the paper said.
The researchers also found that charges were often laid against local officials about a year after allegations about their corrupt appeared online. And of the 680 officials accused of corruption in the data set, those who were eventually charged were referred to nearly 10 times more frequently in the posts than those who were not charged.
The researchers concluded that the central government was choosing to not wield its power to delete such sensitive posts in order to learn about potential threats.
“A complete clean-up of sensitive content can impair the regime’s ability to learn from bottom-up information and to address social problems before they become threatening,” the paper said.
The researchers said the authorities may have allowed these accusations to surface online to tap into the power of bloggers to improve monitoring of local officials.
“Social media in China is likely to increase the power of the central government at the expense of local governments,” they said.
Wu said: “There’s so much debate on the role of social media in countries like China, where power is more concentrated. And a lot of people are sceptical.
“But our research shows social media can provide a lot of quality information.”
The researchers added that even after the government ramped up censorship online and other popular social media like WeChat emerged towards the end of their study period, the number of Weibo posts about sensitive topics like protests, strikes and corruption remained largely unchanged.
Additional reporting by Jun Mai
