How the Chinese Government Fabricates Social Media Posts for Strategic Distraction, not Engaged Argument

Title: How the Chinese Government Fabricates Social Media Posts for Strategic Distraction, not Engaged Argument

Publication Date: January 14, 2017

Authors: Gary King, Jennifer Pan, Margaret E. Roberts

Links: Paper / Atlantic Summary

Very comprehensive paper that digs into the mysterious '50c party' - the supposed army of piecemeal-paid social media posters who engage in online arguments on behalf of the Chinese government, pushing the party line and drawing skepticism about government critics.

Instead, this paper found that the 50c party was mostly government workers, and that they didn't engage in arguments so much as blast the internet with general pro-government rhetoric. This behavior did happen in specific bursts, and particularly right after a riot or other collective action, which implies that it is highly coordinated from the top down.

The researchers conclude that the government's strategy is generally not to interfere with arguments - which could be useful to them as a gauge of public sentiment - but mostly to distract attention away from the arguments and particularly away from collective action, which is the only thing they are really concerned about.

Summary:

• The Chinese government’s censorship strategies are well-known (i.e. the Great Firewall).

• But they also conduct “astroturfing” or “reverse censorship”, posting large numbers of fabricated social media comments. The people hired for this are known as 50c party members, because they are rumored to be paid 50 cents per comment.

• Most journalists, academics, and other social media users seem to agree on what 50c commenters do: engaging others in hand-to-hand verbal combat defending the Chinese government, making specific arguments that support the government and labeling critics as traitors of the country.

• Everyone also agrees that the extent and nature of the 50c party is not well known.

• This study analyzes a recently leaked archive of all 2013 and 204 emails to/from the Internet Propaganda Office. They harvested ~ 40k known 50c posts.

• Found that all of the 50c workers in the sample are government employees, as opposed to the expected ordinary citizens hired for piecemeal wages. It doesn’t seem that the employees were being paid extra, suggesting that the activity is a requirement of their existing job or rewarded in performance reviews.

• The 50c posts were not random or uniformly distributed, but focused into distinct bursts. Examples: posts about veterans and martyrs on Tomb Sweeping Day. Posts about President Xi Jinping’s “China Dream”. Posts immediately following Shanshan riots in Xinjiang, about China Dream, local economic development, etc. These kinds of bursts suggest a strategy of distraction.

• Almost none of the posts fall into the categories “taunting of foreign countries” or “argumentative praise or criticism”. About 80% fall into the Cheerleading category and 13% in Non-argumentative praise or suggestions

• public opinion guidance - official term for CCP policies designed to influence public opinion

• Suggestion that this program is run from the top: right after President Xi Jinping led the first meeting of the Central Leading Group for Internet Security and Informatization, during which he stressed the need for government officials to “have a good grasp of the timing, degree, and efficacy of online public opinion guidance so that online spaces are clear and unclouded”, 156 50c social media accounts were created, well above average.

• The final stage of the research involved a kind of “survey” by creating many fake social media accounts and DM-ing on weibo, asking in positive language if 50c posters were trained in “public opinion guidance”.

• 59% of predicted 50c party members admitted to it.

• Estimate 50c party writes 448 million social media posts each year, half on government sites and half on social media sites.

• Purpose is to 1) shut down arguments using distraction, and 2) divert public attention from actual or potential collective action.

• May be a useful strategy because 1) the actual supportive and critical commentary are useful to leaders in judging local leaders, so artificially changing those would be problematic. 2) Distraction can be the best way of putting an argument to rest. It also reduces anger rather than ruminating. 3) Censorship does anger people, so astroturfing is a way of controlling public opinion without censoring as much.

• One theory about authoritarian governments is that they slow the spread of critical information in order to minimize the development of common knowledge of grievances, which could reduce the probability of mobilization.

• These researchers are suggesting that the Chinese regime differentiates between two types of common knowledge: specific grievances, which they allow, and collective action potential, which they actively suppress. The idea is that common knowledge of grievances is already commonplace, so spreading it isn’t a risk. What is risky, though, is posts with collective action potential.