China’s growing army of paid internet commentators
*With research assistance provided by Maggie Shum
Since 2005, observers of the Chinese blogosphere have noted the presence of users who are paid to support the authorities in online discussions, often referred to as the “50 Cent Party” for the alleged fee they collect for each posted comment. Several incidents in recent weeks have once again drawn domestic and international attention to this effort by the Chinese Communist Party (CCP) to manipulate public debate. Last month, the Propaganda Department in a small county in Hubei Province accidentally posted online a video showing local officials training progovernment internet commentators. The media quickly took interest in the revelation, which stipulated commentators’ duty to guide public opinion in a “constructive” way and engage the internet as “an important battlefield of ideology.” Meanwhile, at a seminar in Beijing at the end of September, the vice minister of public security said that the police should use microblogs as a communication platform to “release correct information and dispel misunderstandings.”
These incidents provide snapshots of a much broader phenomenon that reflects the decentralization and institutionalization of the CCP’s effort to inject the party line into online conversations. Unlike in other countries, the campaign is not an ad hoc effort to preserve the reputation of individual officials or perform short-term damage control. Instead it is a comprehensive CCP policy, accompanied by a vast system of trainings and rewards.
A quick online search of Chinese official websites by researchers with Freedom House’s China Media Bulletin identified over a dozen such training workshops that had taken place throughout China over the past year—including events in Anhui, Shandong, Yunnan, and Zhejiang. Many of them were held in small cities and districts, including in ethnic minority areas like Inner Mongolia and Xinjiang, where party leaders are particularly wary of the internet’s ability to mobilize public protests. Attendance at each training ranged from 50 to 200 people. An examination of the official reports from these workshops indicates that most of the attendees were individuals employed by the government in some capacity, though some trainings likely included private citizens.
Such examples suggest a possible distinction between two kinds of state-supported internet commentator, though the line between them may be blurred in practice:
- The “50 Cent Party” model – Private citizens are paid by the government to pretend to be “ordinary” netizens while in fact promoting the CCP line and seeking, in a nontransparent fashion, to guide public opinion in a particular direction.
- The “Official PR” model – Government employees from a range of departments hone their public-relations skills in order to get the official message out to the public more quickly and effectively. In this case, the commentators’ identity is transparent, but the information conveyed could still be spun or blatantly deceptive. Moreover, there remains the possibility that these government workers might also engage in online conversations under assumed identities, thereby crossing over to something more closely resembling the “50 Cent Party” model.
Of course, there is nothing wrong with a government responding to the views of citizens online or using social-media tools to disseminate information. In fact, many advocates have promoted greater internet access as a means of stimulating such citizen-state dialogue in all countries. But in the Chinese context, the dynamics are not quite so simple.
Whether the progovernment commentary is “planted” or transparent, it is appearing in an information environment that is already heavily manipulated through large-scale filtering and deletions. Freedom House’s 2011 Freedom on the Net report on China details the many layers of the world’s most comprehensive and sophisticated apparatus for censoring online content.
Indeed, accounts describing the operations of progovernment commentators that have emerged in recent years indicate that they are also involved in identifying and recommending content for deletion. Other testimonies highlight the fact that the posts do not only praise or support the CCP and government policy, but also target government critics with negative remarks. Other forms of misdirection involve deliberate attempts to muddy the facts of a particular incident—for example, a false eyewitness can contradict the account of a netizen reporting a case of police abuse.
Though politicians around the world are increasingly using new media to reach their constituents, the kind of deceptive measures employed by the “50 Cent Party” would not be tolerated in a democracy. They would also be unnecessary for a democratic government, since elected leaders have genuine supporters willing to defend them on the internet free of charge, and the population can turn to a free press, rather than self-proclaimed “fellow citizens,” to verify officials’ public statements. Even the worst-case scenario for such a government—a tidal wave of critical opinion online—would simply lead to an orderly rotation of power through elections. In China’s rigid one-party system, the stakes are much higher.
In July 2008, David Bandurski of Hong Kong University reported that there were an estimated 280,000 paid web commentators of the “50 Cent Party” variety. At the time, China was home to about 250 million internet users. Three years later, the number of users has doubled to 500 million, according to official figures. Meanwhile, trainings for government employees preparing to engage in online public-relations efforts have expanded.
The question thus arises: Is there now an army of over 560,000 private citizens being paid by the government to manipulate discussions on the Chinese internet? And, extrapolating from the level of participation in the official trainings, are an additional several hundred thousand government employees injecting CCP-dictated information and views into the blogosphere?
Regardless of its exact size, the phenomenon described here amounts to a propaganda exercise of staggering proportions. Rather than fostering an open dialogue with Chinese citizens, the project is aimed at deceiving them into thinking that public support for the CCP and its policies is much more prevalent than it actually is, and that abuses like corruption and torture are much less common than they are in reality.
It is unclear how successful these commentators have been at “guiding public opinion” online, and their ultimate impact on Chinese netizens’ perception of their country and the rest of the world remains to be seen. Whatever the precise effect within China, because the CCP serves as a model for other economically ambitious but democracy-averse governments, this and other forms of online manipulation are already spreading to other corners of the global internet.
Analyses and recommendations offered by the authors do not necessarily reflect those of Freedom House.
China’s media environment remains one of the world’s most restrictive. As described in Freedom House’s recently released report on the state of global press freedom for the year 2012, the Chinese government’s press restrictions were complex, intricate, ruthless when necessary, and flexible when it suited the leadership’s purposes. At the same time, these controls were subject to pushback from ordinary citizens outraged at the suppression of information about critical events.
Political cartoon depicting the detention of Uyghur students in Egypt, and in some cases their deportation, at the apparent behest of the Chinese government. Image Credit: Credit: Rebel Pepper/Radio Free Asia.
Authoritarian rule in China poses a growing threat to democracy elsewhere.
In the run-up to the 25th anniversary of the Tiananmen Square massacre, the Chinese government has escalated its efforts to suppress any form of commemoration or discussion of the event in the media. But Beijing’s daily attempts to control the news extend far beyond this taboo topic, as an analysis of recent censorship directives reveals.