You are not logged in. LOG IN NOW >

Even For Censorship Savvy China, ICTs Can Cut Through Corruption, Study Finds

BY Rebecca Chao | Friday, May 23 2014

Just how much can China's Great Firewall take? (credit: 阮_先生/Weibo)

In a few years from now, or perhaps it has already happened, mention “human flesh search engine” to a Chinese netizen and they may get glossy-eyed with nostalgia -- the good old days when a digital probe into the life of a politician or wealthy businessman could potentially uncover a trail of corruption: illegally obtained houses, hidden wealth, shady transactions. Now that these searches have largely fallen out of use -- and one can safely assume, due to the intimidation and jailing of those who have spread online “rumors” -- is the fight against corruption lost? A new study conducted by two Taiwanese scholars concludes, perhaps not.

The study, conducted at the National Sun Yat-sen University, is an empirical, numbers-based approach in determining whether ICTs can reduce corruption and improve governance in what the authors say is the most sophisticated country in the world when it comes to censorship.

“In terms of the time interval, speed, and number of stipulations, there is no other country like China in the world,” the authors, Ming-Hsuan Lee and Mon-Chi Lio, write. “The government firmly believes that aggressive and strict control and guidance of public opinion are the only ways to maintain the Chinese regime.” But with all these controls, can ICTs still improve governance and decrease corruption?

The study finds that it can:

The results show that ICT did have a positive and significant impact on the performance of provincial governments in China; the higher the penetration of ICT, the higher the governance scores were. [...] In addition, we found that ICT had two different effects on government corruption: ICT helped expose cases of corruption; ICT also helped reduce government corruption. The first effect may come from the fact that ICT made it easier to report cases of corruption and enhancing the capabilities of procuratorial offices; the second effect may originate from the fact that ICT imposed supervision on the government. Furthermore, among individual ICTs, the Internet had the largest impact on government corruption.

Compared to all other ICTs, the Internet is the most useful in combatting corruption, the study finds: “Comparing the impact of individual ICT variables on corruption reduction, the coefficients of internet are the greatest, suggesting that the Internet had the most significant impact on exposing and reducing government corruption."

The authors go on to speculate that in the absence of democratic institutions that allow for greater citizen oversight, “the Internet may have become a new oversight mechanism and supervised the behavior of the Chinese government.”

Meanwhile, the effect of mobile phones and telephones on corruption appear negligible. The authors explain why:

...the reason for the insignificant impact of mobile phones and telephones may be that the functions of mobile phones and telephones are relatively limited compared to the Internet. Although mobile phones have mobile functionalities and are highly portable, they lack the ability to process information. The functionality of landline phones is the most limited. Therefore, the Internet had the greatest impact on governance performance while mobile phones and telephones had less impact on governance.

Ming-Hsuan Lee clarified in an e-mail to techPresident, however, that the study of mobile phones is limited to the 2004 and 2010 data that they used, so it is not clear how Internet-enabled phones factor into the results; they would have to rerun the empirical tests with the new data in order to see if there have been changes in the effect of mobile phones on corruption. The study also precedes the development of WeChat, a mobile phone app that is surpassing Weibo in popularity and the rise in cheap smartphones appears to be a more recent phenomenon.

A Mobile Future?

Even without the empirical data, stories of anti-corruption efforts abound in the media and might even suggest that there has been a recent switch from Internet to mobile in the fight against corruption.

In November of last year, investigative journalist Luo Changping tried to post information about a corrupt senior economic official named Liu Tienan but Luo's posts were soon deleted. He then turned to WeChat. While it is speculated to also be subject to censorship, particularly in Tibet, Luo found a work-around and used it to publish a series of stories about the corrupt official.

The less public nature of WeChat makes it less likely that users will get in hot water for controversial posts, explains The Economist, and the government finds this “slow burn” spread of information more palatable than the rapid fire of Weibo:

On WeChat, it is usually only subscribers to a public account who will see a post (though such posts may also be viewed on a separate web page), and if a subscriber forwards a post, only that subscriber’s circle of friends see it. Its non-public accounts are even less open. Information on WeChat spreads at such a slow burn that authorities feel they have more control over it. Also in contrast to microblogs, many types of public account (like Mr Luo’s) can send out only one post to subscribers a day, making them much easier for authorities to monitor.

China has also recently cracked down on online rumors and has sentenced its first “rumormonger” to three years in jail for spreading a rumor via Weibo that the Chinese government gave more in compensation to Western victims of the 2008 high speed rail crash than to Chinese victims. Anyone who spreads a false rumor with a Weibo account that has more than 5,000 followers or gets reposted more than 500 times can go to jail.

Hard to Plug the Holes

As sophisticated as China has grown in its ability to censor, it is still unable to fully wipe out all politically sensitive material from the Internet, another reason why it remains a useful tool in rooting out corruption.

A 2009 study of Chinese censorship by Rebecca MacKinnon revealed that sometimes 50 percent or more of posts with censor-trigger words remain online. While “corruption” is one of 50 categories under which officials censor materials, a test revealed that the most rigorous censor deleted only 9 out of 15 or 60 percent of posts while other censors deleted zero to four out of 15 posts.

Jason Q. Ng, who created a computer script to detect and analyze over 500 censored words on Weibo, argues that while China's censorship regime is fairly nuanced, one pattern has emerged: China sometimes shrugs its shoulders at sensitive posts if they do not have a wide audience or make a call to arms. Ng writes in his book, Blocked in Weibo, "However, the censors are not infallible, and it is possible for posts with banned words to escape the censor's eye--so long as they don't gain too much attention or advocate collective action, or perhaps if they're celverly embedded inside images or obscured in coded language"

Ng also notes that Chinese Internet users have developed clever ways to circumvent censorship:

Over the years, in a series of cat-and-mouse games, Chinese Internet users have developed an extensive series of puns--both visual and homophonous--slang, acronyms,memes, and images to skirt restrictions and censors. Such creative usages may still be helpful in evading the censor's eye on Weibo--using a code makes one's post both less likely to get caught in any automatic search filter, and less likely to be found by a human censor later on. Furthermore, Chinese Internet users have mastered the use of irony as protest, reaching the point where emphatically pro-government comments online such as "Socialism is good" and "I have been represented by my local official" are often meant to be satirical. Filtering tools including the ones Weibo uses in its search engine certainly can't recognize such subtleties. In some respects, the filters are "easy" to defeat, emphasizing just how important those human monitors employed by Weibo are. They have the ability to delete individual posts and even entire accounts, which is what happened to the account of Ai Weiwei, the dissident artist.

China's team of human censors is made up of roughly 20,000 to 50,000 Internet police and 250,000 to 300,000 "50 cent party members" who are paid to post pro-government comments.

Then there are those who are working deliberately to end censorship in China, like Charlie Smith who founded the anti-censorship organization and website He claims he has found a nearly foolproof way of ending censorship in China called "collateral freedom." It's an extremely laborious and technical process but it makes censoring one site nearly as impossible as removing your own shadow. For example, last fall, Greatfire placed a version of the censored Reuter's site onto Amazon's cloud service. If China wants to censor that site, it has to either shut down Amazon in China -- a huge economic blow -- or make Amazon remove the site. So far both China and Amazon have done nothing to take down the mirror site.

In the meantime, the Chinese government revealed yesterday, the latest stage in its own private fight against corruption: the removal of "naked officials" -- not the ones that human flesh search engines have been known to unearth through racy photos -- but officials who are deemed a group liable to corruption because they live in China while they send their wives and children overseas. It appears the first victim of this new crackdown is Fang Xuan, deputy chief of the Guangzhou city Party Committee, who was not actually charged with corruption but forced into early retirement. In February, the CCP had issued an internal document to all officials: “Bring back your family, or retire early.”

Personal Democracy Media is grateful to the Omidyar Network and the UN Foundation for their generous support of techPresident's WeGov section.

For a round-up of our weekly stories, subscribe to the WeGov mailing list.