When vicious inter-ethnic violence broke out in Urumqi last year, Chinese authorities flooded the city with security forces. But next came an unexpected step: they cut off internet access across the vast north-western region of Xinjiang. Controlling the information flow was as crucial as controlling the streets, it seemed.
Eight months on, the net remains largely inaccessible in Xinjiang, though officials claim it will soon be restored. The small number of sites that were recently unblocked are heavily censored; only a severely restricted email service is available.
The internet blackout is partly an anomaly, made possible by the region’s poverty and remoteness. It is hard to imagine the authorities gambling with Shanghai or Beijing’s internationalised economies.
But it also reflects the government’s wider approach to the internet: real fear at the speed with which information or rumours can spread and people can organise. And an absolute determination to tame it. The cut-off lies at the extreme end of a spectrum of controls that experts say constitute the world’s most sophisticated and extensive censorship system – and one that is growing.
Google’s decision to shut its mainland search service rather than continue to self-censor has highlighted a crackdown that has closed thousands of domestic websites over the last year and blocked many hosted outside. Casualties included Yeeyan, a community translation website that was running a collaborative experiment with the Guardian, publishing stories in Chinese. Though it was later allowed to reopen, it no longer translates foreign news media. Other measures have included attempts to introduce real name registration and install controls on individual devices via the controversial Green Dam software – though the latter has been seen off by users, at least for now.
“A lot of people are very optimistic that the web will bring us ‘glasnost for China’, but the determination to control it is stronger than ever,” said David Bandurski of the China Media Project at the University of Hong Kong. Few, even those who deal with it regularly, can explain exactly how the censorship system works.
Chinese officials say that most countries control the internet and that Beijing does so according to law. The state proscribes up to 11 kinds of content, which range from spreading obscenity to “disrupting national policies on religion, propagating evil cults and feudal superstitions”.
The difference is not only that China outlaws far more content than other countries, but it does not state clearly what is off limits, why, and who made the decision.
“The special nature of Chinese censorship is that it has no transparency and is very random,” said Wen Yunchao, an outspoken Guangzhou-based blogger better known as Beifeng (North Wind). “When my blog has been closed or deleted or blocked, I have had no notice. After I find out there is no way to petition for its return. No netizens know which department made the decision; no one knows how the system works. We only see the results.”
The first element is the so-called Great Firewall. Internet police bar access to services hosted overseas by blocking URLs and IP addresses and through keyword filteringhttp://www.theguardian.com/world/2010/mar/23/google-china-firewall-censorship-internet. Although many internet users find ways to scale the firewall – such as proxies and virtual private networks – most either do not know how or cannot be bothered.
But the second element is just as crucial: domestic censorship. And here the authorities do not rely on controlling companies – but on making sure that they control themselves. In evidence to a US House of Representatives hearing this month, Rebecca MacKinnon, a fellow at Princeton’s Center for Information Technology Policy, wrote: “Much of the censorship and surveillance work in China is delegated and outsourced to the private sector – who, if they fail to censor and monitor to the government’s satisfaction, will lose their business licence.”
Punishments can also include fines – or having sites closed, without warning. The decentralisation of censorship means that it is highly uneven. The transparency of filtering, content censored and methods used all vary. When MacKinnon tried to post 108 pieces of content on 15 different blog hosts, the most vigilant censored 60 posts; the least, only one.
That is one reason why the system is hard to understand. Another is that it is deliberately opaque; at least one journalist, Shi Tao, is in jail for disclosing a propaganda directive. It is unsurprising that so few people want to discuss the system. But conversations with bloggers, industry sources and experts allow one to build up a portrait of it.
Most sites depend on both mechanised and human observation. Filtering software rejects posts outright or flags them up for further attention, but humans are essential to catch veiled references and check photographs, for example. Sources suggest a huge portal – such as Sina, which runs not only news, but a microblog service and discussion forums – could employ anywhere between 20 and 100 censors.
These people take orders from a complex apparatus that briefs them about what is permitted, orders changes or deletions to content, and punishes them.
At the top are the party’s propaganda office and the information office of the state council, China’s cabinet. These deal with the biggest issues and set out a general approach; and they are replicated downwards through the levels of the bureaucracy.
They have their own monitors who usually work as a single, city-level team . They focus primarily on news-related websites, though this may include blog platforms, for instance. They shape content as well as removing it.
Separately, there are internet police, dealing primarily with security and crime-related issues – though in China these are interpreted very broadly. Usually they order deletion or site closures, rather than issuing directives in advance. It is thought they focus on social websites, such as bulletin boards; and their work veers towards information that may have offline repercussions for police – perhaps complaints that could lead to unrest – and investigations.
These three groups form the bulk of the official censors. Within their ranks dedicated teams watch major sites such as NetEase or Sohu. Like the firms’ own monitors, they work round the clock. Others will run searches on keywords, or check out tips from the public. Increasingly, they are investing in technology such as text mining software.
But there are numerous other entities overseeing parts of the internet – the broadcasting administration deals with videos, for instance. Although they coordinate and cooperate, responsibilities overlap and turf wars can result.
Additionally, each government department or agency keeps an eye on its own patch, with education officials monitoring student discussion forums. Usually these bodies will act via police or information and propaganda officials; sometimes, they call websites directly. It is not clear if they have the right to order deletion of content, but companies usually comply rather than take the risk. “Literally, pretty much every government office – national, provincial, city level – will try to contact you if they have your number,” complained an industry figure, joking that senior staff hate to give out business cards as a result.
Depending on the sensitivity of content and where it appears, orders arrive in different ways. Censors may tell a site to delete content, or tell an ISP to pull the plug. Sometimes a data centre informs a company they have been told to shut a site; sometimes they act without warning or explanation.
Almost always, the information is delivered informally. Authorities will call major players to meetings to brief them; otherwise, they will ring up, send a text message or use online chat services. In a normal week, a major site can expect to receive at least tens of instructions. Smaller sites often rely on word of mouth. But companies usually maintain logs to keep track of these unwritten injunctions and occasionally these leak, offering a rare glimpse into the complex and sometimes baffling decisions of the censors.
One, said to have belonged to Chinese search giant Baidu, included banned sites, off-limits topics and sensitive words or phrases including Aids, migrant workers, reactionary and even Communist party; not problematic in themselves, but in certain combinations.
The censors’ decisions can be baffling – allowing tales of official abuse to remain while apparently innocuous subjects are erased; one list translated by the New York Times this week included an order to delete references to a rare flower. According to Buddhist folklore, it blooms only once every 3,000 years; someone feared it would encourage superstition. “The rest of us have to guess their likes and dislikes,” complained one of those struggling to keep up. Officials do not only prohibit content. They dictate how stories are reported; or if reader comments should be allowed; or set a ratio of “positive” news on sites.
“In the past it was about deleting. Now they do much more[proactively] – saying what we should put on the headlines,” added the industry source. “They are smarter. People open the home page and see tabloid-style or party news and lose interest. They don’t bother to scroll down to the interesting stuff.”
There is one more port of call for Chinese authorities: the individual.”They don’t haul in everyone,” said Phelim Kine, Asia researcher at Human Rights Watch. But “discussion of the Chinese government’s monopoly on power; the need to examine 1989; calls for open elections – those are red flag issues and can get people a knock on the door from the public security bureau”.
Then there are the “grey lines”; topics and discussions acceptable today but perhaps not tomorrow. A summons to an intimidatory chat with police; detention; or even jail could be the result. Amnesty International calculated last year about 30 journalists and 50 others were in prison “for posting their views on the internet” – often, because officials suspected they were seeking to organise others.
But the deletion of content, the closure of sites and the jailing of individuals have a chilling effect that goes far beyond those immediately affected.”Because people don’t know what they are not allowed to say, they kind of guess and take down or stop saying whatever might possibly not be permissible,” said Wen Yunchao. The system’s effectiveness can be judged not by the number of censors, the sophistication of its technology or the information suppressed – but by what it has not done and does not need to do. Li Yonggang, a professor of internet politics at Nanjing university, wrote recently: “In fact, the Great Firewall is rooted in our hearts.”
- OPINION: Google And China Have Similar Ideas About Censorship And Privacy
- China turns to propaganda to right image in virus ‘war’
- Age of Ambition review – an intimate portrait of China
- China turns to propaganda to right image in virus 'war'
- WikiLeaks row intensifies as US makes 'privacy' move against Twitter
- Four die after coronavirus quarantine hotel collapses in China - as officials now say 38 survived disaster following earlier reports that 48 were saved
- China muzzles The Cannon after tycoon's criticism of media clampdown
- China turns to propaganda to right image in coronavirus 'war'
- Ken Glueck: Google Committed ‘Bank Robbery’ with Oracle Case
- Italy's Lombardy seeks tightening of screw to defeat coronavirus
- China, US spar over origin of coronavirus
- China reframes coronavirus narrative, touts Xi's accomplishments despite bodies piling up
Google row: China's army of censors battles to defeat the internet have 1886 words, post on www.theguardian.com at March 23, 2010. This is cached page on Asia News. If you want remove this page, please contact us.