The Filtering Matrix: Reality and the Great Firewall of China



In a recent entry, I argued that those interested in issues of Internet censorship, filtering and surveillance should make an increased effort to demystify the technology being used, be more accurate and thorough when investigating these issues, and avoid speculative scenarios. However, we should not underestimate the filtering and surveillance regime in China: Internet Filtering in China is a reality.

Internet filtering in China is best described as a matrix of control in which technological and non-technological measures intersect at different levels of access to enforce strict information control policies. Such a matrix of controls (combining filtering and surveillance technology with fear, intimidation and imprisonment) discourages most people from attempting to access banned information, let alone actively seek out means to circumvent such controls. Exaggerating China�s power of information control may be mistaken, but underestimating it is worse.

The Filtering Matrix: Reality and the Great Firewall of China

Internet censorship in China has been the subject of much speculation and often extreme exaggeration. There have been reports that “big mammas” monitor every Internet chat room in China and delete content in real time right off of users’ computer monitors while special filtering devices block access to hundreds of thousands of political websites. Claims have been made that foreign search engine results are technically manipulated to return only results favorable to the Chinese government and that there are as many as 80,000 cyber police tracking every internet user’s every move.

Often I have found myself frustrated by exaggerations such as these, and have spent considerable time pointing out misconceptions and correcting mistakes. I have argued that those interested in issues of Internet censorship and surveillance should make an increased effort to demystify the technology being used, be more accurate and thorough when investigating these issues, and avoid speculative scenarios that overestimate what countries like China can do to control information.

However, there are also those who underestimate filtering and surveillance in China, with some going so far as to describe it as a myth. A recent example can be found in an article by Globe and Mail columnist Margaret Wente, titled ‘The e-liberation of China’s youth’. In this article, Wente seriously understates, and even dismisses, the level of Chinese net censorship. She suggests that simply due to the Internet’s availability “government efforts to censor foreign news are futile,” and that “young English-speakers” in China can access all the information they want, including information about SARS and English language news from “CNN, the BBC and The New York Times.” “So much for censorship,” Wente briskly concludes.

I am not sure what methodology Wente employed to test her theories, but she clearly was not exhaustive in her search for banned websites. If she actually bothered to visit the website, http://news.bbc.co.uk/ while in China, she would have found it inaccessible, as is the Wall Street Journal. Likewise, if she had tried to search for the terms “falun” or “falun gong” in any search engine through the Chinese Internet, the result would have been dramatically different than doing so through a computer in Canada. The same goes for human rights, dissident, and religious websites, and scores of others having to do with the independence of Taiwain and Tibet or Tianmenan Square.

How do I know? Rather than sitting down at an Internet caf� for five minutes, my colleagues and I have actually spent several years systematically testing which websites are blocked in China using computers based within that county that are spread across multiple backbone networks and Internet Service Providers. While it may be silly to exaggerate the extent to which China can control information on the Internet, it would plain wrong to say it is a myth altogether. China does censor the Internet, but maybe not the type of information that someone would casually stumble across. Information may also only be censored for a limited period of time; websites may be blocked and unblocked at any given time. The results may also vary regionally or at different levels of Internet access.

Internet filtering in China is best described as a matrix of control in which technological and non-technological measures intersect at different levels of access to enforce strict information control policies. Non-technical measures include self-censorship (such as the “Public Pledge on Self-discipline for the Chinese Internet Industrysigned by Yahoo!; Google’s de-listing of news websites that are blocked in China when its Chinese News portal is accessed by Internet users in China), arrest and intimidation, all of which stifle the type of brash cyber-adventurism of the type those who dismiss China’s censorship practices — often temporary visitors to China — promote so cavalierly.

The non-technological measures are complemented by a variety of filtering mechanisms that target multiple levels of access. For example, domestically developed commercial filtering software (such as Filter King and Net110), similar to the parental control software used here in Canada, is mandated for use in Internet caf�s and is linked to the Public Security Bureau. There are also controls placed by the state on the backbone routing system through which connections to the international Internet are routed, parsing through traffic to disrupt access to specific IP addresses, domain names, or specific keywords.

While English speakers may have access to New York Times and CNN, as Wente apparently did, most countries that filter, including China, focus their efforts on local language content, which is what the vast majority of people living in the country are searching for and will read.

Other filtering mechanisms are employed at different levels of access, though not always uniformly applied. There are efforts to restrict private chat, including evidence of keyword filtering that has been found in certain version of the popular IM client QQ. International blogging sites have been blocked, while domestic blogs sites have been closed and later re-opened. Online forums do have messages with certain content filtered and/or deleted, but not all of them, and not all of the time.

While it is true that Internet censorship and surveillance is not, on a technical level, always 100% effective (especially if one has family or friends outside the country that can set-up software to help get around it) countries like China can and do maintain an effective censorship policy combining filtering and surveillance technology with fear, intimidation and imprisonment. Such a matrix of controls discourages most people � local people, that is — from attempting to access banned information, let alone actively seek out means to circumvent such controls.

Cavalier dismissals of these controls are not only wrong, they blind us to the ways in which our own countries may be implicated in them. Western corporations have played a large role in supplying China with sophisticated filtering and surveillance technologies. For example, a recent Canadian trade mission to China explored opportunities for Canadian companies sell advanced surveillance technology to China, ostensibly for use during that country�s upcoming Olympic Games. Human Rights groups have called on Canada to implement safeguards to ensure that these Canadian technologies will not be used to undermine human rights or disrupt access to information.

Exaggerating China�s power of information control may be mistaken, but underestimating it is worse. Failure to acknowledge the grim reality of Internet filtering and surveillance in China encourages citizens in democratic states to think there is nothing wrong with the sale of security and surveillance technology to a country like China. At the very least, it prevents a factually informed debate from taking place whether we should be doing so at all.

Post a comment.