Last month, Chinese state media published articles commemorating the 30th anniversary of China’s first-ever email: ‘Across the Great Wall, we can reach every corner in the world.’

The email was sent ^[1] from a research institute under China North Industries Group Corporation in Beijing on 14 September 1987 and received by the University of Karlsruhe in Germany at 8.55 pm on 20 September 1987.

Techno-optimists believed that the internet would ensure a free flow of information and ultimately a democratic society in authoritarian states like China. Thirty years on, however, China has instead built a Great Firewall, a vast hardware and software system that aims to prevent access to undesirable websites and censors sensitive content.

While China’s extensive internet censorship isn’t new, there are several myths about how information controls are actually enforced by the Communist Party of China (CPC).

As of June 2017, China’s internet population had reached ^[2] 751 million, which is more than the total population of Europe. The Chinese social media landscape is complex and vibrant. Tencent’s WeChat, China’s most popular instant messaging application, has ^[3] 889 million monthly active users. As of April 2017, Sina Weibo, China’s Twitter-like service, had ^[4] 340 million monthly active users, generating more profits than its counterpart in the West. And the Chinese internet is full of other booming industries, such as the gaming sector, bulletin boards and live-streaming platforms.

Those statistics help to explain why the CPC increasingly sees the internet as the core battleground ^[5] for ideological control and its very own survival. Hence, we’ve witnessed multiple clampdowns on VPNs ^[6] (virtual private networks, which help internet users jump over the Great Firewall and evade censorship), efforts to control narratives in the ideological sphere through the so-called Document 9 ^[7] and other directives ^[8], and restrictions on video-streaming services ^[9] as well as chat applications ^[10], all under the rationale of ‘preventing the spread of illegal information’.

Successful implementation of censorship, which often overlaps with the state’s propaganda agenda, is also a demonstration of the government’s strength in social control. As political scientist Haifeng Huang pointed out ^[11], it’s one of the straightforward tools that’s used to remind people of who’s in charge.

A widely shared perception is that the Chinese government can effectively censor unwanted content in a timely and monolithic fashion. That’s only true to some extent. Analysts tend to overestimate the state’s capacity to impose uniform censorship while underestimating the policy implementation hurdles faced by central and provincial bureaucratic agencies and the extent of the state’s reliance on its symbiotic relationship with internet companies in China.

China’s internet bureaucracy is a two-tiered system. Censorship directives come either from the central government or the provincial one, while high-level policies are made centrally. The management of internet companies follows the principle of territorial management (属地原则, shudi yuanze).

For example, a decision to remove content ^[12] from WeChat should come from Guangdong Provincial Cyberspace Administration since Tencent, WeChat’s parent company, is based in Guangdong. In contrast, general regulations that apply to all internet platforms are drafted by the central Cyberspace Administration of China (CAC).

Below is a list of regulations released by the CAC in a two-week span between August and September 2017.

Regulation	Rule	Impact
Internet Forum Service Management Regulation [13] (互联网论坛社区服务管理规定)	Article 8: Users should be denied service if they do not register under their real identities for online forums and message boards.	Further prevent anonymity on the Chinese internet. Re-emphasise companies’ responsibility to monitor and control content on their platforms. Push the burden of content monitoring and controls further down to individual users. Result in further self-censorship and chilling effects.
Internet Thread Comments Service Management Regulation [14] (互联网跟帖评论服务管理规定)	Article 9: Providers of commenting and posting services should create a credit system where users will receive ratings that determine their scope of service. Severe violators of regulations should be blacklisted and denied future services. Governments should keep a credit file on users.
Internet User Public Account Information Services Management Regulation [15] (互联网用户公众账号信息服务管理规定)	Article 6: Internet users must provide their organisation, national identity documents and mobile phone numbers or be denied service. Article 13: Companies should also set up credit rating systems tied to user accounts.
Management Rules of Internet Group Information Services [16] (互联网群组信息服务管理规定)	Article 4: Providers of information services through internet chat groups and users must adhere to correct guidance, promoting socialist core values, fostering a positive and healthy online culture, and protecting a favorable online ecology.

The implementation of censorship often involves pushing the burden of content monitoring and controls down to the lowest level possible. All internet companies operating in China are subject to laws and regulations that hold them legally responsible for content on their platforms. They are expected to invest in staff and filtering technologies to moderate content and stay in compliance with government regulations. Failure to comply can lead to fines ^[17] or revocation ^[18] of operating licences.

While the state’s reliance on companies’ compliance serves as an effective sword of Damocles, it can result in ‘cracks in the Great Firewall ^[19]’, if it’s in companies’ interests to resist government efforts and provide otherwise censored content to compete with other platforms. It makes a temporary failure of censorship ^[20], whether due to technical glitches or deliberate defiance, possible.

Moreover, there is no clear-cut, easy-to-follow guideline for companies to decide what ought to be filtered from their platforms. In 2010, China’s State Council Information Office published a major government-issued document on its internet policy. It includes a list of prohibited topics ^[21] that are vaguely defined, including ‘disrupting social order and stability’ and ‘damaging state honour and interests’. This regulatory environment pushes companies to over-censor, a phenomenon China expert Perry Link described ^[22] as ‘the anaconda in the chandelier’.

The commercialised process often leads to a decentralised and fragmented application of censorship on the Chinese internet. Those amorphous censorship practices have also caused ^[23] unintended abuses, including China’s ‘black PR’ business ^[24] and companies’ blocking of non-sensitive content out of commercial interest ^[25].

Studies of mobile games ^[26], chat applications ^[27], blogs ^[28], search engines ^[29] and live-streaming platforms ^[30] in China by the University of Toronto’s Citizen Lab and others have consistently found variances in how companies implement censorship. Those empirical studies reveal that while companies may be receiving general directives on prohibited content, they haven’t been given a central list of banned keywords.

Chinese leaders are well aware of the dilemmas in the country’s current internet management system and its management of content via companies. That is why Xi’s administration has taken big steps in reshaping China’s internet control agencies by establishing the CAC and passing the influential Cybersecurity Law.

It’s unclear how effective those internet management reforms will be, but it seems evident that 30 years after it first embraced global cyberspace, the CPC has decided that it’s in the party’s best interests to ‘traverse the hurdle represented by the internet ^[31]’ rather than simply letting its people reach ‘every corner of the world’ online.