Expanding Freedom and Democracy
Freedom on the Net 2017

China

Not Free
13
100
A Obstacles to Access 8 25
B Limits on Content 5 35
C Violations of User Rights 0 40
Last Year's Score & Status
12 100 Not Free
Scores are based on a scale of 0 (least free) to 100 (most free). See the research methodology and report acknowledgements.

header1 Key Developments, June 1, 2016 - May 31, 2017

  • A cybersecurity law passed in November 2016 strengthened requirements for network operators to register Chinese users under their real names and store their information within China (see Surveillance, Privacy, and Anonymity).
  • Evidence emerged of greater censorship on WeChat; several people were also detained in relation to comments shared on the messaging platform (see Content Removal and Prosecutions and Detentions for Online Activities).
  • New regulations sought to limit user-generated news content, adding licensing requirements for all forms of digital news gathering and dissemination (see Media, Diversity, and Content Manipulation).
  • Rules issued in 2017 introduced licensing for virtual private network (VPN) tools, which are used to bypass censorship (see Blocking and Filtering).
  • Activists received sentences of up to 11 years in prison for advocating democracy online (see Prosecutions and Detentions for Online Activities).

header2 Introduction

China was the world’s worst abuser of internet freedom in Freedom on the Net for the third consecutive year. New regulations increased pressure on companies to verify users’ identities and restrict banned content and services. Meanwhile, users themselves were punished for sharing sensitive news and commentary, with prison terms ranging from five days to eleven years.

The government tightened online controls in advance of the 19th National Congress of the Chinese Communist Party (CCP) in October 2017, at which President Xi Jinping, the party’s general secretary, cemented his leadership for the next five years. “Cyberspace sovereignty” has been a top policy goal under Xi, and related legal changes were incorporated into a cybersecurity law adopted in November 2016. The legislation, most of which took effect in June 2017, continued a trend of escalating requirements on internet companies to register their users’ real names, among other provisions. The law also obliges foreign companies to store Chinese user data in mainland China.

The drive to codify what were previously ad hoc censorship and surveillance strategies persisted during the coverage period, with new regulations to license digital tools like VPNs that are used to circumvent website blocking by the centralized censorship apparatus known as the Great Firewall. Other new restrictions targeted citizen journalism, and several sought to prevent websites from republishing “unverified” news from social media. According to regulations issued in May 2017, sites that are not licensed cannot provide any online news and information services.

These rules are taking their toll on civil society. A number of notable domestic websites were closed down during the past year, including Gongshi Wang, a website that sought common ground among different ideological camps regarding democracy and good governance, and Zhongmu Wang, a website serving the Hui Muslim community. At least three website operators in the civil society sector were arrested, including Huang Qi, founder of the human rights website 64 Tianwang, who was detained in December 2016 and later charged with providing state secrets to foreigners.

Dissidents and members of ethnic or religious minority groups received the heaviest penalties for online speech, but ordinary internet users also felt the impact of the increasingly repressive regime. Multiple administrative detentions were used to punish individuals whose posts challenged local or national officials, even in closed messaging groups.

A Obstacles to Access

China boasts the world’s largest number of internet users, yet obstacles to access remain, including poor infrastructure, particularly in rural areas; a telecommunications industry dominated by state-owned enterprises; centralized control over international gateways; and sporadic, localized shutdowns of internet service to quell social unrest.

Availability and Ease of Access

There were 731 million internet users in China as of January 2017,1 according to the China Internet Network Information Center (CNNIC). From December 2015 to December 2016, the number of mobile internet users grew from 620 million to 695 million, accounting for 95 percent of all internet users.2

Though the digital divide between urban and rural areas narrowed marginally in previous years, 72.6 percent of users are based in cities, according to the most recent government figures.3 Penetration rates vary significantly by province, from 77.8 percent in Beijing to 39.9 percent in Yunnan.4 The CNNIC reported that 52 percent of all internet users were male.

The share of internet users connecting through cybercafes and public computers remained relatively constant in 2016, at 16.3 percent and 16.4 percent, respectively.5 Demand for such access points is higher in rural areas and small towns.

The China Broadband Development Alliance reported nationwide average broadband speed at 11.90 Mbps in the fourth quarter of 2016. The highest available rate was in Shanghai, which averaged 14.03 Mbps, while the lowest was in Xinjiang and Tibet, which averaged 9.66 Mbps and 9.27 Mbps, respectively.6 Akamai, which measures access to the global internet, registered slower average speeds (see Availability and Ease of Access: Key Indicators). “Broadband China,” a government strategy issued in 2013, aimed to boost penetration to 70 percent nationwide by 2020 and increase connection speeds to 50 Mbps in cities and 12 Mbps in rural areas, with even faster Gbps speeds promised in bigger cities.7

In July 2016, China Telecom announced plans to gradually eliminate mobile roaming charges to reduce costs for consumers.8 Roaming charges can be double the regular cost per minute of phone use. In a 2017 government work report, Premier Li Keqiang urged telecom companies to remove roaming fees before the end of 2017.9

Restrictions on Connectivity

Nine state-run operators maintain China’s gateways to the global internet, giving authorities the ability to cut off cross-border information requests.10 All service providers must subscribe via the gateway operators overseen by the Ministry of Industry and Information Technology (MIIT).

The government has shut down access to entire communications systems in response to specific events, notably imposing a 10-month internet blackout in the Xinjiang Uyghur Autonomous Region—home to 22 million people—after ethnic violence in the regional capital, Urumqi, in 2009.11 Since then, authorities have continued the practice on a smaller scale. For example, after a knife attack by three assailants reportedly resulted in eight deaths in February 2017, networks in Xinjiang’s Pishan County were cut off.12 A similar outage was reported in a Tibetan area following a reported self-immolation in protest against CCP rule.13 Popular social media applications have been disabled in certain localities in order to “maintain stability.”14

Rights activists and their families are subject to targeted network disconnections. Yang Maoping, sister of imprisoned human rights activist Guo Feixiong, told international journalists that her home broadband service was repeatedly disconnected in August 2016 in a pattern suggesting deliberate disruption, though the provider said there had been a technical error.15 Her access to WeChat Moments, a timeline feature in the WeChat app, appeared to have been blocked during the same period.

Uyghurs, Tibetans, and others who express critical opinions about CCP rule are frequently detained or punished on the pretext that they threaten national security. For that reason, legal provisions that could enable network disruptions to prevent terrorism and protect cybersecurity are cause for concern. Article 84 of an antiterrorism law passed in 2015 introduced fines and detentions of up to 15 days for telecommunications operators and internet service provider (ISP) personnel who fail to “stop transmission” of terrorist or extremist content, “shut down related services,” or implement “network security” measures to prevent the transmission of such content.16 The cybersecurity law passed in November 2016 and effective from June 1, 2017, also provided legal grounds for officials to instruct network operators to stop transmission of certain content to protect public security (see Legal Environment).

ICT Market

In 2011, an antimonopoly investigation accused state-owned China Telecom and China Unicom of abusing their market dominance to manipulate fixed-line broadband pricing, marking the first use of a 2008 antimonopoly law against state enterprises.17 The telecom giants revised their internetwork pricing structures to allow rivals to access their infrastructure,18 and customers can now choose from among many smaller, privately owned ISPs.19

State-owned China Mobile, along with China Telecom and China Unicom, dominate the mobile market. In 2014, the government formally authorized the three major players to set pricing for services according to market forces, resulting in price cuts.20 Private capital was allowed to enter the network leasing business in 2015, and MIIT had issued at least 42 network leasing licenses to private companies by the end of that year.21 In some cities, municipal governments proposed regulations to ensure telecommunication market diversity so that residents in a single community could have a choice of providers.22

Despite the gradual lifting of long-standing market controls, network leasing represents only a small part of the telecommunications sector. Licenses for basic telecommunications services are still effectively monopolized by the three state-owned incumbents, and no other companies are involved in other key services such as public network infrastructure construction.23 In May 2016, state-owned China Broadcast Network (CBN) received a license for basic telecommunications business from MIIT,24 but it would only provide infrastructure and data-transmission services and was not seen as a threat to the three dominant players.25

Authorities exercise tight control over cybercafes and other public access points, which are licensed by the Ministry of Culture in cooperation with other state entities.26 In practice, restrictions can be difficult to enforce. The Ministry of Culture reported 14,000 illegal internet cafés (hei wangba) in operation nationwide as of 2014.27 In November 2014, the Chinese government loosened restrictions on opening new cybercafes, lifting a 2013 rule requiring them to be affiliated with larger chains.28 There were 140,417 cybercafés in China as of June 2016, mostly in second-tier cities.29

Regulatory Bodies

Several government and CCP agencies are responsible for internet regulation at the local and national levels, but the process has been consolidated under Xi Jinping.

The State Internet Information Office (SIIO) was created in May 2011 to oversee telecommunications companies.30 On August 26, 2014, the State Council formally authorized the SIIO to regulate and supervise internet content.31 In December 2014, it launched a new website as the Cyberspace Administration of China (CAC) and Office of the Central Leading Group for Cyberspace Affairs.32 Lu Wei, whom commentators referred to as China’s internet czar, headed the CAC between 2013 and 2016. He was unexpectedly replaced by Xu Lin, a former deputy of Xi Jinping’s, in 2016.33

The CAC reports to the Central Internet Security and Informatization Leading Group, which was formed in February 2014 to oversee cybersecurity and is headed directly by Xi Jinping, making it the highest authority on internet policy in China.34 In December 2014, the leading group took charge of the CNNIC, an administrative agency under MIIT that issues digital certificates to websites.35

In March 2016, Xinhua reported the establishment of the nonprofit Cyber Security Association of China to promote online security.36 It is made up of more than 200 member technology and cybersecurity companies and research institutions, and headed by Fang Binxing, who is recognized as the developer of the Great Firewall.37 In February 2017, the CAC released draft “Measures for the Security Review of Network Products and Services" (wangluo chanpin he fuwu anquan shencha banfa zhengqiu yijian gao) for public comment.38 Article 5 proposed the establishment of a Network Security Review Committee to coordinate network security policy.39

Two regulatory bodies, the State Administration of Radio, Film, and Television (SARFT) and the General Administration for Press and Publications (GAPP), both responsible for censorship in their respective sectors, merged in 2013 to form the State Administration of Press, Publications, Radio, Film, and Television (SAPPRFT).40 The body’s tasks include monitoring internet-based television, online videos, and streaming services. In addition, the CCP’s Central Propaganda Department oversees the ideological inclination of online content.

In March 2016, MIIT announced a draft regulation on domain name management (hulianwang yuming guanli banfa). The regulation requires all domain name holders to go through a real-name registration process, and domain names managed by overseas institutions will not be connected.41 Foreign media noted concerns that the measure could block all foreign websites,42 but MIIT clarified that the regulation only applies to websites with Chinese domain names.43 The draft was still being finalized in early 2017.44

B Limits on Content

The CCP’s Central Propaganda Department, government agencies, and private companies employ hundreds of thousands or even millions of people to monitor, censor, and manipulate online content. Material on a range of issues is systematically censored, including independent evaluations of China’s human rights record, critiques of government policy, discussions of politically and socially sensitive topics, and information about the authorities’ treatment of ethnic and religious minorities. Censorship becomes more intense or expansive during politically sensitive events or in response to breaking news. During the coverage period, user-generated news reports were subject to heightened censorship and punishment. The heavily manipulated online environment still provides more space for average citizens to express themselves or criticize the state than any other medium in China, but digital activism has declined amid growing legal and technical restrictions as well as heavy prison sentences against prominent civil society figures.

Blocking and Filtering

The Chinese government maintains a sophisticated censorship apparatus, including both automated mechanisms and human monitors, to block online criticism of individuals, policies, or events considered integral to the one-party system. Several social media and communication apps are inaccessible from inside China without circumvention tools—and a crackdown on those tools was under way during the reporting period.

The most censored breaking news topics in 2016 were related to the reputation of the party or officials, health and safety, foreign affairs, official wrongdoing, media censorship, or civil society activism.1 There was also consistent and systematic censorship of content addressing long-standing taboos such as the Cultural Revolution, the 1989 crackdown on Tiananmen Square protesters, Taiwanese independence, repression of minorities in Xinjiang and Tibet, and the banned spiritual group Falun Gong. In addition, censors have increasingly blocked international news outlets for reporting on issues like corruption and illicit wealth among high-level officials, especially those with Chinese-language websites. At least 15 global news websites tracked by the nonprofit news organization ProPublica were inaccessible inside China in 2016.2

The government took new measures in 2017 to restrict the use of circumvention tools to bypass blocking and filtering. VPNs offered by domestic or international software developers reroute the user’s traffic through a server outside the firewall to access blocked sites. In 2014, China boasted the largest number of VPN users in the world, according to Global Web Index.3 Circumvention tools are not illegal, since businesses and government supporters also rely on them to communicate and access information beyond China’s borders.4 But authorities have tried to regulate access to the software. Several overseas providers have reported connectivity issues and blocking.5

MIIT launched a campaign to crack down on unauthorized domestic VPNs between January 17, 2017, and March 31, 2018.6 Service providers are now barred from setting up VPNs without licensing, and illegal VPN operations will be subject to closure or blocking.7

Several social media and messaging apps are totally blocked, isolating the Chinese public from global networks. According to censorship monitor GreatFire.org, 171 of Alexa’s top 1,000 websites in the world were blocked in China in 2017 (compared with 138 in mid-2016).8 These include YouTube, Google, Facebook, Flickr, SoundCloud, and WordPress.9

ains in China,” https://en.greatfire.org/search/alexa-top-1000-domains. A number of services operated by Google, including Google Maps, Translate, Calendar, Scholar, and Analytics, were blocked in 2017, according to GreatFire.org.10

Blocking can affect the download pages for related software, or specific functions within an app. The messaging service WhatsApp, which was available in China in 2016, was subject to disruption of its video and call functionality in mid-2017, and later it was almost entirely blocked.11 In March 2017, Pinterest, a popular social media platform for curating visual content, was blocked for unknown reasons, having been fully accessible for many years.12 Most of its content is apolitical.

While the term “Great Firewall” is often used to refer to China’s internet censorship system as a whole, it alludes more specifically to the automated, technical blocking of websites and services based outside China. In some cases, whole domain names or internet protocol (IP) addresses are blocked, with users receiving an explicit message about illegal content. Other interventions are less visible. Over the past several years, observers have documented unusually slow speeds that indicate deliberate throttling, which delays the loading of targeted sites and services.13

Authorities also use deep packet inspection (DPI) to scan for any blacklisted keywords in both a user’s request for content and the results returned. Once the keywords are detected, the technology signals both sides of the exchange to temporarily sever the connection. Such granular control is less noticeable to users because specific pages can be blocked within otherwise approved sites, and because the interruption appears to result from a technical error.14 Returning fake pages, or replacing the requested site with content retrieved from an unrelated IP address using a technique known as DNS poisoning, is another routine method of disrupting access to specific content.

Censorship decisions are arbitrary, opaque, and inconsistent, in part because so many individuals and processes are involved. The impact may vary depending on timing, technology, and geographical region. ISPs reportedly install filtering devices differently, in the internet backbone or even in provincial-level internal networks.15 Blacklists periodically leak online, but they are not officially published. There are no formal avenues for appeal. The government has generally not been transparent about content controls,16 and criticism of censorship is itself censored.17

Content Removal

Websites and social media accounts are subject to deletion or closure at the request of Chinese censorship authorities, and internet companies are required to monitor and delete problematic content or face punishment. The cybersecurity law passed during the reporting period requires network operators to “immediately stop transmission” of banned content.18

SMS and instant messages are also subject to blocking and filtering.19 Nobel Peace Prize laureate Liu Xiaobo’s name, for example, was already blocked from searches on Chinese social media platforms during the coverage period.20 After his death in custody from liver cancer in July 2017, censorship on the messaging app WeChat and the microblogging service Sina Weibo also spiked.21

Gongshi Wang, a well-known site that championed liberal thought, was unexpectedly taken offline in October 2016.22 The owner told the South China Morning Post that he was “uncertain about its future.”23 Gongshi Wang sought common ground among different ideological camps regarding democracy and good governance. Among other domestic websites that were closed down during the past year, Zhongmu Wang, a popular site serving Chinese Hui Muslims, was shuttered in December 2016 along with its official social media accounts due to allegations that it was “spreading extremism.”24 Officials were apparently concerned that the site’s advocacy for better labeling and application of halal standards in China—which is opposed by some Han Chinese groups25 —would encourage conflict.

Domestic social media companies and content providers delete or block sensitive content shared by users. In 2017, one company estimated that 20 to 30 percent of its labor costs were dedicated to “auditing” content.26

Sina’s efforts to manage Weibo content are well documented. Staff members delete individual posts or accounts, often within 24 hours of an offending post, but sometimes long after publication;27 make published posts visible only to the account owner; or personally warn individual users.28 Hundreds of terms have also been automatically filtered from Weibo search results over time.29

WeChat censorship is also increasingly sophisticated. The company no longer informs users when content has been censored,30 and after Liu Xiaobo’s death, WeChat users reported that images depicting him and his wife Liu Xia were also blocked; those sent via private messages were not visible to the recipient.31

In September 2016, President Xi misread the phrase tong shang kuan nong (“facilitating business and deregulating agriculture”) as the similar-looking tong shang kuan yi (“facilitating business by undressing”) during a speech to world business leaders at a Group of 20 (G20) summit held in Hangzhou. The term “kuan yi” was then blacklisted on Weibo and WeChat.32 In another case, censorship authorities issued orders to delete all online references to a newly discovered species of beetle, which a Chinese scientist had named after Xi.33 The Weiboscope project at the University of Hong Kong, which tracks censored Weibo content, reported that the keywords “Cultural Revolution,” “Lei Yang,” “police,” “lawyer,” and “government” were the most likely to be censored in March 2017.

Companies also block entire accounts. In September 2016, the CAC announced that in the first nine months of 2016, a total of 11,459 WeChat public accounts were closed, mainly due to alleged rumor mongering.34

A number of noteworthy Individual accounts were shuttered during the coverage period:

  • On August 1, 2016, the Weibo account of renowned lawyer Chen Guangwu was closed by Sina for unknown reasons.
  • The Weibo account of Yi Fuxian, a demography scholar who had criticized China’s restrictions on family size and reproduction, was closed in September 2016.35
  • In early 2017, liberal economist Mao Yushi’s Weibo account was terminated.36
  • In February 2017, a Weibo account promoting feminism was closed.37

In June 2016, the CAC announced plans for a campaign to purge undesirable content from online comments. Chief content editors from Tencent and NetEase pledged to improve systems for managing user comments.38 On June 28, 2016, the CAC released a “mobile internet apps information service regulation” (yidong hulianwang yingyong chengxu xinxi fuwu guanli guiding), requiring companies that offer digital apps to manage content produced and disseminated by users, effective from August 1 of that year.39

International companies also complied with government requests to remove content. In late December 2016, Apple removed the New York Times app from its regional App Store in China.40 “We have been informed that the app is in violation of local regulations,” the technology company said, without elaborating on the nature of the regulations or how it had been informed, according to the Times.

In November 2016, the New York Times reported that Facebook—which is blocked in China—had “developed software to suppress posts from appearing in people’s news feeds in specific geographic areas,” citing anonymous former employees. The software could enable a partner company in a country like China to monitor users’ news feeds and prevent chosen content from appearing.41

Censors targeted video content as well. In July 2016, the Ministry of Public Security launched a three-month nationwide campaign to purge illegal content from live-streaming websites.42 The alleged purpose of the campaign was to remove obscenity, pornography, gambling-related material, and content that endangers public security.43 In Guangdong Province, a total of 51 websites and 423 live-streaming channels were closed by the police.44

In September 2016, the chief executive of QVOD, which operated the peer-to-peer video-streaming application Kuaibo, was sentenced to 42 months in prison and fined CNY 1 million (US$150,000) on charges of “disseminating pornographic material for profit.”45 Three other executives were also jailed and fined. Kuaibo was taken offline in 2014 during an investigation into pornographic content shared through the site. The court held that the company and its top manager were liable for content posted and shared by users.

Media, Diversity, and Content Manipulation

The digital news media are subject to significant control, and several new regulations limiting user-generated content were passed during the reporting period. Online journalists regularly practice self-censorship. Editors and reporters who post banned content—or content that is critical of the CCP, its high-ranking members, or its actions—risk disciplinary warnings, job loss, or even criminal detention. (News websites in Hong Kong operate with considerably more freedom but have been subject to periodic obstruction.46 )

Websites and social media accounts other than those operated by print and broadcast outlets are not allowed to produce news content by law, although the definition of news is not clear. Regulations in effect since March 2016 clarified restrictions on foreign investment in online publishing, and listed requirements for domestic companies to obtain an online publishing permit. In addition to compliance with censorship rules, the requirements included having at least eight full-time editorial or publishing staff members.47 Some news outlets have been punished for content violations through restrictions on their distribution. In October 2016, authorities issued a two-month suspension that prevented other websites from republishing content produced by Caixin, a relatively independent business publication.48

Starting in mid-2016, the authorities tightened the ban on “illegal” online news content in an apparent effort to reduce the social and political impact of user-generated content:

  • In July, the CAC stipulated that news media should not publish unverified content sourced from social media.49
  • The same month, the local cyber affairs office in Beijing ordered the web portals Sina, Sohu, NetEase, and Ifeng.com to terminate several news and information programs that used unofficial information sources.50
  • On November 4, the CAC published internet live-streaming service regulations (hulianwang zhibo fuwu guanli guiding) to govern live video content.51
  • In December, the SAPPRFT ordered social media platforms such as Weibo and WeChat to cease promoting user-generated news videos.52
  • In January 2017, the Beijing authorities closed 17 websites, many for providing news services without a proper license.53 The website of the Tianze Economic Research Institute, a liberal think tank established by outspoken economist Mao Yushi, was among them.
  • Separately in January, the government ordered the new video website Pear Video to shut down temporarily for publishing news without authorization.54
  • The popular cartoon video site Bilibili told users not to upload news or political content in February 2017.55

Some new measures are likely to give the state more direct influence over private companies’ management of online content. In April 2016, regulators sought feedback from major Chinese internet companies like Baidu and Tencent on a proposal for the state to purchase a 1 percent stake in such firms.56 The proposal was broadened to include video streaming services during the reporting period. SAPPRFT urged popular video platforms such as Youku, Tudou, and Tecent TV to sign an initial agreement with state-owned media groups including China National Radio and China Radio International.57

On May 2, 2017, CAC regulations on managing internet news and information services (hulianwang xinwen xinxi fuwu guanli guiding)58 said that only traditional media or state-controlled enterprises were eligible to obtain a license to gather or disseminate news in any online format.59 Critics said the regulations were unconstitutional and violate China’s trade agreements.60

Online content is subject to extensive manipulation. Propaganda officials instruct internet outlets to amplify content from state media. Since 2005, propaganda units at all levels have trained and hired web commentators, known colloquially as the “50 Cent Party,” to post pro-government remarks and influence online discussions.61 Commentators may report users who have posted offending statements, target government critics with negative remarks, or deliberately muddy the facts of a particular incident.62 Coordinated smear campaigns have been used to discredit high-profile government critics.63 In January 2017, a new analysis of comments estimated that about 1 out of every 178 social media posts is sponsored by the government, amounting to some 448 million posts per year; most of the posts in the study’s sample praised the government to divert attention away from potentially negative stories.64

Much of this activity is conducted on official sites and accounts. In January 2017, the CCP published “Guidance on the use of WeChat for party members” (dangyuan ganbu weixin shiyong zhinan), announcing that party members will be penalized for discussing important government policies without discretion (wangyi dazhen fangzhen) on social media.65

The content manipulation also extends to platforms that are blocked in China. One 2014 study identified over 2,500 “50 Cent” users spreading misinformation on Twitter.66 In 2015, an analysis of the People’s Daily Twitter account found patterns suggesting that many of its followers were artificial.67 Companies also pay for “astroturfing”—positive comments promoting products or services—which further erodes public trust in online content. Commercial commenters are colloquially known as the “internet water army.”68

Local authorities have mobilized ziganwu, volunteer commentators motivated by ideology rather than cash, to promote the government’s image online,69 part of a propaganda strategy to “spread positive energy among society.”70 A document leaked in 2015 revealed hundreds of thousands of “youth league online commentators” in China’s higher education institutions, tasked with swaying students against supposed Western values.71 In 2016, a handful of ziganwu were appointed to a social media division of the Communist Youth League.72 The league’s new media working group uses Weibo, WeChat, and other social media platforms to tailor political messages to younger audiences.73 The CCP has also sought to reach younger readers through new, more appealing state-backed online media outlets such as The Paper, launched in 2015, and an English-language version called Sixth Tone, unveiled in April 2016.

Despite extensive censorship, the internet has provided a forum for discussion and the sharing of information on important social and political issues, including sensitive topics like democracy and constitutional government.74 A certain amount of open debate has allowed officials to monitor public sentiment, debunk “enemy” ideology,75 and conduct internal power struggles, though the space for such online expression has dwindled in recent years.

Civil society organizations involved in social and cultural issues often have a vigorous online presence, while others that are perceived as a political threat are penalized (see Prosecutions and Detentions for Online Activities). Ethnic minorities and persecuted religious groups attempt to use the internet to disseminate banned content, but they remain underrepresented. In August 2016, a research team in Qinghai launched the first Tibetan-language search engine, Yongzin. Some commenters welcomed the platform, but a British Broadcasting Corporation (BBC) report said it mostly returned results leading to state news and video content.76

Many internet users bypass censorship with circumvention technology or creative work-arounds. Humorous neologisms, homonyms, and cryptic allusions substitute for banned keywords, forcing censors to filter seemingly innocuous terms like “tiger.”77 This version of the Chinese internet does not resemble a repressed information environment so much as “a quasi-public space where the CCP’s dominance is being constantly exposed, ridiculed, and criticized, often in the form of political satire, jokes, videos, songs, popular poetry, jingles, fiction, Sci-Fi, code words, mockery, and euphemisms.”78 However, a crackdown on VPNs (see Blocking and Filtering) and advances in keyword filtering raised new obstacles for those seeking to evade censorship over the past year.

Digital Activism

Social media used to be a vibrant space for activism.79 Amid stricter internet controls, however, activism has been gradually waning since 2013.80 The word “netizen”—a translation of the Chinese wangmin, or citizen of the internet—conveys the sense of civic engagement associated with online exchanges, but the term was less common in China by 2015.81 Recent prosecutions of prominent human rights workers have removed important advocates and sources of information from the online environment, and even local activism sometimes resulted in sanctions during 2017. When internet users protested the Henan provincial government’s plans to build four nuclear power plants in the next five years, one was detained for five days in connection with an online article (see Prosecutions and Detentions for Online Activity).82

Authorities remain responsive to public opinion online, though the ultimate impact is limited. In May 2016, for example, Lei Yang, a well-educated environmentalist in Beijing, was found dead after he was detained by local police on charges of soliciting a prostitute. Police said he died of a heart attack, but family members said they saw bruises on his body, sparking an online debate about police brutality.83 In June, under this public pressure, the authorities announced that they would investigate five police officers for possible involvement in Lei Yang’s death.84 However, prosecutors decided not to charge them in December, and students and graduates of Lei’s alma mater Renmin University, one of China’s most prestigious institutions, launch an online petition in protest.85 Lei’s family eventually declined to pursue litigation over the incident.

C Violations of User Rights

A number of criminal laws and internet regulations can ensnare users who post content deemed undesirable by the CCP, and a new cybersecurity law was passed during the reporting period. Criminal charges of subversion, separatism, and terrorism, as well as defamation and “provoking quarrels,” are regularly invoked to imprison citizens for their online activity. Civil society websites were targeted in the latest crackdown on perceived threats to social and public order. Real-name registration requirements undermine users’ privacy and anonymity, and surveillance has increased in ethnic minority areas. Websites, hosting services, and dissidents’ email accounts are routinely attacked by hackers based in China.

Legal Environment

Article 35 of the Chinese constitution guarantees freedoms of speech, assembly, association, and publication, but such rights are subordinated to the CCP’s status as the ruling power. The constitution cannot generally be invoked in courts as a legal basis for asserting rights. The judiciary is not independent and closely follows party directives, particularly in politically sensitive freedom of expression cases. Government agencies issue regulations to establish censorship guidelines. These are highly secretive and subject to constant change, and they cannot be challenged by the courts. Prosecutors exploit vague provisions in China’s criminal code; laws governing printing and publications; subversion, separatism, and antiterrorism laws; and state secrets legislation to imprison citizens for online activity. Trials and hearings typically lack due process.

On November 7, 2016, the Standing Committee of the National People’s Congress passed a cybersecurity law that came into effect on June 1, 2017.1 The law consolidated the role of the CAC (see Regulatory Bodies), which it identified as the principle agency responsible for implementing many of its provisions.2 The law also codified existing restrictions, strengthening self-regulation and real-name registration requirements for internet companies and obliging them to assist security agencies with investigations (see Surveillance, Privacy, and Anonymity) and help implement censorship or, potentially, network shutdowns (see Content Removal and Restrictions on Connectivity).3

Other laws and regulations have implications for online speech. In 2013, the Supreme People’s Court and the Supreme People’s Procuratorate, the top prosecutorial body, formally defined online manifestations of crimes including defamation, creating disturbances, illegal commercial activities, and extortion.4 Criminal defamation carries a possible three-year prison sentence under “serious” circumstances.5 The new interpretation defined those circumstances to cover defamatory online content that receives more than 5,000 views or is reposted more than 500 times.6 Online messages deemed to incite unrest or protest are also subject to criminal penalties under the interpretation.

In November 2015, an amendment to the criminal code introduced penalties of up to seven years in prison for those who disseminate misinformation on social media.7 Separately, the antiterrorism law adopted in December 2015 instructed companies to delete terrorist content or face administrative detention for their personnel, and barred social media users from sharing information about acts of terrorism or spreading “inhuman” images that could promote copycat acts. It also increased pressure on private companies to provide the government with user data (see Surveillance, Privacy, and Anonymity).

Some detentions, including administrative detentions authorized by public security bureaus, do not require approval by a court.8 Chinese authorities abolished the form of administrative detention known as “reeducation through labor” in 2013 in response to domestic calls for reform,9 but individuals can be detained without trial under similarly poor conditions in drug rehabilitation and “legal education” centers.10 State agents also abduct and hold individuals in secret locations without informing their families or legal counsel. In 2012, the National People’s Congress enacted an amendment of the Criminal Procedure Law that strengthened the legal basis for detaining suspects considered a threat to national security in undisclosed locations, among other changes.11

Prosecutions and Detentions for Online Activities

As of December 1, 2016, at least 38 journalists were behind bars in China, 34 of whom worked online, according to the Committee to Protect Journalists.12 Many other citizens have also been jailed for their online activities, including advocates of political reform, human rights workers, members of ethnic and religious minorities, and ordinary users who stir dissent or criticize CCP leaders. A number of long prison sentences were issued during the reporting period, and online speech was also frequently punished with brief administrative detentions. Though the people imprisoned represent a tiny percentage of the overall user population, prosecutions have a chilling effect on activism and encourage self-censorship in the broader public.

As in past years, known dissidents received the heaviest penalties in 2016 and 2017:

  • In June 2016, a court in Hangzhou imposed long sentences for subversion of state power on two prodemocracy activists whose trials concluded in September 2015.13 Lu Gengsong was sentenced to 11 years in prison for activities that included publishing essays on overseas websites. Chen Shuqing, who had published statements about other political detainees before his arrest, was sentenced to 10 and a half years in prison.14
  • In November 2016, Shandong democracy activist Sun Feng was sentenced to five years in prison on the lesser charge of inciting subversion of state power in online posts.15 He had been held for two years before trial.

Activists were also newly arrested for operating websites about civil society and human rights issues:

  • In June 2016, police in Dali, Yunnan Province, detained Lu Yuyu and Li Tingyu, a couple who documented and tallied protest incidents on the Not News website and associated social media accounts. They were charged with “picking quarrels and provoking troubles.”16 Lu was sentenced to four years in prison in August 2017.17 Li was released after a closed-door trial in April, though no verdict was immediately announced.18
  • Liu Feiyue, founder of Civil Rights and Livelihood Watch, a grassroots platform for monitoring human rights issues, was reported missing in November 2016, apparently after being detained by police in Hubei Province. He was charged with inciting subversion of state power in December.19 His case had not gone to trial by mid-2017; prosecutors may have been considering more serious charges.20
  • Police in Chengdu, the capital of Sichuan Province, formally detained Huang Qi, founder of the human rights watchdog website 64 Tianwang, in December 2016.21 He had been missing for several days.22 Huang was accused of providing state secrets overseas,23 but no sentence had been issued as of mid-2017.24 Huang has been imprisoned several times before, including in 2008 after he posted complaints from the parents of victims of the Sichuan earthquake.

Other online speech cases resulted in administrative detentions, which can last up to 15 days. Several resulted from content published in closed WeChat groups. The topics involved ranged from personal criticism of Xi Jinping to current affairs:

  • Wang Wei, a university student from Anhui Province, was held for 10 days for reposting a doctored photo depicting Xi as Adolf Hitler on Sina Weibo in July 2016.25
  • In September 2016, a resident of Sichuan was detained for seven days because he described Xi as a coward in a private WeChat conversation.26
  • Separately in September, a resident of Shenzhen was detained for 10 days on a charge of “disseminating rumors” because she reposted international news coverage of the Wukan village democracy movement on WeChat.27 That month, police had crushed a protest in Wukan, a small Guangdong Province fishing settlement, after an elected village leader was sentenced to prison.
  • In January 2017, Guangdong human rights activist Sun Desheng was detained for 10 days for sharing an image mocking Xi Jinping online.28 News reports said the image showed Xi riding on a flying pig labeled “Chinese Dream.”
  • In February 2017, Anyang resident Wang Shoufeng posted an article online criticizing the Henan provincial government’s nuclear power program (see Digital Activism). Wang was administratively detained for five days for disseminating false information to disrupt public order.29
  • Two internet users in Inner Mongolia were detained that month for 10 days each on charges of “disseminating false information to disrupt public order.” They had expressed dissatisfaction with the local administration in a WeChat group.30
  • Also in February, a resident of Sichuan Province was detained for five days for allegedly disseminating misinformation after local officials objected to his WeChat post about high smog levels.31
  • An internet user in Zhejiang Province was detained for 10 days the same month for allegedly inflating the death toll from a fire in a WeChat group discussion.32
  • In a separate February case, an internet user was detained for five days for allegedly publishing misinformation about the death toll from a chemical explosion in Anhui in WeChat discussion groups.33

At least two cases involving WeChat comments resulted in longer stints in prison:

  • On June 6, 2016, migrant worker Hu Changgen disappeared in Shanghai. More than two weeks later, he was found to be in police custody for publishing comments about the 1989 Tiananmen Square massacre in a WeChat group.34 He spent a year behind bars before being released in June 2017.35
  • In April 2017, a court in Shandong ordered a retrial for an internet user who had been sentenced that month to two years in prison for allegedly picking quarrels on WeChat and QQ, including by using a banned nickname for Xi Jinping, “Steamed Bun Xi.”36

A number of recent detentions have targeted users of VPNs as part of a crackdown on circumvention tools (see Blocking and Filtering). One 2016 example involved an internet user who downloaded propaganda videos from the Islamic State militant group and was charged under the antiterrorism law.37 Another case in February 2017 centered on a Xinjiang resident who used a VPN to reach uncensored news on Radio Free Asia and was detained for 15 days.38

Religious and ethnic minorities face particularly harsh treatment for online activity. Radio Free Asia reported in 2015 that a Uyghur teenager sentenced to life imprisonment in Xinjiang had “simply watched videos on his cellphone,”39 while Tibetan monks received long prison sentences in 2016 for involvement in a WeChat group about commemorating the Dalai Lama’s birthday.40 According to a February 2017 Freedom House study on religious freedom, many Falun Gong practitioners have been jailed for posting messages about the spiritual group or human rights abuses on WeChat or QQ, accessing banned websites, and possessing VPN technology.41

Surveillance, Privacy, and Anonymity

The cybersecurity law adopted in November 2016 strengthens obligations for network operators to register users under their real names, requires that information about Chinese users be stored on servers located within the country,42 and indicates that technology companies may have to undergo security checks.43 International companies said the terms in the law were vague, but many began efforts to come into compliance.44 Implementing regulations that would clarify several provisions had yet to be issued at the end of the reporting period.

Privacy protections under Chinese law are minimal. In the words of one expert, the law explicitly authorizes government access to privately held data, and “systematic access” to “data held by anyone” is a realistic possibility once e-government strategies are fully implemented.45 China’s national identity cards, which are administered by police, are required to be digitally embedded with fingerprints.46 The State Council aims to link credit, social security, and other personal information to these biometric databases.47 Companies and municipalities are also testing “social credit” schemes that will allow an assessment of individuals’ online activities alongside other personal data before they are granted financial credit or access to certain services.48 The schemes have been subject to criticism within China due to privacy and other concerns, but plans to expand the idea continued to be debated over the past year, including some with an ideological slant. In February 2017, CCP magazine Red Flag Digest (Hongqi Wenzhai) suggested the introduction of a credit and review system for significant opinion leaders on social media.49 The central government plans to launch a nationwide, mandatory social credit system by 2020.50

Companies offering web services are required to register users. Businesses must gain users’ consent to collect their personal electronic data, and outline the “use, method, and scope” of the collection, but there are no limits placed on law enforcement requests for personal records.51 Mobile phone users have been required to register since 2010, so providing a phone number is a common way of registering with other services.52 Registration requirements are difficult to enforce, and users have exploited loopholes to evade them,53 but related rules have been steadily tightened.54 On December 28, 2016, for example, MIIT mandated real-name registration for all telecommunications services, effective from February 2017.55

News portals, bulletin boards, blog-hosting services, and email providers have long enforced some form of user registration.56 MIIT requires website operators themselves to register as part of the licensing process.57 Sina Weibo’s reports to the U.S. Securities and Exchange Commission have noted the company’s exposure to potentially severe penalties due to its inability to fully comply with real-name registration rules.58

Cybercafes check photo identification and record user activities, sometimes in cooperation with law enforcement. The Public Security Bureau in Lianyungang, Jiangsu Province, developed a real-name registration app for cybercafés in the city in 2015.59 In March 2017, local news reports said facial recognition technologies had been used to match customers’ faces and ID photos in cybercafés in the city of Xingtai, Hebei Province.60 Hebei authorities instructed public Wi-Fi providers to comply with user registration requirements in April.61

The cybersecurity law introduced new penalties for companies that infringe on privacy,62 but many other laws ensure that companies make user data available to officials. Amendments to the State Secrets Law in 2010, for instance, obliged telecommunications operators to cooperate with authorities investigating leaked state secrets or risk losing their licenses.63 An amendment to the Criminal Procedure Law that took effect in 2013 introduced a review process for allowing police surveillance of suspects’ electronic communications, which the Ministry of Public Security permits in many types of criminal investigation, but the wording about the new procedure was vague.64

Various regulations have outlined requirements for companies to retain user data. In June 2016, for example, the CAC required Chinese app providers to register users and keep user activity logs for 60 days.65 Other ISPs are required to retain user information for 60 days, and submit it to the authorities upon request without judicial oversight or transparency.66

As with censorship, measures that erode privacy disproportionately target groups perceived as threats to the regime. In 2015, the Xinjiang government required real-name registration for Uyghurs attempting to purchase electronic devices with storage, communication, and broadcast features. Stores selling such equipment were required to install software that provides police with real-time electronic records on transactions.67

The authorities justify real-name registration as a means to prevent cybercrime, but experts warn that uploaded identity documents are vulnerable to theft or misuse,68 especially since some verification has been done through a little-known, government-linked contractor.69 In December 2016, millions of personal data records, obtained through an app that allows customers of the State Grid Corporation to pay for electricity online, were sold on the black market by unknown actors.70

Chinese companies have scored poorly in assessments of the user protections incorporated in their technology,71 and there are limits on tools that help internet users conceal their location or the nature of their activity, including VPNs (see Blocking and Filtering). The antiterrorism law passed in 2015 requires companies to offer technical support to decrypt information at the request of law enforcement agencies, among other provisions.72 Regulations for the Administration of Commercial Encryption dating to 1999, and related rules from 2006, separately require a government regulator to approve encryption products used by foreign and domestic companies.73

Direct surveillance of internet and mobile phone communications is believed to be pervasive. One academic study from 2011 reported that when users entered blacklisted search terms on Baidu, their IP addresses were automatically sent to a location in Shanghai affiliated with the Ministry of Public Security.74 Given the secrecy surrounding such capabilities, however, they are difficult to verify.

Intimidation and Violence

Allegations of torture and extralegal harassment are widespread among Chinese detainees, particularly political prisoners and those involved in freedom of expression cases. In 2015, Human Rights Watch reported “physical and psychological torture during police interrogations, including being hung by the wrists, being beaten with police batons or other objects, and prolonged sleep deprivation,” in a review of hundreds of ordinary criminal cases. “Political prisoners … have experienced much of what is described in this report and often worse,” the report said.75 A Uyghur man detained for watching videos on a smartphone was reported to have died in custody in June 2016.76 Nobel Peace Prize laureate Liu Xiaobo died in custody from complications of liver cancer in July 2017, after authorities refused to let him travel abroad for treatment.

Other allegations of mistreatment were made during the coverage period of this report. Lawyers defending Wu Gan, a digital activist arrested for participating in a performance protest in 2015, reported in December 2016 that he had been repeatedly held in solitary confinement and tortured.77

Internet users also risk being held under house arrest. The conditions and degree of confinement can be adjusted arbitrarily over time. Poet Liu Xia, the wife of Liu Xiaobo, has been confined in her Beijing home since 2010, often without an internet connection.78 She went missing for several weeks after her husband’s funeral in July 2017.79 Some groups monitor the number of dissidents under house arrest, but there are no statistics showing how many were targeted specifically for online activity.80

Law enforcement officials frequently summon individuals for questioning in relation to online activity, an intimidation tactic referred to euphemistically as being “invited to tea.”81 Activists have also been forced to travel within the country during sensitive political events, effectively keeping them away from their normal online activities.82

Technical Attacks

Hackers, known in Chinese as heike (dark guests), employ various methods to interrupt or intercept online content with political implications. Attacks known to have originated in China can rarely be traced directly to the state, and much of the activity appears decentralized and uncoordinated. But many attacks employ sophisticated technology, and experts believe that Chinese military and intelligence agencies either sponsor or condone technical attacks on political targets.

Domestic and overseas groups that report on China’s human rights abuses have suffered from distributed denial-of-service (DDoS) attacks, which temporarily disable websites by bombarding host servers with traffic. 83 Spear-phishing, in which customized email messages are used to trick recipients into downloading malicious software by clicking on a link or a seemingly legitimate attachment, is another common tactic. Tibetans, Uyghurs, and others have been targeted with emailed programs that install spyware on the user’s device.84 In September 2016, American cybersecurity company FireEye reported that two Hong Kong government departments were subjected to phishing attacks originating from APT3, a Chinese hacker organization, prior to the territory’s Legislative Council elections. The motive was “certainly” political, FireEye said.85

In 2015, the Canada-based monitoring organization Citizen Lab analyzed a massive DDoS attack targeting the anticensorship group GreatFire.org, along with content that the group hosted on GitHub. “While the attack infrastructure is co-located with the Great Firewall, the attack was carried out by a separate offensive system, with different capabilities and design, that we term the ‘Great Cannon,’” the group reported. Both the technology and the target offered “compelling evidence” of Chinese government involvement.86

China remains a top source of global cyberattacks,87 though those with commercial targets have declined or become harder to trace.88 The degree of state support for commercial attacks is unclear. Other targets have strategic importance.89 In December 2016, investigators said hackers linked to China’s military had seized control of workstations in the Washington, DC–based Federal Deposit Insurance Corporation, which regulates commercial banking, in 2010.90 During the reporting period, Russian media reported that China had reduced cyberattacks against the United States while increasing those against Russia,91 including attacks on military intelligence targets.92

On China

See all data, scores & information on this country or territory.

See More

Country Facts

  • Global Freedom Score

    9 100 not free

  • Internet Freedom Score

    10 100 not free