Beijing Law Review
2013. Vol.4, No.1, 37-41
Published Online March 2013 in SciRes (http://www.scirp.org/journal/blr) http://dx.doi.org/10.4236/blr.2013.41005
Regulating the Internet: China’s Law and Practice*
Haiping Zheng
School of Law, Renmin University of China, Beijing, China
Received November 26th, 2012; revised De cember 27th, 2012; accepted January 4th, 2013
Though internet was not commercially available in China until 1995, it has been growing tremendously
over the years. At the same time, the Chinese government has never ceased regulating or even censoring
internet. This paper provides an overview of the development of internet in China, and the major regula-
tory schemes that have a direct impact on internet speech. Further, it describes some of the specific meas-
ures the Chinese government uses to control the internet: filtering and blocking, imposing liabilities on
private parties, access control, internet “police”, and “guiding” public opinion. Finally, it concludes that
internet censorship does more harm than good.
Keywords: Internet; China; Freedom of Speech
Introduction
China’s internet censorship has drawn much international
criticism. For example, in 2006, the Reporters without Board
included China as one of the 13 “enemies of the Internet” (Re-
porters without Borders, 2006). Then in 2010, Google decided
to withdraw from Chinese market, claiming that Chinese gov-
ernment’s attempts to limit free speech on the web, combined
with other factors, had led the corporation to make such a deci-
sion (Official Google Blog, 2010).
On the other hand, the picture is not that gloomy. By 2008,
the number of internet users in China has reached 220 million,
making China the nation with the largest number of internet
users. Furthermore, due to the very nature of internet, it is often
difficult for the government to control the information trans-
mitted through internet. Despite Chinese government’s efforts
to censor the internet, it is doubtful how much success it can
achieve.
This paper examines China’s internet censorship and its ef-
fects. Part 1 provides a general introduction to the development
of internet in China. Such background information is necessary
for the understanding and analysis of internet censorship in
China. Part 2 introduces some of the regulations that have a
direct impact on internet speech. Part 3 describes some spe-
cific measures (or techniques) the Chinese government utilizes
to control the internet: filtering and blocking, imposing liabili-
ties on private parties, access control, internet “police”, and
“guiding” public opinion. The final part is the conclusion of the
paper.
Development of Internet in China
Although China began to be connected to the internet as
early as 1987, internet was not commercially available in China
until 1995. From then on, internet has been growing tremen-
dously over the years. By 2008, infrastructure has extended
broadband Internet access to 92 percent of townships (Zhao,
2008).
Like many other areas of China’s economic development
since 1978, development of internet was largely driven by the
government. As a result, China’s internet hardware infrastruc-
ture is highly centralized. Currently, there exist nine state-li-
censed Internet Access Providers (IAPs), each of which has at
least one connection to a foreign Internet backbone. All the
IAPs are required to be “at least fifty-one percent controlled by
State-owned companies”. These IAPs, in turn, grant regional
Internet Service Providers (ISPs) access to backbone connec-
tions. All these entities (IAPs and ISPs) are required to register
with the designated government agents. Those who fail to com-
ply with the regulations face the threat of being shut down
(ONI, 2005).
The government’s monopolistic position in internet infra-
structure facilitates censorship. Because all Internet traffic
passes through the nine IAPs, the government can censor the
information flow by adding filtering system “at the gateways.”
Moreover, as Part 3 will show, the government requires ICPs
and ISPs to filter internet content, resulting in severe self-cen-
sorship.
International companies have been playing a significant role
in the development and maintenance of China’s internet infra-
structure. The Cisco system, in particular, has been integral to
China’s Internet development. It specifically implemented the
backbone networks for at least three of China’s nine IAPs.
Western corporations’ such “conspiring” activities have been
subjected to the criticism of human rights activists (Newbold,
2003).
Chinese Government’s Attempts to Control
the Internet: An Overview
Before the introduction of internet, the Chinese government,
under the leadership of the Chinese Communist Party (CCP),
essentially controlled all the traditional mass media, including
newspapers, magazines, television, radio, etc. Unsurprisingly,
the government sought to control the new media even before
internet became commercially available. Over the years, the
government promulgated numerous regulations to control the
*This article is one of the research results of the Project sponsored by China
N
ational Social Science Foundation ( No.12CFX020). internet. This part provides an overview of the regulations that

H. P. ZHENG
have a direct imp ac t on internet s p eec h.
On February 1, 1996, China’s State Council promulgated the
Interim Provisions Governing Management of Computer In-
formation Networks. It prohibits four categories of information
from being produced or transmitted online: information that
would harm national security, disclose state secrets, threaten
social stability or promote sexually suggestive material (art.
13).
On September 20, 2000, the State Council issued the Meas-
ures for Managing Internet Information Services (Measures,
2000), which significantly extended the scope of prohibited
contents. Article 15 of the 2000 Measures provides:
ISPs (internet service providers) shall not produce, reproduce,
release, or disseminate information that contains any of the
following: 1) Information that goes against the basic principles
set in the Constitution; 2) Information that endangers national
security, divulges state secrets, subverts the government, or
undermines national unity; 3) Information that is detrimental to
the honor and interests of the state ··· 6) Information that dis-
seminates rumors, disturbs social order, or undermines social
stability ··· or 9) Other information prohibited by the law or
administrative regulations.
It is easy to see that provisions like this are “vague, confus-
ing and inconsistent” (Li, 2004). Yet similar provisions are
present in many other internet regulations. Indeed, these provi-
sions are so common that many Chinese seem to have “ac-
cepted” them. Few people attempted to challenge the legiti-
macy (or constitutionality) of such provisions. Obviously, such
vague provisions can deter individual citizens from spreading
“sensitive” information that may fall into one of the prohibited
categories.
Internet Censorship and Its Resistance
This Part introduces some specific measures (or techniques)
that China uses to control the internet. Although these measures
in fact overlap with one another, for purpose of clarity, they are
to be discussed separately here.
Blocking and Filtering Systems
The Chinese government consistently blocks the entire do-
main of certain websites that are hard to control, including
some international news sources (i.e., BBC-Chinese), internal
blogger servicer providers (i.e., facebook, blogger), and some
other websites that often post criticism on China’s human rights
and social justice records (i.e., Amnesty International, Human
Rights Watch, etc.) These websites are blocked regardless of
their specific contents, partly because the ISPs of these websites
are unlikely to “cooperate” with the Chinese government in
censoring the internet content (ONI, 2005).
The general trend, though, seems to be that the Chinese gov-
ernment tries to filter specific “sensitive” contents rather than
blocking the entire websites at the backbone level. For example,
before the 2008 Olympic Games, the New York Times website
was entirely blocked. During the Olympic Games, the site was
partially “unblocked”, rendering some URLs (Uniform Re-
source Locates) accessible while others inaccessible. Thus, the
accessibility of a website does not guarantee that all the con-
tents on that site will be available. Typically, the blocked con-
tents are those that are deemed to be “sensitive” by the gov-
ernment. The specific “sensitive” contents change over time.
However, certain contents are regularly filtered: for example,
the Tiananmen Square protests of 1989, the independence of
Ti b et, X inj i ang and T a iwa n , an d t h e Fa l u n Gong movement, etc.
(Zittrain & Edelman, 2003; ONI, 2005).
With respect to filtering technology, China’s technology is
“the most sophisticated effort of its kind in the world”. (ONI,
2005) As early as 1998, the Chinese government began to in-
vest in the notorious Gold Shield software project. The main
function of the Gold Shield software is to censor “illegal” con-
tents. It can pick the sensitive words and block the relevant
content. However, the effectiveness of filtering technology is
unclear. The filtering systems can not foresee all the sensitive
words. In addition, sophisticated internet users can often access
blocked contents through various circumvention technologies.
Moreover, in recent years, as more and more individuals be-
gan to use internet, the resistance against such blocking and
filtering practices also increases. The controversy over the
“Green Dam Youth Escort” (“Green Dam” hereafter) provides
a revealing example. The “Green Dam” software was a filtering
device that was supposed to be very powerful in filtering inter-
net contents. In May 2008, the Minister of Industry and Infor-
mation (MII) spent more than 41 million yuan (about 6 million
US dollars) to purchase the “Green Dam” software from two
companies that had “cooperated” with Chinese government in
the past. The MII then offered the software to internet users for
free downloads. However, few individuals bothered to install
the “free” software (Chao, 2009).
On May 19, 2009, the MII went further by sending a notifi-
cation to computer manufacturers of its intention to require all
new personal computers sold in China after July 1 to pre-install
the “Green Dam” software. However, soon after the notification
was released, there was a surge of online criticism. At the night
of June 30, just several hours before the requirement was to
become effective, MII issued a notice, declaring that the re-
quirement to install the “Green Dam” software was to be post-
poned. Till day, the MII has not re-set the requirement for
compulsory pre-installation of the software.
Controlling the ISPs and the Resulting
Self-Censorship
As mentioned in Part 2, several regulations impose liabilities
on ISPs, blog service providers (BSPs), and BBS (Bulletin
Board System) providers, etc. For example, the 2000 Measures
requires IAPs and ISPs to record the dates and times when
subscribers accessed the Internet, the subscriber’s account
number, the addresses of all websites visited, and the telephone
number used to access the Internet. The ISPs and IAPs must
keep a record of this information for sixty days and provide it to
the authorities upon request. Similar liabilities were imposed on
BBS providers in another regulation promulgated in 2000.
Indeed, the Chinese authorities took specific actions to im-
plement these regulations. For example, on January 9, 2009,
Niubo, a blog service provider, was shut down because it
transmitted “harmful information on political and current af-
fairs” (Wu, 2009). Specifically, the closure was linked to its
status as being the leading domestic circulator Charter 08, a
proposal by Chinese intellectuals to reform China’s politics
(Garnaut, 2009).
Because of the threat of punishment, private entities (includ-
ing IAPs, ISPs and BSPs) often resort to self-censorship. The
following are some of the typical methods that are used by
38

H. P. ZHENG
these e nti t i e s to “ c ens o r ” t h e i nter net. First, like the govern ment,
the private entities also resort to the filtering technology. Some
forum operators have developed their own systems to catch
sensitive words so that they can review the message before it is
posted. As a result, when an internet user attempts to post an
entry which contains a “sensitive” word, he or she would re-
ceive an immediate notice stating “this message can not be
posted because it contains improper conten t ”.
Second, these private entities also employ individuals to
manually delete or conceal “sensitive” posts or comments.
These individuals are commonly known as “internet adminis-
trators”. Their routine job is to spot and delete (or conceal)
posts or comments that are deemed to be “improper”. To help
these internet administrators identify “improper” or “sensitive”
contents, many websites encourage individuals internet users to
report such contents to the administrator by clicking certain
icon.
Finally, if a blog or specific forum becomes too “sensitive”,
the ISP (or BSP) may delete or block the blog or forum. Thus,
Sina.com, China’s most popular BSP, shut down numerous
blogs that are too “sensitive”. Even some international corpora-
tions have yielded to the pressure. For example, in December
2005, Microsoft Corporation deleted the site of a Beijing blog-
ger from its MSN Spaces service (Barboza & Zeller, 2006).
This case drew much international attention partly because it
involves Microsoft, a US-based corporation. It would not have
gained much attention had it been a Chinese corporation.
Like government censorship, “private censorship” is increa-
singly being challenged by internet users. In recent years, there
had been several well-known law suits in which the owner of
the shut-down blogs sued the BSPs. For example, in August
2007, Liu Xiaoyuan, a lawyer in Beijing, sued Sohu.com for
deleting his blog posts. He alleged that Sohu breached the blog
service contract by concealing nine articles he posted on his
blog. These articles, he alleged, were “neither illegal nor ob-
scene.” Although Lawyer Liu was able to file the lawsuit in a
Beijing court, he soon received a court order stating that the
court would not accept the case. Liu sought to appeal. But the
appellate court refused to review the case (Tang, 2007).
While the plaintiffs in most of the cases, like Mr. Liu, were
unable to get their cases filed in the court, Hu Xingdou turns
out to be an exception. Mr. Hu is a professor at Beijing Tech-
nology University. He had a personal website on which he
posted his own articles, most of which were political comments.
He paid certain fees to Beijing Xinwang Corporation, a ISP
which provided the technological support for his website.
However, in March 2007, Mr. Hu received a notification from
Xinwang, informing him that his website was shut down be-
cause it contained “illegal information”. Mr. Hu then filed a suit
in a Beijing court, alleging that Xinwang breached the blog
service contract. More than twenty intellectuals in Beijing,
including some prominent law professors and lawyers, signed a
“public letter” to support Mr. Hu. Partly because of the media
pressure, the court ruled in favor of Mr. Hu, stating that Xin-
wang did not provide any evidence regarding what was “ille-
gal” on Mr. Hu’s website.
Mr. Hu’s victory was rather exceptional. According to Mr.
Hu, he knew it would be impossible to win if he alleges that the
ISP infringed his right to “freedom of speech”. As a tactical
choice, he only alleged that ISP breached the blog service con-
tract. Also, he only asked for a refund of the fees, not the resto-
ration of the website. As such, the court was able to render a
decision without deciding whether Mr. Hu’s free speech right
has been abridged.
Access Control
Currently, Chinese internet users access the internet through
three major channels: personal computers, mobile phones and
internet cafes. The internet cafes are the main access location
for about half of Chinese internet users (CINIC, 2009).
The Chinese government sought to control each of the three
accesses. The first two accesses are relatively easy to control.
An individual accessing the internet through a personal com-
puter, whether at home or at office, can be easily located. Simi-
larly, the mobile phone owners who accessed internet could be
easily identified. The following part focuses on the Chinese
government’s attempts to control citizens’ activities in internet
cafes.
The government’s control is implemented mainly through
two layers of registration requirement. First, every internet cafe
is required to register at a designated local governmental
agency. Because local governments typically limit the number
of the internet cafes in a particular locality, the registration
process essentially involves governmental licensing. The gov-
ernment would only grants the license to internet cafes that
meet certain requirements. The license may be revoked if the
internet cafe does not follow the “rules”. For example, between
June and September 2002, the government shut down 150,000
unlicensed Internet cafes (Hermida, 2002). Till day, police
routinely “raids” internet cafes to see whether there is any “il-
legal” activity going on.
The second layer of the access control occurs at the level of
individual internet users. Internet cafes are required to record
every user’s identity and online activities. Each cafe is required
to keep these records (or “logs”) for at least sixty days and to
provide the records to police upon request. Currently, these
rules are strictly enforced. Thus, one who does not take his
valid identification card with him may not access internet in
internet cafes .
Finally, all cafes are required to install monitoring software
approved by the police. Such software not only monitors online
activities of internet users in the cafe, but also filters certain
“sensitive” information.
Criminal and Administrative Punishment
A lot of individuals have been punished for “illegal” online
activities such as posting prohibited contents. In 2008 alone, it
was reported that China imprisoned at least forty-nine individu-
als for online activities, including several individuals serving
their second period of detention for internet-related crimes
(Reporters without Borders, 2009; China Human Rights De-
fenders, 2009). For example, Liu Shaokun, a school teacher,
was sentenced to one year reeducation-through-labor for post-
ing pictures of school buildings that collapsed in the 2008 Si-
chuan earthquake (Human Rights in China Press Release,
2008).
Individuals may even be punished for sending private e-mails
that contain “sensitive” contents. For example, in 2005, Shi Tao,
a Chinese citizen, was sentenced to ten years in prison for
e-mailing a “state secret” to a New York website editor. The
“top secret” reportedly expressed the Party’s concern about the
possibility of demonstrations occurring on the fifteenth anni-

H. P. ZHENG
versary of the Tiananmen Square protest. Shi used a Yahoo
e-mail to send the information. The case drew international
attention partly because Yahoo “cooperated” with the Chinese
government by providing information linking the e-mail to the
IP address of Shi’s computer (Kerstetter, 2005).
In recent years, as the internet becomes to be used by more
and more individuals, the resistance against government abuse
is also growing. For example, in February 2009, Wang Shuai,
while working in Shanghai, posted a blog entry stating that
officials in his hometown, Lingbao City in Henan province, had
misappropriated funds for combating drought. To the surprise
of Mr. Wang, police from Lingbao arrested him in Shanghai.
Fortunately for Mr. Wang, a report in China Youth Daily (a
newspaper) sparked a heated online discussion. Finally, the
media pressure became so great that the high Party officials in
Henan province issued an apology, compensated Wang for his
eight-day detention, fired the local Party secretary, and pun-
ished three officials who misappropriated funds (Chen, 2009).
Internet Police
In January 2006, the city of Shenzhen introduced two cartoon
characters that appear on all websites or internet forums in
Shenzhen. The cartoons move interactively with the internet
users as they navigate through web pages. In addition to linking
the users to information about internet regulations and inter-
net-crime cases, the cartoons also connect users to internet po-
lice through an Instant Messaging service for the purpose of
answering users’ questions about internet security. However, as
officials of Shenzhen Public Security Bureau informed the re-
porters, the “main function” of the cartoons is “to intimidate,
not to answer questions” (Qiang, 2006).
The intimidation function seems to work. It was reported that
between January and May 2006, the frequency of posting “haz-
ardous information” decreased by sixty percent, and more than
1600 Internet crime allegations were reported through the car-
toon police. Thereafter, the cartoon police were introduced in
many other cities (Xinhua News Agency, 2006).
“Guiding” the Public Opinion
Partially in response to the uncontrollable nature of internet,
the Chinese government attempted to “guide” public opinion by
hiring “internet commentators.” In 2008, a report estimated that
China employed more than 280,000 “internet commentators”
nationwide (Bandurski, 2008). While the government never
explicitly spelled out the qualifications and functions of the
“internet commentators”, media reports suggest that they
mainly perform two tasks: first, “guiding” the internet users
towards “correct” political direction; second, identifying, and
sometimes deleting “harmful” information.
The “internet commentators” originated at Nanjing univer-
sity in 2005: the university recruited students with school funds
to advocate the “correct” line on an online student forum. The
practice soon became popular at different levels of government
and other Party-controlled organizations. For example, Gansu
province, a largely poverty-stricken province in Northwestern
China, announced to recruit 650 “internet commentators” in
2009.
Besides, the Minister of Culture developed Internet com-
mentator trainings. Those who went though the training would
receive a job certification, which would qualify them to serve
as “internet commentators”. There seems to be plenty of job
opportunities for these ‘internet commentators”: not only gov-
ernment hire “internet commentators”, major websites are re-
quired to recruit in-house teams of the government-trained
commentators.
Conclusion
While censoring the internet may have some beneficial ef-
fects from the government’s perspective, it does more harm
than good. There are at least four reasons to conclude that gov-
ernments should not censor internet.
First, although censorship might keep “bad news” from being
released to the public in the short run, it can rarely do so in the
long run. In today’s world, although censorship may make it
more difficult for individuals to find certain “sensitive” infor-
mation, it can not entirely block such information. For example,
many individuals in China can actually use circumvention
technology to access “sensitive” information.
Second, even assume that the government can “successfully”
keep certain information from being transmitted to individual
citizens, such a “success” is not necessarily good. Public deci-
sions based on one-sided information are often problematic,
and may sometime s l ead to disastrous consequences.
Third, such censorship may destroy citizens’ trust for the
government. Individuals tend to suspect the news released by
government. They may think that such news has been manipu-
lated by the government. Such a situation could be devastating
for the government in the long run.
Finally, the financial costs of such censorship are huge. The
Chinese government has spent a lot of money on purchasing or
developing the filtering software, hiring the “internet adminis-
trators” and “internet commentators”, and implementing the
censorship mechanism. It is hardly possible for a democratic
government to spend so much money to curtail citizens’ speech.
REFERENCES
Bandurski, D. (2008). China’s guerrilla war for the web, far eastern
economic review. URL (last checked 16 November 2012).
http://www.feer.com/essays/2008/august/chinas-guerrilla-war-for-the
-web
Barboza, D., & Zeller, T. (2006), Microsoft shuts blog’s site after com-
plaints in Beijing. NY Times.
http://www.nytimes.com/2006/01/06/technology/06blog.html?_r=1&
oref=slogin
Chao, L. (2009). China squeezes PC makers. Wall Street Journal.
http://online.wsj.com/article/SB124440211524192081.html
Chen, J. (2009). Officials punished over land scandal. Shanghai Daily,
29 April 2009.
China Human Rights Defenders (2009). Tug of war over China’s cy-
berspace: A sequel to journey to the heart of censorship. URL (last
checked 9 March 2012).
http://crdnet.org/Article/Class9/Class11/200903/20090319000543_1
4370.html
CINIC (China Internet Network Information Center) (2009). Twenty-
third statistical survey report on the internet development in China.
23 March 2009.
Hermida, A. (2002). Behind China’s internet red firewall URL (last
checked 15 November 2012).
http://news.bbc.co.uk/1/hi/technology/2234154.stm
Human Rights in China Press Release (2008). Family visits still denied
to Sichuan school teacher punished after quake-zone visit. URL (last
checked 29 July 2008).
http://www.hrichina.org/public/contents/press?revision_Id=66556&it
em_Id=6652
40

H. P. ZHENG
Kerstetter, J. (2005). Group says Yahoo helped jail Chinese journalist.
CNET NEWS.COM, 6 September 2005.
Li, C. (2004). Internet content control in China. International Journal
of Communications Law and Policy, 1, 5
Newbold, J. R. (2003). Aiding the enemy: Imposing liability on US
corporations for selling China internet tools to restrict human rights.
University of Illinois Journal of Law, Technology & Policy.
Official Google Blog (2010). A new approach to China. URL (last
checked 16 November 2012).
http://googleblog.blogspot.com/2010/01/new-approach-to-china.html
ONI (2005). Internet filtering in China in 2004-2005: A country study.
URL (last checked 15 November 201 2).
http://opennet.net/sites/opennet.net/files/ONI_China_Country_Study.
pdf
Qiang, X. (2006). Image of internet police: Jingjing and Chacha
Online-Hong Yan. URL (last chec ke d 17 No ve mber 2012).
http://www.chinadigitaltimes.net/2006/01/image_of_internet_police_
jingjing_and_chacha_online_hon.php
Reporters without Borders (2006). List of the 13 internet enemies. URL
(last checked 16 Novem ber 2012).
http://www.rsf.org/List-of-the-13-Internet-enemies.html
Reporters without Borders (2009). 2009 Annual report: China. URL
(last checked 16 Novem ber 2012).
http://www.rsf.org/enrapport57-China.html
State Council of China (2000). Measures for managing internet infor-
mation services. URL (last checked 15 November 2012).
http://www.lehmanlaw.com/resource-centre/laws-and-regulations/inf
ormation-technology/measures-for-managing-internet-information-ser-
vices-2000.html.
Tang, X. (2007). Why delete my blog articles? Chongqing Fazhi Ribao,
29 October 2007. URL (last checked 15 November 2012).
http://blog.sina.com.cn/s/blog_3eba51c501000b7l.html
Wu, V. (2009). Popular blog service provider shut down. South China
Morning Post, 10 January 2009 .
Xinhua News Agency (2006). Cyber police in Shenzhen to curb on-line
crimes. 15 May 2006. URL (last checked 15 November 2012).
http://news.xinhuanet.com/english/2006-05/15/content_4547731.htm
Zhao, Z. G. (2008). Development and administration of internet in
China. URL (last checked 15 November 2012).
http://www.china.org.cn/china/internetForum/2008-11/06/content_16
719106.htm
Zittrain, J., & Edelman, B. (2003). Empirical analysis of internet filter-
ing in China. URL (last checked 15 November 2012).
http://cyber.law.harvard.edu/filtering/china/