In the 21st Century where Internet communications grow rapidly over the world, people can transmit information and communicate across the boundaries in a faster, easier and inexpensive way. With one click, a user can instantly send a message by email to an enormous number of recipients or can post a message onto a bulletin board on the web accessible by indeterminate audience globally. This revolutionized communications brings many opportunities and benefits, but also creates significant risks and tensions facing by the online service providers, Internet service providers, website owners, operators of bulletin boards and individual Internet users in such contexts as defamation and free speech.

The tort of defamation to some extent gives rise to a clash between the freedom of speech and the protection of reputation, and countries have apparently chosen to strike a balance between these interests differently. Some countries like the United States have defamation laws that are more pro-plaintiff while others like the United Kingdom have laws that are more pro-defendant. Given the borderless nature of Internet communication, the issue becomes even complicated with questions of jurisdiction, conflict of laws, and choice of laws.

This paper looks into the key issues respecting the application of defamation laws to the Internet. Given that the laws in effect differ from state to state and country to country, it also attempts to analyze and compare the existing defamation laws and the approach of operating of the defamation laws in different countries including the United States, the United Kingdom, the Hong Kong Special Administrative Region (HK), and the Mainland China, with detailed examination of some cases decided in their jurisdictions.

Historical Background

Prior to the development of the Internet, individual countries had considerable freedom to define the tort of defamation within their own borders. Modern libel and slander laws, as implemented in many Western nations, are generally descended from English defamation law. The earlier history of the English law of defamation is somewhat obscure. Civil actions for damages seem to have been tolerably frequent so far back as the reign of Edward I. There was no distinction drawn between words written and spoken. In the US, laws regulating slander and libel began to develop even before the American Revolution. In one of the most famous cases, New York publisher John Peter Zenger was imprisoned in 1734 for printing attacks on the governor of the colony. It was then established that the "truth" was an absolute defence against libel while the previous English defamation law had not provided this guarantee. As the law continuosly evolves along with the development of the Internet, that freedom has resulted in some more profound differences in approach. For example, Britain provides limited protection to the press and media when they criticize governmental officials, and therefore tends to be more 'pro-plaintiff'.

With the rapid expansion of the Internet and its seemingly endless freedom to disseminate information of all kinds, new concerns and interpretations have inevitably arisen about conflicts between the right to free speech and the legal mandate to protect people being defamed because of the increasing ease and potential harm facing by the defamed. Electronic media in effect poses such problems as to whether the online service providers are publishers or distributors, where should be the most appropriate forum etc. Different jurisdictions may hold different views when applying the basic concepts of the traditional defamation law to the Internet, which at present the countries cannot escape from their further considerations with the laws development and application. As with other areas of the law, we believe the law of defamation will continue to evolve and adjust to changing social and technological realities.

Key Concepts of Defamation

Definition of Defamation

Defamation can be defined as oral or written statements that wrongfully harm a person's reputation. There are two versions of defamation, libel and slander. Libel is when the defamation is written down in a permanent form such as print, writing or film, and slander is when the incident relates to spoken words. In all cases, the harm to reputation must be severe enough to lower the esteem of the plaintiff in the community by subjecting the individual to ridicule, contempt or even hatred ().

Defamation in the Internet can be classified as libel in view of that material communicated on the Internet is published e.g. bulletin boards, emails, chat rooms etc.

Elements of Proof

In order to proof defamation, one has to prove that what was said or written or published was false. The false and defamatory statement, usually of act and not opinion, must be made about another's reputation or business. Direct evidence, innuendo, insinuation or reference may establish this. And the statement must be referring to the plaintiff.

An unprivileged publication is made to a third party. Generally, if the defendant did not intend the publication to be viewed by anyone other than the plaintiff, there will be no liability caused. But it is unlikely that a defendant would be able to argue with this intention as related to publication on the Internet as very few online communications are considered private because of the global accessibility nature of the Internet.

Depending on whether the plaintiff is a public official or figure or the state law, a plaintiff may have to establish some degree of fault or negligence on the part of the defendant. The requirement for a person who is considered as a public official or public figure and who is a private individual is different. For one who is a public official or figure, the burden of proof is more demanding. He/she has to prove the defamatory statement was made with actual malice. Actual malice is the act of publishing or broadcasting the statement with prior knowledge the statement was false or with reckless regards as to whether that was false or not.

Also, the defamatory statement must result in actual or presumed damages. If the defamation is libel, damages are presumed to exist. The permanent nature of a libelous statement has led courts to allow recovery for libel without proof of actual or special damages in considering the ability, especially via the Internet, to distribute widely, and the fact that written words generally require more premeditation than those spoken. If the defamation is a slander, a plaintiff will have to establish actual or special damages unless the false accusation is slander per se where damage is presumed to exist. Special damage means pecuniary loss or loss capable of assessment in money terms. Examples include loss or refusal of employment and general loss of business. A causal link between the slander and the special damage must be established which is not too remote.

Defences of Defamation

Under the common law, there are four main defences to defamation. These are: (1) justification; (2) consent to the publication; (3) fair comment; (4) absolute and qualified privilege; and (5) innocent dissemination.

Truth is recognized as an absolute defence to defamation. If what is written or spoken about the plaintiff is true, or in cases where the statement is published with the plaintiff's consent, the defense is recognized.

For the defence of fair comment to succeed, the comment, i.e. an expression of opinion, must be on a matter of pubic interest that is based on true facts, and is fair in the sense that the comment can honestly be held by a fair-minded person.

In certain situations, one is allowed by law, to speak freely about another without regard to liability for defamation. This defense usually relates to statements associated with the effective furtherance of the operations of the government. Thus, members of the judicial, legislative and executive branches of the government are protected from liability for the publication of false statements as long as these relate to their particular functions. The law is made to recognize the need for these individuals to speak or express freely without the fear of a lawsuit.

Qualified privilege can be raised in respect of reports of the legislative or judicial proceedings or other publications made under a legal, social or moral duty. The defence of qualified privilege will only succeed if there is an absence of malice. In addition, press or the media may have a qualified privilege to inform the public about the activities of the famous which is of public interest.

In some contexts, one who has no knowledge of the defamatory material may employ the defense of "innocent dissemination". In this connection, one will generally be considered as to whether one has exercised all reasonable care in the publication of the offending material.

In the case of Internet publications, difficulties may arise to the operation of some requirements of the defences to defamation because of the unique capability of instantaneously communications across the globe via the Internet. What is considered of public interest or concern in one location may not be of the same interest or concern in other locations. In light of this, a fair of comment may fail where a comment published via the Internet if of only local concern but is published to persons outside the area of local concern.

Liability

The tort of defamation, traditionally, attempts to balance society's interest of free speech and the individual's right to his reputation. The way in which this balance varies from jurisdiction to jurisdiction (while some are more pro-plaintiff, others are more pro-defendant). In traditional media, e.g. print and broadcast media, liability is generally extended beyond the defamer to the "publisher" of the material subject to the defense of innocent dissemination. The idea of holding publishers responsible for libelous material in traditional media has been thoroughly tested and defined in courts in different jurisdictions. Clear precedents for determining who is liable for defamatory statements are well recorded. On the Internet, however, such issues are considerably more nebulous.

In so far as the Internet is concerned, defamation liability has its most important implication in respect of ISPs and websites which host the content of others.

One of the most important issues in cyberlibel is whether ISPs, operators of bulletin boards or website owners are common carriers, distributors or publishers.

Generally, courts have held that a common carrier (telephone or telegraph company) of published or transmitted material has virtually no control over the content of what is communicated over their service. So, they are not liable for the defamation.

Similarly, a distributor of published material such as booksellers, news vendors and libraries, does not exercise control over the content of what is distributed through their service. So, they have no liability for libel unless they are negligent.

On the other hand, a publisher of such as newspapers, magazines and broadcasters, can be liable for defamation. Their liability is grounded in the fact that they can exercise a sufficient degree of editorial control over what they publish.

After all, there are limited cases on the issue of the publisher/distributor characterization and the issue of liability for online defamation, and so limited insight as to what factors a court may consider in order to determine the liability of a particular online service provider for the publication of defamatory material. We can foresee that days are not easy for being a player in the Internet business.

Issues of Online Defamation

Forms of Online Defamation

The world of Internet is informal and instantaneous. There are numerous means for user to publicly express their opinions. These characteristics in a way increase the flow of information and exchange of opinion, but are also dangerous by allowing users to make defamatory remarks freely that they may otherwise be incapable of making. The easiness and anonymity in the Internet may lower the users' alertness. Outlets such as bulletin boards, messaging channels, newsgroups and even e-mail are in fact all subject to the laws of defamation.

Emailing
To send an email or forward an email is just as simple as clicking a button. When a user receive an email which might contains defamatory materials, it is very easy for them to forward it to a number of other recipients. Within a few minutes, this email can be forwarded to thousands of people over the world if other recipients click the same forwarding button. Not only the original sender violates the defamation law, each person who forwarded this email might be also responsible.

Bulletin Boards/ Forums
There are numerous channels for Internet users to voice out their opinion, such as BBS, forums and bulletin boards. These messages can be seen by millions of Internet users and might create an endless discussion by other users. Some messages might contain defamatory content and even arouse hot "flaming", known as online word battle. Though the hosts of these forums usually have a statement or disclaimer which requires the users not to post defamatory and other illegal messages in the forum, there can be thousands of threads posted within a single hour that makes the host impossible to manage and screen.

Chatrooms/ Online Instant Messenger
In a semi-public or a "members only" chatroom, users can freely talk to each other instantly without much regulation. By using online instant messenger such as ICQ and Yahoo! Messenger, people can also chat or exchange opinion, files and URLs instantly. This instant feature can be dangerous if the messages or files contain defamatory content. These messages can flow in a speedy way and hard to predict who will be receiving them.

Newsgroup
There are numerous kinds of newsgroup around the world, using different languages and focusing on different topics. There is no official membership system or regulation and anyone can post their messages in the form of email to a newsgroup. Though there is usually a newsgroup leader acting as a host who is responsible to manage the newsgroup and "kill" the threads that have illegal or off-topic, he/she can only delete the message after it is posted and be seen. They are actually only ordinary users and have weaker control ability than the Internet Service Provider (ISP). The question here is that whether this leader is liable of the defamatory messages posted by other users and which this leader has not noticed.

Improper External Links
In order to enrich the content and interest for the users, many websites post a number of external links. Users can reach these external web pages by a simple click. Some websites even join a "web ring" for promotion. However, it is impossible for the website host to check the content these external pages from time to time. If there are defamatory materials in these linkages, the host of the website might bear the risk for being responsible of spreading these materials.

The Issue of Liability

Alike the case in traditional offline defamation, there are several issues concerning the liability of online defamation. However, the players involved are not the newspapers, book publishers or re-publishers. The relationship between the players in the scope of Internet and the liability of each player are more complicated and new to the traditional law. While the original sender or creator of the defamatory materials should bear the direct responsibility, there are questions about whether other channels that these materials have passed through, such as the Internet Service Provider and the Content Service Provider, should also be liable.

The Original Creator
It is very clear that the person who create defamatory material and spread it in the Internet should be liable. However, the feature of anonymity in the Internet creates difficulty in tracing and investigation. The creator might use false name and email address, even use public computer terminal that does not belong to him, which makes the Internet Protocol (IP) Address meaningless to the investigation.

Internet Service Provider (ISP)
There are different types of ISP which provide different Internet service to users. Some ISP only provides connection and access services (IAP) while some will also provide website hosting and platform services (IPP). Some will also provide content services (ICP) to the users of Internet by hosting a website or portal, in which the content might be created by many other sources such as contracted writers or websites. All ISP have the personal or corporate information of their clients such as their IP addresses, names and mailing addresses. Theoretically, ISP serves as a bridge between the Internet and the end users. It should not be liable when one of its clients violate the law by using their services. The relationship should resemble the inventor of a knife is not responsible for the crime which a killer use this knife to kill someone (Gao, 2002). Some countries such as the United States have also established law that protects the ISP from being sued in the cases involving their clients.

However, given the different services of each ISP, some cases did actually not only sue the original creator of defamatory material but also the ISP. For instance, the role of the ICP looks similar to the publisher of a newspaper of magazine if they have the same editorial control over the content. In fact, the relationship between the ICP and its contracted content provider might be very complicated and there can be another contracted party responsible for the editorial control. Sometimes an ISP will hire a host as an employee or contract an independent company to manage some specific forums or content section, in which the ISP does not have editorial responsibility.

On the other hand, ISP is also unable to censor all the messages and contents posted in their website given the speedy traffic of the Internet and high interactivity. Therefore, when an ISP receive a compliant or warning from the plaintiff of a defamation case, it should seriously consider and decide whether to remove the specific messages or content in order to prevent from being liable of this defamatory content.

The Use of Disclaimers
Many ISPs or ICPs will post a disclaimer regarding the law of their countries in order to clarify their liability concerning the potential legal issues of their website content or services. However, the adoption of disclaimers and user policies to prohibit the posting of defamatory materials on the site may not give a site owner or ISP a watertight defense (Canham, 2002). Though such disclaimers usually state that the ISP or site owners are not responsible for what have been published in their sites or by their clients, they show that there are concerns of regulating the content and thus creating a possible liability of the ISP or site owner for the defamatory content.

The Issue of Jurisdiction

The essence of the Internet is borderless. Once a user posts a message on a website or newsgroup, every Internet user all over the world can read it. Some countries might have more onerous defamation laws while some countries such as the United States have more protection on the freedom of speech and exclude the liability of ISP under certain law. For instance, if a person from Hong Kong posts a defamatory message in a U.K. hosted forum against a company from China and the forum is usually visited by many U.S. Internet users, then where should the plaintiff - the Chinese company sue the person? Is the jurisdiction valid to the law of another country? This is a problem which needs to be tackled in the traffic-speedy Internet world.

Scenario in US

Law and Regulations

Before independence, the U.S. had been using the English Law system and in the 16th century, the authorities began using the law of defamation to punish political criticism against the government (Carter, 1999). Since the United States Supreme Court recognized the First Amendment limitations on the common law tort of defamation, courts have struggled to achieve the proper balance between the constitutionally protected guarantees of free expression and the need to protect citizens from reputation harm. In 1964, the Supreme Court ruled that public officials (later public figures) had to show that an allegedly defamatory statement was published with knowledge of its falsity. Alternatively, the plaintiff must show the defendant published with a reckless disregard for the truth (Flint, 1994).

In order to avoid long and costly trial procedures, US law reformers produced a draft legislation named Uniform Defamation Act in 1991. Though the Act was withdrawn eventually, the suggestions were revived later in the Uniform Correction or Clarification of Defamation Act in 1993.

It is believed that the correction or clarification of a published defamation may restore the victim's reputation more quickly and thoroughly than a victorious conclusion to a lawsuit. Thus the Uniform Correction or Clarification of Defamation Act seeks to provide strong incentives for individuals promptly to correct or clarify an alleged defamation as an alternative to litigation, which is a more simplified resolution. Moreover, the Act provides a uniform set of requirements that assures the media a consistent opportunity to correct or clarify.

The Act not only applies to the traditional media, but also the digital ones. In the Section 2 "Scope", it stated that this Act "applies to all publications, including writings, broadcasts, oral communications, electronic transmissions, or other forms of transmitting information." Unless a good faith attempt to obtain a correction or clarification is made within 90 days of knowledge of the publication, the plaintiff will be limited in any defamation action to recovery of provable economic loss. The 90-day period runs from knowledge of the publication by the requesting party, not from the date of publication (NCCUSL, 1993).

With increasing defamation litigations against the ISP, the U.S. has issued and implemented the Communication Decency Act (CDA) of 1996, which was a reflection of the concern about the liability of ISP. The Section 230 of the Act states that "no provider or user of an interactive computer service shall be treated as a publisher or speaker of any information provided by another information content provider." Thus, it eliminates potential civil liability where an ISP or user leader takes voluntary actions to restrict access to or availability of material that the ISP or the user leader considers objectionable. These provisions protect the ISP from being identified as a publisher in cases which the plaintiff argues the ISP has control or censorship power over the content (Ferrera, 2001).

Cases

Cubby, Inc. vs. CompuServe, Inc.
CompuServe is a company that provides computer-related products and services including a special service named CompuServe Information Service (CIS). CIS serves like an online electronic library which is accessible from a personal computer. Subscribers can access over 150 special interest forums, bulletin boards, chatrooms and databases. One of the forums is the "Journalism Forum" which focuses on journalism industry. In this forum, there is an electronic newsletter named "Rumorville" which was not controlled by CompuServe but by another company named Cameron Communications, Inc. (CCI). The relationship is indirect: CCI was hired by CompuServe to monitor and control the content of "Journalism Forum" according to the standards of CompuServe.

In 1990, Cubby, Inc., a competitor of CompuServe, filed a defamation lawsuit against CompuServe and the publisher of "Rumorville" because the "Rumorville" had published libelous remarks towards Cubby, Inc. CompuServe argued that they had no knowledge of the content of ""Rumorville"" and the allegedly defamatory statements, and that CompuServe was only a distributor but not a publisher of "Rumorville".

The conclusion of the court was in favor of CompuServe. The court agreed that CompuServe had no more editorial control over the "Rumorville" content than does any public library, bookstore or newsstand. It would be also infeasible for CompuServe to monitor every publication in over 150 forums. Therefore CompuServe was not treated as a publisher but only a distributor. The court also stated that the ruling CompuServe as a publisher responsible of monitor control would be an "impermissible burden on the First Amendment" which protects the freedom of speech (Ferrera, 2001).

This case reflects a possible harm for the ISP. On one hand, the ISP can escape liability if it takes a completely "hands off" approach regarding the content. On the other hand, the lack of effort of ISP might also result in bearing the liability of negligence. However, this was an early case involving an ISP in the U.S. which prompted more similar cases afterwards focusing on the liability of ISP.

Stratton Oakmont, Inc. vs. Prodigy Services Company
Prodigy is an Internet service company which operates a network that millions of subscribers can communicate with each other and a number of bulletin boards. Among them is a popular financial bulletin board named "Money Talk". Though Prodigy hired managers and board leaders to monitor the content of publications in "Money Talk", Prodigy had held itself as like a newspaper that monitored and exercised editorial control over the content of the messages in their bulletin boards since its operations started in 1990.

That public expression created a problem in 1994, when Stratton Oakmont, Inc. found some allegedly defamatory statements in "Money Talk" and commenced legal action against Prodigy and the unidentified poster of the statements. The plaintiff perceived Prodigy as a publisher of the statements and the board leader of "Money Talk", who was responsible of monitoring the content, acted as Prodigy's agent.

The rule of court was not in favor of Prodigy. Though Prodigy argued that it was infeasible to review every 60,000 messages posted in their bulletin board everyday, the court indicated that even such review control cannot be complete, it did not minimize the fact that Prodigy did publicly arrogate itself to the role of determining what is proper to post and be read in its bulletin boards. Therefore, the court summarized Prodigy as a publisher rather than a distributor. Furthermore, the court affirmed that Prodigy had directed and controlled the actions of board leader, thus held that the board leader as an agent of Prodigy.

This case reflected the importance of editorial control over content for any ISP. Though Prodigy might have expressed itself as a responsible service provider with well-intentioned monitor policies, it was found guilty because of this same good intention. The role of the board leader created another concern: whether it is an independent contractor or an employee is significant to the ISP as well. These issues aroused much concern and eventually impelled the U.S. Congress to issue the CDA later in 1996 in order to protect the good faith of ISP and freedom of speech.

Zeran vs. America Online, Inc.
Six days after the bombing of the Alfred P. Murrah Federal Building in Oklahoma City, an unidentified person posted a message on an America Online (AOL) bulletin board on April 25, 1995, advertising some "Naughty Oklahoma T-shirts" with features of tasteless and offensive slogans related to the bombing. The advertisement said that interested buyers of the T-shirts could call Kenneth Zeran at his residential phone number in the State of Washington. It was a vicious hoax that Zeran was innocent and did nothing with the advertisement nor the production and selling of those T-shirts.

However, in result of that abhorrent advertisement, Zeran received numerous angry calls from the readers of the bulletin board, some even composed death threats towards Zeran. Zeran contacted AOL immediately and AOL assured him that the posting would be removed but AOL would not post a retraction as a matter of policy. In the next few days, the same anonymous message was posted again and again everyday with even more disgusted slogans and products. As a result, Zeran received more threatening calls. He called AOL repeatedly and got an answer that the account of those messages would soon be closed. Living in a nightmare, Zeran reported the situation to the FBI.

Unfortunately, an Oklahoma City radio station reported the messages and intensified the number of calls to Zeran's home. Not until May 14, after an Oklahoma City newspaper reported the messages as a hoax and after the radio station apologized on air, did the number of calls to Zeran reduced greatly, but still 15 per day.

Zeran finally put the case to trial in April 1996 and asserted that AOL unreasonably delayed in removing defamatory messages, refused to post retractions and failed to screen similar posting thereafter. He asserted that AOL was liable for the distribution of these defamatory materials as they knew them defamatory. The AOL instantly raised Section 230 of the CDA as a defense but Zeran countered by arguing that Section 230 of CDA could not be retroactively applied in this case because his claims occurred prior to the effective date of Section 230, February 8, 1996.

The suit was first put to the federal district court where the court agreed with AOL that the CDA barred Zeran's claims. Zeran then filed appeal to the circuit court. However, the court affirmed that Section 230 applied to all complaints brought after its enactment and Zeran filed his claims 2 months after Section 230 became effective. The court therefore affirmed the judgment of the district court.

Special Issues and Limitations

The three cases aforementioned reflect different complications concerning the lawsuits of online defamation in the U.S. Before the CDA of 1996, ISPs face possible claims and liability that might harm their operations and the freedom of the Internet because of the defamatory content in its website or services. In the aftermath of the enactment of the CDA, ISPs suddenly become totally immune from any liability and lawsuits. This change is not only dramatic but also ironic. In the case of Stratton Oakmont, Inc. vs. Prodigy, we see that an ISP with a good intention might be "unethically" charged. However, in the case of Zeran vs. AOL, we see how an innocent victim of vicious defamatory harassment lost his case. In implementing and enacting of Section 230, it appears that U.S. courts and Congress are more concerned with and dedicated in keeping the Internet free of restraints by immunizing the ISP and encouraging them to self-regulate. However, there are no rules overlooking the self regulation of ISP, then how can we assure the effect these self-regulatory measures and the amount of effort the ISPs make? The victim of the tort of defamation seems to be neglected. The impact of Section 230 involves many ethical issues and should be reconsidered by the U.S. Congress carefully.

Scenario in UK

Laws and Regulations

In the UK, the governing statute is the Defamation Act 1996. The 1996 Defamation Act imposes penalties on the publication of defamatory remarks regardless of its form: libel or slander. Plaintiffs who bring successful actions under the Act are entitled to substantial damages. In order for materials to be considered libelous, the statements must be published and have the effect of lowering the esteem of the plaintiff in the eyes of right-thinking members of society.

Section 1 of the Act establishes a defence, in many cases to Internet intermediaries e.g. ISPs, in respect of material hosted, cached or carried by them, but which they did not create. The section 1 defence is, in effect, a statutory version of the defence of innocent dissemination, modified for modern conditions. It sets out a three-limb test, all of which must be met for a person to establish a defense in defamatory proceedings:

(1) (a) he was not the author, editor or publisher of the statement complained of;
(b) he took reasonable care in relation to its publication; and
(c) he did not know, and had no reason to believe, that what he did caused or contibuted to the publication of a defamatory statement.

As outlined in the Act, "author" means the originator of the statement, but does not include a person who did not intend that his statement be published at all; "editor" means a person having editorial or equivalent responsibility for the content of the statement or the decision to publish it; and "publisher" means a commercial publisher, that is, a person whose business is issuing material to the public, or a section of the public, who issues material containing the statement in the course of that business. In relation to Internet publication, a person shall not be considered the author, editor or publisher of a statement if he is only involved in processing, making copies of, distributing or selling any electronic medium in or on which the statement is recorded, or in operating or providing any equipment, system or service by means of which the statement is retrieved, copied, distributed or made available in electronic form, under section 1(3)(c); or as the operator of or provider of access to a communications system by means of which the statement is transmitted, or made available, by a person over whom he has no effective control, under section 1(3)(e).

It seems that there is an inherent tension between, on one hand, the ISPs' obligation under section 1(1)(b) to exercise reasonable care and, on the other hand, the requirement, to satisfy section 1(1)(a) that the ISPs only be involved in the publication to the extent set out in section 1(3). Self-regulated or responsible ISPs who put in place procedures to detect and remove defamatory material from their websites or discussion forums may therefore be deprived of the protection of the section 1 defence because they cannot be considered as "only involved" in section 1(3)(c) or (e). If the ISPs do not put in place the procedures, they are still deprived of the section 1 defence because they fail to exercise reasonable care.

In determining for the purposes of Section 1 (1) whether a person took reasonable care, or had reason to believe that what he did caused or contributed to the publication of a defamatory statement, the Act has defined in section 1(5) what regard shall be had to: (a) the extent of his responsibility for the content of the statement or the decision to publish it; (b) the nature or circumstances of the publication; and (c) the previous conduct or character of the author, editor or publisher.

Depending on the circumstances, plaintiffs defamed by material punished via the Internet may, in addition to civil defamation law, have a number of alternative causes of action open to them in the UK. Some of the most important alternative causes of action include Negligence, Privacy and Data Protection Legislation.

A negligent misstatement occurs where a defendant, who owes a duty of care to the plaintiff, breaches that duty of care by words which cause the plaintiff to suffer reasonably foreseeable, pure economic loss. This alternative cause of action will generally be most attractive in cases where plaintiffs would be deprived of a remedy in defamation law because false material was published of and concerning them without malice, on an occasion of qualified privilege. Although there is no necessary overlap between privacy and reputation, there will be cases in which defamatory material published via the Internet will concurrently amount to be a breach of privacy. That issues of privacy will also be relevant in defamation actions in the UK by reason of the operation of section 12 of the Human Rights Act 1998. The Data Protection Act 19989 ("DPA") regulates the processing of information relating to individuals in the UK. At the heart of the DPA is the protection of the right to privacy. In broad terms, the DPA imposes obligations on "data controllers" in respect of the processing of personal data. ISPs and content providers will generally satisfy the definition of "data controllers" in the DPA. The negligent publication of a misstatement may also give rise to a cause of action which in the UK, is capable of operating concurrently with civil defamation law.

Cases

Godfrey v. Demon Internet Limited (1999)
One of the few UK cases to date is Laurence Godfrey v Demon Internet Limited (1999) which highlighted the difficulties in applying the Defamation Act to libelous statements on the Internet. It considered in detail the role of ISP's in relation to the Act.

In that case, someone unknown purported to be Mr. Laurence Godfrey, a British physicist, posted defamatory material concerning him to a newsgroup in the US on 13 January 1999. The posting followed a path from its originating US ISP to the Defendants' (Demon) news server in the UK where Demon's customers could read it. On 17 January 1997, the real Mr. Godfrey informed Demon of the existence of the posting and the fact that it was fraudulent and defamatory and requested that the posting be removed. Unfortunately, Demon failed to take action and the posting continued to be available on its news server until it expired in the usual way ten days later.

In its defense, Demon said that it was not the author, editor or publisher of the statement i.e. it was one of the excluded categories. The court agreed. It was merely involved as an operator or provider of access to a communications system through which the statement was made available. However, the court disagreed its fulfillment of the other two parts of the s.1 test that Demon failed to take reasonable care in relation to the publication of the defamatory posting. It was a fact that Demon had been notified of the posting by the real Mr. Godfrey and requested to take immediate and appropriate action, so Demon could no longer say that it had no reason to believe that what it did caused or contributed to the publication of a defamatory statement.

The conclusion was even fortified by the contents of the Consultation Document issued by the Lord Chancellor's Department in July 1995 which said that the defense of innocent dissemination has never provided an absolute immunity for distributors, however mechanical their contribution. It does not protect those who knew that the material they were handling was defamatory, or who ought to have known of its nature. Those safeguards are preserved, so that the defense is not available to a defendant who knew that his act involved or contributed to publication defamatory of the plaintiff. It is available only if, having taken all reasonable care, the defendant had no reason to suspect that his act had that effect. Sub-sections (5) and (6) of the Act describe factors which will be taken into account in determining whether the defendant took all reasonable care.

Although it has been suggested that the defense should always apply unless the plaintiff is able to show that the defendant did indeed have the disqualifying knowledge or cause for suspicion, only the defendant knows exactly what care he has taken. Accordingly, as in most defenses, it is for the defendant to show that the defense applies to him."

The decision in that case was that ISP Demon agreed to settle out of court to pay damages to academic Laurence Godfrey who claimed it had hosted "squalid, obscene and defamatory" material about him.

The case has shown that the legal regulation of the Internet remains a grey area. If an ISP have to remove any offending material from a website simply someone objects to it or claims it be offending because it assumes its all reasonable care to protect itself, it appears that any individual can force an ISP to remove content on demand. It is to some extent unacceptable and undesirable that an offended party should simply notify an ISP claiming the information to be legally defamatory. While in another spectrum, the ISP may possibly be sued by users on the grounds that their right to freedom of expression are violated.

Berezovsky v. Michaels & Others
This case concerned two Russian businessman who alleged that defamatory material had been published in a US magazine, Forbes, and distributed in the UK. Such publication of defamatory material was made on the Internet and in a magazine. Even though Forbes is a US-based magazine, and has relatively few subscriptions in the UK or Russia, Berezovsky and another Russian businessman sued Forbes in the UK. The plaintiffs had decided to sue in the UK rather than in Russia or the US, claiming substantial connections with the jurisdiction through visits, business relationships and, in one case, personal and family ties. They had confined their claims to damages to their reputations in the UK.

The House of Lords said that publication in the UK of a libel that was internationally disseminated constituted a separate tort, and as publication took place where the words were read, an action could be brought in the UK in respect of the publication made there. In addition, given the evidence made available to the court that there were reputation of the plaintiffs to protect in the UK, the UK could be taken as the most appropriate forum.

No doubt that that case was related to the issue of jurisdiction, and was problematic in the sense that the defamatory statement was made by one in one country, on a website hosted in second country but could be accessed anywhere in effect, and the publication occurs in many countries. Complicating the issue may be that the defamed person has a reputation in a third country. The decision of where the suit can be brought somehow depends on how the court in one jurisdiction interprets. In particular, the laws theoretically, can be different from country to country. As concerned, the plaintiff may be encouraged to engage in "forum shopping", choosing the country whose laws are most favourable to him/her.B. Berezovsky v. Michaels & Others

Totalise Plc. V. Motley Fool Ltd. & Another
In that case, Totalise, the UK-based ISP, requested web sites Motley Fool and Interactive Investor International to disclose the identity of the person who used the name "Zeddust" anonymously posted defamatory material to their discussion boards.

The Motley Fool and Interactive Investor discussion boards acted as discussion forums for various companies. Zeddust began posting defamatory comments on Motley Fool's web site. After complaints by Totalise, Zeddust was temporarily removed and then finally banned. Totalise sought disclosure of Zeddust's identity but refused by Motley Fool. Postings began appearing on Interactive Investor's web site. Interactive Investor removed the postings but refused disclosure. So Totalise went to court seeking an order for disclosure of identity.

The defendants attempted to rely on the terms and conditions for access to their websites and section 35 of the Data Protection Act 1998 ("DPA"), which provides that personal data shall not be disclosed except by legal compulsion, and to rely on section 10 of the Contempt of Court Act 1981 ("CCA") which provides that "No court may require a person to disclose¡K the source of information contained in a publication for which he is responsible, unless it can be established to the satisfaction of the court that disclosure is necessary in the interests of justice or national security or for the prevention of disorder or crime.

It was considered that the protection provided to newspapers by the Contempt of Court Act had no application to Motley Fool and Interactive Investor because neither took responsibility for what was posted on their discussion boards nor exercised editorial control, and it was believed that, in any event, disclosure was necessary "in the interests of justice." In view of the potential audience for the postings and the potential harm to Totalise, the judge held that the balance fell in favour of Totalise otherwise one could defame with impunity behind the screen of anonymity by the use of websites on the Internet.

Special Issues and Limitations

Many industry insiders and those civil liberties campaigners e.g. Cyber Rights & Cyber Liberties, believe that the combination of the decisions in the above cases will have a profound chilling effect on publishers based outside the UK who allowed published information to enter the territory, and on the Internet communications all over the world.

In my opinion, the results have brought about a warning to the ISPs all over the world that they need to have special caution in relation to material that can be seen or read in the UK. For the sake of protection, ISP with forums or discussion boards should at least incorporate a term to ensure that the ISPs retain discretion to remove inappropriate or unlawful material because the adoption of disclaimers and user policies to prohibit the posting of defamatory materials on the site may not give a site owner or ISP a watertight defense (Canham, 2002). If an ISP receives a complaint in relation to material found to be offending, the matter should be taken care of without delay, and if appropriate, the material should be removed. Needless to say, changes or modifications to existing laws are urgently needed, in particular relating to how ISPs are regulated. More formal guidelines and procedures e.g. for warning ISPs to take defamatory material down so that material cannot be pulled down from a website simply someone objects to it or claims it be offending. To certain extent, it is not fair to the ISPs who are seen stuck in the middle in such events.

Scenario in HK

Laws and Regulations

In Hong Kong, the law of defamation is governed by the common law, which is modified and clarified to some extent by the Defamation Ordinance (DO). In order to succeed in an action for defamation under the common law, a plaintiff has to establish three elements: (1) that a defamatory statement has been made; (2) that the statement referred to the plaintiff; and (3) the statement was published to a third party (Wright, 2003). Among the three elements, the issue of "published" or "publication" is of concern in relation to the Internet. Publication is important in considering an action of defamation because it relates to how a person is considered by a third party(ies). However, the definition of "publication" can be very wide, in particular when the information is distributed via the Internet. It is sometimes sufficient if only one person has read the defamatory material.

Under the common law, there are four main defences to defamation. These are: (1) justification , i.e. that the words are true in substance and in fact; (2) fair comment, i.e. an honest expression of opinion (not a statement of fact) on a matter of public interest that is based on true facts, being a comment that any fair minded person would be capable of making, however prejudiced his or her views; (3) absolute and qualified privilege; and (4) consent to the publication. One further defence of innocent dissemination is available for a person who is not an author, printer or publisher of the defamatory material.

In Hong Kong, a statutory innocent dissemination is also provided in section 25 of the DO, but unlike the UK Defamation Act 1996, the DO does not specify a defence for a party, such as ISP, who is involved in electronic publications. As far as ISPs concerned, the ISP may make an "offer of amend" if it claims that the words were innocently published. An offer of amend is an offer to publish a correction of the offending words and a sufficient apology to the aggrieved party. In the Internet context, a "sufficient apology" may mean publishing an apology on the ISP's home page or, in the case of a newsgroup, posting such a message to the subscribers of the newsgroups. The offer of amends must be expressed to be specifically for the purposes of section 25 and be accompanied by an affidavit specifying the facts relied on to show that the words were published innocently (Wright, 2003). If the offer of amends is accepted by the aggrieved party and is duly performed, the proceedings cannot be taken or continued against the person who made the offer, but be possibly taken against the author of the defamatory statement. This is rather a practical problem in the Internet communications as the identity of the author is always unknown.

For the purpose of the defence of innocent dissemination, section 25(5) sets out its relevant elements. The words will be treated as published innocently if either of such conditions "(a) that the publisher did not intend to publish the words of and concerning that other person, and did not know of circumstances by virtue of which they might be understood to refer to him; or (b) that the words were not defamatory on the face of them, and the publisher exercised all reasonable care in relation to the publication¡K" is satisfied and in either case, all "reasonable care" must be exercised by the publisher in relation to the publication. At this point, what constitutes a "reasonable care" is not specified but understood that it should not be reasonably expected an ISP to review all the contents or messages the website to which it provides access. In such cases, it will be open for the Hong Kong court to decide. Decisions in other jurisdictions e.g. the decision in Godfrey v. Demon Internet Ltd as discussed above may be a good source as guidance on how "reasonable care" be dealt with in HK.

In view of the defamation law and cases in HK so far, it reflects an immense need for further development, education and application of the defamation law to the Internet in the information age.

Cases

E-silkroad v. Icered.com
The plaintiff, E-silkroad Holdings who provides of e-commerce development services and an on-line trade fair on Hong Kong's Growth Enterprise Market operated by the Hong Kong Stock Exchange, alleged that a number of libelous messages posted on the Icered website between October 2000 and March 2001. Icered is a Hongkong-based website providing a number of forums on which anyone can post messages. The discussion boards are divided up by profession or industries (law, accountancy, banking etc.) and within that by 'threads' (headings posing questions inviting further discussion).

In that case, E-silkroad claimed against IceRed.com who carried messages which were very derogatory of E-silkroad and some of its directors and associates and so affected its share price when it was listed on the Growth Enterprise Market. Apart from seeking damages for libel, E-Silkroad also sought the disclosure of the name(s), address(es), email address(es) and Internet Protocol (IP) address(es) of the author(s). With this information, E-silkroad hoped to identify those responsible.

The fact was that E-silkroad won a court order requiring IceRed.com to disclose the IP addresses of those users who sent the messages. Other than that, the case has not proceeded any further.

Now Icered.com requires users to register with a valid email address before posting message into the forums, but users can still effectively operate anonymously by providing an email address by providing a hotmail or similar nature email account. It is not surprising to see that users will go on their postings anonymously, some of which may give arise to libel and some may not. This is ultimately the essence of the "Internet communications". Because of this, the defamed will find it difficult to successfully identify the "real defamer" and teach him/her a real lesson.

Investasia Ltd. & Another v. Kodansha Co. Ltd. & Another
The central issue of the case was about the "appropriate forum" for proceedings in relation to the publication of defamatory material on the Internet. In that case, the plaintiffs, Investasia Ltd (the first plaintiff) and Seiichi Wada (the second plaintiff) issued a writ against the defendants, Kodansha Co Ltd. (the first defendant) and Hirotoshi Ito (the second defendant) on 20 November 1997. The statement of claims alleged that the plaintiffs were libelled by two articles written in Japanese language by the second defendant and published by the first defendant. Both defendants were in Japan. The first plaintiff was a company incorporated in the British Virgin Islands and carried on business in Hong Kong and elsewhere. The second plaintiff said he was a businessman with business in Hong Kong , Japan and elsewhere and was the controlling hand of the first plaintiff. One of the subject articles was published on the Internet and the other one was published on a magazine which had 157 copies distributed in Hong Kong out of its some 500,000 copies published.

It was concluded that the tort was committed and the damages was sustained in Hong Kong. In coming to the decision, the court has considered and relied on some principles which I think these worth attention to either local or overseas, online or offline publishers. In my interpretation, these are:

Scale of publication
It was no doubt that the number of copies being distributed within the jurisdiction was insignificant as compared to the total number. But the judge's consideration did reflect that we should not rely on such a comparative number, instead, we should consider whether the absolute number of copies concerned is sufficient to determine a complaint about the publication is one of substance. I agree that if there are evidence in any particular case that a person has a reputation in the country to be defamed, or is known there, or traded there, or had a professional or social connections, it may be that the circulation of even a very few copies may cause him very serious harm.

Connections and reputation
Although the first plaintiff was a British Virgin Island company and the second plaintiff did not have a HK Identity card, the considerations should be focused on whether they had a reputation there merits protection. As the Judge considered in that case following the UK case of Berezovsky v. Michaels & Others, the degree to which the plaintiffs had connections and/or a reputation in the subject country was relevant.

The nature of the publication
In that case, one of the considerations taken into by the judge was the nature of publication. As said, if the alleged defamatory material was what one might describe as sensational and juicy, a much smaller size of publication would be sufficient. In my opinion, the consideration is logical but in practice, it may be too subjective.

The recent cases
So far, not many cases about defamation in relation to the Internet are heard. I am not sure why but I do believe there should be various content or messages carried on the website or transmitted via email are subject to defamation. For instance, an email was sent out by a lady resided in Hong Kong which was considered defamatory respecting Fancl, a Japanese brand of cosmetics and nutrients food. It was said that most of the Fancl shops in Japan was closed because of its poor business. As publicized, Hong Kong office of Fancl has taken immediate action against the lady upon notice of the alleged libellous material.

Another case happened in March or April 2003 when every citizen in Hong Kong is worrying about the outbreak of SARs, an email was disseminated widely saying a group of clinical doctors have contracted the deadly virus, SARs. Because of the email, some influenced doctors have reported a financial loss because patients were scared away, and considered taking an action for the alleged email message.

Although we still don't know whether legal actions are eventually taken and the decisions, these happenings do warn us the improvement on the law of defamation with clearer guidelines not only to the ISPs or other online service providers, but also to the netizens of Hong Kong before the situation get worst

Special Issues and Limitations

Undoubtedly, case law on Internet defamation in Hong Kong is lacking, so cases decided in other jurisdictions are much relied on e.g. the UK case of Totalise Plc. V. Motley Fool Ltd. & Another, the UK case of Berezovsky v. Michaels & Others. Problems may arise in view of the borderless nature of Internet communications, and differences in the law of defamation among countries. Taking the case of Esilkroad v. Icered.com as example, if the court took the US position, disclosure of information about anonymous defendant may not be allowed. So one important issue that needs second thought by the HK government or the law reformers is what country and what decision to follow is most relevant and appropriate in events that we don't have the relevant laws.

Also, as said above, the recent alleged defamatory email messages do warn us the urgency for improvement on the defamation law related to the Internet and education of the laws and liabilities to the public including ISPs and netizens.

Scenario in PRC

Laws and Regulations

According to the Constitution of the People's Republic of China, there are laws protecting the citizens from being slandered or libeled. In Article 38 "The personal dignity of citizens of the People's Republic of China is inviolable" under Chapter II "The Fundamental Rights and Duties of Citizens", "insult, libel, false charge or frame-up directed against citizens by any means is prohibited" (PRC, 1982).

In the "General Principles of the Civil Law of the People's Republic of China" adopted at the Fourth Session of the Sixth National People's Congress, promulgated by Order No. 37 of the President of the People's Republic of China on April 12, 1986 and effective as of January 1, 1987 (PRC, 1987), there are also laws concerning defamation.

In the Section IV "Personal Rights" under Chapter V "Civil Rights", Article 101 "Citizens and legal persons shall enjoy the right of reputation" has stated that "the personality of citizens shall be protected by law, and the use of insults, libel or other means to damage the reputation of citizens or legal persons shall be prohibited."

In the Section II "Civil Liability for Breach of Contract" under the Chapter VI "Civil Liability", Article 120 states that "if a citizen's right of personal name, portrait, reputation or honor is infringed upon, he shall have the right to demand that the infringement be stopped, his reputation be rehabilitated, the ill effects be eliminated and an apology be made; he may also demand compensation for losses."

A just few years ago, there are also regulations specifically targeted for Internet control. One of them is the "Administrative Measures on Internet Information Services" (PRC, 2000) adopted at the 31st Executive Meeting of the State Council on September 20, 2000 by Order No. 292 of the State Council of the People's Republic of China. The Article 15 states that Internet information service providers shall not produce, duplicate, release or disseminate information that contains content that insults or defames others or infringes upon the lawful rights and interests of others.

Similar regulations but specifically targeted to online bulletin boards were issued less than one month later. On October 8, 2000, the Ministry of Information Industry issued a new regulation on administration of Internet Broadcasting Bulletin System (BBS) (People's Daily, 2000). In Article 9 of this regulation, it said that no one can release BBS information that insults or defames others or infringes upon the lawful rights and interests of others.

Concerning the penalty of the liable parties, the "General Principles of the Civil Law of the People's Republic of China" has stated a number of methods in Section IV "Methods of Bearing Civil Liability" under the Chapter VI "Civil Liability": the Article 134 states that the main methods of bearing civil liability shall be (1) cessation of infringements; (2) removal of obstacles; (3) elimination of dangers; (4) return of property; (5) restoration of original condition; (6) repair, reworking or replacement; (7) compensation for losses; (8) payment of breach of contract damages; (9) elimination of ill effects and rehabilitation of reputation; and (10) extension of apology. All these methods of bearing civil liability may be applied exclusively or concurrently.

The article also states that, when hearing civil cases, a people's court, in addition to applying the above stipulations, may serve admonitions, order the offender to sign a pledge of repentance, and confiscate the property used in carrying out illegal activities and the illegal income obtained there from. It may also impose fines or detentions as stipulated by law.

Cases

The First Case of Digital Defamation
The case between MAX Computer Station, Inc. (MAX) and an ordinary Beijing citizen named Wang Hung was not only the first online defamation case in China, but also aroused much concerns and discussions among the numerous Internet users in China.

The case can be traced back in 1998. Wang, whose company he worked for bought a computer made by MAX in 1997, had created a personal web page on the Internet and named it as "Indignant with MAX, Protecting the Consumer Rights". There was large number of articles attacking MAX and among one of them even said that the product of MAX was "garbage".

Later in the same year, a magazine and a newspaper reported a story about this article. In result, numerous consumers demanded MAX for refund and the amount reached over 24 million RMB. Thus, MAX sued not only Wang but also the magazine and the newspaper, and demanded compensation equal to 4% of their loss, which means approximately RMB 1 million.

In the first trial on December 15, 1999, the court believed that Wang had deliberately libeled the plaintiff in his website and the two media had also reported defamatory articles against the plaintiff, therefore ruled that Wang needed to compensate RMB 500,000 to the plaintiff and the two media also needed to compensate RMB 240,000 to the plaintiff. Three parties are also needed to publish apology statement to the plaintiff publicly. This was the largest amount of compensation recorded and aroused intense discussions in China (Beijing Morning Post, 1999).

In the second trial in 2000, the court affirmed that Wang needed to stop all kinds of tort of defamation against the plaintiff in the Internet and issue apology statement to the plaintiff in various major Chinese websites. However, the court believed that the economic actual loss of the plaintiff was in lack of evidences, therefore the sum of compensation that Wang needed to pay to the plaintiff was reduced to RMB 90,000. The court also adjudged that the two media that reported Wang's article were misled by incomplete information and therefore they needed not to pay any economic compensation but only an apology (Gao, 2002).

The Haier Case
The nature of the lawsuit between Haier Group, a household electrical appliances manufacturer, and Chen Yi-cong, a 25 year old financial analyst of www.fayhoo.com, a financial news website, was similar to the case MAX case. However, this case had created more discussion concerning the essence of defamation: how to differentiate libelous content and sole opinion?

On July 9, 2002, Chen received the summons from the court. The Haier group indicted him for libeling and infringing the interests of Haier. The indictment said that on March 25, 2002, Chen's article named "The truth of Haier" was posted on the website of www.enet.com.cn, using an anonym. The article was then republished in a financial magazine named "Wealth Today". The plaintiff claimed that this article contained fake and fictional content and requested Chen to delete his article from the Internet, retrieve all the "Wealth Today" issues that contained his article, make an apology publicly, pay all the fees for this lawsuit and a compensation of RMB 300,000 to Haier (Daynews.com, 2002). However, the plaintiff did not sue the magazine and the host of www.enet.com.cn.

It was believed that what triggered Haier to take legal actions might be related to another an article of Financial Times, a UK financial newspaper. On June 25, an article in Financial Times adopted Chen's and another Chinese analyst's point of view. Fewer than two weeks later, Haier had put the case to court.

Chen said that he had never thought of being sued because of this article. He claimed that his article was originally posted in a BBS but his friend reposted it to the "Wealth Today". He did not think that the content of the article was libelous but only his own opinion. He exemplified two sentences that Haier found libelous and claimed that they were not libelous at all. One example is the first sentence of the article: "Maybe Haier was glorified too much and therefore people will become suspicious as usual." Chen claimed that he already used the term "Maybe" and undoubtedly there were people who are suspicious toward Haier. Another example is a sub-header named "Unsuccessful Multiple Strategies". Chen claimed that the word "unsuccessful" did not certainly means "failure" and the meaning of "multiple strategies" was very broad and vague, which meaning could be different subject to different persons.

Unexpectedly, the case was settled only 40 days later. On August 16, 2002, the plaintiff and Chen were mediated by the court and reached a settlement (Sun & Ng, 2002). It was believed that it was because Chen had the better chance in this case. Though this case was over, the impact was still in many Chinese Internet users' heart.

The first case suing the ISP
The plaintiffs of this case are two owner representatives of a residential estate named Beautiful Garden. From June 2001, they had read articles about property in a BBS named after their owner organization. They claimed that three articles were libelous and had insulted them. After negotiation with www.sinoi.com, the BBS hosting service provider, those three articles were deleted gradually. However, the two owner representatives were not satisfied and filed the case to the court, suing the www.sinoi.com for being irresponsible of managing and censoring its website, and not providing the information of the articles writers (Fung, 2001).

The defendant countered with regard to the new regulation on administration of internet broadcasting bulletin system issued by the Ministry of Information Industry in 2000. It claimed that regulation only stated that the host needed to bear liability only when it did not delete or care any illegal messages. In fact, the defendant had no time and ability to censor numerous messages posted to the BBS each by each and they had already deleted the three articles after the plaintiffs' notification.

On March 20, 2002, the court ruled that the three articles did infringe the reputation of the plaintiffs but the defendant did not bear any responsibility. It was because the defendant, served as a hosting service provider, should not arbitrarily delete any messages in the BBS. The court also agreed that the defendant had already taken remedial actions, deleted those three libelous articles. The court overruled the plaintiffs' demand to the defendant for revealing the information and identity of the original writers of the articles, because the time when the plaintiffs raised this request was already over the period of keeping personal information by website.

Though the libelous articles remained unknown, the court in this case overruled the plaintiffs' claims against the ISP, which means a lot to the ISP companies in China. The court accepted the view that ISP should not arbitrarily delete any messages in the BBS when they are not sure about whether any specific message might be libelous (Xinhuanet, 2002). However, how to define whether a message is libelous is another question.

Special Issues and Limitations

Internet use has been expanding at such an alarming speed in China. In 2002, there are 46 million net surfers in China, compared with 8.9 million in 2000. However, the Chinese government still lacks experience in regulating the Internet. Xu Youbai, an official with the China National Net Civilizing Project, says that "as a matter of act, some Western countries have much more detailed restrictions on the Internet. Both the United States and Europe have enacted laws regarding rampant online pornography, fraud and defamation" (People's Daily, 2002).

On the other hand, some regulations have hindered the development and survival of ISP, as well as the digital freedom. For instance, the Internet laws regulating the bulletin board systems issued in 2000 have restricted the content of online chatrooms and bulletin boards. BBS providers and web portals will suffer because they will have to supervise their content more closely. Many struggling Chinese online ventures will not have the financial resources to obey the new laws if they are strictly enforced, and liberal ICP that rely on articles from foreign-based news media to attract readers may be hurt as well (DFN, 2000).

Concerning the anonymity feature of the Internet world, the Chinese law can create a 60-day "gate-keeping" period for tracing the real identity of an anonymous libellant. According to the Article 14 in the "Administrative Measures on Internet Information Services" (PRC, 2000), Internet information service providers engaging in services such as news, publication and electronic bulletin services shall record the information content provided and the time of release of the same, and the Internet addresses or domain names. Internet access service providers shall record information such as the log on times of online users, users' account numbers, their Internet addresses or domain names and the telephone numbers from which they call. Internet information service providers and Internet access service providers shall keep back up copies of their records for 60 days and shall provide the same to the relevant State authorities when they make inquiries according to law. Therefore, lawsuit against anonymous writer should be filed within the 60 days period, otherwise the identity of the writer might not be revealed.

Comparison between Different Countries

Conceptual Framework

US
Freedom of speech is highly protected in the United States by the spirit of the First Amendment. When dealing with defamation lawsuits, the U.S. court differentiates the allegation of facts and allegation of opinion. Allegation of opinion, which means the author only expresses his/her own evaluation on the nature, meaning and values of an incident that already happened, is definitely protected by "Privilege to Opinion" regarding to the First Amendment, in which the concept of true or false does not exist.

Based on this conceptual framework, the U.S. Congress must carefully balance the law of defamation and the freedom of speech. Media which express opinion on public officials have privilege when facing suits of defamation. No matter offline or online, the U.S. legal system does the most to protect the "free" environment.

UK
The UK libel law is somewhat different from the US libel law in view of its approach to the burden of proof. The UK law puts the burden of proof upon the defendant to show that he / she did not commit libel, which is opposite of the UK law that places the burden on the plaintiff to show that the alleged libelous statement did contain actual malice and did cause damages.

The UK takes the view that regulation is necessary at the boundary between freedom of speech and defamation. In addition, the position in the UK can be contrasted with the stricter position being taken by the UK in relation to disclosing the anonymity of defendants. Freedom of speech remains a very sensitive area for legal regulation.

HK
The law of defamation is governed by the common law, which is modified and clarified to some extent by the Defamation Ordinance (DO). Generally, the elements of proof for a plaintiff to succeed in an action for defamation under the common law is more or less the same as other countries like the US and UK. While talking about the overall approach in operating the defamation laws in Hong Kong, it seems to us that its approach tend to be more like that of UK. That can be explained by the facts that the legal framework, not merely the defamation law, has been established from the UK basis, and since 1 July 1997 when Hong Kong was reverted from the UK to Chinese sovereignty, no significant change in terms of its approach was seen.

In light of the decisions and the considerations made to the very few HK cyberlibel cases by the court of Hong Kong, e.g. Investasia Ltd. & Another v. Kodansha Co. Ltd. & Another, and the lack of case law on Internet defamation in H, we can be convinced that the rulings and the future development of HK in the context of defamation will be likely influenced by other jurisdictions, e.g. the UK.

China
In comparison, China has less comprehensive regulations on defamation, especially in lack of concrete conceptual support. Article 38 of the 1982 Constitution states that the personal dignity of citizens is inviolable and any insult, libel, false charge or frame-up directed against citizens by any means is prohibited (PRC, 1982). However, the standard is relatively unclear and the balance of freedom is speech is not carefully dealt with. There is no specifications bout how the law defines the defamation concerning commentaries and opinion.

According to explanation of the Highest People's Court in 1993, the court is the final gatekeeper on how to decide whether an article is defamatory and the crucial factor in consideration is whether the content of the article is "true". This standard is applicable to both online or offline media. The problem is that it is difficult and complicated to prove whether a comment or opinion is "true" or "untrue". A more fine tuned specification to follow is needed.

The Issue of Liability

US
Concerning the liability issue of online defamation, the U.S. congress has made a clear-cut definition in the late 1990s. After the case of Stratton Oakmont, Inc. vs. Prodigy Services Company, which reflected the importance of editorial control over content for any ISP and the possibility that the an ISP with well-intentioned monitoring policies might be sued, the U.S. Congress issued the Communication Decency Act (CDA) in 1996. The Section 230 in the CDA had granted the ISP immune in any lawsuits concerning online defamation. The underlying philosophy of Section 230 is to protect those ISPs that have the good faith to self-regulate and the digital freedom of speech. However, in the case of Zeran vs. America Online, Inc., we also see that there were complicated ethical issues involved. While ISPs are immunized from any liability of alleged defamation cases, there are arguable issues such as what if the ISP

UK
In the UK at Common Law, liability for the publication of defamatory material was strict. There was still publication even if the publisher was ignorant of the defamatory material within the document. Once publication was established the publisher was guilty of publishing the libel unless he could establish, and the onus was upon him, that he was an innocent disseminator. As regards the context of Internet, the liability of ISPs for hosting, caching or carrying defamatory material which they did not create depends on two conditions: (1) whether they are "publishers" of that material for the purposes of civil defamation law; and (2) whether they can establish a defence. There are three distinct sources of law for consideration in relation to the main defences available to them: the defence of innocent dissemination at the common law, the defense set out in section 1 of the Defamation Act 1996, and articles 12 to 15 of the Directive on Electronic Commerce.

Unlike the US, ISPs in the UK do not enjoy a broad statutory immunity from liability for defamatory material hosted, cached or carried by them. Because of the substantial differences between the approaches favored in different countries e.g. the UK and the US, the authorities must be treated with extreme caution because of the global nature of Internet.

HK
At common law, a defence of innocent dissemination is provided in section 25 of the DO. Like the other jurisdictions, a defence is available for a person who is not the author, printer or publisher of the defamatory statement. This defence has been made available to parties, such as postmen, posters who deliver the statement as part of their jobs, or booksellers, librarians and news stand owners who disribute materials without negligence and without knowing their contents.

Unlike the UK's Defamation Act 1996, the DO does not have a specific defence for a party, such as an ISP, who is involved in electronic publications. In Hong Kong, the only protection for ISPs for "unintentional defamation" is provided by section 25 of the DO. Secton 25(1) of the DO provides that a person who has published words that are alleged to be defamatory may make an offer of amends if the ISP claims that the words were innocently published.

In dealing with the issue of innocent dissemination, there is a question - whether the publisher has exercised all reasonable care, to be asked. It should not be reasonable to expect an ISP to review all the content of all the websites to which it provides access, so that issue is more open for the Hong Kong court to decide what constitutes "reasonable care" in any particular circumstances. And the decisions in other jurisdictions will be one relevant source to provide some guidance.

China
In China, the liability of online defamation is not specifically defined. Though the Article 15 in the "Administrative Measures on Internet Information Services" (PRC, 2000) adopted in 2000 states that ISP shall not "produce, duplicate, release or disseminate information that contains content that insults or defames others or infringes upon the lawful rights and interests of others" (PRC, 2000), it has not been clarified that whether the ISP is liable to the allegedly defamatory contents created by its clients. The awareness concerning this issue is not high enough in China and there are not many online defamation cases that can be served as examples. Many plaintiffs did not sue the ISP neither did they require the ISP to take remedial actions. For instance, in the case of MAX Computer Station, Inc. (MAX) vs. Wang, MAX only charged Wang but did not take any legal actions to the ISP that hosted Wang's website. In the case of owner representatives of Beautiful Garden vs. www.sinoi.com, it was demonstrated that the liability of the ISP was not clarified by any specific law but by the definition and explanation of the court. In the midst of speedy Internet growth in China, a more detailed guideline and regulation is needed.

The Issue of Jurisdiction

US
In the result of growth and globalization of Internet, there are more concerns on the increased possibility of U.S. media companies or ISPs being sued for defamation in foreign jurisdictions. This phenomenon is particularly troubling in U.S. because most foreign jurisdictions provide lesser protection to the ISP, or say, in the eye of American, lesser protection to the freedom of speech than the U.S. law. This makes libel claims substantially easier in many foreign jurisdictions than in the U.S.

The problem of foreign online defamation liability is more important for those U.S. media companies or ISPs which maintain substantial business and assets abroad. These companies may be forced to conform with the standards and laws of the foreign jurisdictions, even to some least speech-protective countries in the American's perspectives, such as China.

However, the U.S. courts have done the most effort to prevent such situations. In order to avoid foreign online defamation judgments that might contravene the spirit of the First Amendment, the U.S. courts will not enforce any foreign libel judgments inconsistent with the First Amendment. Though U.S. courts generally enforce foreign-money judgments under principles of comity, they are not required to do so where such judgments conflicts with U.S. law (Wimmer & Berman, 1999).

UK
The place of publication i.e. the place where material is read, heard or seen, is the very heart of the cause of action for defamation. The fact of publication within the jurisdiction of the court is also highly relevant. Matters include whether the plaintiff has a reputation to protect in the UK, whether the plaintiff's claim is limited to publications occurring in the UK, the extent to which publications occurred in other countries in which the plaintiff has a substantial reputation, and the location of the parties and witnesses.

It appears that where material is published by a foreign defendant to a global audience via the Internet, plaintiffs will generally be able to sue in the UK if the material is published in the UK and the harm is suffered in the UK, provided that there is evidence that the plaintiffs have reputations worthy of protection in the UK. The proceedings will likely be allowed to continue in the UK where either or both of the involving parties reside elsewhere other than the UK. The case of Bevezovsky v. Michaels is an example.

HK
The issue of the most appropriate forum for proceedings in relation to defamatory material on the Internet was also considered in Hong Kong in the case of Investasia v. Kodansha as detailed above. In a lack of detailed rules and regulations in operating the defamation law to the Internet, the ruling of the case and the considerations by the judge provide good guidance for all individuals and online publishers as regards to the forum issue. Like the UK, the fundamental considerations are the place of publication and the existence of reputation. So, if the defamatory material is published and the plaintiff have a reputation in Hong Kong to protect, Hong Kong can be the most appropriate forum for the plaintiff to bring a defamation action.

China
In the Civil Procedure Law of the People's Republic Of China adopted on April 9, 1991 at the Seventh National People's Congress (PRC, 1991), Article 238 in the part four "Special Stipulations for Civil Procedures Involving Foreign Interests" states that if an international treaty concluded or acceded to by the People's Republic of China contains provisions differing from those found in this Law, the provisions of the international treaty shall apply, unless the provisions are the ones on which China has announced reservations.

In Article 239, it states that any civil lawsuits brought against a foreign national, a foreign organization or an international organization that enjoys diplomatic privileges and immunities shall be dealt with in accordance with the relevant laws of the People's Republic of China and with the international treaties concluded or acceded to by the People's Republic of China.

It is clear that though China shall follow international treaty generally, the Chinese court reserves the final decision on whether the case judgments involving foreign parties are applied to the jurisdiction of China. Those treaties or jurisdictions that the Chinese government does not accede or contravene with China's own policy will not be applied in China. Since the essence and philosophy of Chinese law is very different from many foreign countries such as the U.S. and U.K, it can be anticipated that complications and arguments will occur in some cross-border cases especially those involving controversial and sensitive issues such as human rights or freedom of speech, which are always in relation to defamation cases.

Future

With a rapidly growing number of computer-communication users, the likelihood of an increase of cyberlibel in the future is inevitable. The only question is we can tell whether the number of litigations will grow steadily or there will be an explosion of lawsuits all at once.

To date, the laws of defamation and the approach in operating the laws are somewhat different from state to state and country to country. Ideally, the establishment of an "international treaty" among nations in the world, accompanied by greater harmonization of the defamation laws will be the solutions to the varies complex and arguable issues such as jurisdictions, choice of laws, forum shopping etc. However, we believe that the likelihood of the establishment of an "international treaty" is impossible in the near future or even such a dream will never come true.

What we wish to see in the coming future is more clear and defined rules and regulations governing the Internet communications among the industry players and individual users, and a more Internet-related version of the defamation laws.

I believe majority of the Internet users may not be aware that they are "publishing" a defamatory statement when they post a message on a bulletin board criticizing someone or corporations. Similarly, they may not anticipate that their liability will be extended to other countries while the message be received and read by someone overseas. Also, I believe majority of the employers may not be aware that they can become liable for an employee's defamatory statements if the employer is responsible for publishing the material. That is particularly true for internal e-mail bulletin board networks where the employer typically owns and manages the entire network. Therefore, the on-going legal debate about the liability of ISPs for the action of their customers can apply to the employer/employee relationship. The question is when this happen to the attention of the employers and the public.

Conclusions

In the last decade, Internet is a booming new communication medium with enormous growth of worldwide applications. In this free online world, people can express their opinion easily, exchange information speedily, and there are no borders or limitations. However, all this freedom does not mean that users can arbitrarily publish materials that harm the reputations of others. Though there is not yet any global treaties or policies that regulation online defamation, there are laws in different countries that prohibit online defamation just as "offline" defamations. The laws of defamation work the same way for each communication medium and every defamatory material on the Internet is governed by the same principles from the law of tort (Canham, 2002).

However, online defamation is far more complicated than offline defamation. One of the major concerns is the issue of anonymity on the Internet. Investigations and tracing need more time and efforts and involve a number of parties such as ISPs. It is also true that in the world of Internet, there are too many channels and outlets that users can violate the law accidentally or intentionally. Simple as sending an email, a user must ensure that what was written or attached was not offensive or defamatory before he or she hits the "send" or "forward" button.

Since the Internet has only developed for around a decade, specific laws concerning online defamation were not made until recently. The same situation is also true in the public awareness on this matter. The awareness of online defamation is high while it is relatively low in other countries. As a result, some countries have specific laws and regulations concerning online defamation while some countries do not and only apply traditional laws to online defamation in the same way as offline defamation. For instance, In the U.S., there are already many cases of online defamation lawsuits and laws that clarify related issues such as liability of ISPs. In China, there are also unexpectedly many cases which reflected the enormous growth of Internet usage. However, the related law in China is incomplete and some controversial issues such as ISPs' liability are not yet covered. In Hong Kong, though Internet development is also blooming, there are not as many cases and references as in China. The public and legal awareness should be enhanced in order to tackle with the possibly increasing cases in the future.

The conceptual framework and philosophy of different legal systems can be very different. As a result, another question arouse due to the borderless nature of Internet communications: the multi-jurisdiction issue. Until a global treaty is created and accepted by most countries, site owners and ISPs should remain cautious that they did not violate the laws of the foreign jurisdiction that their business might have been involved.

After all, the most effective measure to avoid being sued of online defamation is responsible self regulation. As an ordinary Internet user, we should "think twice" before writing about something controversial and remember that it is not totally order-less nor anarchic in the realm of the Internet. As an ISP, we should also take all reasonable efforts to avoid posting or disseminating any defamatory material and ensure that we respond quickly to complaints made about allegedly defamatory materials that appear on the websites we host.

Bibliography

Beijing Morning Post (1999). The first Online Reputation Lawsuit Reaches the Highest Record of Compensation (§Ú°ê­º¨Òºô¤W¦WÅAÅvªÈ¯É½ßÀv³Ð¬ö¿ý).

Canham, J. (2002). Defamation on the internet. CMS Cameron McKenna.

Carter, T.B. (1999). The First Amendment and the Fifth Estate: Regulation of Electronic Mass Media.

Daynews.com (2002). Haier Goes for a Reputation Lawsuit (®üº¸­n¥´¦WÅAÅv©x¥q).

Deturbide, M. (2000). Liability of Internet Service Providers for Defamation in the US and Britain: Same Competing Interests, Different Responses () PohLee Tan. 2001. E-com Legal Guide Hong Kong. (http://www.bakerinfo.com/apec/hongkong_main.htm)

DFN (Digital Freedom Network) (2000). China regulates online news and chats. Digital Freedom Network.

Ferrera, G.R. (2001). Cyberlaw: Text and Cases. (Ch.10, Defamation.)

Flint, D. (1994). The US Too Wants Defamation Reform. Australian Press Council.

Fung, J.P. (2001). The liability for tort of Internet Service Manager (½×ºô¸ôºÞ²zªÌªº«IÅv³d¥ô).

Fusco, P. (isp-planet.com) (2000). The English Are Different.

Gao, Y. (2002). Gao Yun's Internet Law Commentary Series (°ª¶³ºô¸ôªk«ßµû½×¨t¦C).

Her Majesty's Stationery Office (1996). Defamation Act 1996.

HKLII (1999). Investasia Ltd. and Another v. Kodansha Co. Ltd and Another

NCCUSL (National Conference of Commissioners on Uniform State Laws) 1993. Uniform Correction or Clarification of Defamation Act.

People's Daily. (2002). Internet Brings Sweeping Changes to China.

People's Daily. (2000). China to Regulate BBS. (Full Chinese version of the regulation: http://www.itrus.com.cn/repository/policy009.asp)

PRC (People's Republic of China). (1982). Constitution of the People's Republic of China.

PRC (People's Republic of China). (1987). General Principles of the Civil Law of the People's Republic of China.

PRC (People's Republic of China). (1991). Civil Procedure Law of the People's Republic of China.

PRC (People's Republic of China). (2000). Administrative Measures on Internet Information Services.

Richardson, T. (2002). ISPs welcome UK Net libel review.

Roberts, E. (2001). Information on relevant libel/defamation law.

Sun, S.C. & Ng, Q.P. (2002). Why Can this Man make Haier silent (³o­Ó¤H¬°¤°»ò¯àÅý®üº¸¨HÀq)?

Weaver, R. L. (2000). Defamation Law in Turmoil: The Challenges Presented by the Internet.

Wimmer K.A. & Berman, J.A. (1999). United States Jurisdiction to Enforce Foreign Internet Libel Judgments. Covington & Burling.

Wright C. (2003). Internet Law in Hong Kong.

Xinhuanet (Xin Hua News). (2002). Let the Internet be Free First (Åýºô¸ô¥ý¦Û¥Ñ°_¨Ó).