Service providers and liability for digital defamation: Finding the right balance

Defamation suits have always been an occupational hazard for information providers. There was no serious problem when information providers were mainly professionals and big organizations with deep pockets that could take necessary steps to guard against liability or pay up when it came to it. Today, however, with the withering of the Fourth estate and the progressive dismantling of the wall between the journalist and the ordinary citizen, liability for defamatory material published digitally is considerably more nebulous than it is for similar material published in traditional media.

The veritable explosion in the amount of information exchanged and business transacted electronically today and the very important role played by Internet Service Providers (ISPs) and other intermediaries has exposed them to potential legal liabilities that lurk within the ambiguities and uncertainties of the law that applies to digital defamation. The need to define the parameters of liability of these service providers for defamatory messages transmitted through them is crucial if the advances in ICTs are to be exploited to their fullest in the vast, largely untapped global digital marketplace, which is gradually emerging today.

Traditional defamation and the digital environment

Generally, defamation can be defined as the publication of any false statement, which reflects on a person’s reputation and tends to lower him in the estimation of right-thinking members of society generally or tends to make them shun or avoid him (Rogers, 1998: 391). As Prosser points out, defamation is that which tends to injure reputation in the popular sense; to diminish the esteem, respect, good will or confidence in which the plaintiff is held, or to excite adverse, derogatory, or unpleasant feelings or opinions against him (Keeton et al., 1984: 773). This necessarily involves the idea of disgrace. It is a wrong that can be committed not only by words, written or spoken, but also by acts or gestures.

To succeed in an action for defamation, the plaintiff must prove that: the publication was defamatory of him; that it referred to him; and that it was published. On the whole, the law of defamation attempts to reconcile and protect two of the most fundamental and highly valued but often conflicting interests in modern society - the individual’s right to an unimpaired reputation on the one hand, and the free flow of information in society on the other hand (Fleming, 1987: 500). It is an antithesis that is particularly acute in a digital environment.

As regards the first requirement, the plaintiff must set out and plead the actual words or other means used to convey the defamation and show that the substance, purpose or effect of it, lowers his reputation in the estimation of right-thinking persons in society generally or tends to cause him to be shunned or avoided.

The second requirement entails clear proof that the words were published ‘of and concerning‘ the plaintiff. The words must be such that a reasonable person would identify the plaintiff as the person defamed. Where the defamatory statement is made against a class or group, for example members of the medical council, only a member of such a class or group who can show that the circumstances were such that any reasonable person acquainted with him was likely to understand the statement as referring to him as an individual, can successfully sue.1

Finally, there must be proof that the defamatory material was published. Publication is the communication of the defamatory material to at least one person other that the person defamed. In many cases, publication will be presumed without the plaintiff actually proving this, for instance, where the matter is contained in a book or a newspaper. Similarly, the contents of a post card which has been posted, or a telegram which has been dispatched, will be presumed to have been published.2 Generally, everyone who takes an active part in producing or publishing defamatory matter is prima facie liable.

For a newspaper article, this includes not only the writer of the article, but also the editor, printer, publisher and proprietor, all of whom are presumed to have been acquainted with the defamatory material. Mere distributors, such as booksellers, vendors, and other persons who play only a subordinate part in disseminating the material, although also prima facie liable, will have a good defence if they show that they were ignorant of the fact that it contained defamatory material and that this lack of knowledge was not due to their negligence, and also that, there were no circumstances which should have put them on their guard or inquiry.

The distinction between a publisher and a distributor is important because one who republishes defamatory material is as liable as if he had published it originally, whereas one who distributes defamatory material is liable only if he knew or should have known that the material was defamatory. Every repetition of defamatory matter amounts to fresh publication and creates a fresh cause of action.

The requirement of publication in the digital environment is problematic. Because of their global reach, the Internet and other computer networks offer prime opportunities for publishing digital content on a world-wide scale. Publication in this sense includes both the publishing of information that is intended to be read or viewed online and electronically distributing digital content, such as software, sound recordings, movies, photographs and databases for use offline (Smedinghoff, 1996:9). The main problem here is that of determining who, among the various intermediaries through whom the digital information was communicated, would be liable on the grounds of having published it.

Holding the service provider to account

The author of a defamatory message transmitted digitally is obviously the prime target for an action for defamation, but there may be many reasons why the plaintiff may not want to go after him. First, there are many technical devices that the author of a defamatory message may use today in order to disguise or hide his true identity, thus making identification impossible. Second, since such digital messages cross international borders, the author may be so far from the jurisdiction that it might be practically impossible to sue him. Third, the service provider who intervened in the chain of transmission of the message usually has deeper pockets than the author and would therefore make a more enticing defendant, especially since he will also be easily identifiable and within the jurisdiction.

However, if all such service providers and other intermediaries involved in the digital environment were exposed to digital defamation merely because they acted as conduits for the transmission of digital information, this could become a time-bomb that might seriously undermine the commercial growth and viability of this medium of communication. Yet, this is exactly what a strict application of the traditional principles of liability for defamation would lead to. There have been attempts to avoid this rigid translocation of traditional defamation principles in this way.

The US approach

In the United States, many commentators sought to extend existing legal principles to this area by making ISPs and other intermediaries analogous with others such as newspaper publishers and editors, libraries and booksellers, broadcast media and private real property owners (Schlauchter, 1994: 474). Two cases in the early 1990s settled the matter by analogy to existing principles of the law of defamation.

In the first, Cubby Inc.v Compuserve Inc.,3 CompuServe were sued in respect of a message appearing in a local forum hosted by them, called ‘Rumorville USA’. CompuServe had employed a third party specifically to edit and control the content of this forum. The third party posted the information on the Internet once it was edited, with no intervening opportunity for CompuServe to review the material prior to publication. CompuServe argued that they were merely a distributor of information, not a publisher, and should therefore not be held liable. The New York District Court agreed, holding that CompuServe had been acting in a way akin to a news-stand, bookstore or public library, and that to hold it to a higher standard of liability than these distributors would place undue restrictions on the free flow of electronic information.

But in Stratton Oakmont Inc. v Prodigy Services Co,4 on similar facts, Prodigy was sued in respect of comments posted by an unidentified person to a local discussion forum it hosted, the opposite decision was arrived at. The court based its findings that Prodigy exercised sufficient editorial control for it to be considered a publisher on the following:

1. Prodigy required its users to adhere to content guidelines that prohibited the posting of messages that were insulting, harassing, repugnant to community standards or otherwise harmful to maintaining a harmonious online community.

2. Prodigy employed Board Leaders to enforce the guidelines and provided the Board Leaders with an ‘emergency delete function’ that enabled them to remove offensive messages.

3. Prodigy used a software screening programme that pre-screened bulletin board messages for offensive language.

4. Prodigy held itself out to the public as a family computer network, a claim made possible by the use of the above controls.

The court distinguished the case from CompuServe where it was held that CompuServe was not liable because it exercised little or no editorial control over the contents of its publications.

The effect of both cases was to leave ISPs and others in their position in dire straits. To avoid liability and be classified as a publisher, they had to do as little as possible to monitor and edit the content of the messages or other materials that they carry. Nevertheless, the two cases, particularly the decision in Prodigy, was widely criticized since it exposed the providers of such computer services to potentially wide-ranging liability.

In response to these criticisms the United States Congress passed the so-called ‘Good Samaritan’ provision in the Communications Decency Act of 1996 (CDA). The overriding concern of the CDA, as reflected by its long title, was decency on the information superhighway. The Act extends the standards of decency that protected users of other telecommunications systems to the Internet. Section 230, which provides protection to ISPs that take active steps to exercise some editorial control over user postings, counters the ruling in Prodigy. The underlying philosophy of the section is that ‘Good Samaritans’ who undertake editorial duties to remove offensive material should not be penalized for their efforts by being treated as publishers. Section 230(b)(4) of the CDA makes it clear that the objective is ‘to remove disincentives for the development and utilization of blocking and filtering technologies that empower parents to restrict their children’s access to objectionable or inappropriate material’.

According to the CDA, no provider or user of an interactive computer service will be treated as the publisher or speaker of information provided by someone else nor can any provider or user of an interactive computer service be held liable for voluntarily restricting access to or availability of objectionable material, or for making available the technical means to restrict access to such material

In the first case decided under the CDA, Zeran v American Online Inc,.5 an unidentified person posted a message on a bulletin board hosted by the ISP, America Online (AOL) advertising the sale of T-shirts that depicted tasteless and offensive slogans relating to the bombing of a federal building in Oklahoma City. Interested people were instructed to call the plaintiff’s home phone number. He was bombarded with angry and threatening calls, including some death threats. After several protests from him, AOL eventually removed the postings but, as a matter of policy, refused to print a retraction. He sued AOL relying on the traditional grounds that as distributors they were under a duty promptly to remove the postings the moment they knew or had reason to know that it contained defamatory material and they had failed to do so.

The court held that Section 230 of the CDA saved AOL. The court in essence concluded that the immunity of ISPs was the same whether the case was framed as one of publisher liability or distributor liability and it also refused to place any weight on the fact that AOL had notice of the messages and reacted rather tardily. Some critics of this case have argued that the court expanded the interpretation of Section 230 of the CDA beyond what the Congress had intended (Deturbide, 2000). However, all post-1996 cases have followed Zeran in interpreting the CDA as granting United States service providers complete immunity for actions for defamation, whether or not, they undertook any editorial functions 6

The UK approach

In Britain, where there are signs that digital defamation is becoming a serious issue, the approach adopted has been markedly different. The first serious move came with the enactment of the Defamation Act 1996, having as one of its explicit aims the clarification of the defence of innocent dissemination for Internet providers. To some commentators, the Act is considered merely to have fine-tuned rather than actually overhauled the long-standing principles of English law of defamation (Vick and Macpherson, 1997). The Act attempts to take account of the complex difficulties posed by the emergence of ICTs by modernizing the defence of innocent dissemination. In this regard, Section 1(1) provides:

‘In defamation proceedings a person has a defence if he shows that – (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 contributed to the publication of a defamatory statement.’

Stipulations (a), (b) and (c) represent three requirements which a defendant must satisfy to be able to rely on this defence. However, the Act gives rise to almost as many problems as it attempts to resolve. First, read as a whole, the Act places service providers in a Catch 22 situation. If an ISP monitors or edits content, it is likely to be regarded as an ‘editor’ but to get the benefit of the defence, the ISP must only provide Internet access, and not do anything else - not, for example, exercise editorial control or spot-check content. For if they do, it would seem they will be exercising ‘effective control’ over the maker of the defamatory statement (Edwards, 1998). But then, it is unlikely that an ISP which neither monitors nor edits can succeed in proving, as Section 1(1) requires, that it took ‘reasonable care’ to prevent the publication of the defamatory statement.

Second, the Act does not indicate under what circumstances those who make defamatory statements over computer-based communications systems will be considered as being beyond the ‘effective control’ of the operators, a difficult question since it is at least theoretically possible to screen the messages and the users of such systems (Vick and Macpherson, 1997). However, a defendant who cannot rely on Section 1(1) can opt for Sections 2 and 3, which contain a simplified and expanded offer of amends procedure, which enables any person who has published an allegedly defamatory statement to make an offer of amends. If such an offer is accepted, further proceedings against the defendant are barred.

In 1997, the United Kingdom’s first defamation action brought by two companies over defamatory electronic mail was settled out of court. Godfrey v Demon Internet Ltd,7 is the first case brought under the 1996 Defamation Act. Although it was ultimately settled out of court, the judgement on earlier motions filed by the parties does provide some useful insights on the interpretation of the Act. In the case itself, an unknown person posted a message to a newsgroup purporting that it originated from the plaintiff. The posting was described by the court as ‘squalid, obscene and defamatory.’ It invited replies by giving the plaintiff’s e-mail address. The defendant, who hosted the newsgroup, ignored the plaintiff’s requests that the posting be removed. When the plaintiff sued for defamation, the defendant’s defence was that it was not at common law a ‘publisher’ of the allegedly defamatory posting and that even if it was, the defence of Section 1 of the Defamation Act was available to it.

In considering the plaintiff’s application to have the defendant’s defence pleadings struck out as disclosing no sustainable defence, the court held that the defendant could not rely on the Act. This was because the defendant knew of the posting, but did not remove it, and consequently, did not take reasonable care and had every reason to believe that it contributed to the publication of the defamatory statement. Therefore, an ISP cannot rely on the Defamation Act if it was given notice of defamatory postings by the complainant.

The Demon case exposes the uncertainty caused by the fact that the innocent dissemination defence disappears once the ISP receives ‘notice’ that a posting contains defamatory matter. The Act fails to state clearly what actually constitutes such notice and because of this, ‘notice and takedown’ may become too regular a feature amongst ISPs anxious to avoid costly litigation. This procedure can also easily be abused, for instance, by multi-national companies keen to silence public criticism of their activities or products, or even fraudsters interested in making fast money by blackmailing some of the small and financially strapped ISPs (Akdeniz, 1999).

Perhaps the most serious effects of the Demon settlement is that within days of it, several cautious UK companies and individuals moved their sites outside the jurisdiction, especially to the United States servers, whilst others closed down their sites. The settlement also ushered in a new and unprecedented era of self-censorship on the Internet in the United Kingdom.8

Most ISPs routinely require their customers to agree not to make defamatory postings. The attempt to spread the net of those who could be sued beyond the original author of the defamatory material and the ISP or host of the site to others in the United Kingdom shows the gravity of the problems posed by digital defamation in the country.

In 1997, a web site known as ‘Tartan Army’ posted information about the Scotland football team and was sponsored by the brewers, Scottish Courage. Fans were allowed to post directly their views about the game. However, one of the main topics that the fans concentrated on was a discussion about Jimmy Hill, a sports broadcaster about whom various obscene, rude and defamatory comments and jokes were posted in a variety of languages. Jimmy Hill decided to sue not the web site owners themselves, nor their ISP, but Scottish Courage, the sponsor. The latter immediately withdrew its sponsorship of the site. As Lilian Edwards (1998) observes, ’this is a worrying precedent for other sponsors and advertisers on the web, who have next to no control over what is displayed in proximity to their name, and might impede commercial exploitation of the web’ It is therefore not surprising that because of the uncertainties resulting from the Defamation Act 1996, and the scale of the risk of actions for digital defamation, more cases in the United Kingdom are settled out of court or by apology than through litigation (Edwards, 1998).

The position in other jurisdictions

Not many countries have followed the lead of the United States and United Kingdom in enacting specific legislation to deal with digital defamation. A quick glimpse at what obtains in other jurisdictions is appropriate here.

In Australia, the National Conference of Commissioners on Uniform State Laws in February 1992 rejected a draft of the Uniform Defamation Act, which suggested that information retrieval and transmission services, including computer bulletin boards, should be subject to liability as republishers (Nguyen, 1997). Two years later, in a decision that has been widely criticized for its failure to take account of the nature of the Internet, the Supreme Court of Western Australia in Rindos v Hardwick,9 awarded the plaintiff the huge sum of $40,000 in damages and interest without distinguishing between defamation on a computer bulletin board and the more traditional media.

In Japan, the issue of defamation in the digital environment has been left to be regulated by ethical guidelines drawn up by computer network operators (Holland and Meyer, 1996). In Singapore, their fairly strict defamation laws have been extended to cover digital defamation. In Africa, hardly any country, including South Africa with its relatively advanced IT sector, has yet considered any specific legislation in this area.

Conclusions

The question of allocation of responsibility for defamatory matter transmitted digitally is of fundamental importance if the full potential of this medium is to be exploited for enriching learning and promoting global economic growth. A single e-mail message can irreparably ruin a person’s career on a world-wide scale or destroy the business reputation of a multinational company. Should the victim be allowed to rely on the manifestly attractive traditional principle that he can sue any of those involved in the chain of transmission of such a defamatory message? So far, only the United States and the United Kingdom have attempted to enact legislation that purport to address this problem taking into account the peculiarities of the digital environment and balancing the competing interests at stake. Neither attempt provides an entirely satisfactory solution to the matter.

The United States starts from the premise that freedom of speech is sacrosanct and should not be tampered with and in Section 230 of the CDA 1996 creates a fairly permissive liability regime which arguably offers too much protection to the ISPs and little incentive for them to be careful in controlling the transmission of defamatory material. This regime of complete immunity is in sharp contrast to the more cautious approach taken by the United Kingdom legislature in the Defamation Act 1996. This Act adjusts and extends the traditional common law defence of innocent dissemination to ISPs, though its circumlocutory wording leaves the exact extent of protection thus provided fairly vague and uncertain.

Making it too easy for victims of defamatory messages transmitted digitally to sue any of those involved in the chain of transmission such as the ISPs exposes them to potentially open-ended liability because in most cases, they will make more attractive defendants than the sender of the message. This has several drawbacks. It could lead to unnecessary auto-censorship and transform ISPs into defamation police, patrolling the digital environment to clean up at the expense of free exchange of information what they perceive to be potentially defamatory. The mere thought of such censorship will certainly having a chilling effect on freedom of expression. It will also make liability insurance prohibitive and expose ISPs to extortion.

On the other hand, complete immunity takes away the incentive that the ISPs may have to edit, review and limit some of the highly offensive material that is transmitted through them. This is also likely to have a negative impact on the researches into software that could search and block offensive material. There is also a chance that countries with less restrictive laws that grant the ISPs unqualified immunity could become ‘flags of convenience’ in the digital environment.

The main challenge today is that of enacting legislation that carefully and equitably balances the competing interests as well as recognizes the peculiarities of the global digital environment. It is now clear that merely extending existing laws to digital defamation is unlikely to be a satisfactory solution.

Notes

1. Neuemann CC v Beauty Without Cruelty International 1986 (4) 676 (c); Sauls v Henrickse 1992 (3) SA 912 (A).
2.
3. Patrick Motlhwa v The State Crim. App.175/82(Botswana). (Unreported).
4.
5. 766F Supp 125 ( SD NY 1991).
6.
7. 1995 WL 323710 (NY Sup Ct 1995).
8.
9. 958F Supp 1124 (ED Va 1997), aff’d 229F 3d 327 ( 4th Cir 1997).
10.
11. See the following recent cases, Blumenthal v Drudge 992F Supp 44 (DDC 1998); Doe v America Online Inc. 718 So 2d 285 (Fla.Dist. Ct. App., 1998); Ben Ezra , Weinstein and Co.Inc. v AOL 27 Med. L. Reptr. 1794 ( DNM., 1999) aff’d 2000 US App. Lexis 383 ( 10th Circ.); Marianne Bihari and Bihari Interiors Inc. v Craig Ross and Yolanda Truglio 2000 US Dist. Lexis 14810 ( SD. NY., 2000); William Sheehan III v King County No.C97 - 1360 WD (WD. Wash. 1998).
12.
13. 1999 EWJ No. 1226 (QBD).
14.
15. For example, Jim Hulbert’s website which criticized some judges and the Lord Chancellor’s department was taken down following a ‘notice’ it received. The website of the campaign against censorship of the Internet in Britain was removed from its British ISP’s server and re-allocated to a server in the US. See generally, Alan Docherty, ‘ISPs May Bear Content Liability.’ Http://www.wired.com/news/political/0,1283,17953,00 html; and Http//www.guardianunlimited.co.uk/freespeech/artcle
16.
17. No. 940164 ( St.Ct. W.Austl., 1994).
18.

References

Akdeniz, Y. (1999). ’Case Analysis of Laurence Godfrey v Demon Internet Limited’, Journal of Civil Liberties Vol. 4 No.2 pp, 260-267.

Deturbide Michael, (2000), ‘Liability of Internet Service Providers for Defamation in the US and Britain: Same Competing Interests, Different Responses.’ Journal of Law and Information Vol.3; Http://elj.warwick.ac.uk/jilt/00-3/deturbie.html.

Edwards Lilian (1998). ‘Defamation and the Internet: Name Calling inCyberspace.’ Http://www.law.ed.ac.uk/c10_main.html

Fleming, J.G. ( 1987). The Law of Torts 7th ed., Law Book Company Ltd: Melbourne.

Holland Michael and Mayer (1996). ‘Libel on the Internet: AnInternational Problem.’ Http://wings.buffalo.edu/complaw/Comlawpapers/Holland.html

Keeton Page et al. (1984). Prosser and Keeton on the Law of Torts 5th ed. West Publishing: St. Paul.

Nguyen Chi (1997). ‘Defamation on the Internet.’ Http://www.gsulaw.gsu.edu.lawand/papers/sp97/nguyen.htm

Rogers, W.V.H. (1998). Winfield and Jolowicz on Tort 15th ed. Sweet & Maxwell: London.

Schlauchter Eric (1994). ‘Cyberspace, the Free Market and Free Marketplace of Ideas: Recognizing Legal Differences inComputer Bulletin Board Functions.’Hastings Comm/Ent. L.J. Vol.87, pp. 474-485.

Smedinghoff, Thomas J. (ed.) (1996). Online Law. The SPA’s Legal Guide to Doing Business on the Internet. Addison-Wesley Developers Press, Reading, Mass., p. 337.

Vick Douglas, W. and Macpherson Linda (1997). ’An Opportunity Lost: The United Kingdom’s Failed Reform of Defamation Law.’ Federal Communications Law Journal Vol.49. Http://www.law. Indiana.edu/fcli/pubs/v49/no3/v49no3html

Charles Manga Fombad (PhD) is Senior Lecturer in Law at the University of Botswana. He was Senior Lecturer in Law at the University of Yaoundé 2, Cameroon, 1988-97 in which country he was also Visiting Professor of Law at the universities of Buea and Dschang. His main areas of research include media laws, practices and policies, constitutional developments and good governance.