Copyright © 1995 Francis Auburn. First published in Web Journal of Current Legal Issues in association with Blackstone Press Ltd.
*The author would like to thank Adrian Chua for research assistance and the Law School of the University of Western Australia for support.
In Rindos v Hardwick (1994) the Western Australia Supreme Court awarded damages for defamation arising out of an item in a newsgroup published on the Internet. This comment examines the decision and considers its wider implications both for users of the Internet and service providers on the Internet.
A particular concern is that of defamation. Here the example of Usenet News will be referred to, although the problem relates also to other parts of the Internet. Usenet News is run on thousands of computers world wide. There are more than 4,500 news groups. Each newsgroup is a forum in which one can read messages and post about a particular area of interest. Anyone can usually access them and the messages posted are automatically transmitted around the world. Topics range from the professional (such as a large number of groups on computing) through specific interests (related to named countries and culture) through the bizarre (such as a group devoted to the latest sightings of Elvis) and no less than sixty groups on various sex matters (Song-Muh 1994, pp 2-4). The potential for defamation is obviously high. An examination of a range of even the more staid topics reveals that there is much personal abuse, attacks on another’s character and other matters which could well be defamatory.
On 23 June 1993 a message appeared on a world-wide computer bulletin board, inserted by Hugh Jarvis, a United States anthropologist. Although not stated in the judgment, the bulletin board was the ANTHRO-L mailing list, which has participants in the United States, Australia and elsewhere. The message stated that Rindos was "a widely known and respected scholar" and that many recipients of the list were probably aware of Rindos’ contribution to the field of anthropology (Jarvis 1993) and went on to criticise the University. The Jarvis posting asked "what actions can the international community take...Already over thirty international scholars have written on (Rindos’) behalf". The defendant, on 26 June 1993, posted a response to Jarvis on the newsgroup "sci.anthropology". He alleged that Jarvis had denied him access to ANTHRO-L and proceeded to launch an attack on Rindos. The court accepted that words in the message seriously denigrated Rindos’ academic competence and imputed misconduct, both of which were held to be seriously defamatory. "These defamatory remarks were published in academic circles throughout the world". The court also found defamation regarding competence in a copy of a letter sent to a professional association. Ipp J found that Rindos was "well known internationally in academic anthropological and archaeological circles and that he is a person of high standing in those circles" and his standing would suffer a most harmful effect from the defamatory remarks (ibid., pp 8-9). $40,000 was awarded in damages.
Ipp J's belief that the newsgroup "sci.anthropology" to which the remarks were posted was a bulletin board is significant in two ways. Firstly, it is relevant to the question of whether the remark is defamatory. A critical element of defamation is that the words must tend to lower the plaintiff in the estimation of citizens of "fair average intelligence" (Slatyer v Daily Telegraph (1908) 6 CLR 1 at 7 per Griffith CJ). The context in which the words are published is directly relevant to this test. For example, if the expression is made in the context of a harmless joke and is so understood, its defamatory barb may thereby disappear (Fleming 1992, p 527). If the defendant's comments were made in the context of a bulletin board within the understanding of Ipp J, there would be little doubt that the test for defamation would be satisfied.
Unlike the bulletin boards described by Ipp J, messages posted on newsgroups are informal and more akin to conversation than to a conventional letter. For instance, the defamatory posting commenced with "Well, here we have my old mate Hugh Jarvis..."(Hardwick 1993). There is no difficulty in finding strongly worded messages on Usenet. Two randomly chosen recent examples should suffice. A recent posting to one newsgroup included the following: "You still wanna talk about Kennedy right wing scumbag? There it is again! Bigot alert! But you are a scum bag, (person’s name)! You are a bigot...". On another a posting included: "Ha! I find it hysterical when sexist pathetic losers like (person’s name) call women names like "slut" and "whore"". It must be stressed that such highly informal comments are commonplace on Usenet newsgroups. Often one gains the impression that the writer’s reaction is immediate and similar to speech. "They think they are having a private conversation but it is like they are using a megaphone" (Braithwaite 1994). In such circumstances the potential for defamation is high, taking account of the vast numbers of messages sent each day to newsgroups, mailing lists, as email with copies to additional recipients and also noting the very large amount of materials put up on the elements of the Internet. In particular, newsgroups which do not have any intervention by a systems operator to monitor content are particularly open to such problems. Thus the reader of "fair average intelligence" is likely to treat messages on newsgroups less seriously than the messages posted on ANTHRO-L. Hence the question of whether the words, made in this context, are such as to tend to lower the plaintiff in the estimation of citizens of fair average intelligence is less easily answered.
Secondly, Ipp J's interpretation is significant in the assessment of damages for defamation. In assessing damages, the law seeks to compensate the plaintiff for injury to the reputation he previously enjoyed (Fleming 1992, p 597). Thus one factor relevant to the assessment of damages is the gravity of the defamation. This in turn would be determined by the consequences of the attitude adopted towards the plaintiff by others as a result of the diminution of the esteem in which they hold him because of the defamatory statement (McCarey v Associated Newspapers Limited (No 2) [1965] 2 QB 86 at 107, adopted by Mason CJ and Deane J in Coyne v. Citizen Finance Limited (1991) 172 CLR 211). Therefore, the fact that Ipp J did not analyse the informal and unconventional nature of the forum in which the defamatory comments were made in detail means that the quantum of damages awarded might have been larger than it should be since the readers of newsgroup messages are likely to treat the remarks less seriously than if the message was posted in ANTHRO-L. Hence the injury to reputation might have been overestimated.
A further problem of the decision is the estimate of the number of people with access to the "bulletin board". The court stated that "there are approximately 23,000 persons world-wide whose computers have access to the bulletin board in question" (Rindos v. Hardwick, supra p 4). The judgment does not state how this figure was arrived at. With a mailing list one can enter the "review’ command and at least obtain a list of current members. However this does not apply to a Usenet newsgroup such as "sci.anthropology". One might estimate that the professional nature of the newsgroup would restrict the number of persons using it to several hundred. One might examine the current messages, make allowances for multiple postings by one person, and then arrive at the relatively low number of 200-300 for persons making postings. However Hardwick’s posting referred to "professors, lecturers, staff, students, professionals and sundry lurkers and lookers-on" (ibid at p 6). This points out that many people may access a newsgroup without making a posting. However the court’s reference was to persons whose computers have access to the bulletin board. Prima facie this would appear to mean anyone whose computer can access "sci.anthropology". Since the newsgroup is part of Usenet and Network News, anyone who has access to this popular general system has potential access to the newsgroup. Counted in this way potential access is in the hundreds of thousands, at a minimum, and probably in the millions.
Furthermore, one cannot restrict the examination to direct users of "sci.anthropology". It is likely that material related to the Rindos tenure dispute was posted on other elements of the Internet. It appeared in ARCH-L and at least one other mailing list. The inter-connections of the Internet mean that Hardwick’s message on "sci.anthroplogy" might have been accessible through other parts of the Internet such as Veronica, Galaxy, World Wide Web Search Engines and Gopher. Again it is difficult to estimate the extent to which these and other elements of the Internet were used for such access, although the extensive potential use is clear.
The problems in estimating the number of persons who might have read the defamatory remarks is relevant to the assessment of damages. The function of damages is to compensate the plaintiff for his loss of reputation. Hence the extent to which he is held in less esteem is relevant (Associated Newspapers v Dingle [1964] AC 371) and is dependent on how wide-spread the defamatory remarks have been published.
Finally, the fact that Rindos is a judgment of a single judge of the Supreme Court where no defence was pleaded affects its significance. However, Rindos must be seen as of particular importance in marking a successful defamation action in the private arena. Having regard to the enormous potential for such actions, one may well expect further cases. Current developments support this view. Recently a well-known commentator, Brock Meeks, was sued by the Suarez Corporation for allegedly defaming its software package (Sandberg 1994). The case was settled on terms (Meeks 1994). In 1994 Dr. Laurence Godfrey of CERN, Geneva, Switzerland, took proceedings in England against Dr. Phillip Hallam-Baker for alleged defamation on Usenet (Irwin and McGuckin 1994). A student at the University of Western Australia was allegedly faced with a demand for an apology for an item posted on a Usenet newsgroup (van Niekerk 1994). It would thus appear that Rindos could well be the start of a flood of litigation, particularly against persons posting material to Usenet newsgroups.
However, further issues may be raised which did not arise in Rindos.
The liability of these service providers for the distribution of defamatory remarks would discourage the provision of Usenet newsgroups and other Internet services. These adverse consequences on an efficient means of communication may lead the court to apply the test more liberally, thereby allowing such providers to be classified as innocent distributors. Another solution is for these administrators to screen items posted to the extent required to satisfy the standard of care so that they will no longer have reason to suspect that they are distributing defamatory material. This would probably lead to higher costs for the provision of the service. An alternative solution is to arrange for the filtering of newsgroup postings by a moderator (Braithwaite and Carolina 1994, p 20).
The issue was whether the requirements of the Due Process Clause and the Ohio long arm statute had been met (both demanded the same standards). The basic constitutional principle is that of "minimum contacts" complying with "traditional notions of fair play and substantial justice" (International Shoe Co v State of Washington 326 US 310 at 316 (1945)). The particular application of the test required that the defendant purposefully avail himself of the privilege of acting in the forum, the cause of action arise from the defendant’s activities there and the defendant’s acts or consequences caused by the defendant have a substantial connection with the forum (Southern Machine Co v Mohasco Industries Inc 401 F 2d 374 at 381(1968). There was no purposeful availment because Patterson’s sales in Ohio were minimal. The presence of the software on the Compuserve network was entirely incidental to the dispute, so the second prong of the test was not complied with. Finally, there was no more connection with purposeful activities in Ohio than the standard customer dispute. The action was dismissed for want of personal jurisdiction.
If this analysis is applied to a Usenet newsgroup, one difference is that newsgroups are generally personal, whereas in Compuserve v Patterson the context was commercial. Turning to the specific requirements of Mohasco, it will not always be clear whether the defendant purposefully availed himself of the privilege of acting in the forum. However this requirement might be satisfied if the defendant deliberately defamed a named person known to live within the forum, which would be a common situation in such cases. Whether the cause of action arose from defendant’s activities in the forum would not always be clear in inter-jurisdictional Usenet defamation cases, especially if there were no actual proof of publication in the traditional sense. A particular difficulty would arise with the question of the defendant’s substantial connection with the forum. One might assume that a posting to a general newsgroup with many US members is an indication of a connection with the US where numerous members have US addresses. But does this also amount to a substantial connection with a particular US State jurisdiction? For a general anthropology newsgroup one might possibly assume this to be the case for jurisdictions likely or known to have members, such as California, but what of States not known for their interest in anthropology?
The English common law rule appears to be formulated somewhat differently. A stay will be granted where there is some other competent and appropriate forum where the case may be more suitably tried in the interests of the parties and the ends of justice (Spiliada Maritime Corp v Consulex Ltd [1987] AC 460 at 476 per Lord Goff). The court will look for the natural forum with which the action has the most real and substantial connection (The Abidin Daver [1984] AC 398 at 415 per Lord Keith). Relevant factors include convenience, availability of witnesses, expense and residence of the parties. Procedural disadvantages in the natural forum will not necessarily prevent a stay being granted (Trendtex Trading Corp v Credit Suisse [1982] AC 679 at 695 per Lord Wilberforce).
Applying these principles to Usenet defamation, it once again becomes apparent that Rindos presented a simple set of facts and that future cases, with somewhat different circumstances, could well give rise to a stay in English courts. Taking a hypothetical, for a moderated list, the plaintiff might join the moderator as the publisher of the defamation. Add the further hypothetical that the plaintiff had applied for positions in the field covered by the Usenet newsgroup and that there had been applications in a number of different jurisdictions around the world. Senior specialists in the field from each potential employer either post to the newsgroup or state that they regularly read it. The sole connection with England is that one of the potential employers is there. On the one hand it could be argued that England is not the natural forum because none of the three parties have any connection with the forum. However the question then arises whether there is a natural forum. This is an issue of world-wide distribution common to the Usenet groups and other Internet applications and the answer is far from clear.
In Australia the High Court in Theophanous v The Herald and Weekly Times Ltd (1994) 124 ALR 1 held by a majority of 4 to 3 that there is a freedom of expression implied in the Commonwealth Constitution to discuss government and political matters concerning the performance of Commonwealth parliamentarians in connection with their suitability for office. Three of the majority judges said that the publication must be reasonable in the circumstances, with an honest belief in its truth, and without malice. In regard to the possibility of extending this aspect of the implied freedom to other public figures the majority stated that the freedom also includes discussion of the political views and public conduct of persons who are engaged in activities that have become the subject of political debate. The concept of political matters refers to all speech relevant to the development of public opinion on the whole range of issues which an intelligent citizen should think about. Thus what is ordinarily private speech may develop into speech on a matter of public concern with a change in context, emphasis or context.
The possibilities of forum-shopping in relation to Usenet newsgroup defamation becomes apparent from a comparison of these two examples of the public figure doctrine. Take the hypothetical where a message was posted on a newsgroup protesting at the noise caused by the building and operation of a new runway at Ruritania Airport. This issue has been the subject of much public discussion and has received wide coverage in the media. Those defamed in the message include a world-famed academic who is a non-executive director of the Ruritania Airport Corporation (a recently privatised body), the Ruritanian Ministry for Transport and a member of the local council which approved and oversaw the building of the runway. It is assumed that there is no actual malice and that the publisher was unaware of the falsity of the material published and the remarks were reasonable in the circumstances.
The academic is not a public official since the corporation is a private body. In the United States, it is unlikely that the academic can sue for defamation because, although a private person, he is world- famous through his achievements and hence falls within the Gertz extension of the public figure doctrine. Under the Australian doctrine in Theophanous, the defamatory remarks fall within the implied freedom because the subject matter of the remarks relates to the academic's role in a matter which has become the subject of political debate. Thus the academic cannot sue in Australia provided that the publication was reasonable in the circumstances, the publisher was unaware of the falsity of the material published and the publication was not published recklessly.
Both the Ministry of Transport and the member of the local council are public entities and, in the absence of malice, will not be able to sue in the United States. The position is similar in Australia because both are engaged in activities that have become the subject of political debate and the remarks made are relevant to the development of public opinion on an issue which an intelligent citizen should think about. Further, the New South Wales Court of Appeal in Ballina Shire Council v Ringland (1994) 33 NSWLR 680 decided that local councils cannot sue for defamation.
In the United States, a court cannot award damages to a public official for defamation which relates to his official conduct, unless he can prove malice (New York Times v Sullivan 376 US 254 (1964)).
Returning to the Rindos case, take the hypothetical that Rindos sues in the two jurisdictions considered. It is submitted that Rindos' position in the United States is that it is doubtful whether he falls within the Gertz extension to the public figure doctrine. Although he has an international standing in the community of anthropologists, it is doubtful that he can be said to be known to the general public through his achievements. In Australia, the question is whether the subject matter of the discussion was the public conduct of persons who are engaged in activities that have become the subject of political debate. The fact that there was newspaper coverage of Rindos' alleged misconduct in a climate of public concern over such matters means that the matter is one of the whole range of issues which the intelligent citizen should think about and hence falls within the concept of public and political discussion. The difficulty in this argument is that the alleged misbehaviour was conducted in private and hence cannot be described as "public conduct". It is important to note that the Rindos case was decided before the establishment of the public figure doctrine in Australia in the Theophanous case.
Short of terminating the service provided, these news administrators might avoid liability in a number of ways. Firstly, they can screen each item posted to ensure that no defamatory material is allowed to be posted and displayed at their site. Given the volume of postings, this would be impractical. Further, not all defamatory material is readily identifiable. It will be too costly to pay a legally qualified person to screen all the postings. But if the person screening the material is not legally qualified, it is unlikely that this precaution will be effective in avoiding the publication of defamatory material. The magnitude of this problem is evident in the number of defamation actions brought against newspapers despite the fact that each newspaper has sub-editors screening what is a small number of articles in comparison to the number of postings appearing on the Internet. Secondly, it may be possible to arrange for the filtering of a newsgroup by a moderator. Given the international nature of the Internet, this too is impractical. There are problems with who will employ and pay the moderator and how the moderator is to be selected. If the moderator has no legal qualifications, the same problems in identifying the defamatory material arises. This problem may still be present, albeit in a smaller magnitude, even if the moderator or screener is legally qualified. Thirdly, the administrators can refuse to subscribe to newsgroups with known defamatory content. This will not completely eliminate the circulation of defamatory material because a survey of the more staid newsgroups will reveal much potentially defamatory material. However, this precaution will go some way in demonstrating that the administrators, given the measures they have taken, have no reason to suspect that they were handling defamatory material and hence can avoid liability as unintentional distributors. In addition to this, the position of the administrators can be made more secure by randomly checking the postings on those newsgroups subscribed. Alternatively, governments can legislate to protect the news administrators at institutions such as universities from liability for the distribution of defamatory material. This is necessary to protect one of the most efficient methods of communication. Given the slow pace at which governments grapple with technological advances, perhaps the best option is self-regulation and individual restraint.
Braithwaitel, N and Carolina, R (1994) "Multimedia Defamation", International Media Law, March 1994, 19.
Crispen, P D (1994) "Spamming and Urban Legends’ (18 October 1994)
December, J (1994) "Challenges for Web Information Providers" 1(6) Computer-Mediated Communications Magazine.
Fleming J, (1992) The Law of Torts.
Hardwick, G (1993) "Anthropologists and Training Agents" sci.anthropology, 26 June 1993.
Irwin, A and McGuckin, C (1994) "Libel case fuels fears on Internet" The Times, 28 August 1994.
Jarvis, H (1993) "Suppression of Academic Freedom" ANTHRO-L, 23 June 1993.
Kehoe, B P (1992) Zen and the Art of the Internet.
Loundy, D J (1994) "Operator Liability Revisited" 1(3) E-Law IVA.
Meeks, B (1994) "It’s Over" Wall Street Journal, 25 August 1994.
Milles, J (1993) An Introduction to the Internet.
Sandberg, J (1994) "Newsletter faces Libel Suit for 'Flaming' on Internet" Wall Street Journal, 22 May 1994.
Scholes, E F and Hay, P (1992) Conflict of Laws ,2nd ed.
Song-Muh, J (1994) List of all Usenet Groups, 1 March 1994, Part 2.
Stanton K, (1994) The Modern Law of Tort.
Treese, W (1994) The Internet Index, 2 August 1994, No 2.
van Niekerk, M (1994) "Law comes to wild frontier" The West Australian, 25 October 1994.