Australian court denies appeal in landmark ISP liability case

You're right Skippy, maybe the movie industry will finally give up this time.

Last year an Australian judge gave an important ruling in the landmark case Roadshow Films Pty Ltd v iiNet Limited [2010] FCA 24. You can read a full report on the case by yours truly here. iiNet is an Australian internet provider, which was sued by Australian film producer Roadshow Films, part of the Village Roadshow conglomerate. The question at the heart of the case in first instance was whether an ISP is to be held liable for the copyright infringement committed by its customers. The judge delivered an excellent decision in 2010 declaring that while there was direct infringement being committed by iiNet’s customers, the ISP was not secondary liable because it had not authorised such infringement.

Roadshow films and the lawyers for the movie industry appealed the decision alleging that the judge had erred in his interpretation of what constituted authorisation under Australian Copyright law. iiNet of course argued that the judge had been correct and stood by the decision. The Federal Court of Australia agreed with both the defendant and the first judge, and it denied the appeal. The ruling is very interesting because it is a split decision, Emmett J and Nicholas J voted for dismissing the appeal, and Jagot J voted to allow the appeal and have the case sent back to the primary judge.

As with the ruling in first instance, the matter came precisely to the question of whether or not iiNet had authorised copyright infringement committed by its customers. Emmett J encapsulates the issue thus:

“112. It is common ground that there have been primary infringements of the copyrights of the Copyright Owners by use by iiNet users of services provided by iiNet. However, there is a dispute as to the number of primary infringements that have occurred. The conclusion in relation to that dispute may be relevant to any limitation on remedies under the Safe Harbour Provisions, in so far as that limitation depends upon repeat infringement.
113. The principal issue in the appeal is whether iiNet authorised the acts that constituted those infringements. However, authorisation cannot be considered in the abstract. Rather, it must be determined in the light of the specific acts that are said to have been authorised. Accordingly, it is desirable to deal with the issues relating to primary infringement before considering the question of authorisation by iiNet.”

Regarding that important question, Emmett J considered:

“204. iiNet accepts that it had general knowledge of copyright infringement by use of the services provided to its customers. However, it contends that it had knowledge at a level of abstraction that was difficult to act on in any meaningful way. iiNet says that the only evidence of infringement at a level of specificity was the Infringement Notices but that the Infringement Notices did not make clear that the information provided by them was different from thousands of unreliable robot notices received from overseas. It says that, if a positive response was to be expected, AFACT should have explained in detail, when providing the Infringement Notices, how the allegations of infringement came to be made. iiNet says that AFACT was not entitled to keep its method of investigations secret if it expected a carriage service provider such as iiNet to be convinced of the reliability of the allegations of infringement and to act on such allegations. iiNet contends that it was not unreasonable for it to take the position that it wanted an independent third party to attest to the reliability and authenticity of the evidence provided by the Infringement Notices.”

Emmett J then concludes that there is no authorisation:

“257. However, while the evidence supports a conclusion that iiNet demonstrated a dismissive and, indeed, contumelious, attitude to the complaints of infringement by the use of its services, its conduct did not amount to authorisation of the primary acts of infringement on the part of iiNet users. Before the failure by iiNet to suspend or terminate its customers’ accounts would constitute authorisation of future acts of infringement, the Copyright Owners would be required to show that at least the following circumstances exist:
•    iiNet has been provided with unequivocal and cogent evidence of the alleged primary acts of infringement by use of the iiNet service in question. Mere assertion by an entity such as AFACT, with whatever particulars of the assertion may be provided, would not, of itself, constitute unequivocal and cogent evidence of the doing of acts of infringement. Information as to the way in which the material supporting the allegations was derived, that was adequate to enable iiNet to verify the accuracy of the allegations, may suffice. Verification on oath as to the precise steps that were adopted in order to obtain or discern the relevant information may suffice but may not be necessary.
•    The Copyright Owners have undertaken:
◦    to reimburse iiNet for the reasonable cost of verifying the particulars of the primary acts of infringement alleged and of establishing and maintaining a regime to monitor the use of the iiNet service to determine whether further acts of infringements occur, and
◦    to indemnify iiNet in respect of any liability reasonably incurred by iiNet as a consequence of mistakenly suspending or terminating a service on the basis of allegations made by the Copyright Owner.”

In a similar argument, Nicholas J argued that there was no infringement either, and while he commented that he disagreed with the primary judge on various lines of reasoning, the conclusion on authorisation was correct.

Jagot J did vote to allow the appeal. His argument is as follows:

“476. In summary, in this case:
(1) There was a contractual relationship between iiNet and its customers by which iiNet obtained income in return for the provision of internet services.
(2) iiNet knew that the internet was used by customers to download legitimate and copyright infringing content. iiNet also knew (through Mr Malone) that peer to peer software, including BitTorrent, was frequently used for file sharing of films and television shows in a manner that infringed copyright. Indeed, Mr Malone knew that half of all of the traffic on iiNet’s service (in terms of quota or megabytes) is via BitTorrent and a substantial proportion of BitTorrent traffic involves the infringement of copyright.
(3) Through its CRA, iiNet had available to it control over its customers’ use of the iiNet internet service as specified in the CRA’s terms. iiNet also had a range of powers available to it to prevent or avoid copyright infringements including issuing warnings, recording the issuing of warnings against customer accounts, as well as shaping, suspending or terminating accounts.
(4) iiNet received the AFACT notices over a lengthy period, in the same form from the third week. The AFACT notices contained prima facie credible evidence of widespread and repeated infringements of the appellant’s copyright by iiNet customers and users.
(5) iiNet adopted a position almost immediately after receipt of the AFACT notices that it had no obligation to do anything in response. Despite this, it asserted to AFACT that AFACT’s information did not identify customers. While this was technically accurate, it was inconsistent with iiNet’s knowledge that IP addresses were sufficient for it to identify customers and its treatment of customers as responsible for all use on their account in other contexts. It was also inconsistent with iiNet’s position that the reliability of the evidence was immaterial to its position.
(6) iiNet, by its press release, ensured its customers knew that their accounts would not be terminated, irrespective of the terms of the CRA, unless AFACT could prove in court the alleged copyright infringements.
(7) iiNet asked Westnet to change its position on allegations of copyright infringement to be less proactive on the basis that being proactive would harm the interests of internet service providers.
477. These circumstances support my conclusion that iiNet authorised the primary infringements of copyright. In terms of the meaning of “sanction, approve, countenance”, iiNet at least countenanced the primary infringements. It tolerated, indeed permitted, those primary infringements. More than that, by its conduct in all of the circumstances identified, it moved beyond mere indifference to at least tacit approval of those primary infringements.”

I am truly glad that this opinion did not prevail. Knowledge of infringement may be abstract, and also highly unreliable, as the ACS:Law case has demonstrated. It cannot be expected that ISPs will be able to act accurately against infringers, at least in the foreseeable future. Thankfully, we have obtained a good ruling on ISP liability once more, may this streak continue.

Comments 6

  1. I am just wondering if you have actually read and understand the judgment, because I don't see that there is a whole lot for any ISP to be happy about in any of three opinions.

    1. I'm a bit surprised that you would say this. The appeal was rejected, that is a positive result for iiNet by any standard. True, there are some passages in the ruling that are problematic, I found Jagot's arguments particularly troubling, and you will notice that I reproduced a big chunk of his argument. However, this is undoubtedly a victory for ISPs. Had the result been different, I would agree with you.

  2. Technically, you are correct in saying that this is a victory for iiNet, the appeal was dismissed. However, what you seem to have missed is that judgment of Emmett J and to a lesser extent Nicholson J laid down the framework for a judicially created "three strikes" system.

    Emmett J went almost out of his way to tell AFACT what to do if they wanted a finding in their favour. Arguably, this reasoning is obiter dicta, but it is very strong dicta and likely to be followed in lower courts where the cases could be distinguished on their facts. Had AFACT provided notification in the manner which laid down, the finding would have been 3:0 in AFACT's favour.

    The problem is that AFACT now has real options. They can walk away and pay damages, and follow the rule as laid down. Any ISP that wishes to challenge it has to then be prepared to go to the High Court to have it overruled. I understand that you are in the UK and I am not sure how well you understand the Australian judical heirarchy, but a ruling of the Full Court of the Federal Court can only be overruled by the Full Court itself, which is a most unlikely proposition in itself, or by the High Court. Of course, AFACT may appeal and go for a more definitive ruling in the High Court which will bind every court in Australia. However, they also risk a decision that is more restrictive than what was laid down in the present decision. So, why would they appeal? For AFACT (and its backers) $6.5 million is pocket change. They could pay it and then proceed on a systematic program of notifications to every ISP in the counrty following the framework laid down in the current decision. My guess is that they would start with smaller ISPs who are unlikely to be able to sustain the protracted legal campaign that would be required.

    This decision is far from good news for ISPs when you consider the implications of the reasoning of their Honours in the instant case.

    1. Thanks for the Australian background (I'm in Costa Rica BTW). I guess I've become rather pragmatic in my old age, and I took the appeal´s dismissal to be a victory for iiNet.

      I disagree with your opinion (and <ahref="http://www.lawfont.com/2011/03/02/a-few-thoughts-on-iinet-ffc-decision/#more-951">Kim Weatherall's) that this amounts to setting up a three strikes system. The judge talks about potential steps that iiNet could take, but unless I'm reading the judgement entirely wrong, and unless I am missing some intricacies of Australian judicial procedure, this opinion does not establish such a system, it is just part of the reasoning. The important conclusion, at least to what seems to me to be the most relevant point, is that iiNet did not authorise infringement.

      You are correct that there may still be another appeal, but as things stand I still think that the ruling is on the right side for ISPs. Just barely, but still a victory.

      1. Hi Andre,

        You are not wrong in saying that it was a victory for iiNet, they did indeed prevail. But that may not be the case if the circumstances were different. I don't have time to draft a detailed response at the moment. I have exams coming up in a few days time and need all the time I can get. You may find this article of interest in relation to the iiNet case. http://www.smh.com.au/technology/technology-news/

  3. Some minor corrections. Nicholas J not Nicholson J. It is costs that AFACT would have to pay, not damages. iiNet have estimated their costs at $6.5 million

Leave a Reply