1. Introduction

The Framework Convention on Tobacco Control is a major international public health initiative. The Framework Convention is the first treaty negotiated through the World Health Organization (WHO). It entered into force in February 2005 and to date has more than 160 parties.1 The parties have obligations across a wide range of tobacco-related issues, including in relation to price and taxation,2 exposure to tobacco smoke,3 packaging and labelling of tobacco products,4 illicit trade,5 and sales to and by minors.6 In all, the Framework Convention is “the most significant development in international tobacco control” in decades.7

The parties’ obligations also relate to tobacco advertising, promotion and sponsorship – whether delivered by analogue or digital communications media.8 In this respect, the Framework Convention should be of interest to academic researchers of internet content regulation. Within the sphere of the Framework Convention, the WHO offers what may be “an acceptable model of legitimacy for a global system in which territorially-bounded nation states are not the sole sources of authority.” 9 At the very least, the Framework Convention offers an example of the sorts of mechanisms that can be expected to appear in relation to internet content regulation. It is an instance of international cooperation which arises through existing networks that are not generally regarded as closely connected to issues of internet content.

This article outlines some of the key obligations related to tobacco advertising, promotion and sponsorship that arise for parties to the Framework Convention. It then highlights the Framework Convention’s method for promoting international agreement on internet content control, and considers how it underscores the central role of intermediaries in such control. The overall approach offers one possibility for dealing with the challenges of enforcing internet content regulation. The degree to which the approach actually has value – both within the sphere of tobacco regulation and within wider issues of internet content control – will emerge over the coming years as the Framework Convention’s obligations are implemented in domestic law.

The non-legal literature amply canvasses the serious public health issues that are being pursued through the Framework Convention.10 Responsible for one in ten adult deaths, smoking is the leading preventable cause of death internationally and kills up to half its users.11 In addition, tobacco-related illnesses involve major, and in many countries the highest, public health expenditure.12 The health effects in particular are stark. As the US National Cancer Institute summarises:

Smoking plays a key role in the causation of lung, oral, laryngeal, and pharyngeal cancers, such as those of the cervix, pancreas, and kidney, and has a substantial impact on the prevalence of heart disease, emphysema, and pneumonia, among other health problems.13

This is the context underlying the preamble to the Framework Convention, in which the parties recognise that:

[T]he spread of the tobacco epidemic is a global problem with serious consequences for public health that calls for the widest possible international cooperation and the participation of all countries in an effective, appropriate and comprehensive international response.

2. Article 13 Obligations: Comprehensive Ban on Tobacco Advertising, Promotion and Sponsorship

Article 13 of the Framework Convention is the primary provision dealing with tobacco advertising, promotion and sponsorship.14 Article 13 sets out requirements for a comprehensive ban on tobacco advertising, promotion and sponsorship (or restrictions where constitutional provisions prevent a comprehensive ban). It applies to a very wide range of content because of the broad definitions of advertising, promotion and sponsorship under the Framework Convention. It also includes an assertion of jurisdiction by parties to control cross-border tobacco advertising, promotion and sponsorship received within their territory.

These three aspects are outlined immediately below. The focus on comprehensive bans to advertising, promotion and sponsorship, where constitutionally permissible, is supported by academic literature on the effectiveness of complete bans on tobacco advertising in reducing consumption.15 In contrast, non-comprehensive bans generally see expenditure shift to allowed forms of advertising, promotion and sponsorship “which offset the effect of the partial ban so that any net change in consumption is minimal or undetectable.” 16

Article 13.2 establishes the requirement for a comprehensive ban, unless parties’ constitutions allow only restrictions of tobacco advertising, promotion and sponsorship, and sets a five year time-frame for compliance. It also requires parties, subject to legal and technological capabilities, to undertake a comprehensive ban on cross-border advertising, promotion and sponsorship originating from their territory. For parties with constitutional provisions that prevent a comprehensive ban, Article 13.3 sets out obligations to restrict tobacco advertising, promotion and sponsorship. As with Article 13.2, it also requires those parties to restrict or comprehensively ban advertising, promotion and sponsorship originating from their territory, and again applies a five year window for compliance.17

The Article 13 obligations apply to a very wide range of content. Under the Framework Convention, “tobacco advertising and promotion” is defined to mean “any form of commercial communication, recommendation or action with the aim, effect or likely effect of promoting a tobacco product or tobacco use either directly or indirectly.” 18 This appears to be far broader than the common understanding of a commercial advertisement. They would appear to extend well beyond “informational” and “lifestyle” advertising.19 One example of the breadth of this definition is the consideration of “plain packaging” under the FCTC. Plain packaging refers to the material, construction, design colour, text and font used on tobacco packaging (excluding textual and graphical health warnings). It appears likely to be a key future-issue in public health efforts to control tobacco use.20 Similarly, the Framework Convention defines “tobacco sponsorship” as widely as “any form of contribution to any event, activity or individual with the aim, effect or likely effect of promoting a tobacco product or tobacco use either directly or indirectly.” 21 Parties’ initial focus may well be on advertising with a direct consumer marketing focus. However, in the longer term, and through the reporting requirements under the Article 21 of the Framework Convention, it is likely that pressure to deal with a far wider range of content will be seen.22

In terms of territorial reach, through the Framework Convention parties assert their ability to control all cross-border tobacco advertising, promotion and sponsorship that is received within their territory in an equivalent manner to the controls placed on domestically produced content. Article 13.7 states:

Parties which have a ban on certain forms of tobacco advertising, promotion and sponsorship have the sovereign right to ban those forms of cross-border tobacco advertising, promotion and sponsorship entering their territory and to impose equal penalties as those applicable to domestic advertising, promotion and sponsorship originating from their territory in accordance with their national law.

Readers might have noticed that nothing has been said here about one of the key issues for internet content: namely the enforcement of legal controls on extra-territorial actors. The “Achilles’ Heel”23 of enforcement will be returned to below. Readers might also have noted that the language of the above provisions is at quite a general level. That is not surprising given their place in an international convention. The generality is addressed, in part, through the development of guidelines to some elements of the Framework Convention, including guidelines to Article 13.

3. Article 13 Guidelines: Entities Subject to Control

The Third Session of the Conference of the Parties to the Framework Convention – generally known as COP3 – was held in Durban, South Africa, over six days in November 2008 (with in excess of 600 delegates). Draft guidelines for Article 13 had been developed over the preceding two years by a working group of approximately twenty-five countries. The working group had held two face-to-face meetings of three days each: in Finland in 2007; in India in 2008; as well as extensive mediated commentary and drafting before and after these meetings.24 At its Durban meeting, the Conference of the Parties adopted the Guidelines for Implementation of Article 13 of the WHO Framework Convention on Tobacco Control to assist parties implementing their obligations.25 These provide a more detailed level of guidance to parties – with the breadth of advertising, promotion and sponsorship under the Framework Convention meaning that the guidelines cover a wide range of measures.26

As one could expect when considering the control of media content, the guidelines identify different types of entities that have varied involvement in the communication of tobacco advertising, promotion and sponsorship, which parties could seek to make subject to different forms of control:

Responsibility cannot be attributed in the same manner to all entities, as their involvement in the production, placement and dissemination of tobacco advertising, promotion and sponsorship varies […] When tobacco advertising and promotion involve communication, the way in which entities should be held responsible depends on their role in the production and dissemination of the content of the communication and the possibilities they have to control it.27

In relation to the internet, for example, five “principal categories of responsible entity” are identified in the guidelines, according to which “bans or particular obligations should be imposed.”28 Many internet, communications and media-related companies, or other entities would combine two or more of these roles. But the delineation of the roles and the different ways in which they are treated are noteworthy – and relevant to wider options for internet content control. The five categories separate out content producers, content publishers, content hosts, content navigators and access providers:

• Content producers create the content or cause it to be created. These include tobacco companies, advertising agencies, and producers of television programmes, films and games that are distributed online. Content producers should be banned from including tobacco advertising, promotion or sponsorship in the content they produce.

• Content publishers include publishers and entities that select content before it is made available to Internet users (for example, Internet sites of newspapers or broadcasters). Content publishers should be banned from including tobacco advertising, promotion or sponsorship in the content they make available.

• Content hosts are entities that control Internet-connected computer servers on which content is stored, including entities that aggregate content produced by others without selecting the content before they make it available to Internet users (such as social networking Internet sites). Content hosts should have an obligation to remove or disable access to tobacco advertising, promotion and sponsorship once they have received notice of the content.

• Content navigators are entities that facilitate the location of content by users of communications services, such as Internet search engines. Content navigators should have an obligation to disable access to tobacco advertising, promotion and sponsorship once they have received notice of the content.

• Access providers are entities that provide end-user access to communications services, such as Internet service providers and mobile telephone companies. Access providers should have an obligation to disable access to tobacco advertising, promotion and sponsorship once they have received notice of the content.29

The guidelines thus set out three styles of obligation in relation to internet content. First, content bans are applied to content producers and content publishers. Second, obligations to remove or disable access to content are applied to content hosts after they receive notice of the content. And third, obligations to disable access to content are applied to content navigators and access providers after they receive notice. Given the history of Internet liability, it is not surprising that the guidelines explicitly state that parties could limit the third style of notice and obligation scheme for content navigators “to using reasonable efforts to disable access, in light of what is technically possible.” 30 The guidelines do not, however, stipulate in what form notice should be given, leaving that to parties.31

4. Approaches to Internet Content Control

The Framework Convention, and accompanying guidelines, illustrate at least three aspects of existing commentary on Internet content. These concern regulations moving online; the idea of seeking international agreement on standards or imposing digital borders; and a focus on regulating intermediaries that are involved in internet communications, especially domestic intermediaries.

4.1 Regulation is Moving Online

The idea of the Internet as a communications system that exists beyond nation-state control was once commonplace, with a classic example being David Johnson and David Post’s work from the mid-1990s.32 However, maintaining that position can now be seen to “negate … the democratic importance of preventing harm” online, and the idea that the Internet cannot be regulated “no longer has currency.”33 For writers such as Damian Tambini, Danilo Leonardi and Chris Marsden:

As the Internet embeds itself further in everyday life, so too will concerns about content and its consequences … [I]n Europe, and even in the United States, the illusion that the Internet can constitute a ‘free’ sphere separate from social life will fade.34

The Framework Convention illustrates this recognition. As evident in much recent literature, claims to limit regulation of the Internet will lessen “in the face of a large-scale shift of activities onto the internet.”35 Regulation is moving online, with states “now consistently applying traditional territorially based rules to online activity and largely refusing to treat the Internet as beyond their competence.” 36

As Uta Kohl has stated in relation to gambling and drugs:

A failure to regulate the Internet effectively undermines the credibility and effectiveness of the regulation of equivalent offline activity. What is the point of, and how can you justify, a prohibition of physical gambling operations, if similar online gambling operations are beyond the regulatory reach?37

The Framework Convention well illustrates this kind of significant public issue, which underlies at least some attempts to control internet content.

4.2 Seeking Agreement or Imposing Borders

Options commonly canvassed in the literature, in relation to controlling Internet content, involve international agreement with regard to particular content or the imposition of borders online. Agreement is seen as substantively impossible by most commentators for almost all areas of content, and the imposition of borders is seen as being, or potentially being, draconian.

A noted example of the difficulties in obtaining agreement in relation to content is the Council of Europe Cybercrime Convention of 2001 and the approach taken there to racist hate-speech.38 The difficulty of reaching agreement in relation to hate-speech saw the topic dealt with, not in the body of the convention, but in a protocol. Although the United States signed the convention, it did not sign the protocol due to the protocol’s perceived inconsistency with the US First Amendment.39 As Andrew Murray explains, this meant the United States feels under no legal obligation to enforce the protocol.40 Understandably, he concludes that the refusal of the United States to sign the protocol “will fundamentally undermine” its effectiveness.41 Such differences over content and standards are legion:

Tensions over privacy, domain name allocation, freedom of speech and other issues continue to exist between the United States and Europe, reflecting deep-seated unresolved differences in the approach to market, state and society.42

In the words of Michael Kirby (an Australian appellate judge with a longstanding interest in matters of technology and privacy), it often appears that “the regulatory values of the United States inevitably exert the greatest influence on the way the Internet operates and what it may include”43 – at least with regard to the markedly atypical approach taken to the legal regulation of speech under the US constitution.

Even though coordinated international action on matters of content has been rare, as Jonathan Zittrain and John Palfrey note, “there is no inherent reason to believe that international cooperation or governance could not play a meaningful role.” 44 The challenge lies, initially, in reaching agreement. In that regard, the Framework Convention offers a contrast with initiatives such as the Cybercrime Convention. In the Framework Convention an interesting form of agreement was reached. That form, which makes allowances for differing constitutional standards in relation to speech, may have a wider application.

In relation to the concept of regulating borders, existing commentary can over-emphasise the idea of blanket controls, typically referring to high profile non-democratic, or formally democratic, states which famously attempt widespread control of Internet content. Frequent examples include China and Saudi Arabia, but the practices extend to a far wider number of states, as the recent Access Denied study shows.45 Beyond such blanket controls, there are other types of border control such as the borders that are increasingly being constructed by content providers using geolocation.46 This type of “zoning” of Internet sites can take at least two forms. One form involves simply the creation of localised Internet sites for different countries or regions, and the encouragement of users to interact with those sites rather than different versions of the site aimed at other jurisdictions. This approach is undertaken by many large internet entities, such as those providing search, travel and retail services.47 Another option is for sites to be “sealed” to allow access only from particular jurisdictions.48 Notable examples of this approach include digital on-demand audiovisual services, such as that provided in the UK by the BBC’s “iPlayer”; or Channel Four’s “4oD” on-demand service, which use geo-location to restrict access to only those requests emanating from the UK.49 The iPlayer system “recognises IP addresses provided by UK-based Internet service providers.”50 Equivalent rights issues mean that similar approaches are taken in other countries. For example, the Australian Broadcasting Corporation’s “iView” service provides this error message to users with IP addresses outside Australia: “[D]ue to copyright reasons this video program is available for download by people located in Australia only. If you are not located in Australia, you are not authorised to view this video.”51 Equally, the use of geo-location appears central to online advertising practices, in which businesses, not surprisingly, “want their advertisement to be targeted at a geographically relevant audience.” 52The Framework Convention does not directly respond to the growth of geo-location – neither its use in general on the Internet, nor in relation to online advertising. However, the deployment of such technologies, across many and varied types of internet sites and communications, is significant for the degree to which comprehensive bans on tobacco advertising, promotion and sponsorship can realistically be pursued.

4.3 Agreement and Intermediaries

Concerns seen in the literature about reaching agreement and imposing borders are understandable, but under the Framework Convention there has been a great deal of agreement and the controls are likely to be focused on intermediaries in practice – something implicit in the idea of imposing borders warranting closer examination.

In relation to the agreement about content reached under the Framework Convention, studying the implementation process as it develops will show how much more than bare agreement on language is involved in the sort of coordinating task of the Framework Convention. However, the extent of agreement to a comprehensive ban, undoubtedly supported by the serious public health concerns about tobacco, is worth noting because of its scale – more than 160 parties to the Framework Convention to date53 – and because of the form and structure of the obligation. The obligation explicitly allows for parties to invoke constitutional limits to their control of tobacco advertising, promotion and sponsorship. In that situation, parties agree to impose various restrictions instead of a comprehensive ban.54 The caveat provided by this form of agreement has not been enough to see the US become a party, although it has signed the Framework Convention. But once implementation begins to take effect in other countries, this approach may be seen as a reasonably effective side-step around one of the key problems about controlling Internet content that has been raised in the literature – namely, the practical impossibility of achieving internationally agreed standards.

One of the central questions as to whether the approach will be a viable response to problems of international agreement relates to the effects of the flexibility inherent in the form of agreement reached under Article 13. That flexibility means the implementation of parties’ obligations may be flawed in terms of the public health perspective underlying the Framework Convention. Relatively weak attempts to control tobacco advertising, promotion and sponsorship may be made by some parties under claimed constitutional limits. However, the history of public health efforts to combat control – even in the United States where meaningful constitutional limits do exist – suggests that the goal of a comprehensive ban will be drawn on, repeatedly and reasonably effectively, in wider public policy debates about tobacco and public health.55 In that, the Framework Convention will form one element in multifaceted national approaches to this public health problem.

Beyond that probability, there is an additional reason why the form of agreement in the Framework Convention might be seen as having wider value. This is that the broad agreement on a comprehensive ban is likely to be coupled with a focus on controlling domestic entities: those located within the territory of a party that produce, publish or host content; those that navigate users towards content; and those that provide access to content. In this, the Framework Convention illustrates an element that is reasonably longstanding in the history of internet content control – an element which is still gaining in importance. That is, control is less aimed at the level of the device and user to being exercised more at the level of intermediaries through, for example, the notice and obligation schemes listed in the guidelines.56

To return to the key problem of enforcement, it seems parties could claim worldwide jurisdiction in relation to tobacco advertising, promotion and sponsorship. As noted above, Article 13.7 of the Framework Convention at least asserts that in relation to tobacco advertising, promotion and sponsorship.57 Such claims could not be enforced extra-territorially58– although they might underpin agreements by major online intermediaries to zone their content or disable access to particular content for users in particular countries59 (and that approach gains feasibility with the continuing development in geo-location technologies). More significantly though, in parallel to such claimed global ambit, parties can be expected to pay close attention to regulating intermediary conduct within their territories. This would deal more effectively with problems of enforcement and reflect how the technical possibilities for intermediaries to exercise control – as well as the extent to which control is exercised already – are changing rapidly,60 with “[m]ajor technology companies hav[ing] helped to facilitate such restrictions in both democratic and authoritarian countries.”61

If the implementation of Article 13 does develop in this manner, it will illustrate the recognition, in much recent writing, of the key regulatory position of ISPs, and the utility of notice and take-down schemes, which for Tambini, Leonardi and Marsden “is emerging as the key tool in Internet content regulation”62 – notwithstanding the concerns this can raise about private censorship. Such intermediaries may well be “the natural gatekeepers,” 63 and the very widespread international agreement evident in the Framework Convention means that corporate intermediaries may not be such “reluctant gatekeepers” as they can be when implementing a single country’s standards about, for example, political speech64 – especially given that “respectable high-profile online actors and businesses do not want to be seen to flout foreign laws.”65 As Jonathan Zittrain and John Palfrey state:

One of the key findings of our research is the extent to which states cannot do the job of content control alone … Where the state cannot effectively carry out its mandate in these legitimate circumstances, the state reasonably turns to those best positioned to assert control of bits. Often, though not always, the state turns to Internet service providers of one flavor or another.66

5. Conclusion

While the Framework Convention “represents the most significant step ever taken to reduce global tobacco use,” 67 it is not yet clear whether the Convention provides a revolutionary instance of Internet content control. But it is already an important example of emerging attempts to develop control internationally, and it appears to offer several refinements to the existing literature. Those developments could become more evident as parties find ways to meet the obligations in relation to controlling tobacco advertising, promotion and sponsorship that they have undertaken in ratifying the Framework Convention.

In particular, the Framework Convention may offer one approach to issues of enforcement through obtaining very widespread agreement on broad standards and including clear efforts towards local enforcement. Until more is seen about how the Framework Convention is implemented, however, questions remain about how valuable a model it may offer for many other areas of content. But it is certainly an instance of the “significant increase in the range and variety of global governance mechanisms” emerging in relation to the Internet.68 And finally, as Tambini, Leonardi and Marsden note: “[T]hose wishing to see ISPs regulate their content should be careful what they wish for.”69 The Framework Convention may well help more countries to develop techniques to control Internet content, which in turn could prompt challenges in controlling those techniques.

 

---

* Director, CMCL — Centre for Media and Communications Law, and Professor, University of Melbourne Law School. Thanks to Jason Bosland, Research Fellow at the CMCL for assistance with underlying research into the Framework Convention, to Jonathan Liberman and Richard Garnett for discussion and comments, and to Sophie Killen, who conducted earlier research with the CMCL and the VicHealth Centre for Tobacco Control that informs some aspects of this paper.

1 The Framework Convention is also known as the FCTC; current signatories and parties are listed at http://www.who.int/fctc/signatories_parties/en/index.html (accessed 30 Jun 09).

2 See FCTC Article 6.

3 See FCTC Article 8.

4 See FCTC Article 11.

5 See FCTC Article 15.

6 See FCTC Article 16.

7 B Freeman, S Chapman and M Rimmer, “The Case for the Plain Packaging of Tobacco Products” (2008) 103 Addiction, 580, at 580.

8 See FCTC Article 13.

9 D Lindsay, International Domain Name Law: ICANN and the UDRP (Oxford: Hart, 2007), at 31.

10 A brief overview is provided by L O Gostin, “Global Regulatory Strategies for Tobacco Control” (2007) 298:17 Journal of the American Medical Association 2057. See also, e.g., US Department of Health and Human Services, The Health Consequences of Smoking: A Report of the Surgeon General (Washington: US Department of Health and Human Services, 2004) available at http://www.cdc.gov/tobacco/data_statistics/sgr/2004/index.htm (accessed 30 Jun 09).

11 See e.g. World Health Organization Tobacco Free Initiative, “Tobacco Facts” available at http://www.who.int/tobacco/mpower/tobacco_facts/en/index.html (accessed 30 Jun 09); C D Mathers and D Loncar, “Projections of Global Mortality and Burden of Disease from 2002 to 2030” (2006) 3:11 PLoS Medicine e442, available at http://www.plosmedicine.org/article/info:doi/10.1371/journal.pmed.0030442 (accessed 30 Jun 09).

12 Gostin, above note 10, at 2057.

13 National Cancer Institute, The Role of the Media in Promoting and Reducing Tobacco Use, Tobacco Control Monograph Series No 19 (Bethesda, MD: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute, 2008), at 4.

14 The analysis in this section draws on A T Kenyon and J Liberman, “Controlling Cross-Border Tobacco: Advertising, Promotion and Sponsorship — Implementing the FCTC” (Melbourne: Centre for Media and Communications Law, University of Melbourne and VicHealth Centre for Tobacco Control, 2006) available at http://ssrn.com/abstract=927551 (accessed 30 Jun 09).

15 See e.g. H Saffer and F Chaloupka, “The Effect of Tobacco Advertising Bans on Tobacco Consumption” (2000) 19 Journal of Health Economics 1117. That argument, however, has not always satisfied legal analyses of comprehensive bans, under particular constitutional contexts: see e.g. RJR-McDonald Inc v AG of Canada [1995] 3 SCR 199. Even so, advertising, and especially tobacco advertising, is analysed by leading legal scholars as occupying a comparatively weak free speech position, see e.g. E Barendt, Freedom of Speech (Oxford: Oxford University Press, 2nd ed, 2005), at 392-416.

16 National Cancer Institute, above note 13, at 16.

17 See Articles 13.3 and 13.4(e).

18 FCTC Article 1.

19 See e.g. Barendt, above note 15.

20 See e.g. Freeman, Chapman and Rimmer, above note 7 which, among other things, counters legal arguments against plain packaging based on trademark and international trade law: at 585-86. (The authors also note links of at least some of that research to tobacco industry funding, underlining the importance of clear disclosure requirements in academic publishing.)

21 FCTC Article 1.

22 As could be expected with this form of international agreement, the possibilities for enforcement are, in legal terms, relatively weak. Article 21 the FCTC sets out requirements for parties to report progress to the Conference of the Parties, including obligations to provide “information on legislative, executive, administrative or other measures taken to implement the Convention” and information “on any constraints or barriers encountered in its implementation of the Convention, and on the measures taken to overcome these barriers.” The process is bolstered, however, by the recognition in the FCTC of the importance of civil society: see below note 31.

23 U Kohl, Jurisdiction and the Internet: Regulatory Competence over Online Activity (Cambridge: Cambridge University Press, 2007), at 18.

24 I attended meetings of the working group, being appointed as an expert adviser to the Framework Convention secretariat to assist the working group. However, I took no part in the Third Session of the Conference of the Parties, and my comments here are based on separate academic analyses from the activities of the working group, the secretariat and the Conference of the Parties.

25 For the guidelines, see Third Session of the Conference of the Parties (COP3), First Report of Committee A, pages 10-28; available via http://www.who.int/fctc/cop/third_session_cop/en/index.html (accessed 30 Jun 09).

26 Topics dealt with in the guidelines include: retail sale and display; packaging and product features; internet sales; brand stretching and brand sharing; corporate social responsibility; communication within the tobacco trade; sanctions; monitoring, enforcement and access to justice; and public education and community awareness.

27 Guidelines for Implementation of Article 13 of the WHO Framework Convention on Tobacco Control, paragraph 55.

28 Ibid, paragraph 58.

29 Ibid, paragraph 58.

30 Ibid, paragraph 59.

31 It is worth noting, however, that the guidelines anticipate a substantial role for civil society in relation to the FCTC; e.g., the “underlying principles” of the Guidelines for Implementation of Article 13 of the WHO Framework Convention on Tobacco Control, in paragraph 3(7), state: “[C]ivil society has a central role in building support for, developing and ensuring compliance with laws addressing tobacco advertising, promotion and sponsorship, and it should be included as an active partner in this process.”

32 E.g. D R Johnson and D Post, “Law and Borders: The Rise of Law in Cyberspace” (1996) 48 Stanford Law Review 1367.

33 J Zittrain and J Palfrey, “Internet Filtering: The Politics and Mechanisms of Control” in R Deibert et al (eds), Access Denied: The Practice and Policy of Global Internet Filtering (Cambridge, MA: MIT Press, 2008) 29, at 31.

34 D Tambini, D Leonardi and C Marsden, Codifying Cyberspace: Communications Self-Regulation in the Age of Internet Convergence (Abingdon: Routledge, 2008), at 294.

35 I Walden, “Regulating Broadcasting in a Converging Environment: Without Frontiers and Without a Fuss!” (2007) 12 Media & Arts Law Review 423, at 424.

36 Kohl, above note 23, at 11-12. See also e.g. Deibert et al (eds), above note 33.

37 Kohl, above note 23, at 6.

38 See, e.g., the analyses by A D Murray, The Regulation of Cyberspace: Control in the Online Environment (Abingdon: Glasshouse, 2007), at 223-25 and Kohl, above note 23, at 263-65.

39 The First Amendment to the US constitution states, in part, “Congress shall make no law … abridging the freedom of speech, or of the press.”

40 Murray, above note 388, at 225.

41 Ibid.

42 Tambini, Leonardi and Marsden, above note 344, at 18. See also e.g. W J Drake, “Introduction: The Distributed Architecture of Network Global Governance” in W J Drake and E J Wilson III (eds), Governing Global Electronic Networks: International Perspectives on Policy and Power (Cambridge, MA: MIT Press, 2008) 1, who, at 52, notes “the complexity of the issues and variations in national legal systems and traditions make it difficult to reach broad international agreement on either restrictive or permissive measures.”

43 M Kirby, “New Frontier: Regulating Technology by Law and ‘Code’” in R Brownsword and K Yeung (eds), Regulating Technologies: Legal Futures, Regulatory Frames and Technological Fixes (Oxford: Hart, 2008) 367, at 383.

44 J Zittrain and J Palfrey, “Reluctant Gatekeepers: Corporate Ethics on a Filtered Internet” in Deibert et al (eds), above note 333, at 103, 118-19.

45 See Deibert et al (eds), above note 33 and in particular its first chapter R Faris and N Villeneuve, “Measuring Global Internet Filtering”, 5-27.

46 Zittrain and Palfrey note the emergence of filtering by some online publishers of content using geolocation: “Internet Filtering”, above note 333, at 43.

47 See e.g. Kohl, above note 23, at 279. Compare http://www.amazon.com and http://www.amazon.co.uk (accessed 30 Jun 09). The US site provides this message to UK-based users at the top of its main page to encourage them to use the UK site: “Shopping from the UK? Visit amazon.co.uk”.

48 Ibid at 279-80.

49 See http://www.bbc.co.uk/iplayer/tv and http://www.channel4.com/4od (accessed 30 Jun 09).

50 “BBC iPlayer Help and FAQs”, http://iplayerhelp.external.bbc.co.uk/help (accessed 30 Jun 09). The site also notes that “Rights agreements mean that BBC iPlayer television programmes are only available to users to download or stream (Click to Play) in the UK. However, BBC Worldwide is working on an international version, which we will make available as soon as possible.”

51 See http://www.abc.net.au/tv/iview (accessed 30 Jun 09).

52 D J B Svantesson, “Geo-identification and the Internet” (2007) 14 Murdoch University E Law Journal 155, at 155.

53 See above note 1.

54 See above notes 17-22 and accompanying text.

55 As just one example, in four decades since the landmark 1964 report to the US Surgeon General on smoking and health, overall rates of adult smoking in the US have halved: US Department of Health and Human Services, above note 10, at 14. The earlier report is: US Department of Health, Education, and Welfare, Smoking and Health: Report of the Advisory Committee to the Surgeon General of the Public Health Service (Washington: US Department of Health, Education, and Welfare, Public Health Service, Center for Disease Control, 1964).

56 Guidelines for Implementation of Article 13 of the WHO Framework Convention on Tobacco Control, paragraph 58 and see above 29.

57 See above note 21 and text following it.

58 No international agreement exists on recognising and enforcing foreign judgments, and the laws of different jurisdictions varies greatly in their approaches. The agreements that do exist, such as the European Union Brussels Convention, generally focus on civil or commercial matters: Convention of 27 September 1968 on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters. Judgments that are ‘penal’ in nature – that include any criminal sanctions such as fines – will not be enforced under international agreements. This is one aspect that distinguishes regulatory efforts like the FCTC from the often civil law issues surrounding other aspects of Internet content, such as defamation and copyright.

59 Compare, for example, the brief discussion of the voluntary approach of search engines in France and Germany to neo-Nazi content in Tambini, Leonardi and Marsden, above note 344, at 116 and Kohl, above note 23, at 279-80.

60 See, for example, above notes 48-52 and the accompanying text.

61 Drake, above note 422, at 51.

62 Tambini, Leonardi and Marsden, above note 344, at 122. See also, e.g., Joel R Reidenberg, “States and Internet Enforcement” (2003) 1 University of Ottawa Law & Technology Journal 213; J Palfrey and R Rogoyski, “The Move to the Middle: The Enduring Threat of ‘Harmful’ Speech to Network Neutrality” (2006) 21 Washington University Journal of Law & Policy 31. In relation to a focus on regulating ISPs, it is worth noting that neither the analysis presented in this article, nor the guidelines themselves, envisage the kind of ISP-level filtering that has been investigated by the Australian Government in recent years and has created great controversy; see, for example, B Simpson, ‘New Labor, New Censorship? Politics, Religion and Internet Filtering in Australia” (2008) 17 Information & Communications Technology Law 167.

63 L Edwards, “Defamation and the Internet” in L Edwards and C Waelde, Law & the Internet: A Framework for Electronic Commerce (2000) 249, at 263.

64 See e.g. Zittrain and Palfrey, “Reluctant Gatekeepers”, above note 444.

65 Kohl, above note 23, at 279.

66 J Zittrain and J Palfrey, “Internet Filtering”, above note 333, at 45.

67 S Chapman, “Global Perspective on Tobacco Control, Part II, The Future of Tobacco Control: Making Smoking History?” (2008) 12 International Journal of Tuberculosis and Lung Disease 8, at 8.

68 Drake, above note 422, at 23.

69 Tambini, Leonardi and Marsden, above note 344, at 9.