The Australian Experience of World Wide Views on Global Warming : The First Global Deliberation Process

World Wide Views on Global Warming was the first ever global-scale citizen deliberation process, held on 25-26 September 2009 and involving approximately 4,000 citizens in 38 countries. WWViews sought to provide citizens with a voice in the 2009 UN Climate Summit in Copenhagen (COP15) by engaging them in a deliberative process about global political positions on climate change. The process produced clear, comparable results across all participating countries that were given to COP15 negotiators. The Danish Government agencies, the Danish Board of Technology and the Danish Cultural Institute, initiated the global process. Organisers in each participating country ran events using the same standardised process. The University of Technology Sydney, the organisers of the Australian WWViews event, paid special attention to several elements of the process to maximise participation and impact within the local context. This paper outlines the standardised global process used for this deliberative event and describes and reflects upon the tailored approaches developed for Australia. It examines in detail the objectives, processes and outcomes of recruiting and supporting participants and recruiting, training and coordinating facilitators, communications and dissemination of results and specific features of the Australian event. It includes the organisers’ reflections on success factors, challenges and surprises, as well as feedback from facilitators and participants. This paper concludes with a number of critical questions arising from the Australian experience of World Wide Views on Global ∗The authors gratefully acknowledge Dr Annie Bolitho, Dr Lynn Carson, Dr Kath Fisher, Dr Chris Riedy and Nicole Thornton for their advance review and thoughtful comments on this paper. The authors thank all the organisations and individuals who helped to make World Wide Views possible. In particular we are grateful to: the partners in the international World Wide Views Alliance, especially the Danish Board of Technology and the Danish Cultural Institute; the World Wide Views Australia sponsors (University of Technology Sydney, PricewaterhouseCoopers, National Australia Bank, WWF Australia and the Department of Sustainability and Environment Victoria) and other supporting individuals and organisations; the facilitators and event logistics team; and the World Wide Views participants themselves. The World Wide Views Australia core project team was comprised of members of staff of the Institute for Sustainable Futures at the University of Technology, Sydney: Alison Atherton, Amber Colhoun, Jennifer Croes, Jade Herriman, Dr Chris Riedy, Nicole Thornton and Professor Stuart White; and independent contractors: Dr Kath Fisher (Lead Facilitator) and Rebecca Short (Media Officer). Warming that are pertinent for practitioners designing other deliberative forums and particularly anyone concerned about future prospects for global deliberative democracy.


Introduction
On 25 and 26 September 2009, 105 Australians from a diversity of backgrounds as well as thousands of people in 38 countries around the world had their say on climate change action in a deliberative decision making forum.World Wide Views on Global Warming (WWViews) was the first-ever global-scale citizen deliberation process.It was initiated by Danish Government agencies with direct links to the United Nations Climate Change Conference (COP 15) in Copenhagen.WWViews was also endorsed by the Australian Government, which participated in COP15 in December 2009 -the biggest international summit on climate change held to date.WWViews Australia had two broad objectives -to influence the Australian negotiating position at COP15 by making known to Australian negotiators the wishes of Australian citizens regarding the outcomes of COP15, as expressed through the WWViews process; and to raise the profile of citizen dialogue processes as a valuable democratic mechanism.It also sought to consolidate existing experience in deliberative processes and to develop the capacity in process design and implementation for future processes.
This paper describes and reflects upon the running of WWViews Australia.The authors were members of the project team and had various roles in the management, planning and implementation of the Australian event.The paper describes the standardised global process and explains how it was adapted for the Australian event.It examines in detail the objectives, processes and outcomes of recruiting and supporting participants, and of recruiting, training and coordinating facilitators.It also examines the communication processes involved and the dissemination of results and other features of the Australian event.
It includes the organisers' reflections on success factors, challenges and surprises, as well as feedback from facilitators and participants.The paper concludes with a number of reflections arising from the Australian experience of WWViews that are pertinent for practitioners designing other deliberative forums and particularly anyone concerned about future prospects for global deliberative democracy.This paper intentionally focuses on the WWViews process and does not discuss the results from the process in any detail, although these are available elsewhere in full. 1 Subsequent papers will examine the specific dissemination efforts and results of the Australian process (Herriman, White & Atherton, 2011); they will consider the interface of this global process with global decision making (Riedy and Herriman in press) but these discussions are beyond the scope of this paper.

How WWViews fits with other deliberative processes
A range of deliberative processes have been conducted in Australia on various topics (Carson, 2007), including climate change (for example, Kaufman, 2009;NCCNSW, 2009;WA Govt, 2009;Green Cross, 2008;Riedy et al., 2006).However, these climate change deliberation processes have focused primarily on engaging citizens to consider local (WA Govt, 2009), regional (Riedy et al., 2006), statewide (NCCNSW, 2009) or national strategies (Kaufman, 2009).WWViews, on the other hand, focused on the preferred national policy for international climate negotiations.A recent deliberative process (Green Cross, 2008) did address issues of Australia's international responsibilities in the Asia Pacific region, but unlike WWViews it did not focus on global decision-making around issues such as emissions reduction targets.It was also conducted only in Australia, and therefore did not provide comparative information on how other countries see their own responsibilities in relation to climate change.
The WWViews process was a hybrid of several citizen engagement methods, based on the long-running experience of the Danish Board of Technology (DBT) and other WWViews Alliance members (DBT, 2009a).Denmark has a tradition of using deliberative democracy to inform political decision-making (DBT, 2009c).The participatory consensus conference, also known at the 'Danish method' consensus conference (Nielson et al., 2006), or the 'citizens panel' (Guston 1999), was developed by the DBT in the early 1980s and informed by the expert consensus conference model of the US Office of Technology Assessment (OTA) (Nielson et al., 2006).Since 1987, DBT has run participatory consensus conferences in which Danish citizens deliberate on public policy questions, specifically in relation to technology, and their recommendations are provided to the Danish Parliament (Kluver, 1995;Grundahl, 1995).WWViews was similar to this approach in that it involved a group of citizens considering a complex policy issue in a deliberative manner, but the structure, timing and processes used were considerably different.See Table 1 for a summary of some key differences and similarities.
The WWViews process had many similarities with various other deliberative processes that have been conducted at the national and regional levels around the world, particularly those that ask participants to vote on various options (see for example, AmericaSpeaks, 2010a; Fishkin, n.d.).WWViews was similar to AmericaSpeaks 21st Century Town Meetings® and the Center for Deliberative Democracy's Deliberative Polls®, in that WWViews asked participants to vote on various options after reviewing balanced briefing materials on climate change and deliberating in small facilitated groups, without seeking to reach consensus.
Unlike Deliberative Polls®, WWViews did not compare participants' responses to these options with the responses from a random sample of the wider population, and participants did not have an opportunity to question competing experts on the issues (Fishkin, n.d.), although additional 'knowledge' staff were present to answer factual questions based on the information in the briefing materials.
Unlike 21st Century Town Meetings®, WWViews asked participants to vote on a set of questions that were formulated beforehand, rather than developing the questions based on the deliberations amongst participants (AmericaSpeaks, 2002).WWViews also had much lower technological requirements than 21st Century Town Meetings®.WWViews used manually aggregated paper-based voting, and the process could have been conducted without any electronic equipment apart from a television, a DVD player and a dial-up Internet connection to upload the voting results.By contrast, 21st Century Town Meetings® use electronic voting pads to quickly obtain participants' votes on a particular issue or decide which issues to discuss further; and they have laptops with wireless Internet connections at each table that serve as 'electronic flipcharts' to enable rapid identification of key themes arising throughout all the small group deliberations and of relevant questions to be voted upon (AmericaSpeaks, 2010a).
While WWViews differed from other similar deliberative processes in these respects, the key distinguishing feature of WWViews was its scale.This was the first occasion in history that a citizen deliberation process was held on a global level (DBT, 2009a).It was fitting that the focus of this world-first event was climate change, an issue affecting everyone on the planet.
While this was the first global event of its kind, the scale of deliberative processes has been growing steadily around the world, and there are many examples of large-scale national and some trans-national events.For example, the first deliberative poll to be conducted across the entire European Union ('Tomorrow's Europe' in 2007) focused on the future of the EU (Notre Europe, 2007).AmericaSpeaks has engaged more than 160,000 people in large-scale participatory processes since 1995, on issues such as the rebuilding of New Orleans after Hurricane Katrina, health care reform, and childhood obesity.The largest 21st Century Town Meeting® conducted by AmericaSpeaks had 45,000 participants (AmericaSpeaks, 2010b).Also, while the scale of WWViews was global, with over 4,000 participants in total, the process was limited to a maximum of approximately 100 citizens at each of 44 locations rather than several thousand at some AmericaSpeaks events (DBT, 2009b).Consensus sought between all participants on at least some recommendations It should be noted that other large-scale deliberative forums have utilised processes markedly different than those of WWViews.For example, the National Issues Forums initiated by the Kettering Foundation do not ask participants to vote on various options, but rather help participants to do 'choice work' that grapples with the pros and cons of difficult issues, thereby enabling them to reach a more coherent set of views and an understanding of the 'common ground' they share with other participants (Gastil, 1994).This 'choice work' often continues after the facilitated event is over, and indeed National Issues Forums are sometimes conducted as a series of events to enable participants to engage more deeply with the choices at hand (Gastil, 1994).Moreover, National Issues Forums are not initiated by a single organisation and do not aim to roll out a standardised process across multiple sites.Rather, they are initiated by a large variety of different organisations that use prepared 'issue books' but are free to modify the process to suit their constituency and methodological preferences (Gastil, 1994;National Issues Forums, 2010).
Other initiatives have been established to enable global citizens to express their views on climate change, albeit not in a deliberative forum, and they have involved only those who have Internet access.These included a televised town hall meeting with world climate leaders in which questions were formulated and voted upon by online users (YouTube, 2009), and several online petitions, one of which obtained almost 15 million signatures (Avaaz, 2009).

Overview of the WWViews global project and Australian event
Two Danish government agencies, the Danish Board of Technology (DBT) and the Danish Cultural Institute, initiated WWViews "as a response to the emerging democratic gap between global policy makers and citizens, as more decisions become global in scale" (DBT, 2009b).COP15 was a clear example of global decision-making which would affect ordinary citizens from around the world yet was dominated by scientists, politicians and lobby groups.Therefore WWViews was designed to address "issues of immediate relevance to policy-makers" that drew on the voice of citizens at both the national and global levels (Danish Board of Technology, 2009b).
The project had over 50 National Partner organisations (for example, universities and NGOs with a focus on citizen engagement), who worked with the DBT to design the methodology and give feedback on the emerging process.The Institute for Sustainable Futures (ISF) at the University of Technology, Sydney was the National Partner for WWViews Australia, organising the event with support from public and private sector sponsors, volunteers, and participating citizens.
Ultimately, 44 events were held by local organisations in 38 participating countries,6 representing all the major players in climate change politics and many of the world's most populous nations.In total, roughly 4,000 people around the world voiced their opinions (DBT, 2009b).
Every country participating in WWViews followed the same standardised process, including deliberating on the same questions and collating responses in the same way, as specified in the WWViews Process Manual (DBT, 2009a).The reason for having a standardised process was to ensure the results could be readily compared across nations and regions, and easily communicated to policy makers (DBT, 2009b).The project managers from each National Partner organisation attended a training seminar in Copenhagen six months before the event to build a shared understanding of the project, and to establish uniform implementation methods and procedures for addressing cultural challenges (DBT, 2009b).
The global standardised process was designed to be financially and practically feasible to deliver in any country in the world.Therefore, it was not possible to use telecommunications technologies to enable virtual deliberation amongst dispersed participants.Rather, each event brought together participants in one location at the same time.The standardised process required that each event involved roughly 100 citizens who reflected the demographic diversity of their country or region, and who were not "experts on climate change, neither as scientists nor stakeholders" (DBT, 2009b, p.8).
All participants received the same background information from the DBT, including a background reading pack provided to them before the event (DBT, 2009d), and four information videos shown during the event (DBT, 2009g).The information materials were based largely on the latest assessment report published in 2007 by the Intergovernmental Panel on Climate Change (IPCC).While the materials were based on the consensus reached by the IPCC, they also reflected some contrasting views. 7The event was structured to have five core sessions, with individual country organisers able to add locally relevant warm-up exercises and introductory speeches to the program.Many countries ran the entire program over a single day, while some (including Australia) ran the program over a day and a half or longer (in Australia's case mainly to allow for breaks to reduce fatigue and give participants time to deliberate effectively).Box 1 shows the content and structure of the four themed dialogue sessions and the structure of the fifth session, which focused on recommendations.
Participants were divided into groups of 6 or 7 for each of the five facilitated deliberative sessions and they stayed in the same groups for all five sessions.Each of the first four sessions dealt with a theme (relating directly to the themes of the COP15 negotiations), questions and predefined response options.An example of one of the session questions and the answer options is included in Box 2. The fifth and final session had a different structure, with no set theme or questions.Instead, at this time, each small group reflected on discussions over the previous themed sessions and collectively formulated a brief recommendation they would wish to send to their climate negotiators (DBT, 2009a).The large group then came together to vote on which of these recommendations was to be the one put forward by WWViews Australia.
The standardised process was designed to allow participants in each location to express their own views on climate change without being influenced by the facilitators or event organisers, after hearing and considering the diverse views of their fellow citizens.The facilitators (or 'moderators' as the global project materials called them) were instructed not to express their own opinions, and additional staff were also present at each event to help answer any factual questions that arose during the group discussions.These dedicated 'knowledge' staff were instructed to only base their responses on the background information materials without introducing any additional information (DBT, 2009a).To minimise external influences on participants, the session questions were not divulged to them or to the public prior to the event, with each National Partner required to keep these confidential until the day.To ensure that all participants had access to the same information materials, participants were asked not to share additional written information external to the process with other participants during the deliberations.

Box 1 -WWViews program and session structure
Program overview: The WWViews sessions -consistent for all participating countries: Citizen Dialogue Session 1: Climate change and its consequences (1 hr mins) Citizen Dialogue Session 2: Long-term goals, urgency and commitment (1 hr 10 mins) Citizen Dialogue Session 3: Dealing with greenhouse gas emissions (1 hr mins) Citizen Dialogue Session 4: The economics of climate change (1 hr 10 mins) Recommendations Session (1 hr 45 mins including working tea-break) Citizen Dialogue Sessions: Each of the four themed sessions outlined above lasted for between 70 and minutes and followed the following structure: Video (5 to 15 mins) Presentation of the questions for discussion (5 mins) Group discussions (45 mins) Anonymous voting on pre-set response options to questions (15 mins) Recommendations Session: This session took 1 hr 45 mins including working tea-break and had the following structure: Developing recommendations in groups (45 mins) Reading all recommendations (45 mins over tea-break) Anonymous voting on 3 preferred recommendations (15 mins)

Box 2 -Example of questions
Each session (apart from the recommendation session) had pre-set questions for discussion and individual voting.Each session had between two and four questions.An example of a question and the pre-set responses for voting, from Citizen Dialogue Session 3: Dealing with greenhouse gas emissions is included below: Do you think the short-term reduction target for Annex 1 countries should be: Higher than 40%

Between 25% and 40%
Lower than 25% There should be no targets Don't know/do not wish to answer (All questions available at http://www.wwviews.org.au/uploads/wwviews%20questions.pdf ) The voting results and recommendations from each country were immediately posted on a central Internet database, enabling rapid public access to the emerging citizen views.An international forum for virtual information sharing amongst National Partners, also open to participants and the general public, was established in the form of a public blog (DBT, 2009e).The process used a range of technologies, some of them 'high tech' -relying on Internet access and computer technology -and others using more 'low tech' approaches -such as the manual voting and vote tally process (See Box 3 for a snapshot of these).
WWViews National Partners from each of the 38 participating countries were responsible for giving climate negotiators from their own country the results from their citizens' meeting before COP15 (DBT, 2009a).The results from WWViews Australia, including the votes received in response to each question and the recommendations formulated by the participants, were compiled in a full report on the Australian event (Atherton & Herriman, 2009).Additional international results, analysis and discussion are also available in the international Policy Report (DBT, 2009b).Customised data reports comparing the results from different countries, regions and groupings of countries can be produced instantaneously using a special Web tool that can be found at http://wwviews.org/.In addition, the organisers produced a short film about the Australian event (ISF, 2009d), as well as a longer documentary about the background of WWViews and the characteristics of this global process (DBT, 2009c).
Box 3 -Use of technology in WWViews 'High tech' Internet -resources were shared between participating country partners by email.
Websites -project intranet site created by DBT for global event and sharing material between international partners.Individual country websites created by some partners (including Australia) to provide information to participants, facilitators and media.
Social media including Facebook and Twitter -used to connect participants and supporters.Some partners used these more than others.DVD -short information videos were created centrally and used by all participating partners to begin each dialogue session during the event (see Box 1 for information on session structure).
Use of results spreadsheets for vote tally -in Australia voting results were entered into pre-prepared spreadsheets on computer to generate graphs of the Australian results to display during the event.
Comparative results database -a global Web-based results database allowed participating countries to enter their results in 'real time' and for these to be accessible without delay worldwide.
Live link-ups -the Australian WWViews event set up a video call with two other WWViews events (Denmark and England) using Skype, and projected it so that the participants could see each other and interact.

'Low tech'
Process manual -the DBT-produced process manual for National Partners was available in booklet form suitable for printing.
Participant background reading -was produced in hard copy and mailed to participants.
The deliberations were carried out face to face, with participants in the same location.
Voting and vote tallying -both individual vote casting and vote tallying at each table was done manually (pen and paper) before being centrally tallied and entered into the WWViews Web-based global results database (described above).

How the WWViews process was implemented in Australia
As described above, many aspects of the WWViews process were designed at the beginning of the project and documented in a detailed process manual for all participating National Partners and event organisers to follow.At the same time, some aspects of the global standardised WWViews process were not specified in detail, leaving a degree of flexibility for National Partner organisations to devise approaches that were most appropriate in their local contexts.
The Australian organisers, the Institute for Sustainable Futures, had firsthand experience designing and delivering a range of other deliberative processes (for example, CSIRO 2006;Dryzek, 2009;Herriman et al., 2007;Littleboy et al., 2006, Office of Population Health Genomics, 2009;NCCNSW, 2009;Riedy et al., 2006;White, 2001) and were also influenced by the reflections of other Australian practitioners.In the Australian context, and based on the organisers' collective experience, an important consideration was ensuring high-ranking politicians and civil servants saw the process to be credible, legitimate and unbiased.Another consideration was enabling Australian citizens to engage meaningfully in the process regardless of their personal background or financial capacity.These considerations guided important aspects of process design.
The next three sections discuss three key aspects of the Australian WWViews process that the organisers tailored based on their experience and objectives, namely: recruiting and supporting participants; training and coordinating facilitators; and disseminating the results and information about the process in order to influence policy.

Use of random selection and demographic matching
Each WWViews National Partner was permitted to design their own recruitment strategy, which the DBT reviewed and approved.As with the Danish style consensus conference (Grundahl 1995), the WWViews Process Manual specified that the group of participants should reflect the demographic diversity of the population (DBT, 2009a).The Process Manual also gave generic advice on recruitment while allowing for a variety of methods, such as face-to-face recruiting, advertising, snowball-sampling and inviting a random sample of the population (DBT, 2009a).Post-project evaluation carried out through interviews and informal communication with WWViews project managers in different countries reveals that in practice, partners in different countries did use widely varying recruitment methods, for example, advertising, using research students to do the recruiting or outsourcing recruitment to other organisations.
To ensure credibility in the Australian context, the organisers decided that it was crucial that the sample of participants selected to take part in the Australian WWViews event was as representative as possible for a small sample of 100 people (that is, broadly representative of the diversity of key demographic characteristics, while not statistically representative of the population).Any suggestion that the participants were 'hand-picked', or selected for their opinions on climate change, or that 'greenies' were over-represented for any reason, would have significantly undermined the legitimacy of the results.
The organisers rejected methods that could have increased the potential for 'selfselection' (for example, people actively applying to participate because they have strong opinions on the subject matter, leading to a biased sample group), such as advertising, invitation or snowball methods.The organisers opted for a random recruitment method that prioritised representation of demographically defined groups, since this approach was more likely to minimise self-selection (Carson & Martin, 2002).
In order for the WWViews sample to be considered representative and credible to national policy-makers it was particularly important to include participants from all Australian states and territories, as well as a representative mix of urban and rural participants.This led to an early decision that it would only be worth undertaking the event if it was nationwide.In the recruitment process, other important demographic factors included representation of Indigenous Australians and a mix of other characteristics including age, gender, income and education levels.The organisers also specifically excluded people working on climate change in a professional capacity, such as climate change scientists.
The initial recruitment was outsourced and used random direct dialling and demographic matching.After identifying a pool of interested people, 110 people were selected to take part based on achieving the best possible match to quotas in most demographic categories.The group of 105 who ultimately took part was truly diverse, representing a broad cross-section of Australian society, although young people were under-represented, as were people with less educational qualifications.The importance of engaging young people may have been under-emphasised in the WWViews process.Sarkissian, Hoffer et al. (2009) argue that an engaged citizenry "must not exclude any social, cultural or age group" and must promote the inclusion of everyone's knowledge as valid and valuable (p.78).They note that young people often don't become involved in community engagement processes because they find them "irrelevant, a waste of time and boring" and because they do not experience results relevant to their concerns (p.134).
These recruitment results therefore reflect commonly reported difficulties with engaging young people in such processes, and could indicate a lack of attention given by organisers to the issue of how best to engage young people in this process (beyond a general commitment to diverse representation and recruitment through random selection), despite much guidance material being available (see for example Driskell, 2002).

Measures to increase equitable access for participants
To support representation from all groups in society, the organisers tried to cover most of the costs for most of the people taking part in the event.They therefore committed to covering significant costs of participation including flights to Sydney from state capital cities and accommodation in Sydney.At the conclusion of the event, the organisers provided optional contributions toward other expenses involved in taking part, such as public transport and non-catered meals.There may still have been some people who were unable to participate for financial reasons such as being unable to afford time off work or transport from remote areas, but every effort was made to avoid that eventuality.

Extensive participant support and communication before and during the event
In addition, the organisers endeavoured to support participants throughout the process by providing tailored information packs, a dedicated participant support team, and a regular newsletter.The careful planning of supportive processes, with clear instructions, and accompanying information, was key to making the event a success.
A dedicated participant support role was created to deal with all participant queries and needs before, during and after the event.During the event, a participant support desk was open at all times, staffed by the same individual who fielded enquiries before the event, and by the ISF Ethics Officer.The support desk was available to address personal concerns that might have impaired participation, and the Ethics Officer was charged with responding to any ethical concerns that participants might have had (e.g., confidentiality issues).The organisers produced a newsletter seven times between August 2009 and December 2009 (both before and after the event), and sent it to participants, as well as sponsors, facilitators and other stakeholders (ISF 2009i).
The extensive participant and logistical support was directed towards maintaining diverse participation and ensuring empathetic and fair treatment of participants, and during the event it enabled table facilitators to focus exclusively on the deliberation process and the quality of dialogue.Because others were providing participants with practical support, facilitators did not have to be a point of contact for logistical issues.This design feature was added after hearing reflections from several practitioners based on their experiences with other processes which highlighted inadequate participant support as a risk factor in successful process delivery.

Making the process easier for participants by scheduling breaks
The WWViews Process Manual specified a very full one-day event schedule which did not include time for participants to have breaks as a group.Unlike Danish consensus conferences, the standard WWViews process also did not include preparatory sessions for participants in the weeks or months prior to the event (Grundahl 1995).The Australian organisers decided to extend the process so that it ran over one and a half days, to allow for scheduled breaks in which participants could debrief and rest between intense and sometimes emotional deliberations, and to allow time for minor adjustments to the facilitation where needed.
Starting the process a day early also enabled the participants to meet each other and 'break the ice' for just an afternoon, before meeting for the full day, and it allowed space for the facilitators to have a troubleshooting meeting on Friday night to discuss potential facilitation challenges in their groups.The role of facilitators is discussed in more detail later in this paper.
Van Kasteren and McKenna (2006) suggest longer sessions and pre-briefings as ways to overcome the risk of unbalanced representation of views within a process.
Adjusting the schedule to allow longer sessions and briefing participants beforehand may have enabled them to contribute more equally.

Participants' feedback about the process
At the end of the first day, participants were invited to give responses to the questions: "What was something great about today?", and "What is something to consider changing for tomorrow?"Of 103 participants, 20 people volunteered "something to be changed for tomorrow", and 58 people volunteered "something great about today".The feedback was actioned where possible, and at the beginning of the next day's session the lead facilitator explained to participants how the feedback had been addressed.At the end of the second day a detailed 'exit survey' was distributed to all participants.Participation in the survey was voluntary and anonymous, and the questions were almost entirely closed, with pre-set answer options.This same survey was distributed to WWViews participants at many events internationally.The survey included questions about their experiences of the event, their prior knowledge of climate change, their reflections on the results, and their thoughts about the value of processes like this for future policy applications.The survey had a 97% response rate among Australian participants.Some key results are described below.
The feedback from Australian participants was overwhelmingly positive: WWViews was seen as a learning opportunity, a chance to hear diverse views, and a way to have a say on a serious global issue.Participants felt privileged to be involved in this ground-breaking project.They felt processes like this could be used for future policy making (Atherton and Herriman 2009).
Virtually all survey respondents (99%) agreed with the statements, "The recommendation developed by my group reflected a thoughtful and open discussion", and, "In the dialogues, I was able to frankly communicate what I had in my mind".The overwhelming majority (95%) agreed that "I support the recommendations developed in my country".This indicates that the recommendations reflected the views of most participants, and that the process leading to the recommendations was inclusive and based on an open exchange of ideas (Atherton & Herriman 2009).
Additional comments in the open written feedback at the end of day one included: "Everyone remained respectful;" "Interesting and stimulating discussions;" and "Great cross pollination of ideas on global warming."One participant wrote that he/she appreciated "the chance to hear others' opinions and share my own.I'm enjoying the responsibility and privilege."Another noted, "It was a chance to be involved in an environment where people, whether they had convictions of views or scientific opinions, would be able to put them all on the table and talk about them."Feedback included a number of comments specifically praising the quality of facilitation and saying that it promoted open, respectful discussion in the small groups.
Participants' positive experiences at WWViews meant that they were supportive of holding further citizen deliberation processes on other issues, with virtually all survey respondents (99%) agreeing, "It's beneficial to continue dialogue processes such as the WWViews project in the future".
The importance of the recruitment process, with its focus on diversity, is highlighted by the large number of unsolicited responses by participants who reported that meeting and hearing from people from all walks of life was a key feature that they appreciated about the event.Their comments included: "Lovely to meet such a diverse bunch of Australians." "Surprisingly brilliant job of mixing up the cross-section of participants, definitely added to the interest and diversity of discussion." "Meeting people from a range of areas and different points of views has been very insightful and interesting." Nevertheless, a few participants commented verbally during the event that the full multicultural diversity of Australia did not seem to be well represented in the group.The ethnicity variable in the demographic quotas was defined in terms of three categories: Non-Indigenous born in Australia, Indigenous born in Australia, and born outside Australia (Atherton and Herriman 2009).This categorisation may have been too broad, because it did not differentiate between people of many different ethnicities who are born in Australia or overseas.The apparent preponderance of European Australians in the sample may have been due to language or cultural barriers that reduced the response rate amongst people from other ethnicities.A longer time period for recruitment would have been required to address this issue.

Training and coordinating facilitators 01,000
Another aspect of the process that was important for the legitimacy of the Australian event was facilitation.The standardised WWViews process specified that there would be a lead facilitator and that participants would be facilitated in small table groups with at least one facilitator per table.The standardised process allowed some flexibility in terms of who facilitators were and their level of experience, for example permitting adult students to play this role (DBT, 2009a).The process did not allow for alternative facilitation mechanisms, such as enabling participants to self-facilitate by providing them with guidance on how to do so, for example by using a World Café-style process.
Working within the standard framework prescribed by the international organisers, the organisers of WWViews Australia made decisions specifically to bolster the neutrality and quality, and hence the credibility, of the facilitation.They decided that for the Australian process to be credible, a high standard of facilitation would be required.This would give legitimacy to the process from the perspective of participants and would also help to ensure outcomes based on an open, transparent and professionally guided discussion.The organisers therefore opted for experienced professional facilitators, including an experienced professional lead facilitator, who was able to provide feedback on process design and other project decisions that affected facilitation and participant engagement.
The organisers were careful to minimize any actual or perceived bias in the facilitation.First, they specifically excluded facilitators representing organisations that may have been perceived to be biased on the issue of climate change, including the Institute for Sustainable Futures itself, and one of the sponsor organisations, WWF Australia, which is a well-known environmental NGO.In total 33 facilitators were recruited from sponsor organisations (other than WWF) and from independent facilitator and educator networks.
Second, facilitators were provided with detailed information packs, asked to sign Facilitator Agreements and trained in the WWViews process.Each of these measures emphasised the importance of neutral facilitation.Facilitators were clear that their role was to encourage participants to express their views, not to input their own views on climate change.
The facilitators underwent training prior to the event to ensure they understood the objectives and distinctive characteristics of the WWViews process and to ensure that the busy schedule would unfold smoothly.The facilitators were required to play an integral role in vote counting, by tallying the votes from their small groups and bringing these results quickly to a central point for aggregation.
Training events were conducted in Sydney and Melbourne, which also provided an opportunity for facilitators to meet each other and establish the rapport necessary for working well together as a team.The facilitators were encouraged to support each other and share learnings during the event.At the end of the first day a facilitator debrief session was held, where facilitators reflected on their experiences and shared ideas about how to resolve any emerging facilitation challenges.Briefings were held at the start of each day, and as needed -for example, a short session was held to discuss participant feedback that emerged from the Friday feedback forms, and another to discuss how the recommendation session would run.During the event the facilitators had a room that they could access at any time (when they were not needed at their tables), for example to spend quiet time with co-facilitators to revise their approach, plan their micro-processes or check the materials and prepare for upcoming sessions.

Facilitators' reflections on the process
There was limited formal feedback from facilitators at the event.Due to time constraints the organisers were unable to run an extensive debrief process immediately after the event, and instead this time was used to write reports and disseminate results.Systematic debriefing for facilitators soon after the close is something the organisers would recommend for future events.
The following comments are therefore based on anecdotal informationcomments made at facilitator training, in discussions at the facilitator debrief at the end of the first day, by facilitators in online forums, in emails of thanks after the event, in discussions with sponsor organisations who provided facilitators and in four completed post-event facilitator surveys.It appears the vast majority of facilitators found the event to be well organised and the process to be clear and easy to follow.Facilitators enjoyed working with each other and hearing diverse views from participants.
One facilitator commented that "it was a privilege to be in a room with people from across Australia, from all walks of life, who were willing to make this contribution of time and dedication of effort to such an important issue."Another facilitator noted "the fantastic organisation and co-ordination of the event," adding that "having more facilitators than was absolutely necessary meant the workload could be shared and The need for more time to engage in deeper deliberations and move towards consensus was identified by another facilitator.This could have enabled the small groups to cross-fertilise and share ideas with each other, and could have enabled the large group to combine elements of various recommendations into a more holistic and representative recommendation.Nevertheless, "What was achieved within the time and within the structure was impressive."Additionally, the positive feedback from participants can be seen as indicative of the quality and neutrality of the facilitation, which contributed to the actual and perceived credibility of the Australian event.
With the exception of the lead facilitator, who was a member of the core project team, all facilitators provided their time on a voluntary basis and covered their own expenses to attend the event and the training.Some were supported by their employers and others participated independently.The organisers were exceptionally fortunate that highly experienced and skilled facilitators were willing to donate their time to the project.It is unfortunate that events of this nature often need to rely on volunteer facilitators and it is important to acknowledge the dedication and commitment of professional facilitators in voluntarily contributing to the ongoing development of deliberative democracy processes in Australia.Facilitation is an essential, and regrettably often undervalued, role.

Disseminating the results and process
The organisers developed a comprehensive dissemination strategy to influence government decision-makers directly by informing them about the process and the results in face-to-face meetings, and indirectly through stakeholder communication, mass media, social media, and supporting participant outreach.A further objective of the strategy was to support dialogue on climate change in the public sphere, again with a view to influencing policy indirectly.
The strategy also involved providing information about the WWViews process to businesses, to teaching and learning institutions, to professionals from varied fields, to researchers, and to citizens.The strategy had a further stated objective of promoting critical reflection on citizen dialogue processes and supporting the objective of effectively managing relationships and expectations of participants, facilitators and sponsors of the event.
The dissemination strategy consisted of four separate but interlinked strategies: -a political engagement strategy -a communications strategy -a media strategy -a research strategy i) Political engagement strategy for direct engagement with politicians and policy staff: The organisers sought face-to-face meetings with the individuals with whom they most wanted to share the results and process of WWViews and with whom they considered they would have the best chance of securing meetings.These were the people for whom the organisers judged the results and process to be most relevant -namely government climate policy-makers and negotiators, other influential government officials, and politicians with an interest in citizen engagement.To avoid perceptions of bias, and to extend influence, direct engagement with the Australian Government, the opposition9 and the Australian Greens Party was an important element of the strategy.
The organisers made early contact with key politicians and civil servants to inform them about the project and to discuss it in advance of the event itself.As a result, the Federal Minister for Climate Change and Water, Penny Wong, provided a letter endorsing the event and prepared a video message for participants; and Australia's Climate Change Ambassador, Louise Hand, spoke in person at the event.
Unfortunately the organisers' access to politicians and climate change negotiators during their dissemination efforts (October and November 2009) was limited due to preparations for COP15 and the (ultimately unsuccessful) passage through federal parliament of domestic climate change policy.
Nevertheless the organisers successfully arranged several face-to-face meetings which were held after the WWViews event, including meetings with three public servants in the Department of Climate Change (one of whom was a member of the COP15 negotiating team); the adviser on climate change to the Minister for Infrastructure, Transport, Regional Development and Local Government; an adviser to the Federal Opposition Spokesman on Emissions Trading; the Australian Greens Deputy Leader; and the Lord Mayor of Sydney, who chaired a session on citizen participation at the Copenhagen Mayors' Summit during COP15.
In addition to this targeted approach, the organisers sought to extend the impact of the project indirectly by informing as many people as possible in positions of influence about the results and process.The organisers invited politicians and other interested parties to attend the closing drinks function at the end of the event, although only a few accepted the invitation.The organisers also mailed the results report (or a four-page summary) directly to all federal politicians (MPs and Senators), all state government ministers and selected state government MPs, and senior federal and state civil servants, including federal climate negotiators.
As well as being a key principle of effective engagement (reference), past experiences in organising deliberative processes had demonstrated the importance of being clear about the commitment to, and limitations of, any planned dissemination strategy.Understanding the commitment that organisers are making in relation to communicating their recommendations to decision makers is important for their successful participation, and for their satisfaction with involvement in the event.After the event, the organisers provided participants with a reading list of resources about climate change (ISF, 2009e) and information on how to approach their local political representatives, including letter-writing tips and a letter template they could use to write a letter to their local politician if they chose to do so (ISF, 2009f).The information did not advocate a particular position, but rather sought to assist participants who wanted to help to disseminate the results of the process or to further expand their own knowledge.
ii) Communications strategy for indirect influence through existing networks of NGOs, academics, businesses, sponsors, supporters, facilitators and participants: The WWViews Australia website was supplemented by a regular newsletter that went out to participants, sponsors, facilitators and other stakeholders.
Seven of these newsletters were produced over the life of the project.The organisers also used these avenues to distribute a short documentary film about the Australian event, featuring results and participant, facilitator and sponsor interviews.They published it on DVD and on the website (ISF, 2009d).
iii) Media strategy for indirect influence via mass media and social media: The organisers engaged specialist media and communications assistance to increase media coverage and to assist with implementation of the communications strategy.The media strategy included engagement with mainstream media -television, radio and press and use of social networking sites such as Twitter and Facebook (see for example DBT, 2009f).The organisers developed key media messages that were repeated in press releases and interviews.The organisers were successful in obtaining prime time news coverage on the day of the event on a national television channel, national and regional radio interviews, and articles in several leading state-based newspapers (see for example ABC TV 2009;ABC Radio National, 2009;Radio 5AA, 2009;Gordon, 2009;Munro, 2009).
They also achieved substantial regional news coverage by focusing on the personal stories of WWViews participants (see for example Ocean Grove Echo, 2009;Gardiner, 2009;Castlemaine Mail, 2009;Bendigo Advertiser, 2009).The organisers also helped participants to publicise the results in their local media and communities by producing resource materials such as a media release template (ISF, 2009g) and a PowerPoint presentation (ISF, 2009h).iv) Research strategy for indirect influence via research publications: By publishing academic papers and delivering conference papers on the WWViews process and results, the organisers plan to raise the profile of citizens' deliberation in general, to cross-pollinate process ideas with other practitioners and to communicate the views of the world's citizens to a broader audience.
The organisers' experience of the face-to-face meetings with government officials showed that it was very valuable to invest time in speaking to people directly about the project.For future projects, where resources are limited, a strategy of engaging directly with a small number of targeted individuals may therefore be as effective as broad dissemination.
The organisers found that both their NGO and business sponsors were excellent sources of information about accessing political processes, and they had considerable experience and existing contacts within spheres of government (Ikenberg and Petersen personal communication, 2009).The organisers' expertise in planning and designing deliberative events that are independent and unbiased complemented the sponsors' expertise in advocating for political change.However, to maintain a sense of neutrality on outcomes, and to emphasise the 'research' status (and therefore the credibility of the results) the organisers chose to make initial contact with politicians themselves, as the Australian WWViews National Partner.
If the organisers were to arrange a similar event in future, they believe it would be better to try to arrange direct briefings or even to run 'pre-event briefings'.The authors reflect that for an event of this nature it is important that key stakeholders have heard of the event and have a sense of its process before seeing the outcomes.This is especially important when the dissemination process seeks to educate policy makers on the process and its potential as well as the outcomes.
However, ultimately such efforts are at the mercy of changing circumstances and diaries.We further reflect that such a comprehensive and coordinated dissemination strategy would be much less important when the process is commissioned, auspiced or even sponsored directly by those responsible for making decisions about the issues discussed through the process, which in this case would have been the Australian Government Department of Climate Change.

Discussion and critical questions
It is encouraging to see that a deliberative process can be effectively coordinated at a global scale to address a global issue.WWViews demonstrates how ordinary citizens can, despite national, cultural and personal differences, reach agreement on complex, dynamic and plural or 'wicked' problems (Rittel & Webber 1973) when they are given the opportunity to discuss the issue with access to good information.Overall, the results from different countries across all questions show Box 2, regarding a short-term reduction target for Annex 1 countries, on average 89% of participants globally voted for a target of 25% or higher, which was identical to the vote of the Australian participants.People enjoy hearing other points of view, and they can work together to generate thoughtful responses to complex policy issues relevant to all.They are keen to know that their voices will be communicated to decision-makers, and some are willing to contribute their political influence.
The legitimacy of the Australian event was heightened by focusing effort on impartially selecting participants, providing financial and other support for participants, and training and coordinating facilitators.
WWViews advanced the scholarship on deliberation through the wealth of research and publications it gave rise to.Organisers from many participating countries undertook individual research projects based on WWViews, the results of which will contribute to scholarship on the subject within those countries, and where relevant, in international publications.A book on the WWViews experience has been compiled by several researchers involved in the event, including chapters from participating countries and research organisations (Rask et al., 2011).The research on WWViews may be of particular importance in countries where deliberative processes are not common.The results of the process itself provide interesting data on the subject matter, and the post-event participant survey that was conducted in most countries provides useful data on aspects of the process.No doubt this book will be an invaluable contribution to the literature on deliberative processes.
Nevertheless, a number of critical questions arise from the experience of running the Australian event, and from reflecting on the global WWViews process: i) How successful was WWViews as the first ever global-scale citizen deliberation process?
WWViews was successful in conducting a large number of deliberative events simultaneously around the world, with the results immediately available for analysis and comparison on the Internet.The Australian experience of being part of this global process showed that it was extremely useful to have a standardised process developed prior to starting to organise the event.The support structures of websites, handbook, database for reporting, and sharing information with other countries were all invaluable to organisers.In terms of influencing the Australian Government with the results of the Australian population, the organisers found there to be considerable interest in how these results compared to those from other participating countries, suggesting that for a global issue such as climate change when countries are grappling with their negotiating position, information about the views of citizens in other countries is useful.
WWViews was reasonably successful in producing a sense of interconnectedness around the globe.The briefing materials provided to all participants aimed to get them to think at the global scale with an emphasis on global issues rather than the national context.The global reach and significance of the event was highlighted for participants by connecting with events happening elsewhere in the word through live link-up, and by the access to the comparative global results database.This tangible connection with citizens around the world who were involved in the same process created a strong sense of excitement in the Australian participants; the video link-up was one of the highlights of the closing drinks event, and demonstrates the potential for technology to connect geographically dispersed processes in creative ways.The global WWViews process also made use of social media such as Facebook, and participants around the globe could connect with each other through these media.Some participants were also given the opportunity to travel to Copenhagen during COP15.However, the degree to which participants in different countries felt connected to the global process is likely to have been impacted by the steps taken by national organisers to help them make that connection and by their own ability to take advantage of the available resources, particularly as the process was not designed to allow for direct deliberation between participants in different countries.
Certainly, WWViews produced a sense of connectedness for the event organisers through the training event in Copenhagen, regular newsletters from the DBT, and an online discussions forum, as well as the obvious collaborative nature of the entire project and associated research projects.
Danish participatory consensus conferences are not always evaluated (Nielson et al., 2006;Guston, 1999), and with WWViews there was no centralised evaluation strategy in place to assess the quality of the process or its political impact, and nor was there guidance for individual National Partners on how to do so at the local level.It was relatively easy for each National Partner to evaluate the proximate goals of delivering a well-run, credible event but it is much more difficult to evaluate how successful the process was in achieving the ultimate goals of influencing national climate negotiators and the outcomes of COP15.Neither the Australian Partner nor the DBT had funding to evaluate the success of WWViews in achieving its ultimate goals.In Australia, the organisers did not establish clear success criteria for the event at the beginning, or ways to monitor the effectiveness of the dissemination strategy, despite the growing body of work on evaluation frameworks for such processes (Edwards et al., 2008;Rowe & Frewer, 2004;Guston, 1999).It would have been extremely useful, at both the national and global levels, to design an evaluation framework to assess how successful WWViews was in achieving its ultimate goal of enabling citizen views to influence climate change policy.
ii) How meaningful is the comparison of results from different countries?
By ensuring that all participants deliberated on the same set of questions and received the same background information, the standardised global process was intended to make it possible to compare the results from different countries.There was some effort to safeguard the results from being influenced by extraneous local factors, such as facilitators inputting their own views or participants being presented with other sources of information.
However, as this paper has described in the Australian context, there were several elements of the process that were left open to interpretation by the National Partners, including recruitment, facilitation and participant support.We don't know how the results of individual events may have been affected by, for example, different recruitment methods (including advertising) or having facilitators with little or no experience.Moreover, some WWViews events only involved participants from a small regional area rather than the whole country, and the potential for representativeness with a sample of 100 people varies tremendously depending on the size of the national population (for example, contrast two countries that each had roughly 100 citizens participating in WWViews: China, population 1.3 billion, and the island nation of St Lucia, population 161,000).
iii) Can global citizen deliberation processes ever be standardised across local contexts?
Setting aside the variability in the national implementations of the WWViews process, it is worth reflecting on the extent to which a single standardised process can ever be uniformly implemented in very different local contexts.The cultural, social, geographical and technological context for deliberative processes in Denmark is very different to the situation in many other countries that participated in WWViews, and this undoubtedly affected the implementation of the standardised process.
The availability of technology is an obvious limiting factor, with Internet access and audiovisual technology both prerequisites for any global deliberative process.The simple issue of geography and transport is a limiting factor within many countries when participants are required to meet in a single location.The process, designed as it was in an economically wealthy, industrialised nation, was inevitably culturally biased in ways that are not always obvious.For example, written information can be difficult to translate to cultures where the primary mechanism of knowledge sharing is oral tradition.Fundamentally, most deliberative processes require participants to have a certain level of functional literacy so they can understand written background information materials, although the global WWViews process did encourage verbal sharing of background information where required.Linguistic diversity in many countries poses a further challenge for representative participation.Choosing a single language for information provision and deliberation in many countries will necessarily exclude large sections of the population or will require translators to be involved, with the associated risks of mis-communication.Cultural differences affect the practice of deliberation, the understanding of its objectives, and even the segments of the population that are permitted to participate -consider, for example, the challenges of open discussion in societies with fundamental power imbalances, such as unequal gender relations, or countries where political debate is not condoned, and where questioning the status quo could be dangerous for participants.
In Denmark there is some cultural acceptance of deliberative processes and some history of their practice.There is also therefore experience in Denmark of organising such processes.Other countries participating in WWViews did not have this same culture, history and experience, and this is bound to significantly affect the implementation of such a process.Finally and perhaps most importantly, people who are extremely poor, displaced or otherwise vulnerable are likely to be unable or unwilling to participate in any global deliberative process.In countries where a large proportion of the population struggles simply to survive each day, those who do end up participating are therefore less likely to be representative of the general population.
We observe that the many different objectives for such a process can exert tensions -between national objectives and global objectives, and between design decisions intended to enhance deliberativeness and decisions which are directed at engaging decision-makers and evoking a sense of credibility that is culturally appropriate.In this case it is possible that global standardisation increased the credibility of the process among national policy makers (in some countries) but at the same time reduced the capacity for the process to adapt to cultural characteristics (in other countries) and it may also have reduced the deliberative nature of participation.WWViews provided valuable insights for everyone involved on the challenges of implementing a standardised process across so many different cultures.Nonetheless, as a first attempt to do so, it was more successful than any of the organisers had dared to hope for.
However, there are concerns that global deliberative democracy may not impose enough accountability upon participants, and that a greatly expanded form of representative democracy therefore offers a more robust process for global decision-making.One advocate of global representative democracy is George Monbiot (2002), who argues for the establishment of a World Parliament with the number of representatives from each country in the world proportional to that country's population.He believes that randomly selected citizens deliberating together will not necessarily make responsible decisions on global issues because it is difficult to punish them if they make poor decisions (Monbiot & Carson, 2003).Monbiot argues that they cannot lose their job as a consequence, because they are not professional politicians, and the threat of other forms of punishment may repel citizens from wanting to join the deliberation in the first place.
On the other hand, the potential for corruption is greatly reduced when citizens come together for a short time to deliberate on a specific issue (Carson & Martin, 2002).Furthermore, representative democracy has its own serious problems, at least in the way that it is implemented in most nation states today.There is the potential for an elected World Parliament to be corrupted, on an even bigger scale than is evident in national parliaments, by lobby group power, by party politics and by elections based on money and celebrity (Monbiot & Carson, 2003).
Global deliberative democracy can, at the very least, inject more democratic perspectives into the existing global decision-making processes.The deliberations of randomly selected, ordinary, well-informed citizens can almost certainly lead to more democratic decision-making compared to the processes currently operating in the international sphere, where everyday voices are missing.WWViews was an important first step in demonstrating how it can be done.There is much interesting work ahead in refining and improving on the implementation of global-scale deliberative processes, and it should be reiterated that in the absence of an in-depth global evaluation of the process it is difficult to make judgements about the ultimate effectiveness of WWViews.

Table 1 .
Features of different approaches

Table 2 .
Table shows the percentages of the sample who were in each of five age categories alongside the percentages of the general population in those age categories.It can be seen that the percentage of people under 35 years of age in the sample (19%) was significantly lower than the percentage of people under 34 in the general population (35.7%).Comparison of actual participants to the quota for each age category (Data source: Atherton & Herriman, 2009).
Different roles were established within the facilitation team: lead facilitator, roving facilitators and table facilitators.The lead facilitator delivered facilitation training, provided direction and guidance to the facilitation team, and facilitated those sections of the event that involved all 105 participants simultaneously.Twenty-five table facilitators worked mostly in pairs to facilitate deliberations at each table of six to eight participants.There were another eight roving facilitators overseeing the progress of deliberations.The roving facilitators were available to help the table facilitators keep their groups focused.They relayed process questions from table facilitators to the lead facilitator, and they acted as temporary substitute table facilitators when needed.They also collected the tallied votes from each table and brought them to the central point for aggregation.Many processes used in WWViews Australia were selected to help support facilitators and participants, based on their successful prior use in Australia and elsewhere (particularly the US).These included the use of 'roving facilitators' to help troubleshoot facilitation issues; the use of large coloured 'knowledge cards', 'process cards' and 'logistics cards' for table facilitators to indicate when they needed assistance; and the use of a large electronic 'countdown' clock during deliberation sessions to help facilitators and participants manage time.Facilitators were supported by additional staff with roles directly relating to process: there were two dedicated 'knowledge people' 8 who responded to factual climate change questions, two dedicated vote tallying and reporting staff (who tallied votes, entered results into the DBT database and reported back to participants throughout the day on what the Australian results had been), and a time keeper whose role included stage management and liaison with dedicated AV/IT staff on technology issues.
table facilitators could get breaks if they needed it or use it as a technique to introduce a new dynamic to break a 'stuck' table discussion."