Rosemary Bechler (RB): We finished Part One of this conversation with your comment that Gordon Brown’s call for a UK-wide rolling programme of citizens assemblies had ‘muddied the waters’. With interest in and news of ‘citizens assemblies’ springing up on all sides in 2019 – for example the Grand National Debate organised by President Macron in France officially included 12 citizens assemblies, and proposals for them are in key manifestos for the UK general election –­ we need to be clear about the defining features of a successful assembly. Could you take us through some of these do’s and don’ts?

Graham Smith (GS): Citizens assemblies work best when they have a well-defined task, and where the relationship with political decision-making is clear. Participants need to know what they are being asked to do, and in what ways their proposals will be integrated into the political process. There are all sorts of questions about how you actually design a particular citizens assembly, but these two criteria are really critical.

In both Gordon Brown’s call and Macron’s Grand Debate, the tasks seem to be so wide-ranging. They are trying to get the citizens assembly to resolve too many issues at once – all the problems of France, or all those of the UK. For Macron, one assembly was asked to look at tax reform and six or seven different issues. And this was really rushed. The announcement that the assemblies were going to happen and the end of the Grand Debate were only a few months apart. It was too quick and unfocused.

What is happening now with the French Citizens’ Convention for the Climate and the forthcoming UK Climate Assembly is more clarity in their ask of citizens and their relationship with decision making. At the moment, there’s a real danger that politicians are calling for citizens assemblies without having a clear question, rather than starting with the problem they wish to solve and working out whether a citizens’ assembly is the right way to deal with the issue. A defined task is essential. In short, I am not sure what role citizens assemblies really have to play in a more general national conversation.

RB: Gordon Brown said that he wanted to form ‘a new national consensus’ through a rolling programme of citizens assemblies. So the problem is that it is simply not clear how this consensus is meant to come about?

GS: As I said, I worry about a lack of a focused question. It is not clear how Gordon Brown and others see such a conversation working. I have heard the proposal that a number of local assemblies should take place that would then feed into a national citizens’ assembly or summit. This borrows from some ideas about how this might work at a European level – with national citizens’ assemblies feeding into a European assembly. When people speak about this model, they usually suggest that representatives from the more local level will make up the assembly at the higher level.

I worry about this idea. Representatives from the local assemblies will feel mandated by the outcomes established locally. This will undermine the deliberative nature of the summit, because they will feel themselves to be representatives of their locality. What we want from participants in a citizens’ assembly is a willingness to think about the common interest. If people come into it with a strong mandate, the assembly will take on a different character. It’s a representative assembly, not a citizens’ assembly.

A second alternative makes more sense, where local assemblies generate ideas and agendas which are picked up by a national convention that is also selected randomly. Then the representation problem disappears. But, this model may generate a new set of problems: it may be quite difficult to motivate people to engage in local assemblies with no obvious and direct political result. What you are asking people to do is to give up a significant amount of time to make proposals that will then go somewhere else for other people to think about. Usually, the promise to participants is: “Engage with this process and your conclusions will then have the following impact.”

I was involved in the design process for the Citizens’ Convention on UK Democracy that Graham Allan and others initiated. We had some discussion about regional assemblies. We wrestled with the following kinds of problems. Imagine having regional assemblies discussing UK-level electoral reform. Each of them might come up with a different preferred electoral process. Where does the coordinating body go from there, with conflicting mandates from different regions? The stronger the demands of the local assemblies the harder it is for the central body to manage. If you are thinking about a national electoral system, it clearly needs to be a national citizens assembly. An assembly should directly relate to the scale of the issue that it is designed to affect.

RB: Have regional assemblies then been excluded altogether from Graham Allan’s constitutional convention plans?

GS: I don’t think anything is excluded. The whole process is premised on a public conversation that would last a number of months. During that conversation it is possible that regional randomly selected bodies would take place looking at particular issues. But at the moment the thinking is that the conversation would lead to a series of national-level assemblies on specific aspects of the constitution which would themselves have been generated through the conversation. A separate agenda-setting body with large numbers of randomly selected citizens would be tasked with deciding on the emerging issues the first assemblies should look at. This would be a one to two day initiative, compared to the longer citizens’ assemblies.

RB: When you say ‘national’ there, you mean English, or UK?

GS: That’s another interesting question. If the issue is UK-focused then you would want a UK-wide assembly. If it is concerning English governance, then you need an English assembly. You certainly don’t want a UK assembly making recommendations on issues that are devolved to Scotland, Wales or Ireland, or specific to England.

RB: Isn’t this particularly important if we return to the challenge of finding a democratic resolution of the Brexit issue – by which I mean not only resolving the binary issue of the Brexit referendum, but resolving all the constitutional issues which have been thrown up by the crisis following the referendum result.

It seems very clear that the nations of the UK have had almost as little and as unequal an input into this process of decision-making as the English regions. Regions, such as ‘the northern region’ of England have, by some accounts, been much ‘talked about, but not listened to’ across the political class, in ways that actually distort what is at stake for northern voters.

So this remains a delicate and challenging issue – how to have a UK-wide debate that takes into account the different, perhaps shifting and largely unexamined priorities of the regions and nations?

It might be a motivational problem as you say, but do you think if you said to people, please participate in this regional citizens assembly to set the priorities for your region. This is your chance – maybe your first real chance in over three years – to have your voices heard and taken seriously in the Brexit debate, to be listened to and not just ‘talked about’ – don’t you think that could work?

GS: I hear what you are a saying, but what you are proposing seems to be a change in the way we have used citizens assemblies effectively up ‘til now. There is no reason why we shouldn’t use deliberative processes to do the kind of work that you are describing. But I wonder whether citizens’ assemblies are the right approach to the problems you raise. Doesn’t the regional and national conversation you are after need to be much more of a process of mass engagement?

More relevant seems to be the work of organisations like Everyday Democracy in the United States. Everyday Democracy seeds dialogues in communities which, for various reasons, are suffering structural racial injustice. It enables conversations to work towards common understandings, or at least understandings of difference, building capacity within those communities for facilitated conversations. This is a very different process from a citizens’ assembly. It is doing very different work.

It is true that one of the most striking things about the Citizens Assembly on Brexit that we ran just after the referendum in 2017 was how many people said that until that event they hadn’t spoken in depth to someone with a different view from them. That was great for the 50 participants in our Assembly. But it had little or no effect for everyone else around them who didn’t know that it was happening and could not be party to those conversations. They are often still living in their polarised filter bubbles. Set pieces like citizens assemblies could be helpful, but when it comes to national renewal, I must say I think too much expectation is being placed on the citizens assembly model.

The Citizens’ Convention on UK Democracy is a great idea in terms of addressing how we might rethink the UK constitution. But in itself, having a new constitution doesn’t heal a country. It is interesting that in that project separate thinking is going on about the kind of national conversation that is needed, prior to any citizens’ assembly process.

The Citizens’ Assembly of Scotland will be an interesting initiative to watch. It has been asked to consider “What kind of country are we seeking to build?” and “How best can we overcome the challenges Scotland and the world face in the 21st century, including those arising from Brexit?”. I guess this is the sort of approach you have in mind. The Scottish Government has made a significant investment in the Assembly and it will be interesting to see whether it captures public imagination and the kind of impact it has. The questions are very broad and open which is unusual for a citizens’ assembly.

Citizens assemblies are flavour of the month at the moment, but we should be thinking much more fundamentally about the particular problems we face and the institutional designs we need in order to help solve that problem. Not just thinking citizens’ assemblies will solve all our ills.

RB: What many of us who are concerned with the state of our democracies are trying to work out is what does it mean to “go back to the people”, if you rule out a repeat binary referendum that only ends up leaving at least half the population bitterly disappointed? How can you go back to the people inclusively?

Here, the citizens assembly holds attractions that are surely not confined to the ability to take a clearly defined problem and resolve it through deliberation. So let me just try and push back on your question mark over the role of a citizens assembly in a national debate or conversation.

You ask for a clear task, and there are many calls for clarity of outcome in the ongoing Brexit furore. Usually these calls for a clear outcome are uncomfortably closely related to the ‘winner take all’ emphasis in the knock-out battle which is first past the post – a binary tug-of-war in which the strongest wins.

But whatever the result of Brexit, instead of this ‘clarity’, isn’t it the case that what we need is a coming together of the different sides, with the patience to listen to each other, and precisely a will to find a solution that is in the common interests of people across the UK?

Couldn’t a rolling regional and national programme of assemblies, composed randomly but with the proportion of leave and remain voters reflecting the first referendum result, respect that verdict much better by arriving together at a clear set of priority issues for each area. And wouldn’t this really help to inform the parliamentary process and the years of negotiation of the political settlement with the EU that lie ahead.

We would argue that that set of considered judgements would be much closer to ‘taking back control’ democratically than any ‘winner take all’ result could ever be.

This is why we are so interested in the Irish Citizens Assembly, not because it solves everything, but because it really seems to create a space for consensus-building. Actually being able to show that you can belong to a community for a short time and can get things sorted – is very precious isn’t it? It shows people another way of belonging. It boosts our confidence. And this ‘belonging’ is all too rare in a society of fragmented cultures, polarising communication, and isolated people.

Finally, it raises for me this question that we didn’t really go into before: how was it that the Irish Assembly, as a set piece as you say, seemed to help to create a better public debate in Ireland? Is there some relationship between the synergy in the room, the size and cohesion of the constituency which is at stake, the media coverage, which allows it to give people a unique glimpse of a different, more considered way for them to deal with such difficult issues?

GS: I’m not sure I can answer your question about Ireland. You’d need to talk to someone who was more intimately involved in the Assembly and the broader public debate it helped initiate. There is though a difference between the specific question of the constitutional status of abortion and same sex marriage and broader concerns about the ‘future of democracy’. (Although I am sure that for many people in Ireland those two things are connected. ) These are issues where there was public recognition that they needed to be dealt with.

Part of the problem with the broader issues of political identity is that people don’t necessarily think of them as issues that need to be dealt with. You and I might see them that way. But they are not so substantial that people can really grab hold of them in the same way as specific policy or constitutional issues.

RB: Say in the north-west of England, or in Wales, if people were offered the chance – “we need some of you randomly selected within a demographic range, after these three years of Brexit stalemate, to come together now and talk about what really matters to you for the future of your region? ” – do you think both the participants and the audience for that regional or national event would feel that was too insubstantial a question?

GS: No I think that could still work. The issue is – you are asking these people to give up a lot of their time. What is your promise to them? In a sense we had this problem with the Brexit Assembly in Manchester in 2017. What is the relationship to power, to political change, for the participant?

RB: Yes, and as you said, you came up with a very interesting complication there in your outcomes, which was that a set of recommendations that were no good at all as a negotiating base with the EU were nevertheless a very interesting set of recommendations…

GS: Exactly. So what I go back to is this – will you get the kind of engagement you want when there is no clear relationship between the discussions and any political authority? With the Brexit Assembly in 2017, I think we hit a moment when people wanted to talk about Brexit, even though we could not promise a link to political decision making. Can you imagine!

There is a danger of expecting a citizens’ assembly to play far too many roles: changing the minds of elites; bringing divided communities together; addressing longstanding constitutional questions; solving Brexit; encouraging a willingness to engage in open-ended conversations. The pressure is on citizens’ assemblies to solve all our long-standing problems. This is too much! I am cautious about over-extending the model and not getting what we expect out of it. I am certainly not able to employ the power of Jim Fishkin who copyrighted deliberative polling so that he could decide on the circumstances in which it should be used. People will try to use this method for all sorts of things and we will learn more about its breaking point. I am just wondering if this is the right method for the set of problems that you are concerned with.

Other methods may be better suited – like Everyday Democracy’s approach that I mentioned earlier. Also of interest is the G1000 model that has been used in a number of localities in the Netherlands. To confuse matters, it is different from the earlier G1000 in Belgium! In the Netherlands, G1000s bring one thousand people together with an open agenda: what needs to be done in our town or city? I believe G1000s invite around six hundred randomly selected citizens, but also one hundred political officials and civil servants, one hundred people from civil society organisations, one hundred civic entrepreneurs, one hundred from the business community. The organisers talk of “the system in one room”. The idea is that by the end of the day the process generates up to 10 action groups with proposals for what should happen next. It has some random selection in there, some deliberation – but it is very large and very concentrated. We have all sorts of different deliberative models out there that achieve different types of outcomes, not just citizens’ assemblies.

RB: We need to look at more of these. In the meantime, how do you now assess the Brexit initiatives with citizens assemblies that you have been involved in at a parliamentary level?

GS: Towards the end of 2018 and earlier this year, for the politicians and activists who were meeting, the question was – given parliamentary deadlock, what do we do now?

There were a couple of problems with the proposal for a citizens’ assembly at that point. We did not seem to have either the time or the political will across the party divides in parliament for such an assembly. The second problem was that most of the people associated with the idea were perceived as ‘remainers’, or people who had campaigned for remain but were representing leave constituencies. For the citizens’ assembly model to work, we needed it to be embraced by the different sides of the debate. I think there was suspicion that it was a vehicle to get to a second referendum. For those of us who have worked on citizens assemblies, the question was always: Can a citizens’ assembly be a way of breaking this political deadlock? One option for the assembly to consider might have been a referendum, but there would have to have been other options as well.

One of the reasons why quite a few significant figures in this space thought this initiative wasn’t a good idea was because of the pressures it would place on the citizens’ assembly model. Could we really trust the political parties to give it the time and space that would be needed, as we saw in Ireland? It would require politics as usual to be suspended for a period of time while we allowed this process to get on with its work. There was and there remains no sign of that happening.

RB: For that, you would need the newly elected prime minister Boris Johnson to have announced his support for such a process.

GS: That would certainly have made a difference (laughs) and I can’t see it happening. Although Rory Stewart did run on that idea when he stood for the Tory leadership. I didn’t follow his proposal that closely, but at one stage he seemed to be suggesting that he wanted a citizens’ assembly, but if it didn’t come up with the right advice as he saw it, then he was going to ignore it! I am sure this is a misrepresentation of his position, but that is not a good basis on which to set up a citizens’ assembly.

Again, returning to the Brexit debate, there is concern that the moving parts of Brexit are too complicated for a citizens’ assembly. I happen to think you could find a way of dealing with even this toxic political issue if we had the right political conditions. Unfortunately those conditions aren’t there at the moment. I think we’ve missed the boat on that one.

RB: If I am right – both Rory Stewart and Neal Lawson were calling for a Brexit citizens assembly made up of 500 people. Does this inflation in numbers of participants reflect an ongoing uncertainty about whether it is the numbers involved or the quality of the deliberation that is the most important element for a democratic process?

GS: Talking up the numbers is an issue. The largest citizens assembly to date has had 160 participants, in British Columbia. We have seen deliberative processes with 500 people in them but they only last one weekend. Or even a 1000, but only for one day. The larger numbers look more politically significant. But you don’t need to have five hundred people in the room to enable a diverse and reflective deliberation. The larger the numbers the more difficult it is to facilitate the process – and of course it adds to the expense.

The tension between numbers and quality deliberation will always be there. For example, I have been involved in discussions at the European level about how to conduct a transnational citizens assembly. Given the complexity and diversity of Europe, you might actually need at least 500 participants to capture social diversity. But at a national level, you don’t need that.

There may be good reason for having large numbers if you are doing something ideational and short. If you want to generate ideas very quickly, the more people brainstorming the better. But if you want to do detailed policy work, smaller is more effective. My guess is that smaller citizens’ juries outperform larger citizens’ assemblies on particular issues precisely because twenty or thirty people can be much more flexible over how they deliberate and how they develop a culture of collaboration.

RB. I know you have been involved in Extinction Rebellion and debates over citizens’ assemblies in climate change. This seems like a really exciting area at the moment. What do you think of the various national and local initiatives?

No doubt, Extinction Rebellion has helped raise the profile of deliberative practices with its third demand for a national citizens’ assembly on the climate emergency. It was quite mind-boggling to see so many banners and hear so many conversations on the citizens’ assembly, particularly in Whitehall during the October Rebellion. The UK Climate Assembly that will start its work in January would not have happened without XR and the Student Strikes. Some commentators have suggested that the UK Climate Assembly fulfils XR’s demands. But that is not the case. The UK Assembly has been commissioned by six parliamentary select committees and will consider how to achieve the UK’s 2050 decarbonisation target. Its recommendations will feed into select committee deliberations.

XR, meanwhile, is demanding an empowered citizens’ assembly: one that is commissioned by government, has a more radical decarbonisation date and where the government agrees to implement decisions that have supermajority support within the assembly. That would be a game-changer for citizens’ assemblies.

Strangely that is quite close to what is happening in France right now. The Citizens’ Convention for the Climate has been charged with deliberating on how France can achieve 40 percent reduction of greenhouse gases by 2030. But it is empowered to look at more rapid decarbonisation if it sees fit. The main difference with the UK parliamentary approach, however, is that it is sponsored by Macron and he has agreed to implement recommendations with significant support. The question is whether he actually will once they land on his desk.

One of the things I am worried about is that many local authorities in the UK are declaring a climate emergency and running straight to a citizens’ assembly without carefully considering how it is to be embedded in the political decision-making process. I am worried that local authorities are doing this to be seen to be acting, rather than carefully considering whether they are ready to properly resource and respond to the outcomes of a citizens’ assembly. There’s a real danger that poor practice emerges and this undermines citizens’ assemblies as a model of engagement.

RB: Lastly, could you tell me more about the institutionalisation of citizens assemblies which is under way in various parts of the world?

GS: One of the exciting developments at the moment is not just the call for a citizens’ assembly on ‘x’ or ‘y’, but the idea that it should become an established part of our political process. A couple of places around the world have started to do this.

The first is the Parliament for the small German-speaking region of East Belgium. It has instigated a process that started in September. A ‘Citizens Council’ has been formed that is a randomly selected group of citizens that sits for a year. The membership is rotated regularly, but it is a permanent body. The Council takes evidence from government, parliament, civil society organisations, ordinary citizens about what issues need to be dealt with by a citizens’ assembly. Depending on how much resource is available during any given year, the Council can initiate two, maybe three citizen assemblies. The recommendations go to the relevant committee within parliament. What is interesting about this model is that the existing political institutions, the government and the parliament, can’t decide what the assemblies are going to be on. It is a separate body of citizens that is empowered to do that. I think this is really fascinating.

In the city of Madrid, an experiment has been taking place which may not have a long life since the progressive parties recently lost the election. The government had set up an online platform, ‘Decide Madrid’, where people could petition for issues to be dealt with. But they found those petitions were having little effect. So the city administration created the City Observatory, a randomly selected group of citizens who again sit for a year. They are tasked with reviewing the most popular requests that come through the petitions process together with other initiatives which don’t necessarily have that much support but which they think are relevant. Their recommendations go to government and I believe the Observatory has the power to call referendums.

Both of these examples are instances where citizens’ assemblies have been embedded within ongoing political processes. This touches on a big debate which might be a subject for a Part 3 discussion: Shouldn’t we create a sortition legislature and forget about elected institutions altogether? Kick the politicians out and put randomly selected citizens in? The House of Lords has been a particular target for this treatment and openDemocracy has been involved in these discussions.

I actually don’t think that simply replacing politicians is a good idea. To my mind, something is wrong with keeping the institution of the legislature intact, but just changing the people in it. If you are there for four years, whoever you are, you are going to start behaving like a politician. The power of sortition is when it is combined with rotation and a clear task. I would rather see the second chamber dissolved into issue-based citizens’ assemblies that are established as issues emerge. We need to separate out agenda-setting power from decision making.

RB: Not the same people choosing the issues as deliberating on them?

GS; It’s my view that you should separate agenda-setting from scrutiny. It is another important design issue. The reason why I say this is because some people are bound to lose in the agenda-setting process, and they are going to be demotivated as a result when it comes to carrying on with deliberations. Secondly, if you know that you are going to be involved later on when you are going through the agenda-setting, all sorts of opportunities emerge for horse-trading – if you support me on that issue, I will support you on this. Which is precisely one of the reasons why we wanted to get rid of the legislature in the first place! Thirdly, if citizens are in place for too long a period, the opportunities for influence and even corruption by powerful interests is high.

So, I can understand the logic of introducing random selection into the major institutions of our polities. But don’t just remove the elected politicians and dump citizens into the existing institutions. Remember how important rapid rotation was to democracy in ancient Athens. Rethink the structures of the institution as well as who is selected.

I think people are surprised that I am not always arguing in favour of citizens’ assemblies and random selection. A large part of my job as I see it is persuading people not to run citizens assemblies if they do not have the right issues or the resources. Take my engagement with Extinction Rebellion as an example. Some members of that movement were very keen on running a citizens’ assembly on the climate emergency themselves. I was quite vociferous in arguing that this could undermine the movement and their call for an empowered citizens’ assembly. It would not be seen as independent.

Constitutional issues such as the future of the House of Lords or the electoral system are perfect for citizens assemblies. This is because these are the rules of the game. Every political party has its own vested interest in these rules. They should not be deciding on rules that govern their behaviour. These issues should be taken away from politicians and given to a deliberative citizens’ body.