Networking Democracy: can the internet help democracy work better?

OurKingdom is hosting what we believe to be a unique online deliberation on how the potential of the internet can be integrated into a national political process. It is about this question: can participation on the web reinforce representative democracy? The initial phase took place among a group of experts who were asked to set out the problems and summarise what is already known - please look at it here. The first section of it has an email exchange between Anthony Barnett of OurKingdom and Michael Wills, the Secretary of State at the Ministry of Justice - who bravely agreed to an independent debate. Their exchange sets out what we are doing.

We are now opening up the debate for anyone to contribute, here or on your own blogs and websites. The post below is Anthony Barnett's personal summary of the discussion so far. Please read at least some of the discussion before commenting. We have an important opportunity to raise the game of the web in politics - and not just in Britain, we are trying to bring in as much international discussion as we can.

The minister and some of his civil servants at Justice are listening to this conversation, and have already contributed in the initial exchanges as you can see. We would like as many people as possible to blog about this, write about it, and host discussions about it in their own fora. If you do, please link to here and email us ( at jon.bright@opendemocracy.net ). After three weeks, we will aggregate all the discussions that we know about into a final document.

Please also note something that is very important: the starting point of this experiment may be the government's commitment, for which Michael Wills is responsible, to have a citizens summit. But our concern is not with how to make this specific summit work. The Minister's questions are ones of principle: Tories and Lib-Dems and Scot Nats and Welsh and Irish politicians should all be asking these questions for themselves. As the voting system comes under review, how can we widen and deepen democracy by other means as offered by the internet? This is a non-partisan issue that will grow in importance over the coming decades.

Networking Democracy: National conversations on government policy using the internet

Here are some of the lessons I draw from the first stage of our on-line deliberation. A very big thanks to everyone who has participated so far. What follows is not a formal aggregation and taxonomy of different views. It's an overview of the lessons I have learnt.

The aim is to initiate a web-based debate on how a government - that is any government - could integrate the potential of the internet into a national political process designed to support representative democracy. The idea started from a conversation I had with Minister of State, Michael Wills, who is charged with organising a Citizens Summit to make recommendations to parliament on a British Statement of Values. He told me that he wanted to draw on the potential of the web. As you can see from our initial email exchange, he had the courage to ask us to discuss how this could be done - and I want to stress again openDemocracy's OurKingdom team is doing this unpaid by, and independently of, the Ministry.

The nature of the proposed summit caused some nervousness amongst those invited to take part. There were objections to the abstract and general nature of what we have asked everyone to do. What emerges from this is that if a government wishes to draw in popular input it must be clear about what it wants and why. What is the aim of the specific conversation? Who is supposed to be talking to whom - and who is listening? What members of the government are reading the comments even, let alone responding? What is being made of the comments?

The Minister had two specific concerns: how to make sense of large numbers of web contributions (if they occur) and how to ensure both that there is a good input and that it is fairly representative and is secure from ‘capture'.

An important lesson for me is that while the Minister's fears are understandable they may be misconceived. The web is not just a version of ‘The general public'. Voting does indeed disaggregate everyone into private, anonymous individuals, whose ‘x's are then counted. But the web is not just a soup of isolated individuals prone to manipulation by the wicked or the commercial and in need of benevolent guidance from the authorities. Rather, it consists of many groups, networks, communities and cyber-associations, such as regular readers of particular blogs, small and labile aggregations, some very persistent, with over-lapping interests and memberships, often very intelligent and capable of learning. The success of social networking sites like facebook is that they generally use real names and create, precisely, social networks.

To go to the web to assist a national conversation means asking these groups and networks and communities to debate and discuss the issues for themselves, off-line as well, and on their terms, while feeding back within whatever rules are set.

How to use the internet to assist a national debate: my lessons from the initial openDemocracy discussion

1. A government needs to be precise about its aims for the process and must then define from the start what role the internet debate is expected to play.

1. What is the aim of maximising on-line participation? This question is linked to the concept of a citizens summit itself. The thinking behind the concept of a summit is that if a representative group of citizens chosen at random (say 500 or 1,000) deliberate on an issue they can reach a decision that is both wise, because they have listened to experts and debated the evidence, and is perceived as legitimate. (The process has been most systematically developed as ‘deliberative polling' by Jim Fishkin at his Stanford centre). Until we have more experience, on-line participation is unlikely to be used for decision taking. Its use is therefore three-fold:

1. The relationship of online and offline needs to be clearly frameworked from the start. It is a mistake to just ‘ask' for responses, not much will happen if you do.

1. While the numbers the web can bring to a national discussion are large and the relative cost is low, the support and investment in web debate is significant. It needs to be led by people who know about and enjoy the medium and its latest advances. Executive control should not lie with a civil servant who is suspicious of the rabble, just uses email and can't twitter or blog. Discipline and good quality politics are available on-line but is best achieved by those who have embraced its potential.

1. Aggregation. This cannot be done by machines, though they can help. It needs to be done by humans who are given the task of reading the material, use simple categories, report their findings in good prose with all the material open and available for others to assess differently. (This report is an example of just such ‘aggregation' if from someone defiantly one-sided as well!). Aggregation should be in place from the start and not begin when all submissions or comments are in. Those who comment should know their views will be read, assessed and aggregated - this is an incentive.

1. Moderation. Any process needs clear rules, e.g. about brevity, no personal attacks, encouragement of links, clear examples AND engagement with the question being asked. Then moderators can be recruited to ensure these are kept. Personally, I favour pre-moderation: a process that is guarded is more valuable for readers - and there will always be many more readers than commentators.

1. Highlighting. Best comment of the day, wittiest comment of the week, most original, most bizarre, all of this attracts readers as well as encouraging participation.

1. Capture. The first and best answer to concerns about any online process being ‘captured' by groups or special interests is to recruit as many groups and interests as possible to participate!

1. The fear of capture resides in an assumption that any open process of citizen involvement will be a shadow the process of voting. For historic reasons, voting privatises and makes anonymous the individual participant. But online deliberation should be open and joint. This is where the use of online input is not costless. The government needs to approach large off-line associations, such as the Consumers association, the Women's Institute, universities, trade unions, business associations, churches and religious congregations, large NGOs, to ask them to encourage their members to participate. The use of independent polling and the process itself will identify cheating and the mechanical reproduction of unrepresentative views.

1. Personal names. I support the advice of Steve Clift here, and the evidence of facebook. Real names, with area and even postcode, and citizenship identified, should be mandatory - with anonymous contributions being allowed via email to a moderator who also gives the reason for accepting it. I think the use of real names in a public process of this kind is good in itself. We are holding politicians to account and to high standards in the public arena, why should members of the public then hide themselves? We also want to encourage the restraint that public accountability reinforces. There is a clear difference of view on this - a motivated decision will be needed.

1. Multi-media. The use of video, sound, face-to-face debates, should be encouraged, for example with a video wall. The larger aim is to intensify and deepen political talk. It is not that this kind of participation will necessarily influence the outcome in an immediate practical way, but what it will do is assist the positive acceptance of the outcome and widen public understanding independently of the main media (who are likely to slag this off therefore).

1. Invite in as many different organisations and networks, from political parties to churches to online networks. Ask them to run their own debates, train own moderators and encourage members to participate. Use widgets and all other devices: don't isolate the process from the rest of the web, on the contrary. Off-line advertising may help the process gain legitimacy and win the support of the commercial media (perhaps by advertising on their websites and announcing how many have clicked through from them). More important the process should be projected across the web, actively seeking links, ensuring input from blogs, social networking and organisational websites, permitting viral reproduction of comments and comment upon them.

1. Timing: if the internet debate is going to be about something complicated, it might need to proceed in stages and over time with the explicit aim of helping to involve as much of the population as possible. If it is about a relative technical aspect - for example the voting system - then there needs to be advanced web publicity, a clear beginning and end. The web easily gets bored. I would say, over seven weeks with the middle week being a breather week in the middle for the aggregators and moderators to take pause and meet and assess how it is going. If this little experiment is anything to go by the moderators and drivers should do nothing else at all for its duration. The quality of the input is hugely improved by the quality of the caring and moderation.

1. In summary: be clear about the aims and how there will be a policy outcome, have clear rules, work with what already exists on the web and do not try to control the process.

Anthony Barnett, March 2008