Scope of Implementation
Level of Polarization This Method Can Handle


Electoral District Forum

June 15, 2018 Lucy J Parry, Participedia Team
February 7, 2017 Travis Mitchell
Scope of Implementation
Level of Polarization This Method Can Handle

An electoral district forum is an online non-partisan discussion board for the constituents of a single officeholder to interact with their political leaders.

Problems and Purpose

An electoral district forum is an online non-partisan discussion board for the constituents of a single officeholder to interact with their political leaders. Funded and run by ordinary citizens, a district forum is an online community for voters to educate themselves about policy, and for representatives to demonstrate responsiveness to their constituents concerns. It achieves these aims by fostering meaningful dialogue and by reducing everybody's dependence on expensive and agenda-driven commercial media outlets.

Al Gore hosted "interactive townhalls" in 1994 and 1999, and the 2004 Dean and Kerry presidential campaigns hosted lively open discussion boards on their (partisan) websites. However these experiments towards interactive voter engagement were subsequently abandoned, and online participatory democracy has since been a "road not taken" by the political parties of the U.S. 

Related offline, face-to-face models in Participedia are Public Hearings and Direct Representation .

Problem: In modern representative democracies, politicians aren't responsive to their ordinary constituents; instead they respond to pressure from political parties, organized interest groups and the wealthy. This unresponsiveness occurs in large part because the political belief systems of ordinary citizens are generally disorganized and ideologically incoherent. As a consequence, voting is shaped more by social identities than by policy reasoning. The fact that commercial media is expensive and exclusive, and rarely has improvement of the commonwealth on its agenda, greatly worsens the problem. For a good book-length treatment of these issues, see Achen and Bartels (2016).

Purpose: To reform representative democracy to be more educational, inclusive and representative by adding inexpensive, non-commercial conversational democracy elements.

Approach: To create non-commercial, non-partisan, online public spaces for political discourse, to provide an alternative to commercial media (both online and traditional). The goal is to foster political discourse which is educational, inclusive, accessible, unfiltered, non-partisan, conversational and meaningful.

Method: To create moderated and open discussion boards, one forum for each elective office, and then enlist political leaders and constituents to engage with each other there.

Origins and Development

According to the conventional view of democracy, ordinary voters have preferences about what government should do, and their representatives either carry out their wishes or get voted out. 60+ years of academic studies have convincingly shown that this "folk theory" bears little resemblance to what actually happens in modern representative democracies. 

Stephen Coleman has posited that effective representative democracy is (in principle) possible, but that it would require much more information flow in various channels. He offered this appraisal of their status:

  • Government to Citizen (G2C) - takes place largely via the mass media
  • Citizen to Government (C2G) - is limited
  • Citizen to Citizen (C2C) - is the basis of a healthy civil society, but is in decline
  • Citizen to Representative (C2R) - is very limited
  • Representative to Citizen (R2C) - is limited outside of election campaigning

Coleman subsequently surveyed 2,273 UK citizens to characterize what they ideally wanted in their relations with their representatives. He found a strong desire for a connectedness, and that they specifically wanted:

  • To be heard
  • Conversations, not consultations
  • Ongoing, rather than episodic, contact
  • Honest yet not adversarial discussions
  • Representatives to account for themselves

So in abstract theory terms, the underlying problem the model seeks to address is that communication among citizens (C2C) and between representatives and citizens (C2R/R2C) is too little, too infrequent, and not meaningful enough.

Many ideas for participatory politics projects similar to electoral district forums exist - see Participedia itself for some examples - but this specific model has never been tried out before. However, the district forum model has many antecedents, three of which are particularly relevant. 

The Constituent-Representative dialogues which are at their heart can be considered an extension/optimization of a popular internet tradition: the Ask-Me-Anything (AMA). Al Gore's 1994 Interactive Townhall is an early example, and AMAs have since become a mainstream interview format. While mainly associated with celebrity outreach, politicians have also used them to foster voter engagement. Al Gore hosted ongoing and real-time AMAs on his (partisan) campaign website in 1999, Barack Obama did a very popular one on the non-partisan site Reddit in 2012, and 2016 presidential candidate Bernie Sanders received a lot of acclaim for three Reddit ones which he did in 2013-15.

A comparison of district forum discussions and Reddit AMAs is given in Appendix 3; there are many points of similarity. Given this fact, if a district forum is started up then voters should quickly grasp how they work and what they are trying to accomplish.

The Citizen-Citizen dialogues which are a secondary goal have as a direct antecedent the very successful Minnesota E-Democracy listserv discussion group. This pioneering e-democracy initiative began in 1994, and has since evolved into the 50+ forums of which are used by about 30,000 citizens. A 2001 study found that their structure and rules do a good job of satisfying the conditions for a public sphere as set out in Habermas' theory of rational communication. To the best of my knowledge, all other existing online political discourse projects fall far short of this exacting, almost utopian, standard. However, these forums are organized around geographic locations and not office-holders, and to date there has been very little participation in them by political leaders.

The 2004 Howard Dean and John Kerry presidential campaigns hosted lively open citizen-citizen discussion boards on their websites. These forums deviated from the specific district forum model in that they were not hosted on non-partisan public spaces, and there was little direct participation by the candidates or their policy staff.

These experiments towards interactive voter engagement were discarded by later campaigns, and online participatory democracy has since been a "road not taken" by the political parties of the U.S. Joe Trippi, the national campaign manager of the 2004 Howard Dean campaign commented in 2017 about this decision: "It's amazing to me that it was abandoned. And in a very real sense it's how Trump used Twitter." 

Participant Recruitment and Selection

Each district forum is focused on a single elective office. Politician participation is therefore limited to the incumbent representative and to any challenging candidates for the office.

Citizen participation is more open in that anyone can register and get involved. However, many privileges will be limited to district residents. For example, only a constituent will have the ability to “like” a post, and only they will be eligible to propose and participate in conversations with their political leaders. These special privileges will enable constituent views to be separately known, and will afford them greater access to their representative.

Since forum users' real life identities and addresses will be known by the forum administrators, the work required to moderate trolling will be minimized because abusers can be meaningfully banned from participation. (To re-appear after banning trolls would have to commit identity fraud to create a fake account, and very few people will consider this worth doing.) Paid and un-paid operatives that are stating viewpoints without engaging - a common practice in open forums - can also be identified and weeded out. 

The use of anonymous usernames will be encouraged to allow citizens to express their political views without fear of persecution or embarrassment. It may prove useful to allow users to have two accounts, one anonymous to express controversial opinion and another linked to their name to facilitate community organization. Of course, sock-puppetry would be banned.

How it Works: Process, Interaction, and Decision-Making

Electoral district forums are online discussion boards for the constituents of a single officeholder to interact with themselves, and with their incumbent and candidate political leaders. They are intended to provide inexpensive and unfiltered channels of continuous C2C/C2R/R2C information flow which are apart from the mainstream media.

Online civic engagement forums have existed in listserv form since 1994, and discussion boards have existed in modern form since 1999. As a result, the community dynamics of online political forums are well known, as are the problems which can arise. This previous experience tells us that certain design features (such as being moderated) need to be present if political discussion forums are to be successful.

It requires a few paragraphs to adequately describe these features and how they can be implemented in a discussion board. Therefore, the detailed model is presented in Appendix 1 below.

The internet has revolutionized many aspects of society, but so far its impact on constituent-representative communication has been comparatively minor. Significant e-democracy advances have occurred mainly in terms of information provision and delivery of public services. The e-democracy initiatives which are currently being tried are mainly in the liberal-individualist tradition. What have been conspicuously lagging are initiatives incorporating deliberation, the core of democratic governance.

To help remedy this, fostering dialogue is a major goal of a district forum. The importance of unfiltered dialogue, and how district forums make use of it, is discussed in Appendix 4 below. Here I sketch out some ideas for how to structure dialogues, but these are just starting points as the forum structure can be expected to evolve with time. 

Constituent-Representative Dialogues: Forum interactions between citizens and political leaders follow a process in order to facilitate written or verbal dialogues. Because there are many thousands of constituents for every representative, discussion and voting are used to identify the questions and questioners the electorate is most interested in.

  • Constituents start the process for a written discussion by submitting questions in a “Questions Asked” subforum.
  • Constituents discuss and rank the existing questions in the same forum.
  • When politicians decide to engage in a dialogue based on a top-ranked question, they create a new topic in a “Questions Answered” subforum. There they repeat the question and provide their answer.
  • The constituents who asked selected questions (and only they) are able to post follow ups.

These written discussions continue until they are terminated by mutual consent, or the moderator concludes they have outlived their usefulness. Many conversations will amicably end after the constituent and representative have learned exactly where they disagree.

Because there is little time pressure in this discussion format, an unusual degree of scholarship can be exercised. For example, each side can:

  • ask for proof that suspect assertions are supported by the facts,
  • include visual information like graphs, charts and photographs, and
  • ask that the other read and respond to an external document such as an article.

The same two-step process is used to organize off-line verbal discussions to be held via video-conference, tele-conference, or at a meeting hall. Initiating a verbal discussion would begin with a post in the “Discussions Asked” subforum. Ultimately, video or audio of the completed discussion would be posted in the forum.

Citizen-Citizen Interactions: These interactions are more free-form and harder to moderate, and this part of the forum would be fleshed out after the Citizen-Representative section is running. 

Screenshots of the popular (>42,000 registered users) district forum run by the 2004 Kerry presidential campaign show that separate subforums for different communities and politicy topics worked well. The “District Talk” and “Dispatches” subforums in the prototype provide some structure for some activities citizens could do beyond discussing issues: post and answer polls, read and write letters to the community, etc.

The size of the district will impact what is practical and useful here. A small enough district could experiment with participatory democracy initiatives, or work to build up an online civic commons. The admins/moderators of a large district forum would have to work to prevent the same subforums from becoming an unusable cacophony.

Influence, Outcomes, and Effects

An electoral district forum's goal is to create new channels for C2C/C2R/R2C information flow. Put less abstractly, its goal is to facilitate some basic democratic actions which should occur often (in theory) but which currently occur very rarely:

  1. Citizens publicly discuss among themselves what the community preferences for policy are
  2. Candidates inexpensively campaign to tell the community how they propose to implement its policy preferences
  3. The policy preferences of the entire community, not just its wealthiest citizens, are made clear to everyone
  4. The representative communicates her positions and tactics to the community

The activity level of a forum can be easily quantified through pageviews, links to it, shares on social media, etc. The forum's impact on its community will be proportional to its activity level. 

The presumption behind the project is that increased opportunities for C2C/C2R/R2C communication will naturally result in greater citizen engagement with government (see Appendix 5) and more representative governance. The latter would manifest itself by incumbent representatives changing their behavior, or by their being replaced by more populist candidates for their office.


The developer of the Electoral District Forum has put together a fully functional prototype district forum, using the popular and secure open-source phpBB software. An earlier prototype is presented in Appendix 2 below. Using the phpBB software, a small district forum could be hosted very inexpensively (a dollar a day) and without any requirements for programming skills.

Analysis and Lessons Learned

1. One criticism of the district forum concept is that potential participants will be scared off by its novelty. This can be addressed by beginning with a simple, online implementation of a real-life candidate townhall - something every voter is familiar with. After users are comfortable with the site, additional subforums and communication flows can be added.

2. A second criticism is that deliberative politics forums are not intrinsically interesting to many voters, in part because most people are intensely averse to political disagreement. The slow growth of the forums - from ~500 users in 1995 to ~30,000 users in 2016 - is a clear sign that this is a valid criticism.

One strategy to overcome this is to include some simple but useful and familiar services in the more popular liberal-individualist e-democracy tradition to bring users in, such as: 

  • scheduled notifications of elections, and links to ballot information
  • automatic notifications of changes in users' voting registration status or polling place location
  • localized online petitions and polls with participation limited to district residents
  • a central, non-partisan location to learn about upcoming events
  • Voting advice applications for local races

Another popularization strategy is to optimize the district forum's design and management. Active moderation can help make the forums as welcoming as possible, and the inclusion of restricted sub-forums can allow for intra-party deliberation where members of popular social identity groups (such as political party members) can chat amongst themselves.

By far the most promising popularization strategy is to get political leaders participating in the forum, and to publicize the availability of online constituent-representative dialogue. Voters in modern representative democracies are starved for any actual engagement with politicians, to such an extent that Donald Trump's tweets are viewed approvingly as innovative and a mark of authenticity. However, this strategy brings us to criticism three:

3. Successful U.S. politicians enjoy a very high degree of job security, along with the prospect of a lucrative lobbying career after retirement. A third criticism is that many incumbents will refuse to participate since it would highlight their focus on representing mainly the interests of their wealthy patrons.

Yes, in normal times a successful incumbent U.S. politicians will not be interested in participating. However, there are paths to get a district forum started even without the incumbent on board:

  • Challenging candidates interested in a more populist career path may find participation attractive, since it is a zero-cost and innovative way to get their message out there. Introducing the concept to constituents during canvassing should prove an effective way of publicizing the forum and attracting users. Significant forum activity should, in turn, generate pressure for incumbent participation.
  • Occasionally there are open races without a defending incumbent.
  • Advocacy groups may find forum participation a useful way to get their message out to their fellow citizens, even if the incumbent refuses to respond.

Alternatively, other countries have campaign financing traditions which are less corrupting, and have less of a tradition of a revolving door between public service and lobbying. These countries may be more fertile soil for the concept than the U.S.

4. A fourth criticism is that participating in a district forum will hurt, not help, a candidate. The reasoning here is that the candidate's posts will provide material which will be spun and put out in a mainstream media attack ad by a non-participating candidate. Weighing against this scenario happening is that:

  • attack ad fodder is already plentiful since trackers often record what a candidate says at public (and even private) events,
  • written posts are less useful than verbal gaffes for attack ads since posts can be edited for clarity after publishing, and vetted by staff before, and
  • dishonest quotation in an attack ad is liable to backfire since it is so easy for constituents to click over and read the full context of the misrepresented statement.

Additionally, the difficult realities of internet security mean that candidates can increasingly expect to have their private online conversations leaked at a time most convenient for their opponent. Actively and openly engaging with voters online can help inoculate a candidate for when that day comes. 

5. A fifth criticism is that political party leadership will not be receptive to participatory democracy projects which give an additional voice to the ordinary voter, even when these voters are party supporters.

Yes, in principle people-friendly parties should be receptive to such projects because these give it an edge over corporate-friendly parties. In recent practice, however, all parties have been resistant to them - independent of who their base is.

Let's consider the recent history of the Democratic party, not very long ago the party of the ordinary voter. The 2008 Obama campaign embraced the internet fundraising, messaging and mobilization innovations of the 2004 campaigns, but it dropped the participatory democracy innovations and pointedly did not cede any influence on policy to its online supporters during the race. 

After the administration was in office, it went on to pursue many policies which were quite unpopular with its supporters, such as these ten. As a result it acquired a reputation among ordinary voters as being unwilling to listen and "not on their side."

The Democratic party as a whole ended up paying a steep price for this voter disengagement. During Obama's time in office, the number of Democratic governorships fell from 28 to 16, U.S. Senate seats from 55 to 46, and House seats from 256 to 194. About 1000 state legislative seats were also lost, and the state of the party in 2017 has been aptly compared to a smoking pile of rubble

The Democratic party needs to try new directions if it wants to reverse its recent steep decline. New forms of voter engagement are an obvious thing to try. Democratic candidate participation in grassroots-run, non-partisan district forums may be something the party leadership is willing to support, at this moment in time.

6. A sixth criticism is that while it is inexpensive to run an online forum, it is expensive to build up one from scratch. Because very few voters actively seek out information about political issues and candidates, not many will come across their district forums via Google searches. And if they do find it, a big barrier to participation is that people don't want to register with yet another website that requires a username and password for them to memorize. A big advertising budget is usually required to overcome the problem of generating a new online community.

Fortunately, district forum communities don't need to be created from nothing, because the discussion board software can be tightly integrated with the existing social networks. And social networks are ubiquitous - over 85% of U.S. households have at least one active Facebook user, and Facebook is only one of many networks.

District forums take advantage of existing social networks with:

  • Social media sharing buttons for every post. Since many of a person's social media friends are local and live in the same electoral district, sharing will frequently take place. For example, if a forum user likes an asked question and wants to see the representative answer it, sharing the post on Facebook provides an easy way for the user to get it upvoted.
  • Easy registration with an existing social media account. So if a Facebook friend sees the shared post and wants to upvote a question, it takes only one click for them to register to do so.

7. A seventh criticism stems from the fact that academic studies show that only a very small percentage of citizens act as the ideal "policy voter" of the folk theory of democracy - e.g. they work to be educated on the issues, to have a coherent political ideology, to select candidates based on the issues, etc. So what is the point of trying to facilitate policy-related engagement, since most people vote based on their social identities and/or worldview? Why not instead work to effect change using methods (such as advertising) which are known to work with people as they actually behave?

While this may appear to be a devastating criticism, consider these four points. First, one doesn't need many voters to have a viable forum which would have value to its participants. 1% voter participation in a district forum is a realistic goal, and yet also one which would work out very well in terms of producing meaningful conversations. The smallest electoral districts in the U.S. - state legislature house districts - have a median size of about 40,923 voters, and 1% of that is about 400 active users, which is enough to make for a lively forum that will enrich the lives of those 400 people.

Second, when a forum is open to search-engine index bots and integrated with social networks, the impact of a text conversation can go far beyond its two participants. Most internet forums and Q&A sites (like StackExchange) show the number of views which a conversation has had; interesting and informative exchanges typically get hundreds of views by readers who aren't necessarily even registered with the site. So our hypothetical 1% user base may get their dialogues read by 10% (or more?) of the voters of the district.

Third, one doesn't need to impact many politicians or voters to have a beneficial political impact. Convincing just one voter out of a hundred to be an informed "policy voter" instead of a "social identity voter" would be a significant change which would affect the outcome of close elections. And with state legislature politicians, enormous gains in responsiveness stand to be made because citizen engagement levels are currently so low with these offices (Appendix 5).

Finally, it is certainly true that good things can be achieved by working with peoples' social identities rather than their rational preferences - the Don't Mess with Texas anti-littering campaign is an example of this. The problem is that this approach is outcome-neutral, and available to anybody with enough money to fund an ad campaign. As a consequence, policies favoring the wealthy should dominate, which is exactly the situation we are in today. Creating more citizens willing to educate themselves and then vote as "policy voters" is not outcome-neutral - it will result in pressure for policies favoring ordinary people.

8. Andrew Chadwick wrote in 2006 that "...the road to e-democracy is littered with the burnt out hulks of failed projects." A final criticism is that the 1% voter participation goal mentioned above is unrealistic, and if district forums cannot attract enough participants to be viable this concept will also be a failure.

One thing to note here is that Chadwick's full statement was conditioned with the proviso "with the notable exception of some community networks," and that district forums are a form of community network - one based in part on the successful projects Chadwick was talking about.

Additionally, let's look at the numbers from the 2004 Kerry presidential campaign forum. When it was ended after a year of running, it had 42,978 registered members. 122,294,846 votes were cast in the election, showing that about 0.035% of the electorate were active forum participants.

Comparatively, a district forum in 2017 can be expected to attract greater participation because:

  • 12 years later, voters are much more used to doing things online, and more of them have ready access
  • having a candidate or incumbent participating will be a big draw because meaningful citizen-representative engagement is so rare
  • having the forum hosted by a non-partisan citizens' organization (instead of a candidate) will greatly increase the comfort level of undecided voters

Given these considerations, a 1% voter registration level is ambitious, but not unrealistic. 

See Also


  • Democracy, deliberation and design: the case of online discussion forums, by Scott Wright and John Street (2007, pdf)
  • Democracy for Realists, by Christopher H. Achen and Larry M. Bartels (2016, preview)
  • Direct Representation: Towards a Conversational Democracy, by Stephen Coleman (2005, pdf)
  • Extending the Public Sphere through Cyberspace: The Case of Minnesota E-Democracy, by Lincoln Dahlberg (2001, html)
  • Managing Online Forums: Everything You Need to Know to Create and Run Successful Community Discussion Boards, by Patrick O'Keefe (2008, Preview)
  • Re-constructing digital democracy: An outline of four ‘positions’, by Lincoln Dahlberg (2011, pdf)
  • Realising Democracy Online: A Civic Commons in Cyberspace, by Stephen Coleman and Jay G. Blumler (2001, pdf)
  • The Internet and Democractic Citizenship, by Stephen Coleman and Jay G. Blumler (2009, Preview)
  • The Problem of Citizens: E-Democracy for Actually Existing Democracy, by Daniel Kreiss (2015, pdf)

External Links


Appendix 1: Electoral District Forum - Detailed Model

An electoral district forum is a discussion board centered around a single political office. There are many ways such a concept could be implemented, but knowledge of related projects suggests the following features would help it succeed:


  1. restricted to a single electoral district - to narrow focus onto one geographical area and one political office
  2. with clear goals - increasing information flow (C2C, R2C and C2R)


  1. nonpartisan - so everyone feels welcome
  2. moderated - to keep things civil
  3. non-anonymous - so people take their posts seriously, but only forum moderators are privy to posters' real names and e-mail addresses
  4. with disclosure of interests - it must be made clear if someone is posting as part of their paid work
  5. with limits on user posts - so the conversation isn't monopolized by a small number of people


  1. online - so it's easy to participate
  2. mainly text based - to keep it simple, inexpensive, informative and searchable
  3. optimized for searchbots - to make it easy to find content through Googling
  4. registration via existing accounts - the vast popularity of Facebook, Twitter, etc., will allow many (>85%) to become registered forum users with one click - albeit subject to later identity validation
  5. integrated with social media - to foster participation by making it very easy to share posts


  1. with an emphasis on dialogue - to prevent uninformative “defense by talking point”
  2. with dialogue subforums - these will have restricted posting permissions to prevent a cacophony
  3. continuously run - a continuing conversation, not a one-off or an occasional interview
  4. with internal communication systems - to make it easy to communicate privately to other users through message, chat or forwarded e-mail


  1. factual - with fact-checking encouraged since people are not entitled to their own reality
  2. with “like” voting system - so posts can be ranked by popularity
  3. with polls - to quantify what the community thinks
  4. organized with categories and tags - to help find what the community has discussed in the past
  5. with video discussions also possible - because footage can be much more dramatic and viral than text

People Powered

  1. volunteer run and hosted - so it is independent of outside funding
  2. simple - so volunteers can moderate or admin without extensive training
  3. inexpensive to host - to make it possible to be independent
  4. nonprofit - to focus on community serving and not profit seeking

Politician Friendly

  1. fully open only to district residents - since non-constituent opinions are of limited interest to representatives
  2. with a reasonable time commitment - staff can write first drafts of responses, only highly upvoted questions need to be responded to, etc
  3. deliberative - staff can vet posts prior to publishing since there is little time pressure
  4. correctable - posts can be updated afterwards for clarification or correction

A standard, open-source phpBB discussion forum will automatically have features 8, 9, 15, 16, 19, 21, 23, 24, 28 and 29. The remaining ones can be added in the following ways:

  • additional software code: 10, 11, 12, 18 and 20
  • website rules and organization: 1, 5, 6, 7, 22, 25, 26 and 27
  • discussion forum structure: 2 and 14
  • forum moderation and norms: 3, 4, 13 and 17

Appendix 2: Prototype Software

I have put together a fully functional prototype district forum based on the popular and secure open-source phpBB software. This software is able to handle very large communities - the largest one running it has over 100,000 registered users and over 28 million posts.

This screen shot of the prototype's homepage shows one way in which the categories and subforums could be organized:

The basic phpBB board has been extended through the use of approved mods and some custom coding. The prototype incorporates most of the desired features discussed in Appendix 1. For example, the prototype software includes a “like” button for every registered user. Tallying likes allows the best questions, petitions, etc., to get ranked above the less interesting ones.

Appendix 3: Electoral District Forums and Ask-Me-Anythings

District forums can be considered an extension/optimization of a long-standing internet tradition: the Ask-Me-Anything (AMA). The format of an AMA is that any registered user can submit questions and vote to rank others. Reddit is currently the most popular website for this type of interview. The interviewee answers whichever questions he chooses for about an hour, ignoring upvoted questions at his peril. Questions and answers are both short, and there are no follow-ups.

There are many things to like about this new social media form. It is:

  • Online and free to use - so it's easy to participate
  • Egalitarian - no users are privileged above others
  • Unfiltered - no gatekeepers
  • Responsive to good ideas - questions and suggestions are ranked by voting
  • Moderated - which keeps things civil
  • Nonpartisan - so everyone feels welcome
  • Mainly text based - which keeps it simple, inexpensive, informative and searchable

However, while the Reddit AMA format is a good fit for celebrities promoting something, or for Q&As with unusual people, it is a bad fit for constituents wanting to have substantive conversations with their representatives. Here are some important issues:

  • The politician AMAs that have taken place tend to be dull and unenlightening due to the lack of follow up questions.
  • Not many constituents will participate because of the event's brief time window. For example, nobody living in the relevant district submitted questions to this AMA from a candidate for congressional office.
  • Because there is no identity verification, questioners may not be what they claim to be. An account's comment history can't be used to definitively filter out sock-puppets because old Reddit accounts are openly sold to people wishing to influence discourse on the site.
  •  Although in principle there is no gatekeeping on Reddit, in practice it happens all the time. For example, organized users have done it through downvoting ("brigading") since the inception of the site.
  • Because Reddit is a commercial site, the owners do not enforce its stated community standards when these would reduce its profitability. As a consequence, it currently has a problem with white nationalist trolls

For citizen-representative conversations, I believe these important attributes need to be added to the Reddit format:

  • Emphasis on dialogue - to prevent uninformative Q&As where the answer is a talking point, and there are no follow-ups
  • Continuously run - a continuing conversation, not a one-off or an occasional interview
  • More inclusive - let people log in using their Facebook or other social media accounts
  • Non-profit - to allow a focus on community not profits, and to keep separate from Big Media, including Big Social Media
  • Non-anonymous - to keep out corporate and government sock-puppets
  • Covers a single political office - to narrow the focus

When these attributes are included, what results is close to being an electoral district forum as described in Appendix 1.

Appendix 4: The Importance of Unfiltered Dialogue

Minds work much better when they are open. Unfortunately, people have an innate tendency to seek out only information which reinforces their pre-existing views. Furthermore, in those important cases where opinion is bound up with identity, a straightforward presentation of facts will often be rejected without being considered first. And to make matters worse the most accessible information, that provided by the mainstream media, is heavily filtered.

A district forum aims to fight these effects by encouraging participation (active discussion or passive reading) in a nonpartisan forum filled with information and views which have not been pre-filtered by either the user or by the mainstream media.

Can dialoguing on forums result in people learning new things and viewpoints, and perhaps changing their deeply-held worldviews as a result? Certainly, but the process of opening a closed mind is slow, and one estimate is that it takes three or four months of forum participation for users to become receptive to new viewpoints.

Relatedly, wealthy interests have long used a divide-and-conquer strategy of fanning prejudicial resentment to gain the support of people who are hurt by their policies. However, personal contact has been shown to be an effective way of reducing prejudice towards out-groups, even when the contact is computer-mediated. A recent study has demonstrated that reductions in intergroup bias and anxiety occur after only two weeks of online interaction. District forums are one way of harnessing the power of personal contact. 

Appendix 5: Continuum of Citizen Engagement

The question of what level of U.S. citizen engagement is best dates back to the competing visions of founding fathers Hamilton (low levels) and Jefferson (high levels). Katz, Barris and Jain (2013) have sketched out a useful conceptualization of engagement levels: 

  1. Non-engagement: the citizen is law-abiding but otherwise inert in terms of government activity. 
  2. Votes, but does nothing to learn about politics or policy. 
  3. Votes and learns about politics, but gives no meaningful expression to a viewpoint. 
  4. Votes and learns, and also takes actions that reflect or propagate a viewpoint. Examples: discussing politics with neighbors, campaign contributions, contacting elected officials, commenting online. 
  5. Active in suggesting, evaluating and formulating policy. 
  6. Highest engagement: Decides directly on issues. 

District forums are intended to make it easy and enjoyable for citizens to engage at levels 3-4. Previous experiments with citizen assemblies have made it clear that engagement at levels 5 and 6 is not something that ordinary citizens want. 

A important observation is that the level of engagement varies with the visibility of the public office. The majority of U.S. citizens are at levels 2-4 with respect to the presidential election, for example, but level 1 with state legislature and municipal elections.