Who’s who

As you will see, every one of these pools contains an element of chance or random selection, in different ways. 

The Democratic Odyssey is about reinventing together what 21st century democracy can be, a Democracy 3.0 where the ancient wisdom of the lot, or ‘random selection’ is combined with modern technology to supplement our current electoral democracy.

You may ask: How can 500 random chosen persons reflect the social diversity and breath of experience of all those living in Europe today? They can’t. But they can be a mini simplified expression of it.

If you are a member of the Assembly, some will call you a “mini-public”, others “a representative sample”, and others a “randomly selected lot”. All this mean is that you are becoming part of an Assembly of 500 people, who somehow collectively mirror the almost 500 million people who live on this continent. Hence the term “mini”. Some will argue that you offer together another kind of representation, a “descriptive representation” of who Europeans are (see the OECD box in the next page).

For sure, none of you were elected like the members of the European parliament or your city council, and you might think that this would be the real democratic mandate. But don’t stop there. People in our societies can recognize themselves in you. You will not be in charge for long, so hard to corrupt!

Think about the role that “government of chance” has played throughout history in selecting people who make decisions, occupy public jobs or simply gather as an assembly as you will. The ancient Greeks did that pretty well, with the kleroterion machine, except that, ooops, they forgot to include women! In their case, they chose from 10 tribes of Attica where each one of hese tribes brought together people from town, seaside and countryside.

But we do not want to pretend that it is easy to use this lottery approach. Creating a random sample of people is as much art as it is science. Antoine Vergne refers to what we do as an intersubjective concept of sortition.  This is “an approach,” as he tells us, “that puts at its core the realization that we do not have the technology nor the data to achieve an objectively clean sortition. More than that: Maybe we shouldn’t even try.” Instead, at each stage, we have to agree on designing a process that makes sense for in the eyes of the involved parties. At the democratic odyssey, we rely on a network of friends called the constituent network, that is radically inclusive and transparent. And as we encounter new opportunities and obstacles, we think together about the right set of justifications, criteria, and steps that will help create the best prototype assembly possible.

Not starting from scratch:

There is of course plethora of ‘selection by lot’ experiments to draw from at all levels of governance. The EU has even offered its citizens panels during and after the conference on the future of europe (2021-22). But in there is no agreement among civil society actors, practitioners, politicians or scholars on the best way to do this. In fact, no such assembly as ours has ever been tried before. 

OECD Defintion of Civic lottery: A process used by public authorities to convene a broadly representative group of people to tackle a policy challenge. It is based on the ancient practice of sortition, which has a history ranging from Ancient Athens to the Doge of Venice. Today, it is used to select the members in Citizens’ Assemblies and other deliberative processes. The principle behind a civic lottery is that everyone has a more or less equal chance of being selected by lot. There are two stages to a civic lottery. First, a very large number of people, chosen by lot, receives an invitation to be part of the process from the convening public authority. These randomly selected recipients can volunteer by opting in to the lottery. Then, amongst the volunteers, members are chosen by lot to be broadly representative of the public. Civic lotteries aim to overcome the shortcomings and distortions of “open” and “closed” calls for participants, which result in non-representative groups of people who do not mirror the wider population and attract those with the most interest or stake in the issue. (For greater detail, see Chapter 4 in OECD, 2020a.) Source: Evaluation Guidelines for Representative Deliberative Processes, OECD (2021)

This means that the democratic odyssey does not pretend to offer the perfect prototype for what a permanent Peoples’ Assembly for Europe can be. Instead we are embarking on a permanent experiment in order to refine our blueprint by doing rather than in the abstract.

This means that each stage of the voyage and each rejigging of the composition of the Assembly is a very imperfect progressive approximation of what a Peoples’ assembly for Europe can and should be. We will want to learn as we continue the voyage from one city to the next. We are together a laboratory for democracy.

Crucially, the composition and recruitment for the assembly will evolve according to the views on its composition held by the members of the assembly themselves. And  as a radically transparent and inclusive campaign for a standing transnational assembly in the EU, we ask for feedback from the crew at large: all our observers and evaluators, the Odyssey’s constituent network, as well as the media and the broader public.

Where do Assembly members come from? Due to our commitment to minimize carbon footprint, we explore how to  make this a transnational story without flying everyone to Brussels! And we also show that such an assembly does not have to be expensive (we operate with a fraction of what the EU institutions can conjure up for their own panels). In the end, constraints force us to be creative.

As a result, the Odyssey’s ambition is to renew the very idea of democracy across borders by bringing together two spheres of people:

• The transnational sphere is composed of citizens randomly selected through civic lottery from all over the EU: 20 countries as of september 2024, including EU candidate countries. This is the approach that has been taken in the Conference of the Future of European (2021-22) and the panels organised by the European Commission (2023-2024). This sphere conveys of understanding of democracy in Europe which is meant to bring together people who live in every corner of Europe, not only different countries but also in rural vs urban areas. Members from this sphere will also join the city meetings in person during the length of their membership (1 year by rotation).
• The translocal sphere is composed of citizens who live locally in the city where the assembly is hosted, starting with Athens.This sphere conveys of understanding of democracy in Europe which is also about connecting places rural or urban, groups of citizens and decision makers with contrasting experiences. This group of people is responsible for welcoming the transnational pool to their cities or place. Members from this sphere will join the city meetings in person in their own city and send ambassadors in the other cities.

The process of recruitment of all these pools has unfolded through through systematic cooperation between the two foundations in charge (Sortition foundation and QED), in order to make methods transparent and comparable and sometimes rebalance the composition across pools (say adding more women on one side if there were a bit less on the other side). 

We care especially that everyone involved and the broader public understand how this happens on the ground. It is very important but not enough to follow the highest standards of random selection in the field. In doing so the two foundations were also tasked with finding inventive ways of communicating the idea of democratic lotteries. Do watch our recruitment videos here.

A bit of technicalities:

• Both foundation proceeded in the usual two stages (some call it “double stratified sampling”):
  • 1.     1st stage- Interest in participation: A large pool of randomly selected people are invited in person or electronically to participate:  this creates a broad respondent pool.
  • 2.     2nd stage – Participants’ selection (the pool itself): From those who respond yes in stage 1 a second lottery takes place, in order to match agreed demographic targets or criteria.
• On the transnational side (pools 1 and 2), the Sortition foundation recruited 40 members from 20 countries thanks to its local partners on the ground, randomly selecting locations in EU countries and knocking on doors to invite people at or near those locations.  
  
  

  First things first. For pool 1, they first selected citizens 6 European countries (Austria, Bulgaria, Germany, Lithuania, Malta, Spain and Bosnia-Herzegovina), inventing new ways of selecting addresses by lottery such as dice rolls, lottery draws, or dart throwing at a map. And for pool 2, the work was always already done, since they drew on a small existing reserve pool of people who had not been selected to participate in an earlier transnational deliberative event (EU panel on hatred), but had expressed their interest in joining similar future events. This second pool meant we could catch people who care about participation and had been given false hopes. Here was their second chance.

  From these two original pools of respondents, the hard work of extracting an appropriate sample had to be done. In the end, a diverse pool of 40 international participants from 20 different European countries was selected across these two pools of respnded to reflect diversity in demographic factors: their gender, age, education level and views on the EU. Comparing the target population and those ultimately selected (see graph) you can see that the youth is over-represented in order to reflect the intergenerational aims of the assembly. You can also see that the sortition foundation sought to decrease the proportion of “very positive view of the EU” and increase the “very negative” for the finally selected members, since original respondants who say yes tended to be more positive than the population at large. This is of course a very rough way of going about “biases” but we do our best. All to be refined in the future.

• On the local side (pools 5 and 6), the QED foundation in Athens followed a similar method, eg “2-stage sampling methodology”:  
  • Their goal was to provide equal opportunities for every resident of Attica who is 16 years old or older to participate in the assembly.
  • In the first stage, they approached a large sample of residents, asking them to register their interest in participating in the event. They did that by selecting 100 building blocks—50 in the Municipality of Athens and 50 in the other Municipalities of Attica and knocking on doors. In order to ensure equal participation across age groups, door knockers gave respondents a brochure with information on how other members of their household can register on their own.
  • Which doors? If you want to know more, please read the annex below.
  • To go about eliciting circa 500 positive response, QED also contacted people by a email from a large panel of randomly selected people
  • In the second stage, they randomly selected the final participants, matching them alongside the predefined quotas, selecting 60 Attica residents of Greek nationality (pool 5) and 60 Attica residents of other EU nationalities, as well as some candidate countries (pool 6). This pool 6 of non Greeks is the Democratic Odyssey’s way of bringing in the conversation people people can serve as a bridge between the local Greeks and the transnationals traveling in.

This Peoples’ Assembly includes people who have already participated in other transnational or national assemblies: alumni citizens (pool 3). Some in EU panels, others in assemblies organized by our consortium partners like Phoenix. They have already been randomly selected thus preserving our commitment to the lottery ideal. This is one of our main innovations, and one that we care a lot about, as the democratic odyssey hopes to help amplify over time the commitment that these persons bring to the process.

For DO, these members were recruited through open calls to create lists of potential members, and then selected to take part either randomly through dice rolling or according again to criteria like age and gender. As people who have been familiarized with citizens’ assemblies and continue to be involved in various ways (eg WhatsApp groups etc) their experience is precious to DO. Some of them are critical of the follow-up from the EU process, especially for being left out of the loop even if they are told that their recommendations have inspired new EU laws (all while generally very positive about the EU process itself). Moreover, many are familiar with the Democratic Odyssey and have participated in some of the meetings of the Constituent Network. They will provide an important link with the broader participatory eco-system.

By a Peoples’ Assembly we mean to be more inclusive than traditional politics in Europe, including those who are not “citizens” in a formal legal sense and play marginal roles or are not represented at all electoral politics. This is why we created pool 7 for non Europeans, whom we call “global citizens.” These non Europeans can be migrants or refugees living in the host city, or even posted workers in the care sector, the construction industry or ports. They are randomly selected from a pool of respondents to a call again to preserve the philosophy of the role of lottery (this could be done through a criteria in the step two stratification method if the initial pool was big enough).

To build this pool, we launched a call that reached over 2,000 global citizens, with the support of organizations such as Metadrasi, a Greek NGO dedicated to the reception and integration of refugees and immigrants, providing professional training for more than 500 students. We also collaborated with ActionAid, the Greek Refugee Forum, the Athens Coordination Centre for Migrant and Refugee Issues, UNHCR, and the International Organisation for Migration. In response, 60 individuals expressed interest. From this group, we used a quota system to ensure diversity in gender, age, and country of origin. Ultimately, 20 global citizens from 15 different countries—Afghanistan, the Philippines, Pakistan, the Democratic Republic of Congo, Ivory Coast, Ghana, Nigeria, Burkina Faso, Israel, Syria, China, Georgia, Egypt, Iran, and Algeria—have joined the Assembly.

We hope that in the future, the members of the Assembly will give further thought to the challenge of giving a voice, a presence, rights, and ultimately dignity, to those who are completely absent in our democratic debates. Simply absent from citizenship rights, like the refugees and illegal migrants already included in Athens. But beyond, absents like the unborn, as well as non humans, animals, plants, rivers, glaciers. Ours is also a journey of collective exploration of democratic futures, where we hope eventually to blend human and non-human perspectives through an immersive, deep and collaborative series of dialogues.

Perhaps most controversially, the Peoples’ Assembly brings in civil society actors both from global organisations or movements (Pools 4) and from local ones (pool 8). Rather than aiming simply to ‘represent’ European citizens as a whole, we hope to amplify the voices of individuals, movements and collective actors that lack representation in normal institutions and yet contribute immensely and often heroically to public life both locally and at European level. To be very clear, civil society activists do not “represent” their organisation or interest group but are meant to be included in the People's Assembly in their personal capacity because they have valuable knowledge of societal problems and can also speak to them as citizens. As with global citizens, they are picked randomly from a list of respondents to an open call.

To build this pool, we launched a wide-reaching call for civil society organizations (CSOs) to participate in the Assembly. A significant communications campaign was initiated, and we reached out to over 1,000 different CSOs through a targeted mailing. Of course, these respondents registered because they were already engaged and interested, but we still received responses from 83 individuals, representing CSOs from 20 different European countries. From this pool, we randomly selected 20 participants. These chosen individuals work in a wide range of CSOs addressing climate, equality, citizenship, gender, migration, art, culture, and youth support, creating a rich and diverse polyphony of voices and perspectives.

In this way we help construct spaces where mobilised civil society can engage in structured dialogue with other randomly selected people during Assembly debates and rather than as an afterthought.

Will all of the 200 members selected for Athens meet in every port of the Odyssey?  No. Membership of the assembly will grow over time to reflect its itinerant nature, picking up new members as it visits and meets in new cities and places. The members from the transnational sphere (pools 1,2,3,4) will attend the first 3 in-person meetings in 2024-25 and will then progressively be replaced according to the principle of rotation (once a year).   

On the other hand, the local members who belong to the trans-local sphere, will designate emissaries (or ambassadors) to represent each city and the meeting in the next city. But the other local members will not be left behind. Their discussions and conclusions will be summarized and passed on and they will follow the continued discussions through the livestream and through the digital platform.

Selection of building blocks:

• To enhance Athens Municipality’s representation, we proceeded to a stratified selection of 100 main sampling units in the region of Attica. The two strata were: 1) Municipality of Athens and 2) Other municipalities of the prefecture of Attica. We followed building blocks  (given and described by the Hellenic Statistical Authority) as our primary sampling unit. 50 blocks were selected for each sampling stratum to constitute the main sample as well as 15 additional blocks (5 blocks for stratum 1 and blocks for stratum 2 as alternates). For the selection of the building blochs, we relied on the most recent published information from the National Statistical Authority (2011 Census). For each sampling unit, we aimed for 5 positive answers.

Probability of selection:

• The method of selecting our primary sampling units is via random systematic sampling with indirect stratification. The variables selected for indirect stratification were the label of the municipality in which the building block is located and the label of the building blocks. The labels used were those given by the Hellenic Statistical Authority. The probability of selecting each building block was proportional to the number of permanent residents of the building blocks as recorded in the 2011 census.
• In case the number of residents of the selected building blocks is less than the required sample size per main sampling unit (the max positive responses collected per building block) multiplied by 10 (i.e. 50 residents), then 'augmented building blocks’ will be taken as units. During the visit planning process, each building block with an insufficient number of inhabitants will be augmented with neighbouring building blocks to arrive at augmented building blocks of at least 50 inhabitants. The selection of neighbouring building blocks will be done with the following algorithm:
  • If the building block has less than 50 residents, we select the block lying east of and adjacent to the selected building block.
  • If the sum of inhabitants of the augmented building block is still less than 50 we add the one lying north of and adjacent to the original building block.
  • If necessary, we continue in a counter-clockwise direction and move crosswise around the original building block in layers. The transition to augmented OT can also be made in the event that the field investigation reveals that the actual number of inhabitants is less than that indicated in the census of 13 years ago.
• The decision to move to augmented building blocks and their designation is made exclusively by the Director of the field survey department and is passed on to the field investigators. If any of the building blocks selected as the main building blocks do not have any dwellings or cannot be accessed or if there is another important reason, then the augmented building blocks can be used.

Moving through the blocks:

• We start from the corner of the building block that points towards the city center. As the center of the city, we define the central square, the shopping center or the main church.
• Movement on the building block is clockwise.
• Movement on the floors is also clockwise (to the right of the elevator).
• In the first building we start from the top floor, in the second from the bottom, in the third from the top and so on.
• If you completed your movement through the block and have not covered the full number of required contacts (candidate households) or interviews, continue the diagonal block from the starting corner, where again the movement is clockwise. If there is no other building block diagonally, or we have already moved to the one diagonally, then we continue with a priority to the block from the right (otherwise left). That is, we continue from the right on building blocks that enclose the original building block selection.