Sortition, Citizens’ Assemblies, Citizens’ Juries and VoteWrap

By Thor Prohaska • Last updated July 2026

Sortition is the selection of people by lot. Contemporary citizens’ assemblies and citizens’ juries generally use random selection combined with demographic criteria to create a small group intended to resemble the wider population.

Participants are then given information, hear from selected experts, deliberate under facilitation and produce findings or recommendations.

These processes are often promoted as a way to overcome party politics, unequal participation, poor-quality public debate and the practical impossibility of every citizen examining every issue. They can generate useful discussion. However, VoteWrap does not treat a randomly selected group as an acceptable substitute for the political rights of the wider electorate.

Sortition in Athenian democracy

Sortition was an important component of ancient Athenian government, but it was not itself the whole of Athenian democracy.

Athens used selection by lot for many magistracies, for the Council of Five Hundred and for citizens serving in the courts. The method helped rotate public responsibilities, reduce entrenched officeholding and limit the ability of wealthy candidates to purchase election to every position.

At the same time, eligible citizens retained the right to attend the Assembly, debate proposals and vote directly. Sortition allocated many temporary public duties; it did not replace the Assembly with a statistically selected sample that voted on behalf of everybody else.

This distinction is lost when advocates describe sortition as “the original form of democracy.” It was one institutional mechanism within Athenian democracy. The democratic authority of the system also depended on citizens participating directly in the Assembly.

Modern citizens’ assemblies use sortition differently when a small selected group is asked to stand in for an entire population and produce a political recommendation that the wider public has had no equal opportunity to shape.

What citizens’ assemblies can contribute

A well-run citizens’ assembly may provide:

time for participants to examine a difficult issue;
exposure to evidence and competing arguments;
participation by people who do not ordinarily attend political meetings;
discussion outside immediate electoral and party pressures;
insight into how views change after sustained consideration; and
useful recommendations or questions for the wider public.

Those benefits do not establish that the selected group possesses democratic authority over everyone excluded from the process.

The central question is therefore not whether a citizens’ assembly can produce worthwhile work. It is whether its work should inform the democratic community or replace that community’s right to participate.

Roslyn Fuller’s democratic objections

In her presentation Sortition and the Undermining of Democracy at the 2023 Global Forum on Modern Direct Democracy, Dr Roslyn Fuller argued that citizens’ assemblies become undemocratic when promoted as an alternative to open public participation.

Her objections include several distinct concerns.

Participation becomes permission-based

Only those selected by the organisers’ process may participate inside the assembly. Everyone else is excluded, regardless of their knowledge, interest or willingness to contribute.

Randomness may make the invitation process impartial, but it does not preserve an equal right to participate in that decision. A citizen’s political standing becomes dependent upon being selected.

VoteWrap begins from the opposite principle: every verified voter retains the right to participate personally, abstain, observe or choose a representative.

Demographic resemblance is confused with consent

A citizens’ assembly may be balanced by age, sex, location, education, income or other demographic categories.

But someone who shares another person’s demographic characteristics does not necessarily share their beliefs, interests, experience or priorities. They have not been authorised to speak for that person.

Demographic similarity is not delegation.

In VoteWrap, representation comes from an actual relationship of trust. A voter chooses a Bill Representative or General Representative and retains the ability to change that choice or vote personally.

Organisers exercise considerable hidden power

Before participants begin deliberating, other people have commonly decided:

which question will be examined;
how it will be worded;
what information will be supplied;
which experts will appear;
how long participants will have;
how discussion will be facilitated;
which options are considered realistic; and
how the final result will be expressed.

This means political power may move from elected institutions to process designers, funders, facilitators and expert selectors.

Even a sincere process can narrow the issue before the selected citizens see it. A small group receiving common briefing material may also be easier to steer than a large, open and pluralistic democratic community.

Real political conflict can be disguised

Citizens’ assemblies are often expected to produce considered consensus.

But political disagreement is not always the result of misunderstanding. People can have incompatible material interests, values and visions of society. No amount of facilitation necessarily reveals one objectively “better” answer.

Democracy provides a process for exposing and resolving conflict without pretending it has disappeared.

VoteWrap’s purpose is not to manufacture agreement. Its living map is intended to preserve:

support and opposition;
different reasons for the same vote;
minority positions;
disagreement about individual provisions;
changes in position;
abstentions; and
unresolved conflicts.

Consensus is more meaningful when it emerges visibly from this plural process rather than from a requirement that a temporary panel produce a unified recommendation.

Active and informed citizens can be excluded

Citizens’ assembly advocates sometimes celebrate the absence of the “usual suspects”—people who are already politically informed or active.

That risks treating sustained democratic participation as a defect.

Someone who has spent years studying a subject, organising affected people or challenging government claims should not lose political standing because they are considered too committed. Nor should inexperienced participants receive a bounded package of approved information while independent experts and campaigners remain outside the room.

VoteWrap allows newly interested voters, experienced activists, representatives and subject-matter experts to participate without treating any one group as a substitute for the others.

Two common conflations in arguments for sortition

The responses to Fuller’s presentation illustrate two recurring arguments.

“Sortition is democracy because Athens used it”

This conflates a tool used within a democracy with democracy itself.

Athenians used sortition extensively, but they also maintained an Assembly in which citizens voted. The existence of randomly allocated offices does not establish that a randomly selected modern panel should replace the electorate’s political authority.

A jury, public office or administrative council may be selected by lot without making every decision of that body equivalent to a decision of the people.

“A sufficiently large random sample would reproduce the public vote”

A random sample can estimate the distribution of opinions within a population. That is useful for research.

It does not follow that the sampled group has acquired the political consent or authority of the people outside it.

Nor would the result necessarily “mirror” a direct vote exactly. Outcomes can be affected by:

sampling error;
who accepts or declines participation;
participant attrition;
the wording of the question;
the information supplied;
expert selection;
facilitation;
group dynamics; and
changes in opinion produced by the particular process.

Even a statistically excellent estimate remains an estimate of how others might vote. It is not their vote.

The argument also overlooks individual agency. Democracy is not merely a mechanism for predicting the result that the population would probably have produced. It gives citizens the right to participate in producing that result.

The workload problem is real

Supporters of sortition correctly identify a major practical problem with simple direct democracy.

Modern government makes far too many complex decisions for every citizen to investigate personally. People with demanding jobs, caring responsibilities, limited resources or little specialist knowledge cannot continually study every bill and amendment.

A voluntary system can also become dominated by people with greater time, wealth, confidence or political motivation.

VoteWrap addresses this problem without randomly transferring citizens’ decision-making authority to strangers.

Bill Representatives instead of citizens’ juries

VoteWrap creates a body of Bill Representatives around each active bill.

These are not a demographically selected substitute public. They are people who have attracted actual support from voters because those voters trust their knowledge, values, judgment or ability to represent their position.

A Bill Representative’s democratic weight comes from proxies or delegations consciously provided by voters—not from an organiser’s claim that the representative resembles them statistically.

Bill Representatives can:

study the legislation;
explain its purpose and effects;
develop and defend different positions;
identify problematic provisions;
question competing representatives;
consult voters;
respond to amendments;
change their position where warranted; and
help build a clearer map of agreement and disagreement.

The number and diversity of Bill Representatives can grow organically. No central authority has to decide that a panel of thirty, one hundred or five hundred selected people constitutes “the public.”

The role of subject-matter experts

Subject-matter experts are also essential, but expertise should not automatically carry voting authority.

Experts can:

explain technical provisions;
provide evidence;
identify risks and consequences;
challenge factual claims;
model possible outcomes;
answer questions from representatives and voters; and
expose areas in which evidence remains disputed.

Their advice becomes part of the visible deliberative record.

Different experts may disagree. VoteWrap should preserve that contest rather than having organisers privately decide which expert account constitutes the approved briefing.

Experts inform the democratic decision. They do not replace it.

Proxies create a more sophisticated consensus process

The proxy system allows participation to scale without requiring everyone to become a full-time legislator.

A voter may:

vote directly;
choose a Bill Representative;
rely on a standing General Representative;
observe the deliberation before deciding;
change representatives;
reclaim their vote;
alter their position as the bill changes; or
abstain.

This produces something more democratically informative than a citizens’ jury recommendation.

The system can show:

how much direct and delegated support each position has;
which representatives have earned trust;
where consensus is developing;
where disagreement remains;
whether support changes after expert evidence;
whether amendments solve objections;
how communities and electorates differ; and
whether an elected member’s parliamentary vote reflects the verified result.

Consensus is built through open participation, actual relationships of trust and the continuing ability of voters to intervene.

Why VoteWrap does not need sortition

In the VoteWrap model, the useful functions commonly claimed for citizens’ assemblies are already performed more democratically by:

verified voters;
Bill Representatives;
General Representatives;
Community Representatives;
subject-matter experts;
open deliberation;
proxy relationships;
direct voter override; and
the living map of the bill.

A randomly selected jury would add another privileged layer whose members had not earned delegation from voters and whose selection would exclude everyone not drawn.

It would also recreate the very questions VoteWrap is designed to avoid:

Why do these people speak for me?
Who chose their briefing?
Who selected their experts?
Why can they participate when I cannot?
What happens if their recommendation conflicts with the verified electorate?
Why should demographic resemblance outweigh actual trust?

For those reasons, sortition is not proposed as part of VoteWrap’s direct–representative democracy model.

The democratic distinction

Sortition can distribute duties. It can create research samples. It can fill juries or temporary administrative roles. It may help prevent entrenched control of certain offices.

But it should not be confused with the democratic authority of citizens themselves.

VoteWrap addresses the legitimate concerns that motivate sortition—unequal resources, excessive workload, political professionalisation and poor deliberation—without replacing the electorate with a selected miniature.

Its alternative is a continuous democratic network in which every voter retains political standing, representation is chosen rather than presumed, expertise is contestable, and consensus is built from real participation and revocable trust.

Sources and further reading

Related comparisons

Back to The Details

#header_text h2#site_subheading a, #header_text h2#site_subheading{ } @media (min-width: 650px) { #header_text h2#site_subheading a, #header_text h2#site_subheading{ } }