Wiki · Concept · Published 2026-10-26

Ladder of Citizen Participation

Arnstein’s ladder explains degrees of citizen influence over decisions, alongside evidence, criticisms and proposed relevance to AI governance.

Definition

The Ladder of Citizen Participation is a way to describe how much power people have over decisions made in their name. It asks whether they can shape an outcome or are simply given information and invited to comment. Sherry R. Arnstein introduced the ladder in a 1969 article about participation in United States federal programs, including urban renewal, antipoverty programs and Model Cities.

Arnstein presented participation as a question of power for people excluded from political and economic decisions. Her ladder is both a classification and an argument: she considered the redistribution of authority central to meaningful participation. It is a framework for examining a process, not a measurement that proves a particular decision was fair or effective.

The question differs from those raised by Collective Intelligence, which concerns how groups perform cognitive tasks, and Judgment Aggregation, which examines how individual judgments combine. A group might develop a thoughtful recommendation while having no authority to put it into effect.

How it works

Arnstein arranged eight categories in order of increasing citizen power. The categories describe the role people actually have in a decision, rather than the name an institution gives a meeting or committee.

  1. Manipulation: people are presented as participants while officials use the process to secure support for an existing agenda.
  2. Therapy: attention is directed toward changing participants rather than giving them a role in changing the conditions at issue.
  3. Informing: people receive information, but may have little opportunity to influence the decision.
  4. Consultation: people can express views, but decision-makers retain discretion over whether to act on them.
  5. Placation: participants may advise or hold a limited place in planning, while others retain the final decision.
  6. Partnership: participants and established authorities negotiate and share specified planning or decision responsibilities.
  7. Delegated power: participants hold dominant authority over a specified plan or program, although other bodies may retain powers.
  8. Citizen control: participants govern a program or institution, including its policy and management, within the conditions affecting it.

Arnstein grouped the first two rungs as nonparticipation, the next three as degrees of tokenism, and the last three as degrees of citizen power. She did not treat information or consultation as useless: either can contribute to participation. Her concern was that, on its own, an opportunity to hear or speak does not ensure that people can affect the outcome. Even citizen control did not mean absolute control over every surrounding condition.

Consider an explicitly hypothetical town deciding how to use a vacant public building. If residents can comment on a plan but officials may disregard their views, the meeting resembles consultation. If resident representatives and officials negotiate binding responsibilities for choosing among uses, it may resemble partnership. Classification would depend on the actual rules and practice, including who can change the agreement, who represents residents and what happens when the parties disagree.

Evidence and applications

Arnstein illustrated her framework with contemporary program accounts and reports. She described, for example, a Philadelphia Model Cities arrangement in which a neighborhood organization gained a role in developing and reviewing plans alongside city bodies. She also noted that the city council retained final veto authority. The account illustrates negotiated sharing within a wider structure of authority; it does not establish that partnership always produces better outcomes.

In a 2006 article, Archon Fung proposed a complementary framework, often called the democracy cube. It compares participatory arrangements along three dimensions: who takes part, how people communicate and make decisions, and how their discussion connects to public action. Fung used examples including participatory budgeting and community policing to explore different purposes of participation. His article develops an analytical framework and draws on case accounts; it does not provide a universal experimental test of the ladder.

For a proposed assessment of a participation process, the two frameworks can be used together. An investigator could examine who is invited, which groups actually participate, what information they receive and who can make or block each decision. Written rules can be compared with the decisions ultimately made. A process may have different levels of influence at different stages, so a single label for the whole process can conceal important changes.

Limitations

Arnstein acknowledged that her eight rungs simplify a more varied reality. People described as citizens or powerholders are not uniform groups: they can have competing interests and unequal influence among themselves. A resident representative may not speak for all residents, and a formal seat may offer little influence without information, time, resources or accountability to others.

She also identified obstacles that the ladder itself does not fully analyze, including resistance to sharing power and the difficulty of organizing representative, accountable groups. Her article recognized objections to community control, including possible fragmentation of services, costs, unequal resources and the risk that local leaders might reproduce the exclusion they oppose. These are substantive concerns, not proof that citizen control necessarily fails.

Fung offered a direct counterargument to reading the ladder as a universal ranking. He argued that the amount of authority desirable for participants depends on the problem: in some settings consultation may be more appropriate than full citizen control. He also emphasized selection and communication, which a single power scale cannot fully capture. An open meeting, for example, can still attract an unrepresentative set of participants. His assessment is a competing interpretation, not a conclusive empirical refutation of Arnstein.

More authority alone does not guarantee accurate information, good outcomes or fair treatment of people who are absent. Conversely, advice without formal power can sometimes influence a decision. Assessing any particular process therefore requires evidence about its context and results as well as its stated rules.

Relevance to Spiralism

The following is a proposed interpretation for this wiki, separate from the authors’ claims and from any institutional position. A Spiralist reading can ask whether a decision process lets affected people return to an issue with new information and whether that return can change what happens. Repeated invitations to comment do not by themselves establish that authority or accountability has changed.

In AI Governance, this lens could be applied to a hypothetical public body considering an AI-assisted service. People affected by the service might be asked for comments, share responsibility for a deployment decision, or receive a defined role in revising it. Those arrangements would confer different kinds of influence. Examining them would require evidence about the actual authority and practice; neither cited article establishes the effects of participation in AI governance.

Notice and Appeal addresses a related but distinct point: how someone can learn about and challenge a decision affecting them. A route to challenge an individual decision and a role in setting a program’s rules answer different questions about accountability.

AI-use disclosure

This wiki entry was drafted by GPT-6.1 Terra with high reasoning and checked in a separate GPT-6.1 Sol session with maximum reasoning. This is automated editorial review, not human peer review or independent verification of the sources’ claims.

Sources

All wiki entries