Listening to the Public: An Inductive Analysis of the Good Citizen in a Deliberative System

Previous research has explored how deliberation impacts participants and what participants want out of deliberation, but this work has taken place largely in the context of highly structured deliberative events. Increasingly, however, deliberative theorists stress the need for deliberation to be incorporated into multiple forms of engagement, particularly in the context of advocacy and interest group work. This paper utilizes a case study of Community Guides, a hybrid model of participation that incorporates interest groups into deliberative design, to ask what participants want out of engagement and whether their conception of the good citizen aligns with contemporary theoretical norms. Findings suggest that participants want opportunities for public input that center interest formation and recognition, equity, and empowerment. These results suggest a need to better integrate the voices of citizens in normative deliberative theory and research and to more fully explore how the expectations of citizenship might change when advocacy is incorporated into deliberative design.

The deliberative citizen 1 is thoughtful and respectful; they are focused on facts but open to stories and learning from others' perspectives. They contribute to their community by engaging in conversations across difference, joining organizations, and voting. At least this is how scholars of deliberation have framed them (Burkhalter et al. 2002;Fishkin 2009;Knobloch & Gastil 2015;Pincock 2012), but does this description match what citizens want out of civic life?
Contemporary citizens don't often emulate the deliberative ideals scholars have devised for them. They tend to be either polarized or checked out. Many hardly pay attention to politics at all, and those who are highly engaged tend to be highly divided (Abramowitz & Saunders 2008). People with strong political identities not only struggle to find common ground with political adversaries, they 'loathe' them (Iyengar, Sood & Lelkes 2012: 405). In short, most members of the public report attitudes and behaviors that deviate significantly from the portrait of the ideal citizen that deliberative scholars paint.
Critics of deliberative democracy question the very validity of that ideal. Engaged but polarized individuals make effective activists. Activism helps ensure the maintenance of pluralism and the development of interest groups and social justice movements, while deliberative discussion may obscure inequity and privilege the status quo (Mansbridge 1983;Mouffe 1999;Young 2001). In real-politics, however, these two forms of engagement intersect in meaningful ways (Dryzek 2010;Hendriks 2012;Mansbridge 1983;Parkinson & Mansbridge 2012). Social movements can force to the table issues that demand public discussion and create space for deliberation and civic action (Levine 2018;Smith 2016), and deliberative processes often include expressions of self-interest, such as advocacy and storytelling, as forms of expertise (Andersen & Hansen 2007;Polletta & Lee 2006) Together, these trends raise a question for deliberative theory and practice: how should our conception of the deliberative citizen shift when we more fully incorporate advocacy into deliberative design? This paper addresses this question through a case study of a local engagement program that incorporates both deliberation and interest group discussion. It relies on data provided by participants in three iterations of a city-wide Community Guide program. The project asked community members, many of whom were recruited due to their connection to local interest groups, to facilitate deliberative conversations on public policies. Throughout the project, participants were asked to reflect on how the city might improve their engagement efforts. Their responses are used to construct an inductive model of the deliberative citizen in the context of local deliberative policy making. Findings suggest that the public sees participation as a means to express their interests and work toward empowered and equitable decision making. It concludes by calling on scholars to better integrate both interest group theory and the normative democratic ideals of ordinary citizens into the study and design of deliberative democracy.

Constructing the Deliberative Citizen
The citizen plays a central role in every conception of deliberative democracy. Most deliberative reforms call for more public engagement, asking an ever-widening circle of community members to learn about public issues, talk with one another, and develop nuanced and thoughtful policy positions. Because of these requirements, deliberative scholars have produced ample research assessing whether deliberation can produce more deliberative citizens. Such scholarship articulates an image of how citizens should think and act according to deliberative ideals, but it tells us little about whether that matches how the public wants to engage in civic life. Another strain of scholarship has begun to take up this latter question, asking citizens what they like or want from deliberative engagement. The two fields of research at times present a somewhat contrasting picture. Below, I first explore the ways scholars have articulated deliberative citizenship before asking whether this model matches what the public has said they want out of deliberation. This section concludes by asking what we might learn about the deliberative citizen when we better integrate interest group ideals into deliberative practice.

The deliberative citizen
The search for the good citizen is not unique to deliberative theory. The conceptualization of the ideal citizen has undergone several revisions over the centuries (Dalton 2008;Schudson 1998), generally based in the ways that public input is expressed and measured (Ginsberg 1986;Habermas1996;Herbst 1995;Mansbridge 1999). As public officials, political professionals, and scholars develop new ways of soliciting public opinion, they shape what it means to be a good citizen. While the twentieth-century version of civic life-increasingly reliant on polling, bureaucratic procedures, and political professionalsfurther depersonalized, categorized, and aggregated public will (Habermas 1996;Herbst 1995), the emergence of deliberative democracy was intended, in part, to re-center the citizen in the public sphere by emphasizing discursive and collective engagement mechanisms (Habermas 1996). Rather than sitting on the sidelines, citizens under the deliberative model are expected to take an active role in civic life, attending events, learning new information, and finding common ground with political rivals.
Discovering whether deliberation can produce citizens capable of these tasks has been a driving force in deliberative research and is often used as a criterion for evaluating public processes. Deliberative engagement not only requires a specific type of citizen, it should also produce them (Burkhalter et al. 2002;Mansbridge 1999). While originally focused on knowledge gains and opinion change (Gastil & Dillard 1999;Luskin et al. 2002), over the past two decades scholars have expanded their search for deliberation's ' educative effects' (Pincock 2012), arguing that deliberation can produce citizens who are more capable of self-governance.
Deliberative participation can increase factual knowledge about the issue under discussion and encourage participants to recognize pertinent values at play (Andersen & Hansen 2007;Barabas 2004;Gastil & Dillard 1999;Gastil et al. 2018;Luskin et al. 2002). It can make people feel more efficacious and foster community trust, connection, and respect across difference (Hartz-Karp et al. 2010;Knobloch & Gastil 2015;Luskin et al. 2014;Nabatchi 2010). Sustained deliberative experiences can spur increased political participation and heighten the likelihood that members of the public will engage in political conversations, join local organizations, and vote Jacobs et al. 2009;Knobloch & Gastil 2015). Even if one doesn't directly participate, deliberative minipublics can impact individuals who simply hear about or engage with their work. Such secondary experiences can leave one more informed about public policy and more likely to believe they have the power to influence government decisions (Boulianne 2019;Gastil et al. 2018;Knobloch et al. 2020).
In contrast to these more top-down approaches to framing the good citizen, some scholars have turned to citizens to understand what they look for in deliberative engagement. Members of the public seem to want to deliberate (Neblo et al. 2010), and researchers have begun to focus on what motivates individuals to participate in deliberative processes. Those interested in deliberative participation hope to connect with members of their community, (Curato & Niemeyer 2013;Jacquet 2019) and to learn about public issues, both from experts and from other community members (Christensen 2020;Curato & Niemeyer 2013;Jacquet 2019). They often see deliberative participation as an avenue for empowerment-an opportunity to express their voices and to have their input influence public decisions (Christensen 2020;Curato & Neimeyer 2013;Goldberg et al. 2020;Jaquet 2019;Rao & Sanyal 2010), and random-sample deliberation may be most appealing to those who feel most disempowered (Jacquet et al. 2020). For individuals who routinely feel excluded from decision making, deliberative participation can offer a sense of ' collective effervescence,' a 'moment [that] allows individuals with disadvantaged identities… to momentarily discard the stigma of their ascriptive identities and low economic status and to slip into their sacred identity as citizens with equal rights in the eyes of the state' (Rao & Sanyal 2010: 159).

Integrating advocacy
While this research has made substantial progress in identifying the characteristics needed for deliberation and the hopes that citizens take into their deliberative experiences, they have primarily focused on events that hew more closely to the ideal deliberative situation, such as deliberative minipublics. These events ask a demographically diverse sample of participants to come together across difference, learn from experts and one another, and provide input on public decisions (Curato et al 2021;Goodin & Dryzek 2006;Ryan & Smith 2014). Minipublics and similarly structured deliberative processes can play a vital role in enhancing empowerment, but deliberative systems theory reminds us that democracy both includes and requires moments when individuals engage outside of highly structured deliberative events, particularly through interest formation and advocacy (Dryzek 2010;Hendriks 2012;Mansbridge 1983;Mendonça 2008;Parkinson & Mansbridge 2012).
Though deliberative scholarship has largely focused on conversation across difference, some advocate for the inclusion of enclave discussion among historically excluded communities as a pre-cursor for any event that attempts to foster equitable conversations across power differences (Bruneau & Saxe 2012;Karpowitz et al. 2009). Enclave communication can allow communities to engage in interest formation (Hendriks 2012;Warren 2001). Affinity group conversations provide an opportunity for those who have been excluded from power to understand their interests and strategize about effective ways to advocate for their collective needs and goals (Karpowitz et al. 2009;Mansbridge 1994;Offe & Wiesenthal 1980).
Because of this, advocacy and interest groups play an important role in deliberative systems. They help to develop and support enclaves in the work of interest formation and expression and often serve a representative function for identity or interest-based communities (Mendonça 2008). In the larger deliberative system, the integration of advocacy can alleviate existent power imbalances and create more informed and inclusive public discourse by providing opportunities for historically excluded communities to articulate their experiences and interests and push for their inclusion in the public sphere (Mansbridge 1983;Mouffe 1999;Young 2000). Moreover, activism is both effective at creating change (Chenoweth & Stephan 2011;Levine 2018, Young 2001) and mobilizing (Huddy et al. 2015;Iyengar & Krupenkin 2018;Mutz 2002), suggesting that this model of engagement is useful for democratic empowerment and appealing to many who hope to find agency in democratic systems.
If interest formation and advocacy are essential parts of deliberative democracy, then, the notion of the deliberative citizen should be inclusive of the skills associated with those acts. Some studies have already highlighted the ways this might shift what it means to be a good deliberative citizen. For example, a study of Gram Sabhas in India highlights how engagement needs change when advocacy plays a central role in deliberative processes. In these village meetings, participants are given budgetary decision-making power, and an analysis of transcripts of Gram Sabhas indicates that strategic communication and the development of community capacity play central roles in empowering historically excluded communities (Rao & Sanyal 2010). These findings slightly alter our understanding of the good citizen. Though focused on community and open to learning, they should also be effective advocates for their own interests.
This discrepancy may have consequences for how the good citizen is both theorized and enacted. Conceptualizations of the good citizen are important precisely because deliberation is a normative and pragmatic theory. When scholars discuss the effects of deliberative participation, and when they use these constructions to measure deliberation's success, they devise a standard for the good citizen that deliberative processes are evaluated against and built to produce. When those ideals fail to match the goals of the public or their needs for achieving empowerment, deliberative practice may unintentionally reproduce the top-down power structures it is intended to avoid.
The remainder of this paper will be used to explore the meaning of the good citizen from the perspective of ordinary community members working within a system of deliberative democracy that bridges the competing tensions of deliberation and advocacy. It asks how the public wants to engage in decision making and what this normative conceptualization means for how we define and measure the good citizen.

Methods
This paper addresses these questions in the context of a Community Guide program in which community members conduct affinity-group deliberative conversations with friends, family, neighbors, and associational contacts. Below, I provide details of this emerging deliberative institution before relying on an inductive approach to uncover the ways that everyday citizens conceptualize democratic engagement and the good citizen.

Case study: Community Guides
In 2017 city officials in a mid-sized university town implemented an engagement plan that aimed to foster more robust public input in policy development (Blonsley 2018). The program trained a cohort of Community Guides 2 to host small-group conversations about matters of public policy. The city modeled the program after community boards found in other municipalities, with the goal of expanding and diversifying engagement. Since its inception, the project has been used for public engagement processes focused on city infrastructure, environmental sustainability, and health and housing. 3 Though programmatic and contextual changes have been made throughout each iteration, the underlying structure remains the same: community members, acting as either individuals or on behalf of official organizations, receive training in deliberation and facilitation and then host small group conversations within their own communities about the issues in question. During the discussion, Guides collect data about the conversation and return this data to city planners and managers to use in the development of community programs and public policies.
The participants in the program represent a wide variety of community members. Organizers engaged in purposive sampling methods that blend a 'stakeholder model' with 'targeted recruitment' (Ryfe & Stalsburg 2012), attempting to engage both organizational leaders and everyday community members who have particular interests in the issue. The first iteration, focused on city infrastructure, recruited participants primarily through neighborhood associations. Community Guides served as self-selected 'informal representatives' (see Mansbridge 1983: 251) of their neighborhoods, tasked with soliciting the participation of their neighbors. The climate and housing processes took a slightly different approach, focusing on both organizational representatives and everyday citizens. Organizations were recruited to participate based on their connection to the issue or for their work with specific interest or demographic groups. Community Guides in these programs included leaders of non-profit, business, and activist organizations that focused on issues such as environmental sustainability or housing assistance. For these processes, Guides often served in either elected or appointed positions and held more formal roles as community representatives. Community members were also recruited to connect with specific cultural, demographic, or occupational communities, such as members of the BLIPOC (Black, Latinx, Indigenous, and People of Color) community or essential workers. These community members did not act as official representatives of those communities but were a conduit for bringing members of historically excluded populations into the discussion.
The Community Guide program blends purposive sampling with a critical mass approach, which seeks to over-represent historically underserved communities in an effort to prevent tokenization and better enable the articulation of diverse perspectives (Karpowitz & Raphael 2016;Mansbridge 1999). By encouraging affinity group discussion and focusing recruitment efforts on the inclusion of advocacy groups and historically underrepresented communities, it centers the voices of those most directly affected by the issue rather than attempting to mirror the perspectives of the wider population who may not see these issues as pressing concerns.

Data
The research presented herein uses a case study approach to analyze data solicited from program participants across a range of encounters with Community Guides. It relies on Guides' initial applications, their feedback on the programs, and responses provided by the wider community during these conversations.
For each of the three iterations, community members were asked to fill out an application explaining why they wanted to be a Community Guide, what was their experience with community engagement, and who they hoped to engage. Guides also provided feedback during trainings, in debrief sessions, or after their conversations. In addition to insight from the Guides, this research draws on data gathered during the community conversations. Both the sustainability and housing and health questionnaires ended by asking community members how the city could better engage residents. Though the infrastructure process did not explicitly ask for this feedback, participants often commented on these topics when asked for additional input at the close of their conversations.
All data was analyzed using an inductive approach to thematic coding (Strauss & Corbin 1997) to uncover how community members describe the role of the citizen in democratic governance. This method mirrors previous work that has asked practitioners to define deliberation or activism in their own terms (Levine & Nierras 2007;Mansbridge at al. 2006) and can bridge gaps between normative theory and real-life politics (Andersen & Hansen 2007).

Findings
Below, I relay what program participants want out of community engagement and, in doing so, focus on what this means for theoretical and empirical conceptions of the good citizen.

Interest expression
Throughout the data, Guides and participants repeatedly noted that they wanted to have a larger voice in public decision making. For many, this format was a welcome opportunity to have their voice, and the voices of other community members, included in the conversation. As one participant stated, they 'liked the opportunity to be heard.' Generally, participants focused on three interrelated functions of interest expression: (1) interest formation, (2) public recognition, and (3) influence.

Interest formation
Participants saw these conversations as a way to identify their shared interests. They expressed a desire to understand interests within their own affinity communities as well as the ways those interests might diverge or converge across difference. Affinity group conversations created space for in-depth discussion of highly personal matters and an opportunity to understand the breadth and depth of these issues within their communities. Describing their conversation, one Guide said: We also talked about the discrimination and racism they experience in [the city] on a somewhat regular basis. We talked about their concern for their kids being bullied in schools. We talked about how terrible the tap water is in our neighborhood. Many neighbors (including me) have seen black flakes coming out of the faucet on multiple occasions.
By engaging in enclave communication, the participants were able to dive more deeply into their personal experiences and identify specific problems that were broadly shared and should be brought to the attention of decision makers.
In addition to understanding their own community, the conversation allowed participants to think about how their smaller community, who tended to share similar backgrounds and demographic characteristics, related to the wider community. That same Guide continued: Everyone voiced their opinion in wanting to preserve this community to stay a Trailer Park, as there is fear of being displaced when we are annexed into the City. I also asked them about their desire to be more connected within the community... They have many stereotypes towards white people and are afraid everyone is racist. They want to be more connected, as long as people are 'nice'… When coming to America, they thought more people would know Spanish... They have the desire to learn English, although it has been very difficult to do so.
In this conversation, participants begin to highlight shared interests (addressing tap water in low-income communities and ensuring zoning policies were responsive to residents' desire to remain a mobile home community). It also allowed them to identify within, and understand their shared interests with, the wider city community, with which they wished 'to be more connected.' Here, the affinity group interests, such as a decrease in racist behaviors among community members and an increase in access to English-language education programs, would also serve the interests of the wider community by creating greater community cohesion.

Public recognition
Aside from providing space to discuss their collective interests, both Guides and participants saw the opportunity to share their stories as a means for enabling public recognition. Many Guides said that providing an opportunity for community members to share their stories was the reason they applied to be a Guide. As one Guide noted on their application: Everyday, [our organization] interacts with individuals who either do not have housing, who have housing that is currently not stable, or who have experienced one of these realities in the past. These individuals and families have key perspective to share. Some program participants… have expressed interest in sharing their story... Members of the broader community need to hear these experiences, and the community will be stronger when more members of the community who have struggled with housing insecurity are able to effectively share their stories.
For this Guide, sharing stories has a number of interrelated objectives. It provides individuals, particularly those who have generally been excluded from decision making, an opportunity to be part of and recognized in the broader conversation. Stories are also essential for the development of community. As the Guide notes, 'the community will be stronger' when they better understand the full scope of experiences and perspectives held by other members of the community.

Influence
Participants hoped that sharing their perspectives would lead to real change. Guides noted that these conversations have the potential to offer insights into the policy making process that may not have been considered otherwise. As one Guide said, 'I welcome the opportunity to learn more about the community's experiences and shar[e] them with the policy makers, thereby, aiding in the city's effort to finding solutions.' This Guide discusses the sharing of stories as an opportunity both for learning among community members and for policy makers to learn from and respond to community members.
Moreover, participants felt frustrated when solutions did not seem to be thoroughly responsive to public input. During the infrastructure project, participants were asked to discuss pre-developed options that related to different levels of investment. As one Guide said, 'People had ideas for what they wanted but felt constrained by a discussion guide that asked for feedback on three pre-developed scenarios… They want to have actual choices that matter.' When provided with pre-developed options from which to choose, some participants felt that they were left without much power in the decision-making process. As these examples illustrate, community members wanted to share their stories, and they expected their interests to be considered in policy development.

Equity
Another key theme that emerged in the data was a focus on equity. Participants repeatedly expressed a need for the city to widen the conversation and adapt to the barriers they faced to engagement. Together, these trends indicate that an equitable framework for engagement is one in which government officials take responsibility for robust recruitment practices and provide opportunities that meet the demands of both internal and external inclusion. As one participant said, it's 'great to have community involvement-not sure everyone is being included.'

Targeted recruitment
Throughout the data, Guides and participants call for more community outreach. Participants often placed the burden of recruitment squarely on the organizers. They regularly expressed the desire for both more advertising and more personalized outreach. In other words, 'People who want to know about all this don't know where to start.' Participants suggested 'more advertising; posters, mailers' and a greater social media presence. Many participants said they seldom heard about outreach efforts and suggested that their friends and family members were not aware of such opportunities. For some participants, this problem was particularly relevant to historically underrepresented communities. As one participant said, 'I would like to see really effective outreach to smaller, low-income neighborhoods citywide, so all voices are included.' Others suggested more personalized outreach or partnerships with community leaders: 'Get into neighborhoods, organizations, churches, make it personal; Or use community leaders to get information to their folks.' Participants saw affinity groups as places to foster trust and spur participation and suggested that the city should build relationships with those groups to widen the scope of voices included in the conversation. For these participants, lack of engagement wasn't simply a matter of apathy, it was a failure of community leaders to adequately publicize their efforts and build relationships with historically underrepresented communities.

Inclusion
The call for greater publicity was tightly tied with a desire for city officials to make engagement inclusive for diverse communities. For some, this meant bringing engagement to communities. As one participant said, 'Bring the group to where diverse people are-it's hard for us to always have to go to other paces.' For others, this meant acknowledging community members' competing responsibilities, such as childcare or work, that made it difficult for them to attend meetings. One participant put it succinctly: 'People are busy and may not be able to take time for a meeting. Can you go to them somehow?' Community members suggested that events offer childcare or involve young people in more family-friendly activities. They suggested hosting engagement opportunities at festivals, community events, and other places where the public already gathered. Some suggested the need to make engagement more 'fun,' while others said that opportunities should be hosted by their employers during work hours. Several participants requested opportunities that required 'minimal time commitment' or could be completed at home. They frequently mentioned surveys as a convenient way of providing feedback without infringing on their time.
While most of these comments focused on barriers to attendance, a few highlighted the barriers they face once at the meeting. For some, this equated to providing translation of meeting materials, while for others this meant conducting events entirely in languages other than English. One participant requested that engagement efforts 'speak to the community in their language.' Participants and Guides also expressed a desire for engagement designs that were flexible or that could be adapted to the culture of the community. One-on-one interviews were helpful for people who were discussing sensitive topics, while adaptations to community culture helped others feel comfortable. In their debrief session, one of the Guides expressed a desire for less formality in the conversational design: My neighbors didn't respond well to the 'structure' of the discussion, so I created more of a free flow conversation… [The] majority of the women I met with have not had an education. Therefore, they are not used to a 'formal meeting.' I wonder how we could continue to engage their voices while creating a space and dialogue that is comfortable and easy for them to participate in.
The small group setting provided an opportunity to adapt the conversation to meet the needs of the interlocuters and better center the conversation in the culture and language of groups who have been historically excluded.
Many Guides mentioned these concerns in their applications and in the debrief sessions. They saw both the setting and the facilitator as important tools for creating a sense of comfort and openness. As one Guide noted, 'I… have first-hand knowledge of how disparities in a community can affect the overall life of an individual. I would hope to gain a renewed connection to my community, while being a catalyst to change.' This Guide suggests that their own experience would allow them to facilitate a more grounded conversation, one that maintains a tighter focus on the needs of the community rather than getting caught up in traditional notions of expertise that might obscure the knowledge offered through people's lived experiences.

Empowerment
Finally, participants saw a connection between interests, equity, and effectiveness and judged engagement not by its adherence to some normative ideal but by its ability to create real change for the community members it engaged. Participants and Guides often expressed skepticism of city-led engagement efforts and worried that their time was being wasted. They requested that city leaders 'take their ideas and suggestions seriously' and give them 'more weight in decisions.' They wanted ' evidence that having these conversations are actually influencing decisions and making changes.' Such a focus on efficacy was not only a call for empowerment but a requirement for equity. One participant said that the city should Make it easier for lower income people to voice their concerns. Many of us work two plus jobs and have families, so it is impossible to attend meetings. More importantly, find a way for city leaders to take these concerns seriously. Most of us poor people are discouraged away from the conversation because we are not listened to anyways.
For some, this simply meant more transparency in the ways that public input was being used. They wanted the city to 'have follow-up on conversations and allow people to experience how the conversations and their participation has made a difference.' Others got even more specific. Participants wanted to move beyond one-time conversations and requested opportunities for sustained engagement that built community capacity and gave the public more say in the structure of engagement efforts.

Community capacity building
Particularly for Guides, this engagement strategy was an opportunity to widen the scope of who gets to engage the public. This mirrors the calls for more diverse facilitators but additionally notes the Guides' own interests in developing skills in facilitation and community engagement. Some Guides already working in public-facing fields hoped to learn how to improve their own engagement efforts beyond the confines of the program. In their application, one landlord said they hoped to use the program to 'be proactive in listening and looking for win-win solutions for property owners, investors, and community members together.' Guides often shared this sentiment, hoping to bring the skills they learned in the program back to the public outreach they conduct as organizational leaders or professionals.
Others were eager to build connections among community members. One Guide said that they were 'interested in becoming a Community Guide in the hope that I can be a liaison between citizens who wouldn't normally attend meetings with city or county officials, to voice their opinions or concerns.' Participants in conversations across topics often spoke of a need to collaborate with others. As one participant said, 'The city can provide a bridging function between other organizations… Continue public deliberation & other outreach partnerships… be a MATCH MAKER.' Another participant echoed these thoughts, requesting 'more statewide collaboration as a united coalition… Channel more resources to reaching communities of color in a way that works for them.' Such connections were valuable for not only understanding the scope of the specific issue but also for building the community's ability to work together more broadly.

Engagement designed by the public
Data gathered from both participants and Guides suggested a desire for public voice in designing engagement strategies. The question utilized for rounds two and three of the Community Guide program, and which laid the foundation for this research, was developed after a comment made by an inaugural Guide at one of their debrief sessions. They suggested that the process ask conversation participants 'how do we engage more folks from these communities?' This Guide saw such a question as ' a powerful way to reach new folks.' The City agreed that this was a good idea and incorporated the question into subsequent iterations of the program. As the responses above suggest, the public wants a voice not only in policy decisions but also in structuring civic life. As another Guide said at the same debrief, 'Bring community members into the plan design process.' When that question was added, the responses were abundant and robust. Participants offered new ideas for engagement and suggested the need for engagement at multiple points in the decision-making process.

Discussion
In one sense, these findings are predictable. Participants saw interest expression as a means of learning about their communities and influencing public decisions. They wanted equitable engagement opportunities that acknowledged different ways of speaking and barriers to participation. They desired empowerment. This echoes deliberative theorists who attend to power inequities and those who focus on system-wide deliberation (Mansbridge et al. 2010;Mouffe 1999;Young 2000). As John Dryzek (2010: 10) suggests, the realization of deliberative systems should be ' authentic, inclusive, and consequential.' Even so, what participants hope to gain out of public engagement in this context differs somewhat from how the deliberative citizen has been traditionally defined.

Deliberative theory
In their comments, participants in this program indicated that deliberative opportunities should include three components: (1) meaningful interest expression, (2) equitable engagement practices, and (3) democratic empowerment. Though these align with the three pillars of deliberation-learning across difference, inclusion, and decision making-they highlight the need for enclave and interest group discussion to be embedded in deliberative practice. They also push those concepts in the same way that critics of deliberative democracy do: by demanding the theory better address power.

Interest expression and agenda setting
Like studies that focus on what participants want out of traditional deliberative practices (Christensen 2020;Curato & Neimeyer 2013;Goldberg et al. 2020;Jaquet 2019), this study found that community members saw deliberative engagement as an essential site for interest expression. Deliberative intervention too often takes place after the agenda setting phase has been completed; through highly structured processes, individuals are asked to evaluate policies rather than share their experiences, needs, and goals (Fung 2015;Smith 2009;Warren 2009). While these types of events may play an important role in will formation (Curato et al. 2021), helping the public understand their interests in relation to a preproposed solution, the overreliance on these mechanisms without the inclusion of agenda setting functions may inadvertently advantage those who already hold power.
Participants in this process wanted to have influence at an earlier stage, namely through agenda setting. Focusing on agenda setting as a sight of empowerment better centers equity in deliberative design and grants the public more control in democratic practice (Moscrop & Warren 2016;Smith 2009). Some minipublics, such as the Australian Citizens' Parliament and the Ostbelgien Modell (Carson et al. 2013; do give participants the opportunity to set the agenda, but these designs generally require continued involvement from participants over a period of months or even years and take place across difference rather than within enclaves. Such designs can help set the wider community's agenda by pooling concerns across individual interest groups, but it does not offer an opportunity for the enclave discussions that can help historically excluded groups understand and articulate their more specific interests. This program took a different approach. Initial stages of public outreach focused on hearing community stories through enclave discussions. This allowed traditionally marginalized groups to shape the discussion and highlight their needs at the earliest stages of decision making and without committing to a months-long engagement process. For participants, this was linked to empowerment. They saw enclave discussions as an opportunity to be heardby one another and by government officials. Too often, expression is defined and enacted as an opportunity to comment on rather than contribute to. In their work on cross-group dialogue, Emile Bruneau and Rebecca Saxe (2012) refer to this as 'perspective-giving.' While dominant groups likely need to engage in perspective-taking to understand experiences outside of their own, those who have been traditionally marginalized must first have their experiences listened to and acknowledged before they should be expected to listen to and acknowledge dominant perspectives. Deliberative theory would benefit from greater attention to the ways this type of perspective giving may be incorporated into earlier phases of deliberative decision making.

Equity
A second key theme was equity. Though these findings are likely skewed by both the ideological leanings of the community and the pool of participants engaged in these conversations, participants clearly advocated for a model of engagement focused on equity rather than equality. While equality focuses on fairness, equity acknowledges and attempts to mitigate differences in resources, opportunity, and access (Beauvais & Baechtiger 2016). The participants asked for the latter in a number of ways, requesting more outreach to underrepresented communities and better accommodations for individuals facing barriers to deliberation. This has significant implications for the role of representation in deliberative theory and the ways that inclusion may be enacted in practice.
The institutionalization of deliberation has largely taken place through the adoption of minipublics made up of participant samples meant to be proportionally representative of the wider public in terms of key demographics, with Participatory Budgeting being a notable exception (Goodin 2008;Russon Gilman 2016;Smith 2009). Such designs increase equality by giving all members of the polis an equal opportunity to be chosen and often even attempt to move toward equity by slightly oversampling communities that are numerical minorities. While these interventions can play a vital role in empowering public voice and improving the quality of information available to the public sphere, they generally prioritize equality over equity in their recruitment efforts and, as previously noted, often offer limited opportunities for affinity conversations among historically excluded communities.
Community Guide participants expressed a desire for engagement practices that centered equity in design, asking for a critical mass approach to public engagement that focuses on giving more space to those who have been traditionally most marginalized (Karpowitz & Raphael 2016;Mansbridge 1999). Such a demand certainly bends more toward the social justice spectrum of deliberative theory, but in this case at least, it may better match the public's democratic desires. Deliberative scholars should think seriously about both what equitable design looks like in practice and how the goals of equity and equality might be blended in deliberative systems.
The participants' focus on equity was also highly pragmatic, acknowledging that even the most equitable deliberative design may be too demanding for some community members and highlighting time constraints as a key factor in external exclusion. Though deliberative scholars stress the need for both internal and external inclusion (Young 2000), this finding supports the suggestion that formal deliberative processes may be inherently exclusionary even as they attempt to implement equity within the confines of the process (Fishkin 2009;Lafont 2015). Participants suggested ways to address this barrier: making deliberation more fun, incorporating deliberation into the workplace, and offering opportunities for input that require significantly less time.
Scholars and practitioners have begun to think about how deliberation might be made fun, particularly through gamification and role-playing (Gastil & Broghammer 2020;Hassan 2017;Lerner 2014), but the other two suggestions have received less practical focus. Theory and research should consider how deliberation might be better integrated with work and the ways that workplaces or job types may be considered distinct interest groups. Decreasing the time commitments poses a different problem. Like participants in these conversations, survey data suggests that the public would prefer less timeintensive processes (Christensen 2020). The systems approach to deliberation acknowledges the need to bring deliberative practices to mass democracy, but it is less clear how the criterion for deliberation might be upheld in engagement efforts that incorporate the voices of the wider public through less intensive experiences. Traditionally, this type of 'thin' engagement tends to rely on more individualized opportunities and may not be able to offer the same deliberative experiences provided through more resource-intensive programs (Nabatchi & Leighninger 2015). Even so, scholars should continue to consider how deliberative ideals might be embedded into easier engagement opportunities, such as surveys, or the ways that deliberation might be dispersed over time or across community efforts and woven together to enable pseudo-deliberative experiences.

Empowerment
The previous sections discussed how interest expression may be used to influence decision making and the need to address the barriers and constraints faced by community members in designing engagement opportunities. Attending to these issues is central to using deliberation as a tool for empowerment, but participants also highlighted one other way they hoped that deliberation might empower them: by building community capacity.
Much of the literature review focused on the benefits that deliberation can offer for participants. It can increase their knowledge, political efficacy, and sense of community connection and make them more likely to engage in public life (Burkhalter et al. 2002;Fishkin 2009;Knobloch & Gastil 2015;Pincock 2012). These benefits, however, have often been couched in how they help the individual rather than in the ways that they contribute to community. Many of the participants in this process, however, saw individual gains along these measures as a means not for empowering themselves but for empowering others in their community. They saw the development of deliberative skills as a tool for fostering more widespread deliberative engagement, not simply as a mechanism for improving their own decision-making capacity. They described a greater sense of community connection as a tool for collaboration that could amplify the voices of groups who had similar interests but who were not currently working in concert. These findings suggest the ways that participant effects may be better integrated into deliberative systems theory and expanded to explore the cumulative effects of individual change.

Studying participant effects
Though deliberative theorists already contend with interest expression, equity, and empowerment, scholars of participant effects have paid less attention to these components. In part, this is because few studies have focused on engagement efforts that thoroughly integrate advocacy and deliberation. Participants in these processes wanted to acknowledge problems in current governing structures, not simply gain more faith in their power to influence them. They hoped to find community but also identified the discrepancies that kept individuals from feeling like they belonged. They appreciated the opportunity to express themselves but judged the process by whether it was effective and equitable rather than by whether they learned new information.
One thing noticeably missing was a desire to learn from traditional experts. Participants wanted to learn from one another's experiences, but they rarely expressed an interest in learning the data-driven facts that are often used as measures of deliberative success. Recent work on empathy moves in this direction, finding that deliberation can make people more empathetic to others (Grönlund et al. 2017;Suiter et al. 2020), but it does not address whether or what they learned from other participants. Future studies should explore whether participants are gaining new knowledge from input provided by fellow participants rather than focus solely on retention of evidence presented by formal experts. Similarly, participants wanted to be empowered, not simply feel empowered. In addition to assessing changes to political efficacy, research should focus on whether participants learn how decisions are made in their communities and how they might influence those decisions.
This work also highlighted the skills that participants hoped to develop. Participants expressed the need for some deliberative skills already included in the participant effects literature-an openness to learning and a willingness to build community across differencebut they highlighted the need for additional skills that have traditionally been associated with advocacy, such as interest expression and network development. This mirrors findings from Gram Sabhas that highlighted the need for participants to possess advocacy skills and demonstrated the ability for sustained deliberative interventions to create more robust engagement networks (Rao & Sanyal 2010). Together, these studies suggest the need to both include training in interest expression in deliberative design and measure whether deliberative events increase participants' capacity for understanding and articulating their own interests. Similarly, studies might assess the way that participants' personal networks changed as a result of their deliberative experience.
These findings should push scholars to rethink the ways they define deliberative engagement and the effects it might produce. Research should offer ordinary citizens an opportunity to weigh in on what they think a better citizen might look like, and normative theory should incorporate these articulations into their visions of a better democracy. Qualitative studies can allow community members to define the good citizen, and quantitative studies can be redesigned to better reflect changes that are more nuanced than increases on scale measurements or which connect individuals across communities. Survey measures might better gauge things like political efficacy or community faith along a continuum. Community members likely should not be extremely polarized, but they shouldn't place full faith in others who have little concern for their interests. Similarly, some skepticism of public leaders is likely healthy. A simple increase in political efficacy may not be a good gauge of whether an individual has a realistic and politically useful understanding of their governing power. Thinking about collective change, studies of communities' civic capacity may provide insight. This research explores whether communities have the capability and will to enact civic change at both the grassroots and institutional levels (Briggs 2008;Stone 2001). Better linking of deliberative systems theory with studies of civic capacity may help to identify the ways that deliberative interventions may increase a community's capacity to self-govern outside of deliberative events.
In addition to expanding the methodologies and theories utilized in the study of participant effects, future research should explore these issues in other contexts. Communities that have different ideological leanings may produce different results. This study focused on a specific design that takes place in the context of deliberative policy development. As deliberative systems theory notes, engagement takes place across a range of contexts, including electoral contests and mediated debate (Dryzek 2010;Elstub et al. 2018;Gastil 2008;Parkinson & Mansbridge 2012). Community members likely have different expectations for those forms of engagement. Scholars should explore the range of ways participants might act as good citizens and be open to integrating those conceptualizations into their normative ideals.

The Good Citizen
The good citizen that emerges from participants' descriptions of what they want out of engagement is more complex than the model measured through surveys of participant effects, though it does mirror the more expansive vision articulated through deliberative systems theory and studies of what participants want out of deliberation. The good citizen should hold some deliberative skills and characteristics-an openness to learning from diverse perspectives and a sense of agency that spurs action, but they should also be prepared to advocate for themselves and hold accountable leaders who fail to adequately empower the public.
Most noticeably, these findings suggest the need to better integrate power into models of the good citizen. Participants regularly expressed a need to attend to persistent inequities. Though formal deliberative institutions can correct internally some of the power imbalances that exist outside of their walls, those inequities persist. A deliberative process that increases political efficacy may be inadvertently obscuring the inequalities that can be revealed through affinity group discussions. A citizen may be less prepared to act effectively in their community's interest if they develop an illusory notion that those power imbalances can be overcome by deliberative talk. A good citizen may need to be as skeptical as they are hopeful. This paper has asked scholars of deliberative engagement to rethink their understanding of what makes a good citizen, in part by asking citizens what they want out of democratic engagement. Scholars should continue to explore the intersection of advocacy and deliberation and expand their notion of what forms of engagement might produce empowered citizens, whether or not they always adhere to normative deliberative ideals. As deliberative scholars work to institutionalize their theories of democracy (Gastil & Levine 2005;Hartz-Karp & Briand 2009;Smith 2009), the model of the good citizen articulated in those theories and through empirical studies of deliberation's educative effects should align with the ways that citizens want to engage in decision making.

Notes
1 Throughout this paper, I refer to ' citizens.' I follow Robert Dahl (1989) and Russell Dalton's (2008) lead in using this term to denote the specific role that members of the public or demos play in enacting democratic life. This acknowledges the rights of the populous to engage in self-government and highlights the demands and obligations placed on community members to realize self-governance. Though I acknowledge that this term has been used in an exclusory sense to deny civic rights to members of the public who lack legal citizenship status, I use the term as inclusive of all residents within a community who might collectively work to develop and implement decisions on matters that are of public concern, whether or not they possess the privilege of legal citizenship. 2 Community Guides were originally designated as either Community Partners or Plan Ambassadors. Due both to the unwieldiness of the name and the concern that 'Ambassador' indicated a need to advocate for the city's pre-developed plans, the program name was changed to Community Guides during the most recent iteration. That term will be used throughout for simplicity. 3 The author worked on different facets of each implementation. Though I had minimal involvement during the first iteration, I led the trainings for the sustainability program and helped city officials conduct a debriefing session with Guides. My colleagues and I led the implementation of the third round of the program, including recruitment, training, question development, and data collection.