Dissertation Title: Rendering Relationships: Infrastructuring Relational Civic Practice

Abstract:
We are in a moment of civic crisis, in the US and globally. Who responds in the face of social fragmentation, exclusion, and crumbling democracies? Facilitators who bring together groups across differences and enable constructive conversation to foster openness and warmth. Community organizers who reject exclusion and stripping of agency by building collective capacity to create the conditions they seek. Elected officials and other community leaders who increase the scope of participation, co-govern with those often excluded, and enact real material change. What is core and consistent across each of these responders is their centering of the relationship in their civic practice.

In this dissertation, I draw on infrastructure studies scholarship to argue that civic technology is not a neutral substrate but an active participant in shaping what kinds of civic practice are possible. Yet most technology in this space bends toward transactional over relational, driven by economic incentives, political interests, and funding structures that reward mobilization over organizing depth. This dissertation asks what civic technology designed for relational practice would actually look like, and builds, deploys, and evaluates three systems that attempt to do so.

This dissertation offers three empirical studies across three interconnected scales of what I call Relational Civic Practices (RCPs). At the interpersonal scale, I develop and evaluate computational methods for quantifying responsivity, the extent to which participants actively listen to, respond to, and build upon one another, alongside an interactive visualization tool that makes these dynamics explorable for practitioners. At the organizational scale, I present Lumos, a relational organizing data system that supports organizers in strategizing, tracking and understanding relationships, and building capacity codesigned with expert trainers and organizers, deployed at the Harvard Kennedy School and with the Carolina Federation, a multiracial working-class organizing network in North Carolina. At the institutional scale, I present Voice to Vision, a sociotechnical system that connects bottom-up community engagement with top-down planning processes codesigned with the New York City Department of City Planning and community members in Jamaica, Queens, alongside a controlled experiment with 266 participants examining how transparency and personalization in civic decision-making affect perceptions of agency, legitimacy, and community connection.

From this work, I develop and evaluate a framework for analyzing and helping design civic technologies along relational versus transactional lines. A critical distinction emerges: civic systems often scaffold democratic practices without considering or taking into account the organizing capacity communities need for sustained collective power. Together, these studies argue that existing civic technology encodes transactional logics where relational practices are required, and that designing for such practices requires both 1) new infrastructure and 2) new frameworks for evaluation. This dissertation is for the facilitators, the organizers, the leaders, and the civic technology designers who show up to do the relational work that building power and democratic practice requires, offering them new tools, new frameworks, and new evidence that designing otherwise is possible.

Committee members:
Deb Roy, MIT Professor of Media Arts and Sciences and Director of the MIT Center for Constructive Communication (CCC)
Christopher Le Dantec, Northeastern University Professor and Director of Initiatives in Digital Civics
Eric Gordon, Boston University Professor of the Practice, Journalism and Director of the Center for Media Innovation & Social Impact