Sustaining Peace Project

For policy makers, civil society, and local communities around the world, specifying and achieving “sustainable peace” has too often proven to be elusive. This is in part due to challenges comprehending peace in complex societies as well as to a fragmented understanding about the conditions and processes conducive to sustainable peace. Research about peace has primarily studied the pathologies of war, violence, aggression and conflict – and peace in the context of those processes. Few efforts have been devoted to studying peace directly as a positive state. In addition, research and practice relevant to peace are typically rooted in specific disciplines while interdisciplinary approaches are limited. As a result, the complexity, multidimensionality, dynamism, and sustainability of peace are not well understood, contributing to a lack of coherent, measurable, and implementable policy agendas that effectively sustain peace.

Our Sustainable Peace Project, grounded in dynamical systems theory and informed by historical and anthropological evidence indicating that humans are fundamentally cooperative beings, seeks to:

• Advance conversations about peace with academic experts, policy makers, and local stakeholders;
• Bridge the gap between the academic understanding and practical applications of sustainable peace by providing policy-relevant tools; and
• Advocate for a more comprehensive and fundamental understanding of sustainable peace.

Our Current Work

Nestled within the Sustaining Peace Project, the Peace Speech Project investigates how language shapes peace and conflict in societies. Using machine learning, natural language processing, and insights from peace studies, we analyze linguistic patterns in news and social media across high- and low-peace countries. Our team has studied over 900 million articles from 20 nations, revealing consistent differences in how stories are told in more peaceful versus more fragile contexts. These insights inform our machine learning classifier, trained on global peace indices, which can track shifts in peacefulness of communication in real time.

We are currently developing an interactive app that provides feedback to journalists and readers about the levels of peacefulness in their news content. The app is designed as a research tool to explore how language can encourage more constructive public discourse. This project brings together data science, qualitative research, and human-centered design to explore how communication can actively support more peaceful societies.

See more below from our 2025 Data Science Day poster presentation. Learn more about our research, collaborators, and tools on our full project site.