Participatory approaches are increasingly used in transboundary river management to address complex social-ecological issues. Social learning, facilitating interactions and developing relational capacities, is crucial in these approaches. While the social learning discourse has advanced, relational features like trust and shared group identities remain under-researched, particularly in emergent (as opposed to deliberative) processes. This study addresses this gap by examining how trust and shared group identities shape emergent social learning and transboundary river basin cooperation in the Zambezi basin. The research questions explore how these relational features shape learning spaces (interaction, deliberation, reframing) and how these spaces, in turn, influence transboundary cooperation processes.

The study reviews literature on trust, social identity, communities of practice, and social learning. It defines social learning as a change in understanding situated within wider social units or communities of practice through social interaction. Different perspectives on social learning processes and outcomes are discussed, highlighting the lack of consensus and the need for a comprehensive approach incorporating relational features. The concept of "learning space" is introduced as an arena for interaction, deliberation, and reframing, bridging the gap between deliberative and emergent processes. The literature emphasizes the impact of relational features such as trust (competence, integrity, intuitive) and shared group identities on social learning processes and outcomes. The study uses Hartman's model of trust and Wenger's concept of communities of practice to operationalize these concepts, focusing on the attributes of shared group identity (joint enterprise, mutual engagement, shared repertoire).

A single case study approach was used to analyze transboundary basin cooperation processes in the Zambezi basin, chosen for its active river basin organization (ZAMCOM), long history of cooperation, and accessibility through an ongoing research project (DAFNE). The study analyzed three phases of cooperation (1940s-1960s, 1970s-1980s, 1990s-2017). Data collection involved literature review of key documents, in-depth interviews with diverse actors (18 interviews across eight riparian states and regional bodies), participant observations of basin meetings (10 hours across three meetings), and informal conversations. Purposive sampling was used to select interviewees meeting specific criteria (involvement in basin management, belonging to a community of practice, key roles in basin management). Data analysis used open and concept coding in MaxQDA, identifying themes like cooperation shaping, trust building, and identity markers. The analysis involved comparing and contrasting emerging patterns to develop a deeper understanding of the dynamics of trust and shared identity in social learning processes. The study acknowledges limitations, such as the small sample size of interviews, suggesting further research with wider sampling to enhance external validity.

The study found that trust and shared identities in the Zambezi basin initially stemmed from cultural and historical connections, pre-dating formal collaboration. Early cooperation (1940s-1960s) was driven by shared experience in anti-colonial struggles and commitment to regional integration, fostering a sense of shared identity rooted in collectivist values (Ubuntu). Subsequent phases saw trust development influenced by regional organizations (SADCC/SADC) and the Zambezi Action Plan (ZACPLAN), creating structural learning spaces for deliberation and joint action. The establishment of ZAMCOM in 2004 marked a significant shift, providing a coordinating body and further strengthening shared identity through common goals and shared practices in basin management. However, the study also highlights challenges. Initial national interests and priorities sometimes clashed with regional objectives, delaying ZAMCOM's establishment and impacting trust, particularly integrity and intuitive trust. Transparency issues regarding data sharing and the perceived benefits of regional cooperation caused some tensions and hindered trust among actors. Nevertheless, through ZAMCOM's institutional structures, protocols, and regular forums, trust gradually evolved. The study emphasizes the role of learning spaces (process, structural, physical) in fostering social learning, with ZAMCOM providing a crucial arena for interaction, deliberation, and reframing of basin management issues. The establishment of data-sharing protocols and other regulations facilitated cooperation and enhanced trust. The study differentiates trust development across phases, tracing the evolution from intuitive trust (based on shared culture and history) to integrity trust (through joint participation and commitment) and eventually competence trust (supported by institutional structures and shared power).

The findings demonstrate the interconnectedness of trust, shared group identities, and learning spaces in shaping transboundary cooperation. Trust is essential for collective action and learning, and its evolution across phases highlights the influence of shared identities and learning environments. The decline in trust during specific periods emphasizes the importance of transparency and addressing power asymmetries. The study supports the view that shared identities are crucial for fostering trust and impacting interactions within learning spaces. The development of trust from intuitive to integrity to competence trust reflects a process of deepening engagement and shared responsibility. The role of structural learning spaces (such as ZAMCOM) in facilitating social learning through providing structures and platforms for participation and engagement is reinforced. The case study highlights the importance of nurturing shared group identities and leveraging them to achieve collective goals. The study's findings contribute to a deeper understanding of social learning processes by explicitly incorporating relational features and utilizing the learning space framework.

This study advances our understanding of social learning by emphasizing the critical role of trust and shared group identities in emergent processes. It shows how these relational features shape learning spaces, influencing transboundary cooperation. While leveraging shared identities stimulates cooperation, it's not a guarantee; the dynamic nature of relational features necessitates ongoing efforts. Structural learning spaces are crucial for institutionalizing commitments and practices, but their success is strongly tied to pre-existing trust and shared identity. Future research could explore other relational features (e.g., power) in social learning and test the framework in other contexts and with larger sample sizes. The findings offer valuable insights for practitioners in transboundary water management, underscoring the need to foster trust and shared identities to improve cooperation.

The study's primary limitation is the reliance on a single case study and a relatively small sample size for interviews. This limits the generalizability of the findings. While the study provides in-depth analysis, the conclusions might not be directly transferable to all transboundary water management contexts. Further, the data collection focused mainly on the Zambezi basin and may not fully capture the nuances of other regions or types of transboundary cooperation.