When the Canals Run Dry: New Institutions and the Collective Governance of Irrigation Systems in Tajikistan

Thumbnail Image



Journal Title

Journal ISSN

Volume Title

Repository Usage Stats



I present a study of how water users, namely farmers, choose to participate in new institutions for irrigation governance and how these new institutions contribute to irrigation infrastructure conditions. Institutions are the sets of working rules or rules in use for manage natural resources. In the process of decentralizing irrigation management since the late 1990’s, the government of Tajikistan has created over 400 formal Water User Associations (WUAs). WUAs are non-governmental organizations, which aim to increase the participation of local water users in the management of irrigation systems. Despite significant governmental and international organization efforts to establish new WUA institutions, the degree to which water users participate in and adopt WUA institutions in Tajikistan in new WUAs remains uncertain. I explore the following research questions in this dissertation: 1) How do water users participate in new WUA institutions in Tajikistan? 2) How do new WUA institutions in Tajikistan affect irrigation infrastructure conditions? 3) How do the contextual features of a locale affect the adoption of new WUA institutions in Tajikistan?

I used a combination of qualitative and quantitative methods and theories of institutional change from the literature on governing common pool resources, as well as the disciplines of sociology, political science, and evolutionary economic geography. I collected data in Tajikistan in 2015-2016. Qualitative data consist of field observations and water user focus groups in four WUAs in southwestern Tajikistan, and semi-structured and narrative interviews with key actors involved in the development of new WUA institutions across three 100-kilometer rural-urban study sites in Tajikistan. Quantitative data include structured interviews with 159 WUA conducted in the same three rural-urban study sites.

First, I examined how four preconditions for self-organization (trust and reciprocity, common understanding, water user and WUA autonomy, and prior organizational experience) drive water user participation in irrigation infrastructure maintenance activities. I conducted a qualitative comparative case study of four WUAs using data from my field observations, focus groups, and semi-structured interviews. The four WUA case studies were selected based on water availability and estimated levels of trust among water users that were obtained from structured interviews. Results showed that the preconditions for self-organization were positively associated with farmer and WUA contributions to irrigation infrastructure maintenance activities. However, water user participation in maintenance activities was not associated with better irrigation infrastructure conditions.

Second, I investigated how new WUA institutions predicted the participation of water users in irrigation governance and irrigation infrastructure conditions. Using the Institutional Development and Analysis (IAD) framework to pose my hypotheses, I examined how collective choice arrangements, monitoring, and sanctions affected water user rule compliance, water user participation in maintenance activities, and irrigation infrastructure conditions. I performed ordered logistic regression analyses on data from structured interviews with 159 WUA leaders. Results revealed that collective choice arrangements, rather than monitoring and sanctions, positively predicted water user compliance of WUA rules. Water user compliance of WUA rules, rather than water user participation in maintenance activities, was positively associated with irrigation infrastructure conditions. Implications of the findings suggested that collective choice arrangements and rule compliance play a critical role in facilitating irrigation infrastructure conditions, yet monitoring and sanctioning rules have not been fully established in new, yet evolving WUA institutions in Tajikistan.

Finally, I studied how the contextual features of a locale affect water user adoption of new WUA institutions. I hypothesized that water users are more likely to adopt WUA institutions when WUA service areas are located close to urban centers, have greater frequencies of interactions with state officials, and have a limited history of irrigation practices. Using data from 159 structured interviews with WUA leaders, results from ordinal logistic regression analyses illustrated that WUA adoption was associated with endogenous variables that affect water users’ choice to adopt WUA institutions, such as WUA service areas’ distance to urban centers and the dependence on food production as a means of supporting livelihoods. In addition, WUA adoption was also associated with exogenous variables, such as the frequency of government officials’ visits to the locales where the WUA service areas were located, as well as household consumption of food products from beyond WUA service areas. I conducted a narrative analysis based on narrative interviews to corroborate these findings.

Broader implications of my dissertation revealed that water user adoption of new WUA institutions was contingent on local contexts and levels of trust, reciprocity, and common understanding amongst all actors in Tajikistan’s irrigation sector. Water user and WUA autonomy were important for sustaining WUA institutions beyond the initial WUA formations. Yet, some level of financial and technical contributions from the government of Tajikistan and international organization were necessary for maintaining larger irrigation infrastructure, especially since the scales of infrastructures that water users inherited from the Soviet Union did not match smaller scale WUA irrigation infrastructure maintenance efforts. Finally, existing institutional frameworks and literatures for studying common pool resources did not sufficiently capture the characteristics and evolution of new institutions (i.e., institutions as young as one to thirty years), especially since most common pool resource studies have focused on characterizing long-enduring institutions (i.e., institutions greater than 100 years). I highlighted some key features of new institutions for natural resource systems: 1) a certain amount of time for learning and adaptation is necessary for formal rules to promote normative behaviors among resource users (i.e., monitoring and sanctioning rules); 2) credible commitments and communication foster trust in and common understanding of new institutions for resource governance; 3) in the early stages of developing and implementing new institutions, resource user autonomy and participation contribute to the ongoing use of those institutions; and 4) geographic, historical, and social contexts can influence resource users’ considerations and incentives regarding whether new institutions are worth pursuing. In my concluding chapter, I emphasized the need for further study of the characteristics and evolution of new institutions for natural resource governance beyond current frameworks of well established, long enduring institutions.






Hannah, Corrie (2018). When the Canals Run Dry: New Institutions and the Collective Governance of Irrigation Systems in Tajikistan. Dissertation, Duke University. Retrieved from https://hdl.handle.net/10161/17456.


Dukes student scholarship is made available to the public using a Creative Commons Attribution / Non-commercial / No derivative (CC-BY-NC-ND) license.