While co-management arrangements may link government management agencies directly with local resource users, this connection is rarely an "institutionalized, partnership of equals." In the case of caribou co-management in northen Canada, the work involved in creating bridges between the knowledge and govenance structures of traditional aboriginal communities, and regional and national scale Canadian management agencies, is formidable. Just as aboriginal communities have resisted attempts to have govenance powers unilaterally devolved from the Canadian govenment rather than their pre-existing rights and responsibilities recognized, so have communities resisted efforts to have their traditional knowledge defined and co-opted by outside forces. This thesis has outlined these lines of power struggle and resistance, but is primarily focused on questions of social learning. When and how have comanagement efforts managed to create the space for double loop learning, Iearning where aII participants have the opportunity to question their own assumptions about what they know about human-caribou systems? Are co-management arrangements succeeding in building environments where individuals and organizations, often with different views on data interpretation, social values, conservation principles and governance, can come together to make decisions? This exploration is rooted in the author's opportunity to live and work with the Lutsël K'é Dene First Nation, where the project examined a number of themes including: the historical background of the early exchanges between Dënesçtine peoples in the Northwest Territories and the Canadian government, and the role of co-management in Iocal empowerment in a community with regular and direct experience with barren ground caribou. The thesis then tums to the links between traditional knowledge, community institutions and organizations and the means communities use to ensure that these links are not broken when information is shared with organizations outside the community. A preliminary investigation of Lutsël K'é elders and hunters knowledge of changing caribou movements explores local concepts of natural versus unprecedented changes in barren-ground caribou migrations. Finally, the thesis looks at the role of trust in caribou co-management systems, trust between people and in the knowledge employed to make management decisions. The objectives of the thesis were to look at how cross-cultural differences can be negotiated in the co-management of barren-ground caribou herds, to examine how community-based caribou monitoring can be implemented, and to identify the mechanisms that create links between westem scientific knowledge and traditional ecological knowledge and their applications to co-management. The thesis found that without first understanding that traditional aboriginal barren-ground caribou hunters, biologists and government policy-makers express the complexity and uncertainty of existing knowledge of caribou systems in different ways, it is not possible to begin to negotiate cross-cultural understandings of caribou herds. Community-based caribou monitoring must ensure that traditional caribou hunters do not solely bear the consequences of management decisions, but also build opportunities to benefit from and take responsibility for management actions. This is reflected in recent efforts to establish community-based caribou monitoring programs by the Beverly and Qamanirjuaq Caribou Management Board and within its recently revised management plan. This thinking is also reflected among those communities participating in the Bathurst Caribou Management Planning Committee. The creation of linkages, not the integration of scientific knowledge and traditional knowledge, will be possible only when discussions of the differences between knowledge systems move beyond naÏve dichotomies that make unfair comparisons between whole cultures and the specialized technical knowledge of specific members of a culture. Without venues where biologists and aboriginal caribou hunters and eiders can share their knowledge in direct and regular interactions, the social learning involved in linking knowledge systems is not likely. There are signs that current efforts to create multiple caribou monitoring techniques requiring collective understanding of the signs of feedback from caribou populations, may allow this kind of interaction where it has not systematically occurred before. The thesis found that while addressing imbalanced power relationships between comanagement participants is key to examining the success of decision-making processes, the conditions for double loop learning to occur (where participants examine their assumptions) do not necessarily follow from balanced power relations. The capacity for double loop learning is dependent on the building of a "safe" environment where a nonthreatening dialogue between participants can take place. The capacity of comanagement organizations to act as catalysts bridging differences in scale and knowledge systems are influenced by rapid outside changes as well as rapid population or cultural changes. However, developing the capacity to picture change and cope in the face of change is dependent on alternative management systems that allow innovative learning environments. Double loop learning cannot be achieved without trust between participants where multiple perspectives and learning traditions can be respected.