IntroductionParticipatory forms of evaluation aim to produce a range of empowering outcomes and impacts, including increased community capacities in planning and conducting evaluations and broader stakeholder participation in decision-making. Other outcomes include improved communication and trust among stakeholders, and constant improvement of initiatives in ways that meet community or client needs. In her extremely positive assessment of participatory evaluation, Diez (2001: 907) suggests that this methodology can be a useful tool to 'mobilise communities for regional action, empower local agents and enhance learning capacity'.Many researchers argue that building community capacities and fostering empowerment are more effective ways of achieving sustainable community development than programs and success indicators imposed by outside experts (Harrison, 1998;Mobbs, 1998). They point out that outside experts usually have limited knowledge and understanding of the particular context, needs and issues of a community. Local solutions to achieving sustainable community and economic development are therefore seen as important outcomes of a capacity building approach to evaluation. Nevertheless, evaluation methodologies which explicitly aim to be participatory raise many complex theoretical, methodological and ethical issues, as Gregory (2000), McKie (2003), Rebien (1996) and others point out. They include issues related to stakeholder representativeness, the greater level of time, energy and resources required to conduct evaluations and develop trust, the potential for dependence on the facilitator/professional evaluation consultant, and for conflicting agendas and perspectives of various stakeholder groups to hinder success (Gregory, 2000;Papineau and Kiely, 1996;Rebien, 1996). Rebien (1996) also notes the conceptual weakness of the concept of participation, while the contested 2 concept of 'empowerment' is often used in uncritical ways. Indeed those advocating participatory approaches to research and evaluation often make idealistic or naïve assumptions that community participation will automatically lead to empowerment.Thus, while power is a central issue in participatory forms of evaluation, it is often ignored (Gregory, 2000). Vanderplaat (1995: 85) suggests that even the more critical models of evaluation have failed 'to deal, in any meaningful way, with the concept of relative power, or more specifically the unequal distribution of discursive power, a central construct in empowerment-based social programming'. As McKie highlights in her recent paper on the concept of 'rhetorical spaces' (McKie, 2003), there is a need to focus on the communicative and relational dimensions of participatory evaluations which can affect their outcomes in unintended ways. This requires rigorous analysis of both the intended and potentially unintended or negative impacts and effects of participatory evaluations.This paper presents outcomes of a detailed analysis of the empowering and, at times, disempowering impacts of the imple...