Community-based natural resource management (CBNRM) has been a pervasive paradigm in conservation circles for three decades. Despite many potentially attractive attributes it has been extensively critiqued from both ecological and sociological perspectives with respect to theory and practice (for example Leach et al. 1999; Berkes 2004; Fabricius et al. 2004; Blaikie 2006). Nonetheless, many successful examples exist, although an equal number have seemingly not met expectations. Is this because of poor implementation or rather a generally flawed model? If the criteria and conditions for success are so onerous that relatively few projects or situations are likely to qualify, what then is the value of the model? The questions thus become: how and what can we learn from the past theory and practice to develop a new generation of flexible, locally responsive and implementable CBNRM models, and what are likely to be the attributes of such models?