Forest edges are sometimes presented in the scientific literature as one of the causes of the degradation of biodiversity, linked to the fragmentation of forest areas, and sometimes presented as species-rich habitats. This apparent paradox makes it difficult to define land management strategies that are favourable to the preservation of biodiversity and the associated ecosystem services. An analysis of the literature on the subject shows that this paradox is the result of confusion between several ecological effects improperly grouped together under the same expression "edge effect". Fragmentation-related edge effects concern the level of habitat fragments and their species composition, whereas local edge effects concern portions of edges at a much finer spatial scale. Moreover, these effects are not expressed at the same moments in the spatial dynamics of edges. Fragmentation edge effects concern the consequences of a change in the characteristics of forest areas, and thus the appearance of edges, whereas local edge effects concern edges that are already in place. This analysis therefore shows that the creation of new edges through fragmentation is often the cause of a degradation of biodiversity at the level of woodland fragments, whereas existing edges, after a certain period of time, may harbour a locally rich biodiversity. This analysis should lead to a more precise definition of edge effects in order to reduce confusion.