Between 1925 and 1980, landscape ecology underwent important changes through the gradual imposition of the view from above, through the uses of aerial photography. A new concept emerged, “the smallest unit of landscape,” also called ecotope and land unit, expressing a direct visual grasp of the landscape. This article compares the view from above as introduced and promoted by geographers Carl Troll and Isaak Zonneveld, with its (problematic) history vis-à-vis a school of ecology, i.e., plant sociology, led by Josias Braun-Blanquet and Reinhold Tüxen. This school’s internal struggles with balancing the physiognomic gaze (at the ground) and numerical methods are discussed. In comparison, the geographers based themselves on the mechanical objectivity of standardized aerial surveys, whereas the plant sociologists relied on their subjective expert judgment of plant recognition together with the structural objectivity of their numerical methods. An important communality of both schools was their inductive building of a landscape from its constituent landscape fragments. Landscape fragments were identified through abstraction and categorization, emanating from a taxonomical style of science.