None of the contemporary scientific disciplines studies humans as humans. Science partitions the phenomenon of being human into different 'slices', i.e., different epistemic objects, and does so according to its disciplinary structure. Through this epistemic fragmentation the human being itself disappears as an epistemic object of science. This paper analyses whether the epistemic fragmentation of 'man' in the contemporary sciences is something we should condemn or not. 1 shall argue that we should not, at least not in principle. Fragmentation can be of epistemic value and it can fail to be so. To show this, the paper compares separation and integration, understood as two complementary scientific research strategies, with respect to their heuristic fruitfulness. The main example for separation will be the hardening of the divide between nature and culture at the beginning of the 20th century. The paper ends with a note on what a philosophy of science perspective can and should contribute to an anthropological turn in philosophy.