The initial, anthropocentric view of the deep ocean was that of a hostile environment inhabited by organisms rendered lethargic by constant high pressure, low temperature and sparse food supply, hence evolving slowly. This conceptual framework of a spatially and temporally homogeneous, connected, strongly bottom-up controlled habitat implied a strong constraint on, or poor incentive for, speciation. Hence, the discovery in the late 1960s of high species diversity of abyssal benthic invertebrates came as a surprise. Since then, the slow-motion view of deep-sea ecology and evolution has speeded up and diversified in the light of increasing evidence accumulating from in situ visual observations complemented by molecular and other tools. The emerging picture is that of a much livelier, highly diversified and more complex deep-sea fauna than previously assumed. In this review we examine the consequences of the incoming information for developing a broader view of evolutionary ecology in the deep sea, and for scavenging amphipods in particular. We revisit the food supply to the deep-sea floor and hypothesize that the dead bodies of animals, ranging from zooplankton to large fish are likely to be a more important source of food than their friable faeces. Camera observations of baited traps indicate that amphipod carrion-feeders arrive within hours at the bait which continues to draw new individuals for days to months later, presumably by scent trails in tidal currents. We explore the different stages of food acquisition upon which natural selection may have acted, from detection to ingestion, and discuss the possibility of a broader range of food acquisition strategies, including predation and specializations. Although currently neglected in deep-sea ecology, top-down factors are likely to play a more important role in the evolution of deep-sea organisms. Predation on amphipods at baits by bathyal and abyssal fishes, and large predatory crustaceans in the hadal zone, is often observed. Finally, we develop hypotheses regarding the effects of past, present and imminent anthropogenic activities on scavenger biomass and how these can be tested with the most modern tools.