Aims Although the impacts of alien plant invasions on native communities are well documented, the mechanisms that underlie these impacts are poorly resolved. Little is yet known as to why invasive alien plants have greater impacts than native species on their neighbouring species within a community. We investigated two potential mechanisms by which invasive plants may harm neighbouring species: resource competition and novel compounds of the invaders. Locations Field removal experiment in France and Switzerland, and common garden experiment at the University of Fribourg, Switzerland. Methods We first conducted a field removal experiment to compare the response of neighbouring communities to experimental biomass removal of two alien invasive vs two dominant native plant species in two sites of each target species (thus eight sites in total). In addition, we conducted a common garden pot experiment to compare the effect of eight triplets of closely related target species (24 species in total), each consisting of an alien invasive species in Europe, a dominant native and a random native species, on the growth of a grass community under the addition or not of activated carbon (AC). Results (1) Field removal experiment: in the control plots, we found that the biomass of native species explained a substantial amount of the variation in biomass of the neighbouring communities, but the invader's biomass did not. Moreover, the neighbouring community recovered from removal of the dominant native species, showing a significant increase in biomass, but did not after removal of the invaders. (2) Common garden pot experiment: we found a larger effect of invasive species than native species on the growth rate of the grass community, but this difference disappeared when AC was added. Conclusion Both experiments suggest that the impact of some invasive species is not driven by resource competition, but by other mechanisms, most likely by novel weapons or novel plant–soil interactions.