Protein aggregates such as Lewy bodies have done much for the scientists in the field of neurodegenerative diseases: They have highlighted the affected cell populations and they have trapped the mutant disease protein. Instead of a good reputation, however, protein aggregates have received incriminations, because they are consistently seen at the site of crime. Reviewing the arguments, crucial evidence has become known that (a) the specific neuronal pathology precedes the appearance of protein aggregates in mouse models of disease, (b) the neurodegenerative disease in patients occurs with comparable severity when visible protein aggregates remain absent, (c) the neurotoxicity in vitro is best reproduced by oligomers, not polymers of the mutant disease protein. Is it feasible that protein aggregates are inert byproducts of the disease protein malconformation, or that they even represent beneficial cellular efforts to sequestrate the soluble toxic disease protein, together with normal or aberrant interactor proteins? Whatever the answer will be, one positive role of protein aggregates seems clear: In contrast to earlier speculations that random cytoplasmic proteins are trapped within the aggregates, scientists now believe that the composition of the Lewy body reflects the network of interactions between crucial players in disease pathogenesis, such as the PARK1, PARK2 and PARK5 protein.