In the last 10 years, the discovery that antibodies to citrullinated proteins are highly specific for rheumatoid arthritis has led to a model of pathogenesis that ties together the genetic and environmental risk factors for susceptibility and severity of disease. The authors propose that the chronic inflammation is driven by two phases of an immune response. The first phase is the priming of autoimmunity, which may occur many years before the onset of disease and is caused by environmental factors, such as smoking and infectious agents, in the context of disease susceptibility alleles. This may occur in sites outside the joint, such as the respiratory tract. The second phase is the induction of arthritis, which is associated with the generation of citrullinated proteins within the joint, which is then perpetuated as the erosive disease by a local chronic immune response. The identity of candidate synovial citrullinated antigen(s), whether fibrin, vimentin, α-enolase, collagen type II or others yet to be described, may be the key to the pathogenesis of the destructive disease characteristic of rheumatoid arthritis. There is emerging evidence that citrullination may already be modified by established therapy in rheumatoid arthritis, but more specific inhibitors of deimination may provide new agents for future treatments.