Accessible space is a necessary component of urban protest. Little research, however, has examined the spatial evolution of protest activity over time. Much of the existing research emphasises the legal right to protest, however, less effort has been made to examine how micro-contexts may facilitate or impede dissent. This research focuses on how the built environment of cities can serve as either attractors or detractors of protest using a unique geocoded sample of 6217 protest events taking place in New York City between 1960 and 2006. I use a spatial count model to examine the relationship between the built environment and protest intensity. The results point to significant shifts in where protests have occurred over time. Protests become increasingly spatially concentrated, with a disproportionate amount of activism taking place on or in close proximity to privately owned public spaces. Spaces in close proximity to powerful organisational or institutional targets also experience heightened protest activity. Overall, I show that the built environment, and the social relationships creating it, powerfully influence where dissent occurs. This is consistent with the advent of neoliberal policies directing urban spatial restructuring, which have brought about a process of structural funnelling for protest, ultimately making events more likely to occur in spaces that are hostile to mobilisation.