Did breakthroughs in core processes during the Industrial Revolution tend to generate further innovations in downstream technologies? Here a theoretical model examines the effect of a political shock on a non-innovating society in which there is high potential willingness to cooperate. The result is regional specialization in the innovation process by degree of cooperation. Tests with a zero-inflated Poisson specification indicate that 116 important innovations between 1700 and 1849 may be grouped into three categories: (1) General Purpose Technologies (GPTs) tended to be generated in large states with standardized languages following transition to pluralistic political systems; (2) GPTs in turn generated spillovers for their regions in technologies where cooperation was necessary to integrate distinct fields of expertise; (3) however, GPTs discouraged downstream innovation in their regions where such direct cooperation was not required.