Investigations into comparative constitutionalism, particularly authoritarian constitutionalism, are hampered by the fact that, for the most part, we lack a theory of constitutional evolution. We can tell where a constitutional system stands on an authoritarian–liberal spectrum in the here and now, but we have little idea about how to tell if or how it might evolve in the future. Using China as a case study, this chapter suggests what we might look for in trying to evaluate the evolutionary potential of an authoritarian constitutional system. It proposes that (one of) the keys for this potential is a rarely investigated aspect of constitutionalism which, following Michael Mann, it will call ‘infrastructural power’. Infrastructural power lies in the routinization of governmental functions; it can be catalyzed through such processes as functional differentiation, bureaucratization and professionalism. The basic claim is that such routinization of this sort promotes what is often called ‘positive freedom’. As people continually exploit this freedom, they can trigger – even in authoritarian systems – a dynamic of ‘runaway legitimation’ that progressively and spontaneously constrains the prerogative capacities of politics.