Human-engineered systems have long required overrides that prevent the unchecked execution of system procedures from resulting in undesirable outcomes. In the age of complex engineered machines whose failure can predictably kill those using them, this has led to the increasingly sophisticated design of killswitches, failsafes, and overrides. Indeed, the historical emergence of these words marks the advent of exponential increase in complexity and risk in human systems. But killswitch functions are older, and more central to governance of human affairs than buttons or triggers on modern complex systems. Killswitch governance protocols can range from fully automated to entirely human-driven and from centralized to distributed, and feature a wide range of costs and benefits. In a world whose organizational processes are increasingly automated and distributed, killswitches are increasingly central to protocol design due to their role in ensuring true distribution of governance authority, and their concomitant vulnerability to special interest capture and attack. In this essay, we survey the history and current state of the art of killswitches for the benefit of protocol designers.