The ‘Dictatorial Peace’ logic is one of the most innovative improvements to the ‘Democratic Peace’ approach in international relations. The elaborated analyses of the evidence have allowed the inference that authoritarian regimes might face similar constraints or impulses when deciding whether to increase belligerence or not. However, this assumption is valid after the classification of these non-democracies in several categories according to some specific institutional characteristics. Thus, an authoritarian leader who faces a coordinated elite willing to punish him after wrong foreign policy decisions might show – like democracies – more reluctance to escalate in a crisis than personalist regimes. I test this logic in one particular case, the ‘Beagle channel (1977-1979)’ crisis that compromised the Chilean personalist regime, headed by Army General Augusto Pinochet, and the Argentinian military junta presided over at that time by Rafael Videla, the commander-in-chief of the Army. How valid were the dictatorial peace assumptions in this particular case? Did the institutional constraints in these two former authoritarian regimes – leader and audience – reacted as expected after the crisis unfolded? In the following pages I briefly summarise the dictatorial peace debate, focusing on the models proposed by Fearon (1994), and Weeks (2008, 2012). After that I discuss the ‘Beagle channel’ crisis underlying the relevant aspects of this theoretical framework, and conclude by arguing that while is possible to detect some elements described by the model – for instance, the late international audience involvement, which increased the costs of war – other factors that permitted the backing down are not sufficiently examined, particularly the absence of a homogeneous position within Argentina’s military junta.