It Takes Two to Tango

In a large fraction of all cases where a state faces an immediate military threat, it chooses to respond without resort to the threat, display, or use of military force. We seek to account for this phenomenon of nonmilitary response and to assess the utility of this strategy. We present a series of hypotheses to account for nonmilitary response, based on the literature on reciprocity, and test them using cases of interstate conflict involving Latin American states since 1816. Target states are less likely to respond militarily when the issues at stake are not highly salient, when the initiator's threat does not involve the actual use of military force, and when the target is substantially weaker than its adversary. A militarized response is more likely when the target state has employed a similar strategy in past conflict against the same adversary, regardless of the outcome of that past confrontation, and when the target is preoccupied with a civil war. No significant effect appears for contiguity or for the target state's preoccupation with another militarized dispute. Regarding the consequences of nonmilitary response, responding with militarized means did lessen the likelihood of defeat in a confrontation and increase the prospects for a successful outcome, but this advantage was small. From a more long-term perspective, nonmilitary response served to decrease the likelihood of future confrontations against the same adversary and to increase the length of time until the outbreak of the next confrontation. There thus seems to be a trade-off in terms of the outcome of the confrontation between the slight short-term costs of responding militarily and the longer term benefits of helping to postpone or avoid future conflict.