Ending violent international conflicts requires understanding the causal factors that perpetuate them. In the Israeli-Palestinian conflict, Israelis and Palestinians each tend to see themselves as victims, engaging in violence only in response to attacks initiated by a fundamentally and implacably violent foe bent on their destruction. Econometric techniques allow us to empirically test the degree to which violence on each side occurs in response to aggression by the other side. Prior studies using these methods have argued that Israel reacts strongly to attacks by Palestinians, whereas Palestinian violence is random (i.e., not predicted by prior Israeli attacks). Here we replicate prior findings that Israeli killings of Palestinians increase after Palestinian killings of Israelis, but crucially show further that when nonlethal forms of violence are considered, and when a larger dataset is used, Palestinian violence also reveals a pattern of retaliation: (i) the firing of Palestinian rockets increases sharply after Israelis kill Palestinians, and (ii) the probability (although not the number) of killings of Israelis by Palestinians increases after killings of Palestinians by Israel. These findings suggest that Israeli military actions against Palestinians lead to escalation rather than incapacitation. Further, they refute the view that Palestinians are uncontingently violent, showing instead that a significant proportion of Palestinian violence occurs in response to Israeli behavior. Well-established cognitive biases may lead participants on each side of the conflict to underappreciate the degree to which the other side's violence is retaliatory, and hence to systematically underestimate their own role in perpetuating the conflict.