Beheadings get people’s attention. This week, the India-Pakistan ceasefire along the Line of Control in Kashmir broke down in dramatic fashion. The Line of Control serves as the de facto boundary between India and Pakistan in the disputed territory of Kashmir, the product of the first India-Pakistan war of 1947-1948 with a few adjustments in the 1972 Simla accord. Despite a ceasefire since 2003, small skirmishes have been fairly common in the last decade and, while this one was more gruesome than most, I would not be surprised if the ceasefire is restored in the coming days or weeks with no long-term consequences. But the episode does appear to illustrate Robert Jervis’s spiral model fairly well. So before getting into the more grisly elements of the episode, let’s review theory and then turn to its application.
Jervis sketched out the spiral model as one pathway to conflict, where insecurities feed on one another, generating conflict even when states are not inherently aggressive. Imagine a situation in which one state, let’s call it India, attempts to make itself more secure through an action that intentionally or unintentionally makes another state less secure. When faced with this circumstance, a second state, let’s call it Pakistan, makes a countermove to increase its security, which in turn makes the first state (India) even less secure. This basic security dilemma might be exacerbated by psychological tendencies. Decision-makers know that they are taking actions for good reasons, and do not understand that their actions might be viewed as dangerous by others. When the other state reacts, the initial state views its reaction as a sign of perfidy or aggression, rather than a product of its own initial moves.
Normally at this stage, hardliners in one or both of the states demand “firm” or “strong” retaliatory steps to demonstrate that unacceptable activity by the adversary will not be permitted. And now a tit-for-tat spiral is underway. And these spirals can be difficult to control even if there is just a little “noise” in the “signal.” What if states have imperfect information or if they sometimes take action out in their phalanges about which central leaders remain unaware? It becomes very difficult to modulate tit-for-tat responses. Almost thirty years ago, Downs, Rocke, and Siverson demonstrated that even a little misperception of the adversary’s action or a just a bit of incomplete control of one’s own actions can lead to very bad outcomes for actors that are just trying to pursue benign tit-for-tat strategies (by which I mean they will cooperate so long as the other side plays nice).
Ok, so that is what the theory tells us.
Last Sunday, January 6, the Pakistan military announced that one soldier was killed and another injured in what it described as an Indian raid on a Pakistani post along the Line of Control that separates Indian- and Pakistani-administered Kashmir. The raid was followed by cross-border shelling by both states, a relatively frequent occurrence along the Line of Control despite the ceasefire. India, for its part, denied that a raid had occurred, though it had admitted to taking some undefined steps to respond to what it said was a Pakistani ceasefire violation prior to January 6. Pakistan, India claimed, had “initiated unprovoked firing” and whatever occurred on January 6 was just India responding to the Pakistani provocation. The Indian response had been calibrated, Indian spokespersons assured the Indian media. Then on Tuesday, January 8, the Indian Army reported that two of its soldiers were killed in an ambush on the Indian side of the Line of Control, and one of the soldiers was beheaded, and his head was apparently carried back by the Pakistani intruders. The Indian defense minister called the episode a “serious provocation” and “inhuman.” The opposition BJP party said the Pakistani attack was “an unprovoked act of aggression” and called upon the Indian government to respond “firmly.”
All of these occurrences took place along one stretch of the Line of Control. On Thursday, January 10, the Indian media provided an account of the whole episode that was much more complicated, with guilt dispersed across Indian and Pakistani actors. The reports suggest that the Indian military had become aware last fall of gaps in their ability to control movement along this portion of the Line of Control. They decided to construct new bunkers along the Line of Control that may have been in violation of the 2003 ceasefire. Pakistani troops protested the move—first verbally and then with harassing fire. The local Indian commander obtained permission to take action to try to stop the harassing fire. As one senior government official told Praveen Swami of the Hindu newspaper, “Let’s just put it this way. There was no formal permission to stage a cross-border raid to target [the Pakistani post]. However, in the heat of fighting, these things have been known to happen.” An Indian intelligence official similarly told Saikat Datt of Daily News and Analysis, “We believe that this was a local action purely in retaliation [for] the raid our troops carried out in the Uri sector.”
Obviously the decades-old spiral model is not the only theoretical tool that sheds light on this episode. The role played by media institutions in both countries, but particularly in the less institutionalized Pakistani democratic milieu, touches on a literature on nationalism, media, and mass opinion (see Van Evera 1994, Gagnon 1994-1995, Snyder and Ballentine 1996). The role of public pronouncements and opposition statements seems consistent with a literature on audience costs and the process by which democracies generate resolve (see Fearon 1994 and Schultz 1998). But those are other posts for another time.