I was reading Myra Macdonald’s excellent piece on the Siachen conflict and this portion jumped out at me:

Yet Siachen was a single-minded battle for territory. Soldiers in both countries told me that “not 1 inch of land” could be ceded to the other side. And the result has been a fight with many casualties and little gain, employing World War I-style trench warfare at 18,000 feet.

Siachen is the largest of a number of glaciers — giant, rubble-strewn, potholed, cracked open by crevasses — that slide down from the jagged peaks of the Karakoram Mountains into the snow-filled valleys below. Nothing grows there; no animals can live there. The region lies on the outermost rim of the old kingdoms of Jammu and Kashmir. Since the times of the maharajah, it has been a place of myth. Only the bravest of explorers dared to go there, and those who did traveled in awe of the mountains. It’s hard to imagine how anyone could try to own them.

She goes into the history behind how the two militaries came to occupy the glacier, and it’s worth reading the whole thing.

The thing that struck me about this is how perfect an illustration of the simple prisoner’s dilemma this is. Both Pakistan and India would much prefer they “cooperate”, if by “cooperate” we mean “mutually get the hell out of there”. But such an outcome is impossible because the individual incentives to defect are simply too powerful; neither side wants to run the risk of ceding the territory to the other, and as such, remain locked in this suboptimal status quo.

Note that very few territorial conflicts can be strictly thought of in this way. The reason Siachen is unique (or close to unique) is that the only reason the players want it is because the other side claims it. There is no intrinsic value to the territory in question. No oil, no lebensraum, no minerals, no nothing.

Nice to look at. Not necessarily worth fighting over. Photo: EPA/Khaqan Khawer

When it comes to other territorial conflicts, the incentives are different. Presumably there is something about that territory that is valuable. Unlike the Siachen prisoner’s dilemma, a state would want such territory, and not want to let go, irrespective of what the other side is doing.

The distinction can be thought of this way: in the case of Siachen, both sides would rather not have it, and just let it be a no-man’s land — if they could convince the other side to do so as well. In the case of random valuable territory X, both sides would rather have it. This makes those conflicts more akin to the game of chicken than prisoner’s dilemma.