I was out today so I missed an epic rant by Ben Denison about the lousy arguments people use to argue that the Balkans had conflict because of artificial boundaries. The good news is that we have both Storify and the Storify master, Kelsey Atherton, which means we have this.
The only things I would add to this are:
Boundaries are damn near always "artificial", given that the world did not emerge spontaneously with maps with handy dandy lines dividing people. Politics among and within groups of people produced these magical lines. Some of these lines appear more natural than others, but the reality is that most are quite artificial and yet .... conflict is rare. While most war does involve territorial disputes, most territorial disputes ... do not involve war. Ooops. There are many, many quite artificial boundaries over which there is currently little conflict. But we don't notice the non-events.
Which gets us to another basic reality: the artificiality of borders is pretty constant--borders don't change much--but conflict varies. And basic social science says you cannot explain some thing that varies with something that is constant (and vice versa).
Indeed, For Kin or Country (the new paperback version with a new intro is coming out this summer) started with the puzzle that there was both more and less violence in Eastern Europe than we expected. The least legitimate, most artificial border in Europe? Probably that separating Moldova from Romania as it was drawn by Molotov/Ribbentrop but really Stalin/Hitler.
Anyhow, if anyone says that a conflict happens because of an artificial boundary, you have my permission to run screaming from the room.