Sub-national treasure

0.5 hours writing, 4.5 hours researching, 0.5 hours administrivia.

Greetings from the NY Public Library. I brought an Ethernet cable so I could plug in directly from within the spectacular third floor reading room. If you've ever seen the movie The Day After Tomorrow, it's the big room with the chandeliers and picture windows that the refugees in New York congregate in. Couldn't think of a more inspirational place to write, it's just beautiful....

I continue to investigate the idea of doing a sub-national comparison. Today's main insight was that while La Violencia was widespread, it was really concentrated in the Andean region of the country. The Atlantic and Pacific coasts, the southern region on the Ecuadorean border, and the Amazonian region in the south were really not affected. I'll confess my ignorance here and say it's really not clear to me yet why the Atlantic Coast in particular was relatively immune to the violence. The Amazonian region, and to a degree the southern region near Ecuador, that makes sense: they're not densely inhabited. But the Atlantic Coast has 3 major cities and a large cattle-ranching presence: why didn't it experience conflict? This is something that's so seemingly taken for granted in the sources I've read that I'm almost embarrassed to ask the question, "Why not there?"

In thinking about a sub-national comparison, I need to get a clearer sense of the question that information will help me answer. If I'm trying to show that variation in the type of security forces (politicized vs. militarized) shapes the types of armed challenges the regime characteristically experiences (insurrection vs. coup), then I'd ideally want to compare two places with different types of security forces and show that they experience different types of armed challenges. However, there's a question of units of analysis: if I'm interested in regime stability at the national level, what will a sub-national comparison tell me?

Well, for one thing, national-level directives are carried out incompletely in Colombia. The difficult geography made national-level coordination - of the state, of political parties, heck even of the road system - difficult for longer than you would expect. For the time period I'm looking at, when urbanization was in progress, it's plausible to think that some states within Colombia would be more politicized than others in their security forces. But this may be mixing up de facto with de jure, and design is supposed to be about de jure.

Anyway, an interesting problem to continue grappling with tomorrow. Even if sub-national comparison doesn't end up being a feature of my dissertation, I needed to get immersed in the details of La Violencia anyway, so this is as good a lens as any through which to do that.

One last thing: I was looking into historical maps of Colombia that the Public Library has in its collection - luckily, there seems to be a good amount of continuity in the shape of the states in the two time periods I'm looking at, and these maps will help me substantiate that - when I came across this, which led to this. Omigod, this is a movie just waiting to be made. "E. Forbes Smiley III": can you imagine a better name for a guy who stole dozens of rare maps worth hundreds of thousands of dollars from six libraries and then sold them on the Internet to pay for his house on Martha's Vineyard? Ari, I'm 2 for 2 in the past week on fantastic projects that someone needs to snap up the rights to, pronto. A cut would be nice, but I'll take a "Special Thanks" in the credits....