To an extent, all historical/geopolitical/megapolitical/etc analysis suffers from this. The process is (a) observe a bunch of facts about history, geography, e

To an extent, all historical/geopolitical/megapolitical/etc analysis suffers from this. The process is (a) observe a bunch of facts about history, geography, economic policies, politics or whatnot (b) explain the present (or future) state as a consequence of these facts. If you're inclined to observe certain facts or emphasize certain fields (say history), then your conclusions reflect that. The "method" of analysis is hopelessly prone to biases, from professional biases to ideological or chauvinistic. As the article shows, ideological (anti slavery) can affect the geographic (soil erosion/smallholdings) affecting and being affected by other things. "Why is Haiti Poor?" isn't, perhaps, as important to answer as "How can Haiti become wealthy." You need a less comprehensive theory for that. If our theory-building practices are free-style and bias prone, IMO it's better to minimize theory zize.