6 ideas
Analysts explain government action through implicit conceptual models that decide which questions get asked, which evidence looks relevant, and what counts as an answer. Swapping the model changes both the story and what 'explained' means, so the lens has to be chosen explicitly and not inherited by default.
Treats the government as one unified actor that sets goals, weighs options by their consequences, and picks the one with the best value. You explain an action by rebuilding the calculation that would make it a sensible choice. That makes it powerful for inference but prone to inventing coherence and intent that were never there.
Government action comes out of large organisations running their standard operating procedures and pre-set repertoires, not deliberate choice. Leaders can only choose among the options organisations already have ready. What looks like strategy is often routine carried out on autopilot, which explains odd details better than intent does.
Government decisions are political results of bargaining among players in positions who hold different goals, stakes, and power. The outcome is often something no single player intended.
Soviet crews in Cuba built missile sites in the standard layout used at home and did not camouflage them until after U-2 flights spotted them. That gave away a supposedly secret deployment. A rational-actor reading struggles with the blunder, but organisational routine explains it: units did what their manuals prescribed.
ExComm's choice of a naval 'quarantine' over an air strike came out of shifting coalitions, personal influence, and the Air Force's admission that a 'surgical' strike could not be guaranteed. The Navy's own procedures for running the blockade then clashed with the President's wish for close control. The chosen option and how it was carried out came from separate processes.