This sort of stuff doesn't require the politicians in question to change their minds. Think of it as a two-fold selection effect. First, the candidate who alrea

This sort of stuff doesn't require the politicians in question to change their minds. Think of it as a two-fold selection effect. First, the candidate who already holds views of which the defense industry approves, receives campaign contributions therefrom and is therefore more likely to win. Second, I would think that defense companies are hardly likely to donate to a politician of the Ron Paul stripe when they are in a race against someone more ... conventional. As rayiner pointed out elswhere in the thread, correlation does not imply causation. This mechanism may have deleterious effects on our political discourse and decision-making, but it's hard to call something that doesn't involve a politician changing his mind as a result of money "bribery". This mechanism is a systems problem, it seems to me, not a problem with individuals. Now I'm sure this is far from the only thing that's going on. (In particular, I think I remember Lawrence Lessig talking about how money may or may not buy a change of views, but it surely buys access. You pick up the phone for the guy that contributed $LEGAL_LIMIT to your campaign.) Moreover, I haven't ruled out that the money changed their minds---just provided an alternate hypothesis. Nonetheless, this toy model provides a not-totally-implausible explanation for things we observed, without invoking bribery. Edited for style.