You have well encapsulated the major problems with the US approach (essentially "It's all about us...") in developing nations. I've seen it happen in half a dozen countries, every flaw you cited.

We need to stop trying to do this because we just flat do not do it well and usually do as much harm as good. Your point that development should not -- truly, cannot -- start until the security problem is resolved is spot on. I've watched us waste millions doing that stuff before the situation was resolved. Goos news is that we sometimes get to re-do it several times...

Military forces do military things -- development is not a military thing. You can use the military force to do that but it will do a mediocre job at best and too frequently, will just do a really poor job. I am familiar with all the arguments for 'why' the Armed Forces 'must' do this, to include the 'first responder bit. As I said, I've watched it many places -- and we have NEVER done it well, thus I believe most of those arguments to be fallacious and simply varied repetitions of 'that's the way we've always done it.' May be correct but that doesn't make it the best solution.

What is that old saw "If you're in a hole, stop digging..."

As Bob Killebrew said in his Blog post on the front page, "What we're doing now isn't working..." Too true. Yet we keep trying. I'm still wading through the over wordy and so far not terribly coherent offering from the great thinkers at CNAS (also on the the Front page) but I really get the impression that too many people think we should just keep doing things that do not work...