John,
In Iraq, the interviews would have been conducted under my concept by Petraeus and Odierno with, my preference, Crocker participating.
I had to think about that one. I know that DIV CDR's, BDE CDRs and even BN CDR's often do interviews, review files to snatch up in bound leaders who fit their mold, pull along leaders they know they can count on to strengthen their team.

I suppose leaders of JTF HQs, Sub-Unified CMDs, and the COCOMs do likewise within their HQs, but I have not really thought of their influence as the operational CDR on the command slate based on conditions. To have the effect you propose, I think you'd have to catch it at the CSL before they went into the ARFORGEN cycle.

Marc had mentioned merit (which I'm in favor of), but I'm not sure this represents merit exactly, but more along the lines of temperament, or qualification. I am in favor of putting round pegs in round holes, but I think here the issue is the same as fully meeting operational requirements in general (as opposed to filling them). As part of the Generating Force the leader development and education process should also reflect the range of operational requirements we anticipate and prepare leaders for those challenges in a given position or rank - and the promotion and CSL processes should provide outputs which reflect what we value (I've heard plenty of stories from 05s and 06s I respect who sat on boards that this is not always a merit based system)

The problem I think has been effectively communicating operational requirements, qualifying and quantifying them, so that we establish the rationale for changing our systems. After all leadership is in itself the enabling capability for all our other functions. Assuming we could get that right, and have the will to change what we value to reflect all the operational requirements for leadership such that we identify, advance, and retain leaders who demonstrate they meet those requirements, then the issue of leaders who did not understand the requirements of their position should be minimized.

I know that is theory, and does not account for external influences to the processes, but I think that is the underlying issue. I see this as an institutional shortfall. Note- I did not say failure because some of our processes have made possible advances in leadership - we just may not know which ones, or be able to assign a value to them - which means we might accidentally adjust the wrong thing. I believe we can do better, and one of the projects we're working on (I mentioned it on the SFA as an Individual Capability thread) may have some merit in this vein.

Until then, commander's at all level should probably be prepared to scrutinize the potential and demonstrated performance of their subordinate leaders and adjust them based on their judgment to better support the mission.

No low hanging fruit here - just hard work.

Best, Rob