Clemenceau remarked with some justification (particularly after the shoddy performance of the French military, and particularly its senior leadership, during the 1914–18 battles) that "war is too important to be left to the generals. There is a big difference between "war" and "battle," though. Exhibit A: Desert One (the abortive attempt to rescue the hostages in Tehran, which failed due to micromanagement from Washington); Exhibit B: McNamara's idiotic micromanagement of "appropriate" targets in Vietnam. Exhibit C: never mind, I think I've made my point.
Professor Bainbridge's post this morning on "new" management theories at the DOD had me singing "Everything Old Is New Again." (That is a truly horrible event to consider; I may have perfect pitch, but I cannot sing.) When I was on active duty, newfangled "management theories" that were to be applied across the board came in about every third year. I suffered through writing mission and vision statements; I suffered through "TAC Brown" paint schemes; and more other horrors than the Geneva Convention allows. Professor Bainbridge is rightly concerned about adopting a kaizen (continuous change) management strategy for the military. My own favorite remains HP's famous "MBWA" (management by wandering around)—which works pretty well in a rear-area peacetime unit, but I hate to think what it would be like in combat.
Where this really misses the point is that war is not battle. Politicians can, and should, manage war. One of the few core concepts that Clausewitz espoused that retains much persuasive power in strategic thinking is that "war is merely politics continued by other means." Perhaps battles would be less bloody if we put the politicians in the front lines. But we don't; and for precisely that reason, the actual management of the actual violence should be left to the professionals—the military officers. We can use help in managing the civilianized aspects of keeping a military, such as ensuring proper accounting for pay, and so on; but "cost-effectiveness" is not a valid or sound objective for military operations when costs are all translated to dollars.
Then there's the inherent contradiction in kaizen as it relates to the military. Kaizen contemplates producing a better product by a continuous process of improvement at all stages, traditionally in management theory research, design, production, and marketing (stretching back, as Professor Bainbridge notes, to the early 20th century), with lots of interaction among them. I'm sorry, but I have a great deal of difficulty with the concept of marketing death and destruction, or with the interaction between the production of death and destruction and research on death and destruction. <SARCASM> I'm sorry, did I offend your sensibilities? </SARCASM> The end result of any military must focus on death and destruction. The politicians tell us whom to kill and what to destroy; but it's up to us to do it. Particularly since most politicians—even a few assistant secretaries of defense I could name—wouldn't recognize the Geneva or Hague Conventions if it slid up their legs and bit them in the ass.
What Rumsfeld's initiative really seems to be doing is concentrating more and more decisionmaking authority at a higher and higher level, because only at those top levels can one observe (let alone influence) the "interaction" among marketing ("kills commies three different ways!") and production (M–16, anyone?). Not only is that megalomaniacal; it just asks for more corruption in procurement and everything else, because fewer people have to be in on the secret to make it work. <SARCASM> Of course, I would never suspect any high government official of corruption. We got rid of all of that when we fired Vice Admiral Poindexter. </SARCASM>