The institution which is the US Army is generally considered monolithic and hidebound. I tend to agree with that, and in reading Swift Elusive Sword, I started thinking about the force structure here. Boyd's main contribution to warfare was the invention of the OODA loop, and the successful commander is the one who manages to "get inside" his opponents OODA loop. Large, technologically constrained organizations will tend to lose engagements because the OODA loop of such organizations is too long, and the enemy can quickly respond and re-maneuver to take advantage of hesitations in the opponent. Now, by technologically constrained I do not mean ones without technology. In fact, it is typically the opposite which holds true. The more technology a force has, the more it is constrained by that same technology. One of the oft repeated mantras of the COIN disciples is "infantry, infantry, infantry." Computers, updates, reports, etc., will actually constrain a force from responding quickly.
The best officer I ever served under called all of the NCOs in a room before our deployment and told us "I will always explain the orders I give, and why I am giving them, and I expect you to do the same with your men." The counterpart to this, of course, is that sometimes orders will be given without time for explanation, but having known that there was a reason before, you will know that there is a reason for the quick orders, too. However, the unspoken element of this is directly applicable to COIN. By knowing the purpose underlying the orders being given, all of us NCOs, and our soldiers, were aware of the overall intent of the commander, and thus we were able to take our own initiative to solve problems and respond to situations without having to ask for orders at each point. Although this is generally an element of US Army doctrine, I have known many commanders who simply put out orders without explanation, leaving the lower officers and NCOs to attempt to interpret and request guidance in unfamiliar situations. This adds to the time it takes that lower officer to complete his OODA loop.
Another problem is the transmission of information inherent in the typical Army battalion/brigade. As it was put by one soldier here: "information goes up the chain, it never comes back down." We have our S2 (intelligence) units compiling information and reports collected from lower levels, and then maybe providing reports to the battalion/brigade commanders, who use that information to design missions. Well, if you really want to minimize friction, speed up your OODA loop, and run an efficiant organization, those reports need to be distributed down to the very bottom level. I remember a number of stories from fellow soldiers who had just gotten back from Afghanistan where they had gone to a village, spoken with people, and only months later found out that they had been having dinner with a high value target. I'm sure it happens in Iraq, too. I think it actually happened with our team a few times.
Now, this all gets to a point which I had intended to start with, but luckily this isn't an academic paper, its just a blog: our current force structure in Iraq is still based on a maneuver warfare, big Army war. We have slowly shifted to COIN, which is a good thing, but we are still organizing our units in traditional ways. We have individual brigades, who are given an area. Then each battalion is given an area, then each company in the battalion is given a smaller area. Sounds all well and good, except that you have Armor battalions, infantry battalions, Engineer battalions, light infantry, heavy infantry, mechanized infantry, etc. I can't tell you how many tankers I've met who complain that they have to go out on foot patrols and fight as infantry, when all they want to do is ride in their Abrams/Bradleys and lay down fire. Now - even though we are much "lighter" in a COIN fight, I think there is still a place for seriously heavy firepower.
Now, each brigade tends to be a combined arms organization. Some light forces, some heavy forces, some armor, etc. But why do we stick this at the Battalion level? This is left over, as near as I can tell, from the days of Mass Warfare, large squares of infantry arrayed against others, and the chess moves of trying to outmaneuver the opponent so that you're heavy infantry hits his light infantry, you're cavalry hits his heavy infantry, etc. And you did this on a large scale. But here, in Iraq, with an enemy which shifts as quickly as the stock market, none of that paradigm holds true. We need to completely break away from the battalion/brigade focused doctrine we currently have.
Abu Muqawama has a blog entry that touches on this, in terms of brigades as advisor teams. And a wonderful analogy with ice cream. To continue the analogy, I would say that if a "scoop" represents a battalion, we need to move away from a Baskin Robbins paradigm and into a Cold Stone Creamery paradigm. We need to mix up our "scoops" at the very basic level - since we know that its the soldiers who are on the streets that will win the COIN fight, those soldiers are the level at which the combined arms approach needs to be addressed. A platoon of light infantry, a platoon of heavier infantry (Bradleys/Strykers), and an assortment of other elements (engineers, MPs, etc.) Since COIN is so often modeled off of SF, look at how a SF team is organized: one/two people have specific areas of expertise, rather than having a "weapons team" and a "commo team" or some other unresponsive system. We should be doing the same for our maneuver units.