McChrystal's Response on Force Size Poses More Questions Than it Answers
From Ackerman’s coverage of McChrystal Testimony:
Even Further Clarification on Afghan Security Force End-Strength « The Washington Independent
During today’s House Foreign Affairs Committee hearing on Afghanistan, Rep. Gene Green (D-Texas) asked if an increase to 400,000 Afghan soldiers and police — that figure stands at a little under 200,000 total now — will be enough.
Gen. Stanley McChrystal replied that a “detailed analysis” using “basic COIN doctrine” to determine that 600,000 police and soldiers were necessary in a country of Afghanistan’s size. (That acknowledged a criticism made by counterinsurgency critics that the need is too great for the U.S. and Afghan capabilities.) “But the insurgency is not in the entire country,” he continued, so “we were able to reach what was necessary a better long term endstate.” That led to an estimate of “about 400,000,” — clarifying that that remains, sort of, his goal – with a breakdown of 240,000 soldiers and 160,000. That total, however, is “not a hard number at this point but a goal we work toward and adjust accordingly. As McChrystal said earlier, he will “re-look that every year” to determine what’s realistic.”
This is an interesting argument. So, the guideline comes from studies of past insurgencies. You figure out what the smallest ratio of forces was for success, and what the largest ratio was that still led to failure, and between those two is the minimum forces you need. It isn’t that easy, of course, because the demarcation line is not perfect (i.e. they are some cases of success with lower force ratios than exist in at least some failures). But at least the methodology is reasonably transparent.
Anyway, we’re nowhere near those numbers in Afghanistan. And General McChrystal is arguing that we don’t need to get there anyway because:
“But the insurgency is not in the entire country,” he continued, so “we were able to reach what was necessary a better long term endstate.”
But look, virtually no insurgency is in the entire country. In almost all insurgencies there are large segments of the country wholly untouched by war. And there are almost always large swaths that are securely under government control. The methodology used to derive the 3-24 recommendations of 20 per 1000 already ASSUMES implicitly that the insurgency is not in the entire country.
For McChrystal’s argument to make sense, he would need to argue that the insurgency in Afghanistan is geographically more limited than the average past insurgency used to baseline the force requirements. I see no reason to believe that is the case, and certainly I have not seen anyone systematically demonstrate that point.
The stronger argument that McChrystal sometimes makes is that they’ve actually done a micro-regional analysis and have in fact determined that they can allocate the necessary forces to every significant spot. And that may be true, though this analysis remains classified and hence is the opposite of transparent. But if that is the case, it is not an application of 3-24, but rather a clear deviation from the guidance there.
