Here’s a good summary:
…The greatest growth [in the American presence in Mexico] involves intelligence gathering, with Homeland Security and the American military flying manned aircraft and drones along the United States’ southern border — and now over Mexican territory — that are capable of peering deep into Mexico and tracking criminals’ communications and movements, officials said.
In addition, the United States trains thousands of Mexican troops and police officers, collaborates with specially vetted Mexican security units, conducts eavesdropping in Mexico and upgrades Mexican security equipment and intelligence technology, according to American law enforcement and intelligence officials.
It’s counterinsurgency because there’s a number of armed groups working against the state’s legitimacy that have successfully undermined the state’s infrastructure and governing ability; dealing with that therefore means beating the armed groups by building up the state’s authority again.
As with most counterinsurgency efforts, quite a bit of skepticism applies: about the methods, sure, but the point of it all, too. Start with the methods: there’s the usual concerns about who we’re training. In Afghanistan the question is whether we’re just training and arming mercenaries; in Mexico the question is just the same—or whether these police and troops are actually in the drug cartels’ payrolls. That’s one of the ground-level practicality questions. Zooming out to the bigger picture: why is it we expect to this to work? The Times includes the de rigueur throwaway line that the U.S. lawmakers don’t trust their partners, and I’d be surprised if that feeling weren’t shared by some of the people on the ground. Because if you’re doing counterinsurgency, it means the state itself is compromised, and therefore by definition the partner is unreliable—if not corrupt, then incompetent or impotent. Why do we expect to be able to build up a state singlehandedly, particularly with such a limited investment? Available experience indicates it takes a very long time with a lot of resources, something we rarely have and don’t have now. There are cultural reasons to expect such an intervention to be ineffective: Mexico’s constitution specifically prohibits armed foreign agents in its country, and I’d imagine that keeping armed Americans out is at least part of the point. Intervening might well undermine the constitution and thereby undermine confidence in the government, thereby moving people away from the government and perhaps to the drug cartels.
The biggest picture question is why we’re doing it this way at all? The drug cartels have power in the first place because of money, money that they mostly get from U.S. drug sales. Since it’s hard to restore the state’s authority in the absence of brutal methods, it will inevitably be a slog and perhaps the correct response is to attack the cartels at the root. Unfortunately digging up the roots is a considerably more problematic task than most might assume: yeah, the cartels make a lot of money from marijuana, so legalizing or decriminalizing it would drain their revenues. But most drug cartel money is made from non-marijuana drugs, which are drugs that many people are less comfortable with the legalizing/decriminalizing route, understandably so. It’s a tough problem; counterinsurgency, which seems to offer an advanced way of dealing with advanced bandits, seems like an appealing solution. Ultimately, though, counterinsurgency seems to promise goals that we may not want using methods that may not work.