Archive | Iran RSS feed for this section

Is American Statecraft More Effective Than We Realize?

6 Dec

I’ll second Minipundit’s praise for this Robert Kagan column that says, in the wake of the new NIE,  a military strike against Iran is effectively off the table and that we should open direct talks with Tehran.  This also got me thinking: Does this mean that American statecraft (from sanctions to international agreements and talks) is more successful at changing a state’s behavior than we’ve previously thought?  We went into Iraq (at least in part) because many experts and members of the Bush administration thought that the sanctions regime was broken (and it was, at least in part) and, therefore, it was likely that Saddam had reconstituted his WMD program.  Of course, after invading we determined that Iraq had no new WMD.  And if you had asked me (or many of the same experts) about Iran’s nuclear program, I would have said that they were working on a bomb, but that they were years away from having a working weapon.  But it now seems that Iran froze its nuclear program in 2003–problem solved deterred.

So why do we underestimate American influence? My shot-in-the-dark guesses:

  1. Everyone assumes sanctions regimes are ineffectual at best, especially when there’s a compelling realist case to be made (like national security) on the part of a non-ally.
  2. We’re bad at interpreting past failures (e.g. North Korea) and thus generalize about the efficacy of statecraft.
  3. The availability heuristic causes us to overemphasize threats.
  4. The political and human costs of failure (e.g. allowing an unfriendly state to develop a nuclear weapon) are so great that officials want incontrovertible evidence that the threat has been eliminated or deterred, and thus they rely on solutions that appear to give them this level of certainty.

I’d also argue that there’s a modeling problem for judging state (or state leader) behavior, but I’m woefully undereducated when it comes to international relations theory, so a generalized sense of “modeling bad!” doesn’t really say much.