Ben Wittes argues that the Faisal Shahzad case proves nothing either way:
To say, as many liberals and the administration are now saying, that Shahzad proves that the criminal justice system works in high-stakes counterterrorism crises is very silly. It proves at most that the criminal justice system sometimes works in such cases, a point that nobody serious doubts. The government has three big interests when it makes an arrest in a crisis setting like this one. It wants short-term intelligence. It wants to incapacitate an extremely dangerous person, and it wants to punish him. These interests can conflict at times, but they are not inherently in conflict. And when a suspect both talks freely and pleads guilty, the government’s interests align so neatly that the case simply doesn’t test the system. Any system is going to look good when faced with a suspect who wants to cooperate.
Wittes makes a number of other good points, but I think we have to be careful with where we draw the line of "seriousness" here. On the one hand, Wittes is right; no one worth taking seriously doubts that the criminal-justice system is at least sometimes capable of handling terrorism cases. On the other hand, the Republican Party, recently seriously declared its intention of trying to prevent the administration from ever using the criminal-justice system to try terrorism cases in a very unserious document called the Pledge To America, and they have a serious chance of acquiring a congressional majority in November.
I'd also grant that the Shahzad case, as an isolated incident, doesn't "prove" the criminal-justice system works in terrorism cases, although it certainly supports that proposition. I'd argue that the hundreds of terrorism convictions in civilian courts since 2001 proves the criminal-justice system works, particularly relative to the military-commissions system, which has a paltry four convictions to show for the last eight years, all of which may now be in jeopardy. As Wittes notes, military commissions are "off limits" for citizens like Shahzad anyway, and "enemy combatant detention has always been a last resort -- notwithstanding current Republican mythologies."
One last thing: There's a difference between arguing that the administration "got lucky" because Shahzad's bomb didn't work -- we all got lucky -- and arguing that the interrogators "got lucky" because he talked. The interrogators' job was to get Shahzad to talk. We should be used to it by now, but there's something outrageous about conservatives who want to see terrorists tried in the military-commissions system suggesting that the folks at the FBI just have no idea how to do their jobs.