Under Fire.

I wanted to take a moment to discuss the drama surrounding the way that targets have been acquired and neutralized overseas in theatres such as the Middle East and Africa. I've talked about one method used to select targets based upon their behavior patterns, or "signature strikes." The accuracy and efficacy of these strikes remains to be seen, but that's not what I'm looking to discuss right now. We have information on another type of assassination program.

This program uses what is called "signals intelligence." Targets are acquired due to digital footprints picked up off of electronic devices. The example I would like to use today is Operation Haymaker, launched in Afghanistan, using a support and information network stitched together from around the globe. The numbers in Operation Haymaker are eye-opening. According to the Pentagon, 39 targets were neutralized. Roughly 219 innocent persons on the ground were taken out. That's a ratio of 1 to 5 or 6.

During one 5 month period of this undertaking, the accuracy of the strikes was estimated to only be 10%. That's absolutely horrible.

I am looking for something in particular when it comes to drone strikes: raising the threshold for strike clearance by increasing the level of corroboration amongst different types of intelligence that we gather on a daily basis. This should happen before any strikes are green-lit to reduce the possibility (probability) of collateral damage. Human intelligence should not be discarded here as well. In short, the current benchmarks are not acceptable.

We simply have to do better. While we don't have access to all of the information that we would like, I believe that we can be effective in pursuit of our mission while at the same time minimizing the problems that we might create on the ground in the process.