Intelligence Sharing: Squaring the Circle

The New York Times reports that the United States is depending on other countries to obtain counterterrorism intelligence:

"The United States is now relying heavily on foreign intelligence services to capture, interrogate and detain all but the highest-level terrorist suspects seized outside the battlefields of Iraq and Afghanistan, according to current and former American government officials. The change represents a significant loosening of the reins for the United States, which has worked closely with allies to combat violent extremism since the 9/11 attacks but is now pushing that cooperation to new limits."

This move has the advantages of reducing the need of the US to detain and interrogate people itself, and drawing on foreign intelligence services that have better language skills and contextual knowledge useful for obtaining such human intelligence.

At the same time, though, it exposes the US to these partners' actions. What if they abuse the rights of detainees? Do a poor job of interrogating them? Allow some to escape? Gather useful intelligence, but fail to share it with the US? Demand more guns or money from the US as compensation for sharing the intelligence they do collect? All of this has happened before, and the problem is likely to get worse as the US comes to depend more heavily on others.

What to do? One solution is for the US to exercise more direct control over its partners. It can use its leverage to try to insist they respect rights, learn effective interrogation techniques (although I'm not sure that the US intelligence community is the best teacher on either of these points), strengthen security at detention centers, etc. In the language of principal-agent theory, the US is increasing the principal that sets (intelligence collection) goals, and its partners are (potentially shirking) agents. Some direct control by the US one way it can monitor for and punish shirking. And there's new evidence that this is happening. In congressional testimony last week, Chair of the Joint Chiefs Admiral Michael Mullen noted that Pakistan--whose intelligence service is hugely valuable to the counterterror effort, and hugely successful at shirking--is changing personnel at a rapid clip. He also points out that he, meaning the US, is keeping a close eye on this process and seeking to ensure that the new faces at the ISI are willing and able to effectively cooperate with the US.

The establishment of such a hierarchical relationship might reduce shirking by intelligence partners. But it's not cost-free. The biggest cost is that the US will take on more responsibility for the mistakes and abuses that its partners make in the name of counterterror. A shiny new ISI reformed to look more like the CIA will still make mistakes (the CIA has made mistakes, right?). But now the US may bear some of the blame for these mistakes, to the extent that a reformed foreign intelligence service got reformed by its American counterpart.

This is so interesting that someone should write a book about it....oh, someone has.

Congress and Torture: He Said, She Said

The drama about what and when Nancy Pelosi know about the treatment of detainees raises a larger question: Why did Congress do such a poor job investigating the bad things done in the name of the War on Terror?

I cannot think of a major abuse that was exposed by Congressional investigation. Torture, Abu Graib photos, warrantless wiretaps, extraordinary rendition to countries practicing terrorism, abducting that Muslim cleric in Milan, waterboarding, that dead guy who they put on ice--all of these were revealed in the press, not by Congress. How come?

Pelosi implies that her intelligence briefers misled her. This may be the case, and it's certainly true that the current system--under which, for example, on an handful of top Congressional leaders learn the details of sensitive programs in oral briefings--makes oversight more difficult. But certainly not impossible. A motivated Congress could give itself the authority to undertake real investigations.

A better explanation is that Congress was not motivated to find out what was going on. Remember, the Republicans controlled both Houses until early 2007, and had little to gain from exposing abuses that took place under the authority of a Republican president. Republicans did, for example, limit Congressional efforts to fully investigate warrantless wiretapping, and delayed a Senate Intelligence Committee report on the use of intelligence prior to the invasion of Iraq.

Consider the counterfactual, though--if the Democrats had controlled Congress, would they have launched full-throated investigations? I doubt it. Sure, some members, including some pretty powerful members, would have pushed for such investigations. But I wonder if the leadership would have seen this as in the party's interest. Republicans would have no doubt countered that thorough oversight would weaken counterterrrorism efforts (although they would not phrase this so politely) at a time when Democrats did not want to appear "weak" on national security (remember how John Kerry "reported for duty" in 2004?). Better, from this perspective, to give the Republicans the rope to hang themselves. If these abuses seemed to stop terrorism, highlighting them would put the Democrats in the position of criticizing apparently successsful (if nasty, but mostly to foreigners) security measures. If they did not work, producing more terrorism overseas or at home, then the Republicans could be criticized for implementing policies that failed.

The Democrats clean sweep last November changes the dynamic, though. Now it is the (Democratic) White House that is telling some (Democratic) members of Congress to go slow on investigations. I doubt this will work. These members of Congress want to uncover Bush era abuses. This does not directly threaten Obama (although it makes his job a bit trickier), so the White House might not push back very strongly. And now that the Democrats do control the Hill, they can use the legislature's institutional powers to launch more intrusive investigations. So I suspect that many of the administration's attempts to get Congress to go slow will fail, and that Congress might now play its proper oversight role--better late than never.

More Deep Thoughts on Torture and Terror

In part in response to an earlier post discussing my research with Jim Piazza on the relationship between torture and terrorism, Joshua Tucker of The Monkey Cage ponders this question:

My original thought was that good social science research that shows that torture does not extract useful intelligence information would be the final nail in the coffin in any public argument in support of torture. But what happens if one of us gets access to the relevant data, does the empirical analysis, and then discovers the opposite: that torture does lead to useful intelligence information. What do you do then? Sit on the results? Would any political science journal publish such a paper? How would that look in a tenure review? (“Right, she’s the one who said torture was valuable…”).

Which leads to another question: should social scientists by engaging in research where we only want to share the results if they come out in one particular direction? I personally believe US national security is harmed by the use of torture in any form by our government, so I would welcome good empirical findings that provide added weight to arguments against the use of torture. But despite that goal, should I actually engage in research if I’m not willing to accept (or publish) findings to the contrary? Do we need some sort of social science code of ethics that sets certain research topics off limits? (e.g., something equivalent to doctors refusing to work on projects about devising more effective/painful instruments of torture.) Or is that an automatic affront to intellectual freedom?

For what it's worth, here's the short reply I posted yesterday:

This sort of question is of deep importance for social science. I would still urge publication of the results, though, for two reasons. First, the other arguments against torture you mention are pretty powerful. Finding that torture “works” in the sense that it provides valuable information would be only one victory for those favoring torture, and they lose all of the other battles. Second, I imagine that this problem comes up often, and that any important question might yield an answer with which one disagrees on ethical or moral grounds. What if an Americanist finds that negative advertisements “work”, or a scholar of international relations finds that preventive war “works” for the state that initiates it? These would be important if unpleasant realities. They might be valuable, though, for those opposed to torture, or negative ads, or preventive war, if they identified the conditions that facilitate each of these actions. Opponents could use this knowledge to advocate for more effective policies for ending torture, for example.

Country Reports on Terrorism

The US State Department today released its annual Country Reports on Terrorism 2008. The entire report is available here as a (quite long) .pdf file. You can also access individual parts of the report here.

Some initial reactions:

• Most of the report is devoted to detailing terrorist activity in most countries of the world, the evolution of specific terrorist groups, and details on state sponsors of terrorism. Pretty dry stuff unless you happen to be interested in a few groups or countries for whatever reason.
  
• This mass of detail is preceded by an introductory chapter that outlines trends in terrorism and the main lines of US counterterrorism policy. It's pretty short, though, which is either a good thing (if you want a quick overview) or a bad thing (if you want policy details).
• As further evidence that I am a boring person, I found the most interesting part of the report the "Annex of Statistical Information". It's got a nice discussion of how difficult it is to collect reliable information on terrorist attacks, perpetrators, and motives. It also suggests that summary measures, such as the number of terrorist attacks worldwide in 2008, are a pretty poor indicator of how terrorism is changing.
  
• Given the media's treatment of swine flu, though, I doubt many reporters will pay attention to this good advice, and expect to see TV news readers tonight starting with something like "According to the State Department, the number of terrorist attacks worldwide in 2008 declined by over 20 percent from the previous year".
• We can also expect fans of the Surge in Iraq to claim that this is further evidence that it worked. That is, if swine flu does not prevent them from getting any attention at all.
• The report includes a summary of a letter by Gary LaFree, the Director of the START Center at the University of Maryland and one of the world's best known researchers on terrorism. LaFree summarizes some of the recent developments in terrorism databases and makes some interesting suggestions about how these could be improved. A longer version of the letter is at the end of the National Counterterrorism Center 2008 report.
  
• My colleague Greg Weeks did a double-take on reading why Cuba is a classified as a state sponsor of terrorism; read his post here.