Hostages and Prisoners

I’ve been impressed by the number of questions I’ve fielded in the past few weeks from students, colleagues and media alike about whether the United States can and/or should pay ransoms or exchange prisoners for Americans held by various groups overseas. (I discuss the issue in short clips here and here.) Why did we exchange prisoners to rescue Bowe Bergdahl, but refused to pay ransom for James Foley? Is it illegal to pay ransom to these groups, or just a bad idea? Is it really a bad idea?

In the interest of consolidating some answers on a topic that raises a complex cluster of issues, I thought it worth summarizing some of them here – first on the topic of ransom for hostages taken by terrorist groups, then on the topic of prisoner exchanges more broadly. The upshot: It may well be the right policy decision in an individual case for a government not to pay ransom to a terrorist group, but the broader, categorical statement that “we don’t negotiate with terrorists” is neither historically accurate nor strategically wise.

Paying Ransom to Terrorists

Let’s start by distinguishing whether we’re talking about ransom demands that terrorists make of companies and private individuals, and ransom demands made of governments. There is and long has been, alas, a sizeable industry of terrorist and other organizations taking Americans or others hostage and then seeking large sums of money from their families or employers in exchange for their freedom. Indeed, companies with substantial overseas presences often take out insurance for such contingencies, and there is a developed private sector market in firms that help negotiate the resolution of such crises. Could the United States ban private Americans and companies from giving in to terrorists’ ransom demands? Certainly. Indeed, there are a host of domestic and international laws on the books – U.S. material support laws among them – that squarely prohibit paying money to terrorist organizations. And families and companies that pay ransoms notwithstanding these laws are in principle at least at risk of prosecution. On the other hand, to my knowledge the United States has never used these or other laws to criminally prosecute private actors who pay ransom to terrorist groups in such circumstances. Why not? Perhaps it seems like a classic circumstance of duress. Perhaps it is because the U.S. government has no interest in turning families like the Foleys (or large corporate oil interests) into criminals for doing what seems completely understandable and rational in extraordinary individual circumstances. Whatever the reason, the existence of this unfortunately active international market has implications for arguments about whether and what effect it has for governments to refuse to pay ransoms or, even more broadly refuse to “negotiate with terrorists.”

Then there are ransom demands made by terrorist groups of governments – in part what seemed to be at stake in the recent horrifying beheading of American James Foley by ISIL. The U.S. government has taken the policy (as opposed to legal) position that it does not make “concessions” to terrorists, and this Administration (like many U.S. political leaders) has argued against the payment of ransom to terrorist groups. The policy argument here is in the first instance one of deterrence: paying money to these groups will only encourage the practice of hostage-taking. More, ransom payments can be a key source of income for many of these groups, funding future operations. Yet not every government has adopted the current U.S. policy. On the contrary, European governments and others have paid ransoms under similar circumstances through direct and indirect channels. (It should also be noted of course that there are some high profile examples of the U.S. government taking a rather different approach, most famously in President Reagan’s administration with the sale of arms through intermediaries to a terrorist group holding several Americans hostage.) And it is easy to imagine the contrary position; while it is one thing to reject all ransom demands in the abstract, you try telling the parents of an about-to-be-beheaded child that it’s all just a matter of policy. So who’s right?

In my view, the answer can depend enormously on the particular circumstances of cases that can be dramatically different. Take the deterrence argument, which seems logical on its face, especially when it comes to groups (like the Somali pirates who attacked the Maersk Alabama) whose prime motivation is monetary. Still, the statistical data one would expect to find if terrorists were in fact deterred from seizing some countries’ nationals by those countries’ non-payment of ransom (and encouraged to seize more hostages by the ransom-paying countries) is far from dispositive. A recent, widely cited New York Times article deeply critical of European governments who have paid ransoms to Al Qaeda to recover their citizens suggests that the statistical correlation is there, but (a) the correlation is not apparent in their own statistics (Germany, which has paid ransoms, has reported fewer of its citizens taken hostage by Al Qaeda than has the United States in the past 5 years); and (b) correlation is not causation, and the article provides no evidence for the latter. Why might one not find either? One can imagine a host of explanations (terrorists are more opportunistic than strategic in hostage-taking choices, country reporting may be incomplete, etc.), but for now let’s take one: monetary gain may not be the but-for motivation of all hostage takers. While Al Qaeda, for instance, has made hostage taking a primary source of income, ISIL’s recent gains have left it extraordinarily flush with cash. More to the point, ISIL has political motives for kidnapping Americans that are unlikely to be affected by the payment of money in any particular case. In the end, ISIL described Foley’s execution as in retaliation for recent U.S. airstrikes in Iraq. ISIL’s predecessor terrorist organization in Iraq, led by Abu Musab al-Zarqawi, was also fond of beheading prisoners, including American Nick Berg, beheaded on video in 2004. His execution, according to the group, was in retaliation for the U.S. torture and abuse of prisoners at Abu Ghraib. Are such groups deterred by the payment or non-payment of money? Not at all clear.

Whatever the decision about ransom per se – and if the deterrence argument is weak, the argument against directly funding terrorist groups remains compelling – the point here is that the outcome in a given case will and should depend a lot on whether the demands are made of the government or of private actors, on who is taking the hostage, and on what they really want. That means that ‘negotiating’ with such groups through intermediaries or otherwise – in the sense that discussions may include finding out as much as possible who they are and what they want – can be valuable to us under any circumstances. It can be an important source of intelligence and insight into the thinking of our adversaries. Even more instrumentally, it can buy time while a range of options are considered. In other words, “negotiation” may sometimes be an important, self-serving tool in our tool box. It makes no sense to take it categorically off the table in all circumstances in advance.

Exchanging Prisoners

This brings us to the rather different circumstance posed by the Bowe Bergdahl case (in which a member of the U.S. military held prisoner by the Taliban was traded for several Taliban prisoners held by the United States at Guantanamo Bay) earlier this year. What to do about prisoners taken by an adversary with whom we say we are at war? Here, the history is clear.

In all of the major wars of the 20th and 21st centuries in which U.S. detention operations are now concluded – World Wars I and II, Korea and Vietnam, the 1991 and 2003 Iraq Wars – conflicts during which the United States held hundreds of thousands of prisoners in total, the imprisonment of enemies held pursuant to wartime authorities has always come to an end, and the resolution of these detentions has always been handled by the executive branch. Among the most common mechanisms for the resolution of detention: executive agreements providing for prisoner exchange, transfer or release, negotiated with a wartime enemy, often through a third party intermediary. For all the recent controversy surrounding Bowe Bergdahl, such arrangements are, from a historical perspective, prevailing U.S. custom.

To be clear, this is not meant to excuse the Obama Administration’s failure in the Bergdahl case to comply with Congress’ requirement that it be given 30 days notice before the transfer of any Guantanamo prisoner. As the GAO recently concluded, setting aside any arguments the executive might have that Congress’ restriction was unconstitutional, it seems apparent the Administration violated that law. It is, however, noteworthy that Congress has not imposed any such restriction on the exchange, transfer or release of prisoners, during or after the period of armed conflict in any of the previous conflicts over the course of the past century.

Should these historical examples matter? Isn’t the U.S. conflict with Al Qaeda and the Taliban different from all these others? In some respects yes. In key respects no. The notion of returning prisoners to a homeland of violent political instability, for example, is not new. We returned prisoners twice to post-war European nations whose economic, political, and state security systems had been decimated by what were then the most destructive wars history had ever known. Neither is it the case that we would never release prisoners who still harbor violent intentions toward the United States. In World War II, among the first prisoners released were those Nazis whose enmity was “most hardened” against us (principally because they were not good sources of prisoner labor). Nor can it be contended that we would never release prisoners as long as they have ideological bretheren with whom they might again affiliate in re-engaging the fight. We returned thousands of communist prisoners to communist nations – for a half-century our most feared, most hated ideological opponents – at the height of a half-century long war that was “hot” (in Korea and Vietnam) almost as often as it was cold. And to be clear, we returned prisoners not only to state enemies, but to non-state enemies as well. In Vietnam, we at times unilaterally released prisoners held by the Vietcong – the communist insurgent group in the south that would under today’s usage almost certainly be described as terrorists. We did this by taking a calculated risk that any short term tactical burden we might bear by some individuals’ release was outweighed by the long term strategic benefit to the United States of acting, and being seen to act, in a manner consistent with the prevailing law.

When and how such exchanges may be carried out in the existing conflict with Al Qaeda and the Taliban depends on a host of factors – legal and political – that I address at length here. For present purposes, perhaps it is enough to note that negotiation with our most abhorrent adversaries has long been a central tool of U.S. policy. It is contrary to more than a century of American practice to suggest otherwise.