I’ve long been interested in the occasional anticooperative effect of law — the tendency of the threat of criminal punishment to sometimes discourage cooperation with the legal system (even though the deterrent effect usually tends to encourage following the law).
Max Boot points to the same effect of international criminal prosecution (a topic that international human rights law scholars have discussed in the past):
Hillary Clinton claims that Moammar Qaddafi may be exploring exit options. Count me as skeptical. The problem is that we don’t have a whole lot to offer a dictator in exile….
Qaddafi … has committed war crimes such as the bombing of Pan Am flight 103. He knows that if he leaves power he could wind up in the dock at the International Criminal Court.
The ability of the international coalition or the Libyan opposition to make a deal for his abdication has been complicated by the Charles Taylor precedent. Taylor was the former president of Liberia who left office in 2003 as part of an agreement that allowed him to escape into exile in Nigeria. But Interpol promptly issued an arrest warrant for him and in 2006 Nigeria handed him over to the UN’s Special Court for Sierre Leone. Eventually he wound up in the custody of the International Criminal Court in the Hague where his trial continues to drag on….
[I]n return for getting Taylor into court, we are making it more difficult to depose other dictators. Qaddafi has every incentive to fight to the death and take a lot of people down with him….
As Boot points out, the threat of such an anticooperative effect is by no means always a reason against criminal punishment. (We don’t decline to punish rapists or robbers, for instance, just because the risk of punishment may increase the incentive for the criminal to kill the victim and thus eliminate a witness.) But it is a reason to seek some solution to the problem, especially when one possible consequence of the anticooperative effect — here, of a refusal to cooperate with a possibly win-win deal offered by the government — might be the death of many thousands in a protracted war. Meting out justice to murderers is an important goal, but not the most important goal.
Boot, for instance, suggests a procedure for granting immunity, something akin to the American Presidential pardon. (In fact, one value of pardons and amnesties has historically been the possibility of ending a civil war by offering most rebels a reason not to fight to the death.) I’m not an expert on the subject, but I’m inclined to think that’s probably a good idea.