Now that initial air strikes to disable Qaddafi's systemic air-defense capabilities have been concluded, the United States, the United Kingdom, and France should stay within the bounds of the two explicit missions of protecting civilians and enforcing the no-fly zone set forth in U.N. Security Council Resolution 1973. This means that military operations to "take out" Qaddafi or topple his government and to engage in close air or naval missile or gunfire support of rebel forces advancing against Qaddafi's forces in open terrain (as opposed to rebel forces defending civilian population centers) should be foreclosed. Rebel forces engaged in combat are not themselves civilians under the customary laws of war, and, accordingly, there is no mandate in the UN SC resolution to help them win a civil war. In other words, the United States and its allies should make a good-faith attempt to make sure that military intervention is limited to the explicit terms of the Security Council resolution, despite a lack of assurances that there will be regime change in Libya or that Libya will continue to exist as one country. Why?
Limited but continuing U.S. military operations in Libya are sure to be unpopular both with humanitarians who see the removal of a dictator who kills his own people as the real goal and with realists opposed to humanitarian interventions independent of concrete U.S. national security and political interests with U.S. military assets stretched thin in other conflicts. To use a metaphor, humanitarians can reasonably claim that it makes no sense to save the lamb from slaughter by taking away the butcher's knife, only to give it back to him and walk away. And realists can reasonably claim that it makes no sense to try to save somebody else's lamb from slaughter when we have our own flock to protect and limited resources to do so.
But I think both views, though reasonable, lose sight of why it was important for us to intervene militarily in Libya: the defining American interests in freedom and the rule of law. Even the most hardened realist would not deny that the United States, as a country dedicated to democracy and individual freedom, should act to defend people in other countries who are being killed by a dictator. The much harder question is to decide which cases merit U.S. military intervention. My own view, after much soul searching, is that it is nearly impossible to make such judgments as to whether Saddam Hussein is worse than Muammar Gaddafi, or whether one genocide is worth intervening in and another is not. The difficulty lies not only in the impossibility of deciding between such terrible cases, it also lies in the fact that even with the best of intentions, perceptions of national self-interest will inevitably skew which cases are chosen for military action. In light of both difficulties, the best option is to rely on the international legal process--the extreme difficulty of getting a UN Security Council resolution in light of the veto power of any single permanent member--as a procedural filter for determining when to intervene. In other words, the difficulty of getting a Security Council resolution authorizing military measures protects against self interests masquerading against humanitarianism while remaining true to the humanitarian principle in a small subset of cases (which admittedly may not be the worst cases in terms of human suffering).
A corollary of this procedural solution is that the United States and its allies should engage in a strict interpretation of "all necessary measures" authorized by the UN Security Council resolution. On this view, the opinions of the countries that abstained rather than vetoing the resolution, and the Arab nations who supported the resolution, should be given significant weight. I intend this approach as "legal" (akin to using original intent or meaning to interpret a domestic constitutional text) rather than politically pragmatic (to address accusations of anti-American and Western European meddling in domestic affairs), because the normative upshot is a commitment to the rule of law, not political expediency. And a commitment to the rule of law is an essential part of the American identity. Additionally, as a matter of U.S. constitutional law, U.S. military actions in Libya ordered by the President without specific congressional approval stand on a much firmer footing after the issuance of the Security Council Resolution (see, e.g., the Korean War) but it logically follows that the case for congressional authorization strengthens with the scale of military operations.
Consider two examples of how this understanding might be applied going forward. First, in my view, whether air and naval missile strikes on command-and-control networks and radars and surface-to-air missile sites distant from sectors where civilians were at risk fell within the Security Council's twin mandates resolution was a subject of reasonable disagreement at the time of the initial allied air attacks. But the reaction of Arab nations, China, and Russia to the initial air and naval attacks would seem to indicate that the original meaning of the Resolution did not go so far, and so the United States and its allies should exercise restraint in such future attacks, even if it means additional risks to pilots.
Second, the United Nations are not necessarily bound to the terms of any unilateral ceasefire by Qaddafi, but if the preponderance of the evidence indicates that a declaration of ceasefire has been communicated to Qaddafi's local commanders on the ground, that his warplanes have been grounded, that his artillery, tanks and troops are pulling back from sectors in which they might be engaged by rebel forces, and that, in general, he is seeking to comply in good faith with a decision to end civilian casualties (for example by treating civilian wounded and restoring water, power, and journalist and NGO access to rebel areas), then the United Nations should likewise refrain from military strikes and other exercises of armed force.