One of the perennial problems of international criminal law is defining terrorism. The entire discussion was stalled for decades by the slogan, "one man's terrorist is another man's freedom fighter," and hence stalled by the wars and ideology of decolonialism and the wars of national liberation. Even though that era is largely over - it survives, really, only in Palestinian liberation ideology - the problem of definition remains intractable at both the abstract level and the practical level.
In practice, the question of the definition of terrorism arises each time, for example, a media outlet - newspaper, the AP, CNN, NPR, the BBC, what have you - refers to an attack, for example, on Iraqi civilian election workers as an attack by "insurgents." Terrorists, apparently, is too loaded a word for the main stream Western media, seeking polite neutrality. In the case of some especially politically freighted media, such as Reuters, the locutions seem aimed at making a definite political statement that the "insurgents" are not "terrorists," rather than some (morally indefensible) attempt to maintain a supposed "neutrality."
That's at the practical level. The theoretical problem can be stated this way. Leaving the media aside, there is general agreement among Western elites - including the functionaries of the UN and the elite international bureaucracies - that the core definition of terrorism must encompass attacks which target civilians directly. That, presumably, is the easy case, because it is always indefensible. It is part, for example, of the suggested definition of terrorism in the new report to the Secretary General by the High Level Panel, at paragraph 164. A number of treaties, resolutions, declarations, and so on have been formulated which adopt as a definition, in one fashion or another, the prohibition on direct attack against civilians from such sources as 1977 Additional Protocol I or other law of war sources.
Well, "always indefensible," anyway, except if you are among the numerous Islamic clerics (too many unfortunately in what passes for the "mainstream" of the Muslim world) for whom direct attacks on civilians, such as the terror attacks on Israelis, are acceptable because, being the weaker party, you supposedly have no other means of striking back against the oppressor. They treat such terrorism as a kind of extreme case of military necessity (and perhaps even a version of Walzer's "supreme emergency" thesis) - supposedly giving you the "necessity of nature," as Hobbes or Thucydides might put it, to attack civilians. That the prohibition in the law of armed conflict of direct attack against noncombatants is not accepted even in principle by a significant portion of influential opinion in the Muslim world must give great pause.
Still, one might wonder why these Islamic clerics should bother thinking any further about the moral implications of attacking civilians, however, given that not even the leading Western newspapers and media, by applying terms such as "insurgents" rather than "terrorists" (even to attackers such as those who over the weekend assassinated civilian Iraqi election workers), trouble themselves to apply the term to acts that even Kofi Annan and the High Level Panel would call terrorism. Those who attacked and summarily murdered the civilian election workers are "terrorists" even on the purely neutral criterion that you are a terrorist if, among other things, you use means and methods which directly target civilians. This is not the rocket science of ethics. The inability of writers and editors in the media to recognize even that much as plain terrorism and to call it by its name speaks to a particularly intractable and vicious brand of multiculturalism, politically correct sensitivity gone morally amok. They are intellectuals, Camus might have said, who deploy the weapons of their language in the justification of murder and then call it neutrality. (You can read my view of the moral poverty of discourse of neutrality here, in the last sections of the article on humanitarian neutrality.) But, of course, flogging the NYT, the Washington Post, CNN, NPR, or the BBC is flogging a dead horse - recall, after all, the BBC reporter, a few weeks back, describing the tears welling up in her eyes (and not tears of joy) at the death of Arafat.
The theoretical problem even beyond the question of civilians is that this definition of terrorism is too narrow. It does not include, for example, IRA attacks on British soldiers or policemen, whether in northern Ireland or elsewhere. In effect, what the IRA said was that it was in a war, and it would abide the general rules of combat - i.e., aiming at what it said were "combatants" in that war, and not aiming at what it said were "noncombatants." This meant that, according to its own criteria, its fighters were "combatants" fighting other "combatants" and - a crucial corollary - collateral damage, including death and injury of bystanders, would be measured according to the traditional law of war standard of proportionality between military necessity and harm caused. And of course, if they were in a war, and they otherwise met the requirements to be combatants in, say, article 4 of the Third Geneva Convention, then all that would be quite true.
But according to the British government and everyone else - leaving aside the looney left -the IRA was not a legitimate party to a legitimate conflict. The IRA had no right to be fighting at all, and it did not matter if they conducted themselves according to the laws and customs of war - any death, injury and destruction they caused was simply criminal and terrorism, because they were not entitled to fight. They did not benefit from the so-called "combatant's privilege" - the legal privilege of aiming at other combatants without the results of the fighting, or proportionate collateral damage resulting from the fighting, being accounted to them as criminal; a legal privilege matched, however, by the disability of being a target of someone else on the other side. The reason they did not benefit was simple: they did not fight in a legal and legally recognized war. Hence the people they attacked, including the soldiers and policemen, had no reason to think they were liable to attack and (if done for political purposes with the intent of influencing political policy) those deaths and injuries are no different for purposes of describing the attack as "terrorism" than if they had attacked civilians.
(The High Level Panel report carefully elides this question - referring specifically in par. 164(d) to definitions of terrorism that cover only "civilians and noncombatants" while also referring in par. 164(b) to earlier anti-terrorism treaties that arguably have broader scope. It is yet another reason why the report is and should remain a nonstarter.)
As soon as the definition of terrorism moves - as it must - beyond the question of civilians and noncombatants, however, it also moves beyond the jus in bello question of simply means and methods of warfare - what you aim at - to the jus ad bellum question of whether you are entitled to be fighting at all, i.e., aiming at anything or anyone at all. After all, if the mafia in the United States simply decided to announce that it was fighting a war and entitled to international law protections, no one would pay its claims any attention. The next step would be for it to announce that it was fighting for political goals, but still no one would pay any attention - it would not be regarded has having the political legitimacy to fight at all.
The Geneva Conventions seek to address this problem by saying, in effect, that if a conflict walks like a duck, and talks like a duck, then it must be a duck. If the objective level of conflict and fighting becomes widespread enough, rises high enough, to start looking like a war, then the rules of war apply. This has the virtue of seeking to maintain the strict separation of jus in bello from jus ad bellum - it has the disadvantage, however, that no clear practical line separates one level of fighting or attacks from another.
In any case, even if the fighting rises to a level high enough to trigger the rules of war, but the conflict is a purely internal one, it is covered only by the minimal protections of Common Article 3 of the 1949 Geneva Conventions, and not the full body of treaties. Common Article 3, note, does not prevent the legitimate government of a state from trying its insurgent for treason and insurrection and murder and anything else that might be a crime arising from armed conflict - it confers no prisoner of war protection or combatant's privilege on rebellion, only minimal due process protections and protection against summary execution. And there must be a certain level of conflict before even Common Article 3 applies; the point about terrorism is that even if the acts are carefully limited to what, in a legitimate war, would be legitimate military targets, they rarely rise to the level of conflict necessary even to invoke Common Article 3 (although the ICRC has had a long and dubious pattern of lowering the bar in its pronouncements on the status of conflicts).
When, therefore, the USS Cole was attacked, that attack could be viewed as either terrorism or a surprise attack by a new enemy engaging in war. The view of the US and nearly all sovereign states (unsurprisingly) was that the attackers had no right to be engaged in armed conflict at all, and thus although, if it were a war, it would have been exemplary in that it involved no civilians or collateral damage to civilians, that did not matter. Surprise attack by a legitimate adversary is no violation of international law, but since there was no legal right to attack these people at all, any attack on them was the same as though civilians had been attacked. This is how one treats it as terrorism. However, reaching that conclusion requires a conclusion about a core issue of jus ad bellum - viz., who has, in traditional just war terminology, the "right authority" to engage in war? This is not a question the ICRC or other guardians of jus in bello are qualified to answer, because it requires a judgment about jus ad bellum that these institutions are supposed deliberately to eschew.
I accept the position that past a certain level of fighting, seen on an objective, purely behavioral basis - sustained control of territory, for example, large numbers of fighters, and other such indicators - the rules of war apply, if only the minimal provisions of Common Article 3. Short of that, however, or if, in the case of transnational Islamist actors who attacked the USS Cole, one catches someone and has to decide what to do with them, then the question of who may legitimately make war and who is not will not go away in the question of defining terrorism. Jus in bello will not answer it. It is a question of right authority. Sovereign governments, pretty obviously, would prefer to limit it largely to, well, sovereign governments. The difficulty, going back to Locke at least, is that the leading democratic states - the US, France, even Britain - arise out of civil war and acts of rebellion and secession. It is what, after all, the Declaration of Independence is all about - justifying the use of force when you are not actually a sovereign. Drawing that line in today's world is no easier.
Yes, ruling out attacks on civilians and noncombatants is easy. Ruling out attacks on the armed forces of a state, however, against something that, in a state of legitimate war, would plainly be a legitimate target depends fundamentally on saying what parties are legally entitled to fight, and that is far harder. But unless one is willing to do so, it is not possible to capture what we mean by terrorism, because nearly all of us intend the definition of terrorism to include far more than just attacks on civilians.
This, the United States should take careful note, is not something on which the political actors in the world - sovereign states, the UN, NGOs, all the rest - will be able to agree. It is possible to imagine near universal agreement in ruling out attacks on civilians and noncombatants. It is not possible to imagine agreement, even agreement between the US and its European allies, on what parties are entitled to fight. And because there will be no agreement in moral and political principle on such an issue, it is a foolish and dangerous thing for the United States to pursue treaty discussions that would limit its scope of action with respect to what it would see itself morally and politically bound to regard as terrorism - including the attack on the USS Cole and the treatment of its perpetrators as terrorists. Lacking a principled ground of agreement on the outer boundaries of the definition of terrorism - even if there is agreement on the inner boundaries of no targeting civilians - the United States would be well advised to stay out of treaty negotiations over its full definition that will inevitably be weaker than what the US must regard as minimally required.