Ethics and Military Force: The Jus in Bello

Edited transcript of remarks, January 2002 Carnegie Council workshop on European and North American perspectives on ethics and the use of force.

In this short presentation, I would like to describe the international humanitarian law (also known as the law of war or armed conflict) principle of distinction, paying particular attention to those aspects of "future war" likely to challenge humanitarian lawyers and war fighters charged with its application in combat. I label these features of armed conflict "stressors"; they "stress" the principle by making it difficult to apply in practice.

The law on distinction is set forth in the 1977 Protocol Additional I to the Geneva Conventions of 1949. Most participants at this seminar come from countries that are Party to the convention. Indeed, all NATO States except Turkey and the United States enjoy Party status. Although the United States has signed Protocol I, ratification is unlikely under the Bush administration. However, US opposition derives primarily from provisions of the agreement other than those relating to distinction. In fact, the US views the bulk of the text on distinction as a restatement of customary international law; it therefore sees itself as bound by the principle pursuant to that separate body of law. Given the breadth of the Protocol I regime, I will refer to its provisions when discussing distinction.

The core principle of distinction is found in Article 48 of Protocol. It provides that "Parties shall at all times distinguish between the civilian population and combatants and between civilian objects and military objectives and accordingly shall direct their operations only against military objectives." In armed conflict, distinction plays out in two separate ways. The first is the prohibition on the use of indiscriminate weapons, whereas the second involves the indiscriminate use of weapons that are capable of distinction.

Indiscriminate Weapons

The proscription of indiscriminate weapons is articulated in Article 51 of Protocol I. An indiscriminate weapon is a weapon that is, in lay terms, tough to aim. More precisely, it is one that has difficult-to-control effects. Examples include biological weapons in which the spread of toxins cannot be controlled and, perhaps, persistent airborne chemical weapons. To illustrate the nature of indicriminancy in the context of weapons, consider cluster bomb units (CBU), i.e., bombs that dispense multiple individual bomblets. Although characterized as indiscriminate by many, the US position, one correct as a matter of law, is that because CBUs can, in many circumstances, be directed with great effect against specific targets, they are not indiscriminate. For instance, CBUs could be used discriminately against fielded forces located far from the civilian population, as in those deployed in a desert or mountainous region. A second, very politically charged example is anti-personal mines. The United States, again correctly, urges that a mine is not an indiscriminate weapon, for it could be used, as an example, in a marked area or designed so that it self-neutralizes after a set period. As with the CBUs, the issue is whether it is being used indiscriminately; allegations that it is indiscriminate ab initio are quite simply red herrings.

Because very few weapons are inherently indiscriminate, the topic is not a major focus area of international humanitarian law. I expect this to change as new war-fighting technologies are developed and fielded.

One likely stressor is the development of computer technology capable of attacking computer networks. These computer network attacks (CNA) are able to cause enormous human suffering or physical damage. Examples range from shutting down air traffic control systems or disrupting financial networks to altering data in computerized medical records. In that attacks are typically targeted against a particular database or network, most CNA is extremely precise. However, it can also be highly indiscriminate. Perhaps the best example is launching a computer virus designed to spread randomly through an enemy's network. The fact that it can spread so broadly is its attraction, particularly for those facing a computer dependent foe. This attraction is heightened by the fact that the technology to mount CNA attacks is relatively cheap and the know-how to conduct them is widespread; as "war on the cheap," CNA invites use. The problem is that the interrelationship of 21st century military and civilian networks means that such viruses (and other forms of attack) will inevitably reach the latter systems in ways uncontrollable by those who mount the attacks.

A second stressor is terrorism, for in many cases the very raison d'etre of a terrorist act is to injure or kill civilians. After all, for terrorists, attacking the population is a form of force-multiplier because doing so draws enormous attention to their "cause." The tragic events of September 11th in New York, Washington, and Pennsylvania are perfect, albeit criminal, examples. Use of an indiscriminate weapon, such as a biological, chemical or cyber weapon, would only enhance the intended effect of a terrorist attack; indiscriminancy heightens the population's sense of helplessness, thereby increasing the terrorists' incentive to employ them.

Indiscriminate Use

The second aspect of distinction is indiscriminate use, which is using a legal weapon in an indiscriminate way. Indiscriminate use has three separate components.

1) Discrimination: "What is a legitimate target?"

2) Proportionality: "Is it legitimate to attack this legitimate target given the unintended but foreseeable harm likely to civilians or civilian objects?"

3) Duty of care. "Even if a target is legitimate, and even if it can be struck proportionally, what are the requirements to exercise care in attacking the target?"

Let us examine each of these in turn.

a) Discrimination

Article 51 of Protocol I provides that "the civilian population as such, as well as individual civilians, shall not be the object of attack;" moreover, campaigns that are designed to terrorize the civilian population are prohibited. To determine what a civilian is, it is necessary to refer first to the third Geneva Convention of 1949, which sets forth five categories of legitimate combatants:

• Armed forces;
• Militias that are part of the armed forces;
• Regular armed forces of a party to the conflict that is not recognized by the other side. An excellent current example is the Taliban military forces, given the lack of recognition of the former Taliban government by, inter alia, the United States;
• Volunteer corps or groups or resistance movements that are not formally part of the armed forces, but are involved in the conflict. To enjoy combatant status, they must meet certain criteria: be under the command of a person who is responsible for them, be distinguishable from the civilian population, carry their arms openly, and conduct their operations in accordance with the law of armed conflict. This last requirement is somewhat flexible, for elements of every military force violate the law of war at some time during a conflict; every member of a force cannot possibly be controlled absolutely; and
• Levée en masse, i.e., members of a civilian population that rise up spontaneously to fight an invader.

In 1949 this was a relatively innovative definition of combatant by virtue of its inclusion of resistance and volunteer forces. Previously, they were not considered combatants. However, such groups were quite active during World War II. Tito's Partisans are perhaps the best example. The French resistance, on the other hand, generally would not have complied because they did not wear uniforms or distinctive emblems.

Following such conflicts as Algeria and Vietnam, in 1977 the international community recognized the fact that many involved in the post-WWII conflicts failed to meet the 1949 criteria. For instance, the Viet Cong, albeit directly involved in the conflict in a major way, were not encompassed in the scope of the Geneva Conventions. Therefore, to reflect the new reality of armed conflict, in Protocol I the standard for combatant status was lowered. Henceforth, such groups were to distinguish themselves in an attack unless impractical, in which case they were only required to carry their arms openly during each engagement. The United States strongly opposed this relaxation of the standard for two reasons, both related to the principle of distinction. First, failure of such groups to wear uniforms or otherwise distinguish themselves as participants in a conflict presents a force protection dilemma. Simply put, soldiers will not readily know against whom to protect themselves. Second, the relaxed standard endangers the civilian population because soldiers are more likely to fire on civilians whom they cannot easily distinguish from those involved in the military action.

Combatant status is important for three reasons. Most notably, a person's status determines whether he or she is a legitimate target -- both combatants and "illegal combatants" (those engaged in the fight but not complying with the combatant requirements) may be directly attacked. Others may not. On the other side of the coin, an individual's status determines the activities that he or she can engage in during an armed conflict. Combatants participating in an international (as opposed to internal) armed conflict cannot be punished for engaging in hostilities against members of the enemy's armed forces. Finally, combatants enjoy POW status and treatment if captured during an armed conflict. Others do not, at least once a tribunal (pursuant to Article 5 of Geneva Convention III) determines that an individual does not meet the requirements of combatant status.

Given current events, a very topical issue is the status of Al-Qa'ida and Taleban forces. First, what is the status of terrorists unconnected with the activities of any State? As a general rule, they are not combatants because they are not members of an armed force and do not otherwise meet the legal requirements of combatant status. They are also not illegal combatants because the issue of combatant status only arises in the context of international armed conflict, of which terrorism standing alone is not a part. So, when terrorists operate as "non-State actors" they are simple criminals. Absent a connection to any government, therefore, those responsible for the attacks of September 11th committed murder in the states of New York and Pennsylvania and in the District of Columbia, participated in activities prohibited by treaty- such as hijacking - and conducted crimes against humanity, i.e., systematic attacks on the civilian population.

Assuming, solely for the sake of analysis, that the Al-Qa'ida terrorists were not directed by any State, could the US and other countries respond by attacking them? The answer is yes. To begin with, an argument can be made that pursuant to customary international law (and common law) principles, self-defense can be conducted against anyone. The attacker need not be a State. It would, of course, be cleaner as a legal matter if the self-defense were permitted by Article 51 of the UN Charter, which allows for individual or collective self-defense in the face of an armed attack. One quandary is that the Charter governs relations between member States. Though there is a growing tendency to view the Charter as reaching relations with individuals (as in the case of human rights), some would assert that Article 51 is limited to defense against States. However, the prevailing view would now appear to be that Article 51 can include terrorist attacks, a view supported by post 9/11 Security Council resolutions citing the right to self-defense and NATO's invocation of Article V of the North Atlantic Treaty, which provides for collective defense in accordance with Article 51. By this interpretation, the US response would be judged against the three requirements for defensive action: 1) necessity, i.e., using armed force in self-defense only as a last resort; 2) proportionality, i.e., using only that amount of force necessary to get the attacker to desist; and 3) imminency, i.e., that the attack be on-going or imminent. With regard to the final criterion, 9/11 is only the most recent of a series on ongoing attacks - the bombing of the USS Cole, the bombings of the US embassies in Africa, and so on. Thus, that criterion is met. If one of the three requirements is not present (or is no longer present), the matter becomes one of international law enforcement.

And what of the status of the Taliban? The Taliban was a de facto government, which is not recognized by the US. Indeed, it was recognized only by a handful of States. However, as a matter of law that the actors have not been recognized is irrelevant; the issue is whether the group is de facto "running the country" (or most of it). Therefore, for our purposes, the sole issue is whether the Taleban engaged in an armed attack against the United States, thereby allowing the US to respond in self-defense.

Although the Taliban did not actually carry out the attacks themselves, did they constructively do so through the instrument of Al-Qa'ida? This depends on the nature and extent of the relationship between the two groups. The International Court of Justice addressed a similar issue in the 1986 Nicaragua case. In its judgment, the Court held that an armed attack must be understood as including the dispatch of armed bands or irregulars into the territory of another state, even if they are not your own armed forces. On the other hand, the ICJ also held that simply providing weapons, logistics, or other support to the insurgents, while wrongful, did not amount to an armed attack under Article 51. Along the same lines, the International Criminal Tribunal for the former Yugoslavia (ICTY) held in the Tadic case that the critical issue in determining whether acts can be attributed to a State is the extent of subordination to the State of the group committing the act. Even more recently, the International Law Commission, in its Draft Articles on State Responsibility focused on instruction or directing or controlling persons or of groups of persons as the threshold for State responsibility.

As can be seen, a relatively high legal standard has been set for determining when the activities of a non-State actor are appropriately attributable to a State. Mere acquiescence to the presence of terrorist on one's territory, or even their use of it as a base of operations, would not appear to be sufficient. Whether or not the Taliban ties to Al-Qa'ida are robust enough is a question of fact. Absent public release of all relevant material, it is difficult to offer any definite conclusions on the issue. However, if the facts did reveal sufficient ties, it would be entirely appropriate to attack the Taliban in self-defense pursuant to Article 51, and the Taliban military forces would be combatants in an international armed conflict, albeit ones that may have committed war crimes.

At this point, the analysis on Al-Qa'ida forces would shift. If the Taliban are so tied to Al-Qa'ida that they constructively committed the armed attacks, then the relationship would be reciprocal. As an instrument of the Taliban, the Al-Qa'ida members who carried out the attacks would be illegal combatants. Not being members of the "armed forces of Afghanistan," and not otherwise complying with the requirements of combatant status but nevertheless participating in an international armed conflict, would merit this status. The major difference would be that they would be entitled to POW treatment (but not status) pending a formal determination of their status. Moreover, as illegal combatants, any attacks conducted against military forces would be wrongful ab initio.

Twisting the scenario one last time, if the Taliban are not sufficiently directive of Al-Qa'ida, but are merely acquiescing in their presence (for whatever reason), what is the appropriate course of action for victim States? The law on this issue reaches most noticeably back to the Corfu Channel case, heard before the International Court of Justice in the late 1940s. Basically, the standard that has emerged from that case and similar situations is that if non-State actors are using a territory as a base from which to launch attacks, the victim State must ask the State from whose territory the attacks are occurring to take whatever action is necessary to halt them. If it does not, the victim State may cross the border in an act of self-help to eradicate the threat; the State on which this operation is conducted may not interfere so long as the self-help is necessary and proportional. So if the relationship between the Taliban and Al-Qa'ida were too tenuous to merit an attack in self-defense on both groups, the US would have been perfectly within its rights to cross into Afghanistan to defend itself against Al-Qa'ida's ongoing attacks. Armed Taliban resistance to the self-help would have been wrongful and amounted to an international armed attack against US forces. In that an international armed conflict would then be taking place, the status of the Taliban and Al-Qa'ida would shift (generally) to combatants and illegal combatants respectively.

In addition to the requirement to distinguish civilians and combatants, there is also the obligation to distinguish between civilian objects and military objects. Specifically, Article 52 of Protocol I provides that "civilian objects shall not be the object of attack" and that "attacks…[must] be limited to strictly military objects." A military objective is defined in the article as "those objects which by their nature, location, purpose or use make an effective contribution to military action and whose total or partial destruction, capture or neutralization, in the circumstances ruling at the time, offers a definite military advantage."

This is one juncture at which the Americans and their European friends disagree. The lead humanitarian law manual in the United States is the US Navy's Commanders Handbook on the Law of Naval Warfare. It includes as a military objective those objects that are war sustaining. By this definition, industrial and other economic entities are legitimate targets because they sustain the war effort. Many other States would take a narrower view, urging that the law is unambiguous -- in light of use of the terms "military action" and "military advantage," the attack has to have an immediate effect on the enemy with whom one is engaged.

A number of stressors exist in maintaining the force of the principle of discrimination. One is the reality of asymmetry. Unfortunately, we are witnessing the growth of disaffected groups that lack the means to directly take on a State's armed forces. Some have concluded that to defeat the State, they must strike at a center of gravity other than the military. This belief often leads them to attempt to bend the State to their will by targeting the citizenry directly -- an asymmetrical attack. Particularly concerning is a growing trend towards mass casualty attacks. Of course, an attack on even a single civilian flies in the face of the principle of discrimination.

The same dynamic operates on the State versus State level. As some countries develop high-tech, highly impressive militaries, and others fall further and further behind, the latter have no chance of meeting the former's superior forces on an equal basis. If you cannot viably face an opponent on the field of battle, you will have to find another way to fight or give up the fight. This massive gap between capabilities is an incentive for ignoring the principle of discrimination.

Compellance strategies represent a further stressor. Such strategies are designed to compel an opponent to do something (or not to), rather than defeat him outright. Of special concern are operations intended to motivate the population to pressure its leadership into compliance. Military forces may not represent the center of gravity in many limited conflicts. The Kosovo case is an apt example, for Milosevic showed little propensity to bend to the will of NATO as his forces were progressively whittled down. In response, one NATO commander suggested "shutting down the lights of Belgrade" in the hope that the resulting discomfort would cause the population to pressure Milosevic to desist from his slaughter of the Kosovar Albanians. However, targeting electrical assets of a metropolitan area simply to affect a population amounts to the targeting of civilian objects (no definite military advantage).

Another stressor is the tendency to "outsource" many military support activities. As armed forces downsize, there is a trend towards contracting civilians and civilian firms to perform functions that used to be performed by military personnel. One example is the use of civilian firms to perform civil engineering functions, such as maintaining runways. Should conflict commence, civilians conducting these activities will find themselves at risk from hostile action against the bases at which they operate. More to the point, some experts have opined that civilians who perform essentially military functions are themselves legitimate targets, rather than merely civilians who may be incidentally injured during attacks on traditional military targets. Whatever their status, there is little question that range of those endangered by hostilities will expand as civilians take on duties previously within the purview of military personnel.

Before turning to the principle of proportionality, it should be noted that I do not view "military necessity" as a principle of humanitarian law distinct from discrimination, though many military manuals do. As typically articulated, military necessity provides that the destruction of property is only permitted when "imperatively demanded" by the necessities of war. If an objective makes an effective contribution to military action and the destruction thereof yields a military advantage, then in most cases an attack would be imperatively demanded.

That said, the concept of military necessity is useful in a non-legal sense. Determining necessity requires one to define the situation, which is certainly determined by the intent of the actor and the context in which the attack will occur. As to the former, consider a very simple illustrative example, the burning of a field. Is the field to be set ablaze in order to create a clear line of fire designed to enhance the defensive situation or instead to deny food to the civilian population? The latter intent demonstrates a lack of military necessity. Correspondingly, it would violate the principle of discrimination.

Moreover, context is relevant. Consider the likelihood of occurrence. For the sake of analysis, assume that the intentional release of oil into the Persian Gulf by Iraq during the Persian Gulf conflict was intended to foil an amphibious assault by Coalition forces (in all likelihood it was not, but assume it was for the sake of this illustration). In assessing the imperative military necessity of the resulting damage to the Persian Gulf ecosystem and oil reserves, the likelihood of the amphibious attack must be estimated. Had such a landing been wildly speculative, it would have been unnecessary quite aside from issues of whether the releases were proportional or attacks on a protected object (the environment). Along the same lines, one must make an effort to estimate the likelihood of success of an attack? Returning to the Gulf War, assume, arguendo, that the Iraqis set the Kuwaiti oilfields ablaze to complicate Coalition air operations by reducing visibility. In fact, the burning oil did present problems to Coalition aircrews. Thus, if this was the Iraqi intent, it would have been militarily necessary given Coalition air operations; to violate humanitarian law, then, the acts would have had to violate a principle of humanitarian law other than discrimination, such as proportionality or specially protected objects (the environment).

Before turning from discrimination to the second aspect of distinction, proportionality, the law of non-international armed conflict deserves mention. The law I have been discussing is the law of international armed conflict; it consists of norms governing State-on-State hostilities. However, most armed conflicts today are internal in nature, styled "non-international armed conflicts" in humanitarian law parlance. The body of law governing such conflicts is miniscule - one article in each of the four 1949 Geneva Conventions (Common Article Three), the 1977 Protocol Additional II to those conventions, and a limited degree of customary international law. The limited reach of humanitarian law into internal conflict should come as little surprise. After all, it represents external limits on what sovereign States may do to their own citizens within the confines of their own borders. States resist efforts to tie their hands except when doing so yields tangible benefits.

Most relevant here is that these cited instruments do extend protection to civilians. Unfortunately, in a non-international armed conflict it is often quite difficult to determine who is and is not a civilian; indeed, their status may shift back and forth as the conflict proceeds. The Institute of International Humanitarian Law is presently sponsoring a project to set forth the law applicable in such conflicts, one in which I am participating. Although the project is far from complete, the drafts currently being circulated reject the term "combatant." In international armed conflicts combatants are entitled to use force against opposing combatants and military objectives. Rebel groups, by contrast, are criminals in the eyes of the State and are in violation of its domestic criminal code. To label them "combatants" might appear to legitimize their activities. Instead, in the law of non-international armed conflict the objective is to highlight the special protection enjoyed by those not participating in the hostilities. In other words, violence against them deserves to be prohibited in both domestic and international law.

To resolve this quandary, the project has tentatively adopted the term "fighters." A fighter is a member of the armed forces of a party to the conflict or one otherwise taking part directly in the hostilities. The definition would include, for instance, someone who is attacking an opponent, conducting sabotage, delivering ammunition, or serving as a spotter for artillery. Fighters are not entitled to the special protections under the law of non-international armed that civilians enjoy. Finally, I must hasten to point out that the nature of the cause the insurgents are fighting for, and the extent to which it is just or not, is completely irrelevant to the question of discrimination in non-international armed conflict. The normative objective is protection of the innocents.

b) Proportionality

Set forth in Articles 51 and 57 of Protocol I, and unquestionably a component of customary international law, proportionality is a principle of distinction that prohibits any attack expected to cause a loss of civilian life (or injury) or civilian property damage that is "excessive in relation to the concrete and direct military advantage anticipated." The principle applies when civilians or civilian objects are not the intended target, but injury or damage to them is likely during the course of an attack on a legitimate military objective.

This balancing test is extraordinarily difficult to conduct because it requires quantifying and comparing dissimilar values. For instance, in practical terms, how is one to measure the suffering caused to civilians during an attack on a command and control facility against the military advantage of disrupting enemy communications? The "value" of an object depends on many varied factors -- the circumstances of the conflict, one's culture, one's historical and experiential base, etc. For example, Germany is very concerned with environmental damage during armed conflict and has considered the topic deeply; other State's practically ignore the environment as a victim of armed conflict. Value may even be temporally determined. Again using the environment to illustrate, the environment is an issue of great concern today, but that was certainly not the case at the time when most humanitarian law conventions were drafted.

A second issue is likewise problematic. In an attack that may be expected to cause the loss of civilian life or damage to civilian objects, what does "expected" mean? How do you calculate the likelihood of collateral damage and incidental injury? What about harm that does not amount to physical damage or suffering? Consider a new method of warfare currently in the spotlight, computer network attack. Many computer network attacks can cause great inconvenience without actually causing injury or physical destruction. For instance, it would be a relatively simple matter to alter bank account data. Given the new capability to cause enormous levels of non-physical "suffering," does proportionality govern this reality?

Finally, when performing the proportionality balancing test, is military advantage resulting from an individual attack to be compared against the incidental injury and collateral damage from that attack, or is the military advantage accruing to the overall campaign as a result of the attack the relevant standard? Upon signing Protocol I, many countries articulated the former position as one basis for their willingness to become Party to the treaty; yet many other concerned players, for example Rights Watch, adopt the former position in their assessment of the proportionality of a particular attack.

Before turning to the stressors to the principle, it is useful to point out several false dilemmas with regard to proportionality, particularly the zero-military causality issue. It is often claimed that militarily advanced States, particularly the United States, seek to conduct warfare in a casualty-less manner…and that doing so violates the principle of distinction. In fact, zero-casualties are an enormous military advantage, particularly in an era of warfare where one highly trained individual in control of advanced weaponry can be deadly effective against an enemy. It is this military advantage, combined with destruction or damage to the objective, which must be weighed against the collateral damage and incidental injury. War fighters have always sought to minimize casualties to their forces; nothing has changed. The extent of casualties is simply one of several military advantage components weighed in the proportionality balancing.

Another false dilemma is the problem of reverberating effects. It has most frequently surfaced in the context of attacks on electrical grids serving both military and civilian facilities. Notable examples include the attacks on Iraqi and Yugoslavian grids during Operations Desert Storm and Allied Force. Recall the charges of medical care being hindered, refrigeration and heating shut-off, etc. Should the second, third, fourth, and fifth tier effects of an attack on a legitimate target be considered when calculating proportionality? Of course they should. Nowhere does the law distinguish between the direct and indirect, or short and long term, consequences of an attack. Yet, the more interconnected and interdependent societies become, the greater the reverberating effects of many attacks, and the more difficult it will be to assess them in advance of an attack. Despite this fact, it simply does not matter whether incidental injury and collateral damage is a first or fourth tier effect. It must be included in the proportionality calculation.

A final false dilemma is that of human shields. Many war fighters mistakenly believe that if the enemy uses human civilian shields, the shielding civilians do not count in the proportionality equation because the enemy is in violation of humanitarian law. That view is, quite simply, wrong; the use of human shields does not relieve an attacker of the responsibility to take civilian injury and death into consideration when assessing whether the target may be attacked. Humanitarian law is generally intended to protect non-participants, not ensure a fair fight. One caveat is merited. If the human shields, as in Operation Desert Fox against Iraq, shield a legitimate target of their own free will, it would not be unreasonable to argue that because the action is volitional, they are now taking a direct part in hostilities. This would result in a status similar to that of illegal combatant; they are taking a "direct part" in hostilities and thus, like the illegal combatant, their injury or loss does not constitute civilian suffering.

Like discrimination, future war poses a number of potential stressors to the principle of proportionality. Consider the transformation of battlefields into battle spaces where combat is virtual and non-linear. In wars of the future, hostilities will be four-dimensional -- war will be everywhere. No longer will civilians be separated from the battle area by limitations in the range of weaponry. Even today, attacks can be launched from continents away with pinpoint accuracy. Unfortunately, at least in the context of preserving the principle of distinction, this means that targets will increasingly present themselves in areas where civilians and civilian objects are collocated. The inevitably of targeting errors (e.g., attacks on ICRC warehouses during Operation Enduring Freedom) and mechanical problems in weapon systems will certainly place protected persons and objects at risk.

Future war will also evidence a growing inability to distinguish combatants from civilians. Given the exponential increase in surveillance and reconnaissance capabilities, as well as other advances that render the battlespace increasingly transparent, combatants have a growing incentive not to identify themselves, lest super high-tech war fighters "ping" them at will. Indeed, much of the controversy of the treatment of Al-Qa'ida and Taliban prisoners surrounds the issue of their failure to distinguish themselves.

Many other similar stressors exist. For instance, off-the-self technology is increasingly being acquired in order to minimize procurement costs and many civilian facilities are now used for military purposes due to the expense of building purely military ones. If military planners rely on Microsoft word processing technology, does the Microsoft plant become a legitimate target? What about a dual-use airfield? Even if they are valid targets, issues of proportionality certainly arise, issues that seldom surfaced in the past. Or consider over-the-horizon, remotely controlled weapons. Although stand-off weapons offer attacking forces greater safety, and though such weapons are highly accurate, ultimately if you cannot see the target, if you remove the human link, mistakes will be made. Even the new non-lethal weapons are a form of stressor. They make it thinkable to use weapons in situations where military forces are intermingled with civilians. In the past, combat forces may have hesitated to use force in such circumstances and, for instance, pulled back. With non-lethals, the proportionality calculation shifts dramatically.

The prognosis is not all poor. Obviously, precision weapons make it easier to discreetly and surgically strike targets and use less explosive force to destroy a target. This is an era when a missile can "count" the number of walls it penetrates before a small warhead explodes to destroy the intended target, and that object alone. Improved bomb damage assessment capabilities generated by the increasing transparency of the battlefield also facilitate distinction. Knowledge as to the precise level of damage to a target means that fewer re-strikes will need to be conducted to ensure it is neutralized. In terms of distinction, the increase in alternatives to kinetic weapons is particularly welcome. For instance, consider the April 1999 bombing of the Radio Televivizije Srbije (Radio-Television Serbia) headquarters during Operation Allied Force, the attack that formed the basis for the Bankovic litigation in the European Court of Human Rights. The affair focused great attention on the question of whether media facilities are valid targets; the case's dismissal on jurisdictional grounds in December 2001 has only served to continue the debate. Through computer network attack, the controversy surrounding attacks against broadcasting facilities need not even surface. Without causing injury or damage, CNA can simply shut off offending broadcasts, thereby rendering the present "dispute" moot.

c) Duty of Care

Protocol Additional I requires an attacker to do everything feasible to verify that a target is legitimate. It also requires using methods and means of attack, such as smart weapons if available and militarily sensible, which minimize incidental injury to civilians and collateral damage to civilian property. In particular, an attack must be cancelled if an attacker realizes that it is not legitimate or that the resulting collateral damage or incidental injury will be disproportionate. Additional, but only to the extent militarily and practically feasible, civilians must be warned of an impending attack

One stressor on this duty is information dominance. An objective of any campaign is to blind the enemy. Modern technology makes this possible to an unprecedented degree. But to the extent you can blind the enemy, and I can tell you we do this quite effectively, you also deprive the enemy of its ability to distinguish between a legitimate and illegitimate target and to accurately assess proportionality. In many cases, therefore, by blinding an enemy, you may place your civilians and civilian property at greater risk.

Another potential stressor is relativity. If a wealthy country can afford weapons that can carry out attacks with great precision and with confidence in target legitimacy, must it? Are poorer countries to be held to a lesser standard of distinction because they cannot afford systems that permit them to, for instance, identify a target with nearly unquestioned accuracy? Disparate standards invite disintegration of the normative regime because one side will feel victimized by the law.

Future Trends

Of course, speculating as to the future is always risky. However, allow me to conclude by identifying five likely trends in the relationship between future war and international humanitarian law. First, there will be increasing technological ability to discriminate between military objectives and protected objects/persons due to advances in weapons guidance and battlespace transparency, a very positive consequence of future war. However, the increasing ability of high-tech forces to blind an enemy in order to achieve information dominance will diminish the enemy's ability to achieve distinction. Second, I expect to see a growing tendency to directly target civilians and civilian objects, both due to the spread of terrorism and as a result of capability gaps that will drive "have-not" forces facing advanced militaries into asymmetrical strategies and tactics. Third, because of technological capabilities, advanced militaries will see their humanitarian law mission planning focus shift from proportionality issues to those surrounding their duty of care. As proportionality becomes increasing easier to achieve due to the ability to conduct surgical attacks, the key question will be whether or not the attacker exhausted the resources available to avoid incidental injury and collateral damage. Moreover, despite the fact that I have labeled the issue of reverberating effects a "false dilemma," they nevertheless present very practical duty of care issues. For instance, should computer experts, engineers, or biologists become part of the mission planning process in order to assess reverberating effects? What is the appropriate level of care in ferreting out possible second, third and fourth level effects of an attack? Fourth is relativity. I predict that humanitarian law will move towards an "if you can, you must" standard. The problem is that the "haves" in this "haves/have nots" dichotomy are going to focus on fairness, or the perceived lack thereof (rather than the desire to protect civilians and civilian objects, the underlying essence of humanitarian law). "If he doesn't, I won't." We have already seen States hesitate to join humanitarian law regimes because of concern that other Parties will ignore normative standards when the time comes to apply them. Fifth, war will almost certainly extend into new areas such as space that are presently not fields of combat. This is certainly the case with regard to cyberspace, a fact that is already presenting troubling implications for the principle of distinction.

Finally, what is to be done regarding the stressors I have highlighted? Although it is not my purpose to comment on the ethics of future war, I would suggest that ethical precepts could prove invaluable in assisting humanitarian lawyers, policy-makers, and combat leaders to interpret the existing law in the context of new methods and means of warfare. Moreover, ethics can assist them in determining how best to fill normative lacuna in humanitarian law that will become increasingly apparent over time. Thus, I enthusiastically welcome this initiative by the Carnegie Council on Ethics and International Affairs to create a dialogue between the legal and ethics communities. Such a dialogue is essential as we contemplate, and prepare for, 21st century conflict.