The effort to perceive distinctions and to draw lines between different actions, plans, and characteristics lies at the core of practical choice, whether in legal, religious, or moral contexts. While an area of uncertainty or dispute about difficult cases will always remain, such an effort can reduce it considerably. (1)

      To see how questions of line-drawing arise in practice, consider, once again, the French sabotage of the Rainbow Warrior. Those who sponsored this venture might agree that it is generally wrong to sink foreign vessels in peacetime and to take human lives in the process. But in this instance, the French officials might argue, the assault was a clear case of self-defense. In their view, members of the Greenpeace organization threatened France's military preparedness by their insistence on observing its nuclear tests. The government had tried every legal means of stopping such observations, to no avail. And while the action presented serious risks to those on board the vessel, no violence toward any individual had been intended, given the warning blasts meant to frighten all on board into disembarking.

      These arguments might well appear less persuasive to the officials who planned the sabotage if environmentalists were to take similar liberties with the property and the lives of French citizens. That question took on practical significance a year later, when Paul Watson, reportedly expelled from Greenpeace in 1977 and the leader of another environmental group, the Shepherds of the Sea, claimed responsibility for sinking two of Iceland's four whaling ships. He argued that his followers had used no explosives and been careful to avoid any loss of life. They had begun opening key valves in the ships only when they had ascertained that the entire crew had left each ship; and they claimed that such destruction of property did not constitute violence. Illegal acts, Mr. Watson is reported to have said, are justified to stop environmental abuse, as long as they defend and do not endanger living things: but "people tend to have more respect for private property than for the sanctity of life."

      The Greenpeace organization, though its representatives have long urged an end to whale hunting, denounced such methods of aggression, whether to property or to life. We see, then, three different lines drawn to define what one can rightly do to those regarded as adversaries on a political issue: one line rejects all action meant to injure life or property; a second rejects posing direct risks to life but not destroying property; and a third allows placing lives at risk as well as destroying property. It is possible to envisage any number of additional cases incrementally more or less violent and/or destructive than the attack on the Rainbow Warrior or of the Icelandic whaling ships. Likewise, one can imagine different degrees of justification for the acts, ranging from self-defense in an imminent national emergency to sheer aggression for its own sake.

      Line-drawing, in these cases as in many others, is both unavoidable and open to different forms of error and bias. States, like individuals, define themselves and the integrity they strive for in part through the lines that they draw in principle and the degree to which they honor these lines in practice.

      That task is never entirely completed; new cases and new circumstances arise to test boundaries that once seemed clear. To take an analogy from medicine: up to this century, it was usually easy for doctors to state whether a person had died or was still alive. With the advent of modern technology allowing them to keep a patient's heart beating even though the brain may have ceased functioning, physicians and legislatures have had to draw new lines with respect to when death has arrived. In most cases, doctors can still diagnose death as readily as in the past; but there are now different types of borderline cases where complete certainty eludes them and a few where sharp moral disagreement persists. Similarly, technology has made it possible for states to threaten and to spy upon each other in new ways; and the questions raised by these developments have brought disagreement as intense as that associated with the end of life.

      How might policymakers best approach the task of establishing standards and making choices, given these difficulties? It helps to start out with a shared framework of moral constraints and to formulate clear-cut cases to which all agree that the constraints apply; from such a basis, it is easier to explore remaining differences. For this purpose, one can envisage certain examples of clearly unjustifiable or, on the contrary, clearly justifiable instances of violence, deceit, breaches of faith, or secrecy, as benchmarks.

      Terrorism, for instance, is clearly unjustifiable from the point of view of such a framework of constraints. One need not attempt to draw lines with respect to acts of threatening or engaging in violence against civilians, often randomly, so as to spread terror and thereby to further political aims. Such acts are wholly on one side of the line --the side that should be ruled out. Ruling out terrorist practices, however, will call for more than the customary denunciation, the more so as many leaders publicly reject such practices while engaging in them, subsidizing them, or tolerating them in secret. Partisanship leads some to justify terrorist action by those on their own side on the ground that they are freedom fighters while castigating their opponents as terrorists pure and simple. Yet all who place bombs on buses or planes, say, or strafe health clinics or schools, must be seen as terrorists, however dedicated they may be to the cause of freedom. Governments likewise engage in terror if they resort to such methods at home or abroad, no matter how noble the purposes they claim to serve.

      Disinformation is another practice that can serve as a benchmark when it comes to violating a fundamental moral constraint. For while one may debate where to draw the line when it comes to different items of deceptive propaganda, depending on how clearly they are intended to deceive and the degree to which they depart from the truth, no such debates need arise over schemes of disinformation; they are planned from the outset to mislead, whether by means of planting false information in the press, forging documents, or inducing officials to give false testimony.

      Terrorism and disinformation, and related forms of violence and deceit, not only wreak scatter-shot injury on victims; they also hurt the credibility and the reputation of governments and organizations known to engage in them, damaging the atmosphere in which nations must try to work out responses to the overriding collective threat of war. Once set, such examples of disrespect for law and morality linger long after particular incidents have ceased to dominate newspaper headlines. Such policies invite retaliation and imitation on the part of adversaries and add to public distrust, especially among citizens made complicitous against their will.

      With such benchmarks in mind, it becomes easier to separate clear-cut cases from the rest and to determine where, along such different dimensions as that of the degree of violence or the innocence of the victims, uncertainty and disagreement arise. Much of what human beings do--governments as well as private groups and individuals--is perfectly consonant with the four constraints and thus unproblematic from those points of view. Of the types of conduct that do violate the constraints, a small number are within the range of justifiable breaches, while most are clearly illegitimate.

      As for cases where there is more disagreement even in principle about how to evaluate them, what might be entailed by deliberating about them? In earlier writings, I have suggested a three-step procedure for weighing an action or a practice fraught with conflict: first, to ask whether there is an alternative course of action that will achieve the aims one takes to be good without breaching moral constraints; second, if one sees no such alternative, to set forth the moral arguments thought to excuse or to justify such breaches, along with possible counterarguments; and third, in weighing them, and as a test of the first two steps, to ask how such arguments would fare if defended in front of an assembly of reasonable critics.

      How does this three-step procedure apply in the case of the Rainbow Warrior or the activities of the Shepherds of the Sea? Those responsible for both actions argued that they chose to act as they did only after having exhausted all lawful means at their disposal. But even if such a claim is accurate, it is not, by itself, sufficient to justify peacetime assaults on civilians or their property. If government officials explained that they had tried every lawful means of raising funds needed to support a friendly regime in distress, and that they had turned to extortion for this worthy cause only when all else failed, they would not expect to persuade many.

      Nor does the moral deliberation called for in the second step justify the French action or that of the Shepherds. France's long-range aim of enhancing national security by preventing observation of her nuclear tests might justify restricting access to the test site but not sinking ships and taking human lives. The Shepherds argued that their assault was called for in defense of innocent lives otherwise at risk--an argument also advanced by those who place explosives in abortion clinics. Among the many groups likewise invoking the defense of innocent lives or national security for their violations of clear-cut moral standards, few go beyond such rationales to examine the impact of their actions on those whom they injure indirectly, on themselves as moral agents, or on their societies. Nor do they consider the moral relevance of inviting, through their conduct, further retaliation by their adversaries and imitation by yet other groups with different agendas, equally passionately held, or the cumulative effect on the climate of distrust within or between communities.

      In making such arguments, while shunting aside all but the narrowest considerations, those planning to violate moral constraints often take for granted that these constraints should nonetheless hold for others. Yet the question, What if everybody did that? is in part meant to expose the inadequacy, in the absence of further arguments, of such self-serving exceptions.

      The reasons offered in support of violations are especially likely to remain inchoate and partisan so long as they are not open to challenge from the outside. The third step, the test of publicity, is meant to provide such a challenge. It helps counteract biases, errors, and ignorance--and thus the misjudgments they bring about--by asking how justifications and excuses would hold up in open debate and what would happen if, as often occurs, those responsible for planning, authorizing, and carrying out clandestine actions were exposed to public criticism.

      When it comes to government policies in a democracy, the test of publicity requires full accountability to the public or to their elected representatives. By making one's arguments explicit and subjecting them to inspection and criticism, the test challenges private biases, errors, and ignorance, and allows the stretching of perspective so crucial to moral choice. Such sensitivity is never more important than when distrust between adversary groups is strong and partisanship risks skewing their judgment.

      The three-step procedure may turn out not to be needed in clear-cut cases. And it can rarely bring a fully satisfactory resolution to the most difficult conflicts--least of all to what have come to be known as "tragic conflicts," in which the available alternatives are all dismal from a moral point of view. In Mortal Questions, the philosopher Thomas Nagel asks about such a situation: "What if the world itself, or someone else's actions, could face a previously innocent person with a choice between two morally abominable courses of action, and leave him no way to escape with his honor?" (2) An officer may be ordered to torture terrorists to learn where they have concealed the bombs with which they are threatening to blow up a city; a government leader may have to choose between violating a binding treaty and risking full-scale war.

      But such conflicts are rare. Good government can make them rarer still. Most of the time, government officials as well as private individuals know quite well when injunctions such as those not to kill or to lie bear upon their actions. As a result, I shall argue that we can get off to a good start with a strategy for peace merely by cutting back on activities that clearly violate the most fundamental moral standards--rejecting all support for terrorism, torture, assassination, disinformation, cheating on arms agreements, and the mining of harbors and international waterways, for instance -- and by encouraging policies that restore, rather than damage, confidence between nations.

      Even with respect to marginally legitimate activities -- say, ones that come close to involving treaty violations -- it matters that governments bend over backwards to avoid them when at all possible. Such acts are peculiarly open to misinterpretation and to charges of hypocrisy. They create precedents, habits, and policies that can lead to more flagrant abuses. They then combine with the clear-cut violations to increase distrust of a government among other nations and to add to the cynicism and discouragement among its own citizens. The more discredited a government has become, the greater are the difficulties that it will experience in regaining respect. For all these reasons, those who advocate or take part in practices that risk violating fundamental moral constraints should assume that they confront a heavy burden of proof.

      Putting into practice a framework of moral constraints does not call for agreement on all questions still in dispute. If it did, the urgent work for peace would never begin. Rather, it calls for agreement about clear-cut practices and for working one's way toward resolution of those about which there is still widespread disagreement. It is in such practical ways that the strategy for peace can be undertaken. Focusing on the aim of collective survival, cognizant of the resources it must mobilize and of the obstacles it confronts, holding to the framework of fundamental moral principles that restricts some means and encourages others, such a strategy can serve to unite many disparate efforts to reduce the threat of war.

(1) Thinkers in the great traditions of practical moral inquiry--among them Confucians, Stoics, and commentators on Christian, Jewish, Buddhist, and Islamic ethics--have developed methods of sorting through moral problems that can illuminate many of the difficulties that still confront us today. They have discussed the ways in which particular choices and ways of leading one's life interact; considered the role played by definitions and different forms of justification; explored the nature of analogies and the relationship between principle and practice; and worked out methods of line-drawing in difficult cases of conflict that distinguish between marginally different definitions, circumstances, and degrees of justification.