1. Non-ideal Theory With and Without Guilty Parties
Non-ideal theory concerns how we should proceed when the moral sphere breaks down. Its general goals are to maintain the moral sphere to the degree possible, restoring it when it has broken down and preserving it to the degree possible from future breakdown. What specific rules or principles would these general goals imply?
One such principle has already emerged in the arguments thus far. When the moral sphere breaks down, one should “restrain or stop the guilty parties (those whose plans of action are moral sphere-breaking), not the innocent”—where “restraining or stopping the guilty” would mean interfering with or thwarting to the degree that one reasonably can those who would break the moral sphere in the pursuit of their plans of action and ways of life.
But this requirement leads immediately to a question that has not been previously addressed: Must there always be a guilty party when the moral sphere breaks down? The answer, it turns out, is no; and this answer forces us to delve more deeply into the nature of moral sphere breakdown and the requirements of non-ideal theory.
When moral sphere breakdown occurs, it may simply be because there are conflicts of interests between those involved, such that all parties cannot have all their desires, interests or purposes fulfilled, even though none is yet guilty of using others as mere means. “The ends of human beings are many,” as Isaiah Berlin has aptly said, “and they often come in conflict with one another.” Two nations, for example, might have reasonable claims to the same piece of land. Or one man, let us say, wants to practice the trumpet all afternoon while his neighbor, who works at night, wants to sleep. Or a husband wants to play golf on his day off, while his wife wants him to mow the lawn.
Strictly speaking, the moral sphere has broken down in such situations, because the desired ends or purposes of both parties involved cannot be realized, no matter what is done. Yet neither side may yet be guilty, if no one has thus far decided to “impose his or her will on the others, whatever the desires, interests or purposes of the others may be.”
It would be different if one nation decided to take the land by force, or one neighbor went next-door with the intention of destroying the other’s trumpet. These would be moral sphere-breaking plans of action. Such disputes may come to that, if other options fail. But prior to reaching such a point, the parties merely have conflicts of interests, all of which cannot be satisfied.
What does the account of moral sphere breakdown have to say about conflicts of interest of these kinds, when there is as yet no identifiable guilty party? For the parties involved in such conflicts (say the trumpeter and his neighbor) to “treat each other as ends” would mean “doing what they can do to maintain a moral sphere in which each is treated with openness respect by the other.” This in turn implies that each should “do what he can to allow the other to pursue his way of life and realize his desired ends” (the trumpeter to practice, the neighbor to sleep) to the degree possible in resolving the conflict.
The desired solution, in other words, would be for the parties involved in such conflicts to attempt to reach a compromise to which both can voluntarily (i. e., knowingly and non-coercively) consent. That would be a solution in which neither party “imposes its will on the other, or makes the other do what it wants, whatever the desires, interests or purposes of the other may be in the situation.” The trumpeter might agree to practice only during hours when the neighbor is not sleeping. The two nations might agree to a negotiated settlement in which the disputed land is divided between them; and the husband may consent to mow the lawn if he can play golf some other day. The parties would thereby “treat each other as ends” (by respecting their interests and purposes) in resolving the conflict Carly in the sense required by the Ends Principle.
Alas, we know that many real-world conflicts of interest are not so ideally resolved. Often it is not possible to split the disputed objects, as it is for a piece of land or pieces of a pie. And even when that is possible, it is not always possible to reach compromises to which all parties can agree.
The well-known biblical story of King Solomon and the baby claimed by two women illustrates this point. The King’s suggestion that the baby be cut in half was not meant to be a compromise, but was rather a clever means of revealing the true mother. Even where compromises are possible, parties involved in the conflicts often cannot resolve them face to face; emotions run deep or suspicions may exist that others are not bargaining in good faith.
To deal with such problems, individuals and societies usually resort to “second-best strategies” in order to resolve conflicts of interests—third party arbitrators, judges, juries, majority vote, choosing by lot, and other procedural methods, private, legal and political. These second best strategies will often produce less than ideal solutions. Yet they can be ethically justified, from the point of view of the Ends Principle, to the extent that they attempt to depart as little as possible from a moral sphere in which all persons are treated with openness respect when all cannot realize their desired ends to the degree that they would wish, no matter what is done.
Consider, for example, majority vote as a method of settling conflicts in social and political contexts. A vote of the majority will produce an unhappy minority that does not have its desired ends or purposes fulfilled; so it is far from an ideal solution. But democratic theorists tell us that in many real-world situations, majority vote is the fairest procedure one can devise for settling many disputes. It is so to the extent that it respects the voice of each party equally in deciding which view will prevail, when one view must prevail over others, no matter what is done. Among the caveats added by democratic theorists, of course, is that the majority vote is arrived at under fair voting conditions—no fraud or manipulation of the results to favor one outcome, no tampering with voter machines, and so on. Any such fraud or manipulation would involve one group “imposing its will on others, whatever the desires or interests of those others might be in the matter” —and not doing so to restore the moral sphere, but to break it.
2. Non-ideal Theory Continued: Life-boat Scenarios
Many other examples further illustrate the application of these principles to moral sphere breakdown situations without a guilty party. Particularly instructive are examples of so-called “life-boat scenarios,” which are familiar in discussions of ethics.
A ship has gone down at sea and ten persons find themselves on a life-raft that can support only eight. Or, five men on a disabled airplane have only four parachutes, none of which will support more than one man. The moral sphere has badly broken down in such situations, since the desires and purposes of all parties cannot be realized, no matter what is done. (We assume they all want to live.) Yet no one of the parties is guilty, if the disasters were accidental and no person involved was responsible. The world has simply turned rotten for these people through no fault of their own.
Consider the five men on the disabled plane with four parachutes. What should they do? What they might do, of course, is fight over the parachutes, thereby settling the matter by force. But, while it may come to that eventually, the question is: What “ethical” or “moral” options are available to them before force prevails? For guidance here we must look again at the requirement for treating all persons as ends in such moral sphere breakdown situations without a guilty party: “Do what you can to maintain a moral sphere in which each party is treated with openness respect by the others, when the ends or purposes of all parties (to live in this case) cannot be fulfilled, no matter what you do.”
A number of possibilities are suggested by this requirement in the present case. Suppose, for example, that one of the five men were to volunteer to sacrifice himself and become a hero to save the others. Such sacrifice is not required, since respecting each in the sense of openness means respecting their freedom to choose. Yet our intuitions suggest that such a decision to sacrifice on the part of one of the men would lead to an ethically favorable solution. If so, why?
Well, note that if one of them did voluntarily take this heroic step, there would be a significant sense in which the conditions of a moral sphere would be restored to a degree among them, despite their tragic circumstances. For all of them would now be “able to pursue their chosen ends or purposes without interference or coercion by the others.” The four who would now have their parachutes would be able to survive and the one who had chosen to sacrifice would have freely choosen to take this heroic course. Since the volunteer’s choice was genuinely voluntary and uncoerced, no one would be “imposing his will on the others.”
Note an interesting lesson here: Heroic acts like this one may restore the moral sphere to a degree possible when the world has deteriorated to such an extent that maintaining the ideal of openness respect to all is at its most difficult. Heroism of this sort goes beyond the call of duty—it is a “supererogatory” act, as ethicists say. Yet it may serve the ethical ideal in many circumstances by restoring and preserving the moral sphere to a degree when that ideal sphere has badly broken down.
Suppose, however, that in this case of the disabled airplane, the heroic solution is not available. No one volunteers to sacrifice. What then? A suggestion that naturally comes to mind is that, time permitting, the men should draw lots to see who will get the parachutes. In the absence of other ethical options, this suggestion has merit. But, if so, why?
Though someone is going to die unwillingly, if they all want to live, choosing by lot would mean that all will be treated equally in deciding who it will be. By deciding in this way, the men would therefore come as close as they can to maintaining a moral sphere among them in which all are treated with openness respect by the others in these dreadful circumstances. For each would have an equal chance of having his purposes fulfilled, when it is impossible for all to have their purposes fulfilled. Centuries ago, Aristotle noted that choosing by lot would be the fairest solution to some difficult social and political problems that could not be fairly resolved in any other way. Our reasoning supports this claim.
But now suppose, finally, that in this downward spiral of events, the men on the airplane do draw lots and the loser refuses to abide by the outcome. What then? It appears they’ve now reached a stage at which force is all that remains. If the lottery loser tries to secure one of the parachutes by force, the other four are justified in subduing and restraining him. What justifies them in doing so, however, is that they have reached a point where, by following such a procedure to decide who would survive, they have done as much as they can do in the circumstances to maintain the ideal of a moral sphere in which all persons can be treated with openness respect by all others.
By choosing in this way to use force only as a last resort, the men on the airplane are following the ethical requirement which the retreatants arrived at in part 3 as a result of their reasoning—not to forceably impose one’s will on others except as a last resort.
This constraint, which the retreatants’ imposed on themselves by taking an attitude of openness respect to all persons, can now be seen to be related to a general principle of non-ideal theory: In non-ideal circumstances of moral sphere breakdown with a guilty party or not, it is the persistent striving to maintain to the degree possible a moral sphere in which all persons can be treated with openness respect by all others that guides one in the search for the ethically correct solution.
3. Levels of the Moral Sphere: Might and Right, War and Peace
These results can now be summarized in a simple diagram.
The inner circle (1) here represents the moral sphere in which all persons can be treated with openness respect by all others. Circle 2 represents moral sphere breakdown without a guilty party, including conflicts of interest, while circle 3 represents situations of moral sphere breakdown with a guilty party. Finally, level 4 represents the point where ethical options have been exhausted and force prevails, as when the lottery loser on the plane refuses to abide by the result and must be subdued by the others.
Level 4 represents the most morally degraded of conditions. It is, as 17th century philosopher Thomas Hobbes aptly called it, the “state of nature”—a state of war of “all against all” in which no moral laws obtain. No further circle is drawn around 4 because it is not strictly speaking a moral sphere at all, as Hobbes noted. Whereas in the inner three circles, we retain a measure of respect for the ideal moral sphere and attempt to restore it, at level 4 the world is reduced to a point where we have lost it altogether.
Imagine the diagram this way: As we move outward from the center, the moral condition of the world gets worse. But the ethical aim remains the same—to try to restore and preserve as much of the original ideal of a moral sphere in which all persons can be treated with openness respect by all (circle 1) as is possible in deteriorating circumstances, until we no longer can (level 4).
It was noted in part 2 that pacifism might be the right view if the moral sphere never broke down, if we were always in circle 1. We may now add that the opposite doctrine of “might-makes-right” might be correct in many circumstancess, if a Hobbesian state of nature always prevailed, if the world were always at level 4. But we are not always at either of these two levels.
When the moral sphere breaks down, it is not the case that we are thrown immediately into level 4 where “anything goes” or “might makes right.” To the extent that we continue to be guided by the ideal of a moral sphere in non-ideal circumstances, moral constraints emerge as we have seen between levels 1 and 4. These constraints, such as restraining the guilty (those whose life-plans are moral sphere-breaking) rather than the innocent, and seeking the fairest resolution of conflicts, are ones we can seek to satisfy before the state of nature prevails.
4. Minimal Force, Warfare and International Conventions
Viewing things this way suggests another principle of non-ideal theory that must now be added: “When the moral sphere breaks down, use minimal force to restore and preserve it.” The assailant in the alley whose life-plan broke the moral sphere should be restrained. But neither the passers-by nor the police have the right to shoot him through the heart, if he can be subdued with less force. What justifies such a requirement?
The answer is that by using minimal force to restore and preserve the moral sphere, we accord to the guilty parties as much residual recognition in the form of openness respect as we can, consistent with the fact that we must restrain them. Doing this is a further implication of the general goal of non-ideal theory: To depart as little as possible from the ideal of a moral sphere in which all are treated with openness respect when that ideal cannot be perfectly realized. For similar reasons, prison inmates retain some rights and should be treated with some measure of the recognition required by treating them as ends, even as they have forfeited the full measure of such recognition.
Consider as further examples the rules of just warfare embodied in international conventions, such as the Hague and Geneva conventions. One can see operating behind these familiar conventions of warfare the two rules we have identified for moral sphere breakdown at level 3 of the diagram: “Restrain the guilty, not the innocent” and “Use minimum force to restore and preserve the moral sphere.”
The first of these rules is exemplified by Geneva conventions against harming innocent non-combatants and indiscriminate saturation bombing of civilian populations, among many other restrictions. The second rule is exemplified by rules protecting prisoners of war or mandating medical care for captured enemy soldiers, among many other requirements.
Both these rules (“Restrain the guilty, not the innocent…” and “Use minimum force…”) are consequences, as we have seen, of striving to maintain the moral sphere to the degree possible in non-ideal circumstances. Innocent noncombatants are to be treated with as much recognition of their desired ends (staying alive, not being harmed); and combatants are to be treated with as much residual recognition of their desired ends as is consistent with the needs of battle.
But what happens if both (or many) parties are guilty? What if both of two warring nations are guilty of being overly aggressive and harming the innocent in settling their dispute or if different factions in a civil war are both brutalizing each other and innocent civilians? These are more extreme cases of level 3 breakdowns of the moral sphere, that is, breakdowns with more than one guilty party.
The goal that is consistent with our reasoning in such cases would be to try to restore the ideals of fair conflict resolution like those of circle 2, where no party is yet attempting to impose its will on the other by force. In other words, attempt to return, if possible, to a level closer to the moral center (circle 1)—in this case, from level 3 to level 2. Not surprisingly, fair conflict resolution is more problematic and difficult when there are many guilty parties than when none is yet guilty. But the goal is the same—resolution of conflict that respects all parties in the sense of openness to the degree possible.
With this goal in mind, familiar methods suggest themselves. One might try to talk the warring parties (nations or factions in civil war) into a voluntary cease-fire, to cool down emotions and then try to get them to sit down with each other at the negotiating table. If face-to-face negotiation fails because of distrust or lack of bargaining in good faith, one might try third party arbitrators or judges (that is, second-best strategies).
These are common diplomatic procedures. But we can also recognize them as ethically justified efforts of non-ideal theory, however difficult they may be to carry out in practice. For they all involve staying as close as possible to the ideal of a moral sphere in which all parties are treated with openness respect to the degree possible under extremely difficult conditions.
If each of these successive efforts fails, the situation will edge closer to level 4 where force may become the only option. At a certain point, other nations may have to stand back and let warring factions fight while taking what steps they can to protect innocent non-combatants. Or, better, but still far from ideal, other nations might try to impose a cease-fire by force, if they have the power to do so without unreasonable costs and without causing more bloodshed than would otherwise occur. This would especially be the case if such a cease-fire were a prelude to subsequent negotiations between the warring parties themselves.
Imposition of force by other nations in international relations is fraught with practical and long-term political dangers, as we all know from the history of the past century; and it should be engaged in with great caution. Yet it may often be ethically justified, if its aim is to restore a moral sphere with minimal force and it does not lead to greater harm to innocents and bloodshed than would otherwise occur.
Similar practical and political obstacles attend humanitarian interventions within nations where one group is brutally persecuting others, even though humanitarian interventions can also be ethically justified by similar reasoning, if they do not lead to greater harms to innocents than would otherwise occur. But it should not be surprising that moral sphere breakdowns with many guilty parties would be more difficult to resolve than those without guilty parties. Yet the ethical goals remain the same, even in such degraded conditions—to try to restore and preserve the moral sphere to the degree possible.
5. Non-violence and “Experiments With Truth”
Such goals also tell us something interesting, in conclusion, about the ethical significance of non-violence movements, which have been associated in the 20th century with figures such as Mohandas Gandhi and Martin Luther King. We know from experience that humans are often prone to assume too quickly that the moral sphere has broken down and prone to leap to violence too soon as a means of resolving conflicts. We often use more force when less force will do.
But how can we know how much force is needed unless we are willing to experiment with less? The significance of non-violence movements is that they may teach us how to experiment with less. It is noteworthy that when Gandhi undertook his efforts to make the British end their control in India without resort to violent revolution, he referred to his practices of non-violence as his “experiments with truth.” Satyagraha was the Sanskrit term he used for his non-violence movement, which literally means “truth-force.”
This idea of practical “experiments with truth” is especially important for the argument for non-ideal theory developed here, in which the goal is to maintain a moral sphere to the degree possible in an imperfect world. If we need scientific experimentation for technological progress, then why not moral experimentation for moral progress? What would be meant by “moral progress” in this context, in contrast to technological or other kinds of progress, is expanding the moral sphere (circle 1) so that it encompasses more areas of life—so that more conflicts can be settled by persuasion rather than by coercion.
One final lesson emerges here. It is instructive to think of the moral sphere in this way as the sphere of persuasion, in contrast to coercion, which must often come into play when the moral sphere breaks down. But saying this would not be quite accurate, if the category of persuasion included the manipulative persuasion that is so common in modern societies through political spin doctoring, public relations, propaganda and other means. For the goal of such manipulative persuasion in its many forms is usually to “impose the persuaders’ wills on others,” and often in doing so, to use others as a “mere means” to the persuaders’ ends.
A useful expression to contrast such manipulative persuasion from the kind of persuasion intended in the moral sphere is the Quaker expression friendly persuasion. This is a kind of persuasion that respects the values and purposes of others, allowing them to freely choose to assent or dissent from their own points of view. By saying moral progress lies in expanding the moral sphere, so that more conflicts are settled by persuasion rather than coercion, it would be friendly persuasion in this sense that is meant.
When viewed from this perspective, non-violence movements, like those of Gandhi and King, are among the most important moral experiments of modern times. They contribute to moral progress by showing that we often use more force when less will do, so that non-coercive methods can work in areas of life where they were never previously thought possible. The moral sphere is thereby expanded beyond where it had been before.
6. Summary: Moral Sphere Theory
To sum up, non-ideal theory deals with moral sphere breakdown and how we should proceed when the moral sphere breaks down. It may be viewed in terms of the earlier diagram of nested circles, which describe levels of moral sphere breakdown, with and without guilty parties.
Principles of non-ideal theory emerge at the different levels: “Restrain the guilty, not the innocent” and “Use the minimum force necessary to restore the moral sphere,” when there are guilty parties (level 3). And where there are conflicts of interest, but as yet no guilty parties (level 2) “Try to find as fair a compromise as possible, to which all parties can voluntarily consent.” Or, where such compromise is not possible, “Try to find second-best strategies that will respect the interests, concerns and purposes of all parties to the degree possible.”
All such principles of non-ideal theory are guided by a unifying goal: Striving to depart as little as possible from the an ideal moral sphere in which all persons can be treated with openness respect by all others (circle 1), when one must depart from this ideal to some degree. It is this unifying goal that makes non-ideal theory an ethical theory rather than a collection of ad hoc pronouncements.
This same goal also makes non-ideal theory a natural extension of ideal theory since both follow from the Ends Principle. To “treat all persons as ends in every situation and no one as means only” requires maintaining the moral sphere to the degree that one can in the varied situations. This means not breaking it where it obtains (ideal theory), and doing what one can to restore and preserve it when it has broken down (non-ideal theory).
 Berlin 1965: xl.
 Hobbes 1958, chapter 13.
 Gandhi, Selections in Somerville and Santoni eds., 1963: 500-510.