Double Effect, Double Intention, and Asymmetric Warfare
The following set of statements is inconsistent. The statements cannot all be true. This is a problem for just war theory because each of them is either a claim to which the theory is committed or is obviously true.
(1) Some wars are justifiable.
(2) Civilians must not be attacked in war.
(3) Military operations in general cannot be carried out without civilians being attacked.
The first statement is a basic
assumption of just war theory, specifically, of jus ad bellum, and is the basis of the theory’s distinction from
pacifism. The second is a partial statement of the principle of discrimination,
central to jus in
The usual way of solving this trilemma is the doctrine of double effect. After showing how the doctrine avoids the inconsistency, I will discuss some problems with the doctrine and consider a revision of the doctrine offered by Michael Walzer’s, the idea of double intention. Then I will criticize and elaborate on Walzer’s revision and explore some implications of all this for contemporary asymmetric warfare.
One approach to avoiding the apparent inconsistency in the trilemma is to claim that there is an equivocation in the term “attacked” as it appears in (2) and (3), such that, once the equivocation is removed, consistency is achieved. “Attacked” may have two meanings: in the course of military activity, people are attacked when
(i) the attackers intentionally harm them; or
(ii) the attackers harm them, intentionally or not.
To avoid equivocation, “attacked” must have the same meaning in (2) and (3). If we use meaning (i) in both, then (3) is clearly false, since military operations can be carried out without the attackers intending to harm civilians. If (3) is false, then the trilemma is avoided. Thus, a solution to the trilemma is to understand statement (2) and the principle of discrimination in terms of the first meaning of “attacked.” Statement (2) then becomes:
(2’) Civilians must not be intentionally harmed.
This is an understanding the principle of discrimination that avoids the trilemma. This is the doctrine of double effect. Military actions can have two kinds of effects involving harm to civilians, harm that is intended and harm that is not intended. Intentional harm to civilians is prohibited, but unintentional harm to civilians is not.
The distinction between intended and unintended effects can be understood in different ways, and it is important to appreciate how it is understood by the doctrine of double effect. Consider this division of an action’s effects:
I. Unforeseen effects (resulting, for example, in mistakes and accidents);
II. Foreseen effects:
a. Effects the actor seeks to bring about (the goals of the action), as well as effects that are the means to the achievement of the goals;
b. Effects of the action consequent to the achievement of the goals; effects that the actor does not seek to bring about, but which she recognizes to be inevitable or likely byproducts of the action.
The basic division is between foreseen and unforeseen effects. An effect is foreseen when the actor believes that there is a significant possibility, though not necessarily a certainty, that the action will bring it about. If the actor has no such belief about an effect of the action, then the effect is unforeseen. In terms of this division, which effects are intended and which not? Type I effects, unforeseen effects, are clearly unintended. In contrast, type IIa effects are clearly intended. The question is about type IIb effects. Some philosophers and legal scholars regard type IIb effects as intended, though they are often distinguished from type IIa effects by referring to them as indirectly or obliquely intended. But, in terms of the doctrine of double effect, for the sake of understanding (2’), type IIb effects are not viewed as intended. They are often referred to as effects that are merely foreseen.
The doctrine of double effect is not, however, an adequate solution to the problem. Interpreting the principle of discrimination in terms of the doctrine does not do enough to constrain military activity in the risk of harm it imposes on civilians. The doctrine of double effect is too lenient; it does not capture the extent to which combatants should seek to avoid harming civilians. Not only should combatants not try to harm civilians; they should try not to harm them. Taking care to avoid harm to civilians must mean more than simply not intending to harm them. To see that this is the case, consider a counterexample to (2’). It is an implication of (2’) that a city may be leveled in order to destroy a military command post in that city. After all, the attackers were not trying to harm the civilians, but simply trying to destroy the command post. If the intention behind the attack were to destroy the command post, then the deaths of the civilian residents would not have been intended, merely foreseen. Such an implication casts (2’) into doubt. In fact, the reductio ad absurdum of the doctrine is the idea that all it requires of attackers it that they go through the mental exercise of “focusing their intention” on whatever military aspect of the target may exist.
response to such a counterexample, defenders of the doctrine of double effect
might point out that the destruction of the city, though perhaps permissible
under (2’), would not be permissible under the other main rule of jus in
Michael Walzer has proposed an interpretation of the principle of discrimination that corrects this difficulty in the doctrine of double effect. As a revision of the doctrine of double effect, he proposes that we think of the principle of discrimination in terms of a double intention. Intentionality should relate to both kinds of effects, not simply in the first kind of effect. The second kind of effect, the “merely foreseen” effect, should involve an intention to reduce the risk of civilian harm, “a positive commitment to save civilian lives,” more than the negative commitment involved in avoiding an intention to do civilian harm. Combatants must act with an intention to reduce the risk of harm to civilians. Under this interpretation, the principle of discrimination then becomes:
(2’’) Civilians must not be intentionally harmed, and positive efforts must be make to reduce the risk of harm to them inherent in the military action.
One characteristic of this positive commitment is that it will often involve an increased risk to the combatants. But, “if saving civilian lives means risking soldier’s [sic] lives, the risks must be accepted.” But what more precisely does (2’’) require?
Before addressing this question directly, consider first some criticism of Walzer’s account of the principle of discrimination. One critic, J. G. Fleury, denies the need for an interpretation of the principle of discrimination more stringent that the doctrine of double effect. Walzer’s notion that combatants should assume greater risk “conflicts with military logic and the psychology of command.” He asserts: “Nobody has a right to ask more from soldiers who are putting their lives on the line. The principle of double effect, which stipulates that the death of non-combatants is morally acceptable only when death is an indirect effect and not an end or a means in itself, provides the moral guidance necessary in such circumstances.” What is his argument for this? “Certainly, soldiers do not have the same positive duty to protect innocents among the enemy population, as they have to protect their own population, although they have an obligation not to harm innocents intentionally regardless of their nationality.” Apparently, the doctrine of double effect is sufficient to satisfy the combatant’s duty because the civilians who are at risk of harm from his actions are enemy civilians, not fellow citizens.
Fleury’s argument should be rejected because it assumes that the moral status of civilians results from their being enemy civilians. Rather, their moral status, their right not to be attacked, results from their status as human beings, irrespective of their nationality. The principle of discrimination is based on the idea that enemy civilians are protected because they are innocent, in the sense that they are not causally responsible for the threat of harm posed by their nation’s military forces. They are just as innocent in this sense as the combatants’ fellow citizens. Walzer asserts: “The structure of rights stands independently of political allegiance; it establishes obligations that are owed, so to speak, to humanity itself and to particular human beings and not merely to one’s fellow citizens.” We should not follow Fleury in rejecting Walzer’s revision of the principle of discrimination on the grounds that combatants do not owe the same respect to enemy civilians as they owe their own.
The key feature of the idea of double intention is that a military action that carries civilian risk must include an intention to reduce that risk. But what, more precisely, does it mean to intend to reduce civilian risk? This is separate from and prior to the question of how much the risk should be reduced. In combat situations, there are military objectives that the combatants seek to achieve. For any military objective, there is a set of alternative means available for attempting to achieve that objective, and the combatants must choose among these. One of these alternatives is best in a military sense, that is, it is the most likely to achieve the objective at the least military cost. Let me refer to this alternative as the optimal alternative. All other members of the set of alternatives will be suboptimal alternatives. Normally, many, if not all, of the alternatives will pose some risk to civilians. Thus, we may divide the suboptimal alternatives into two groups, those that pose a greater civilian risk and those that pose a lesser civilian risk than the optimal alternative poses. The latter group of alternatives may be referred to as the lesser-civilian-risk alternatives.
With this stage setting, we may offer the following characterization of part of the idea of double intention.
The idea of double intention requires that combatants choose, in respect to every military objective, a means to attempt to achieve it that is a member of the set of lesser-civilian-risk alternatives for that objective.
This is what it means to intend to reduce the risk to civilians. We may note in passing that the fact that the idea of double intention requires the choice of an suboptimal alternative shows why, as Walzer asserts, having a proper second intention generally entails an increase in combatant risk. A suboptimal alternative is more militarily costly, and combatant risk is one of the chief forms of military cost.
Now to the next question. How much does the idea of double intention require that civilian risk be reduced? Or, how much should the risk to our combatants be increased to bring about a reduced risk to their civilians? Each alternative in the set of lesser-civilian-risk alternatives decreases civilian risk to some extent. So to ask how much civilian risk should be reduced is to ask which of those alternatives may be chosen. Presumably, the idea of double intention will not tell us which particular alternative should be chosen. Rather, it will tell us which members of the set of lesser-civilian-risk alternatives may be chosen and which may not be chosen. The idea of double intention divides the set of lesser-civilian-risk alternatives into permissible choices and impermissible choices. The impermissible choices are the alternatives that do not lower civilian risk sufficiently. If the idea of double intention did not indicate that some members of the set of lesser-civilian-risk alternatives were impermissible, then the tiniest bit of positive effort to reduce risk to civilians, corresponding to nothing more than the tiniest increase in combatant risk, would satisfy the requirement for the second intention. This would allow a form of moral disingenuousness in adherence to the idea of double intention similar to that provided by the idea that adherence to the doctrine of double effect requires only that combatants focus their intentions on the military aspects of their objectives.
What does Walzer say about how much the risk to civilians should be reduced?
When first introducing the idea of double intention, he says that the second intention should be to reduce the risk to civilians “as far as possible.” But this cannot be correct because reducing the risk to civilians as far as possible would involve an open-ended increase in the risk to combatants or an abandonment of the military objective. Walzer recognizes this, saying later, “there is a limit to the risks that we require” that combatants take to reduce civilian risk. But what is that limit? How much should civilian risk be reduced and, correspondingly, combatant risk be increased? In answer to this question, he makes three basic points.
First, he claims that the extent to which civilian risk should be decreased varies from case to case, depending on “the nature of the target, the urgency of the moment, the available technology, and so on.” There is no way to specify in advance how much decrease in civilian risk and increase in combatant risk is required. It depends on the circumstances.
Second, he proposes that the extent to which civilian risk should be decreased is set by the right of enemy civilians to have combatants exercise “due care” in their military operations. This right, he suggests, is analogous the right of citizens in the domestic sphere that their fellows exercise due care in the amount of risk of harm their actions impose on others. This domestic analogue is presumably the notion of a right of due care that is elaborated in negligence law. The second point elaborates the first point that the required decrease in civilian risk varies from case to case; for, what due care requires varies form case to case as well.
Walzer’s third point, like his second, makes use of the domestic analogy. In war, the standard of due care owed civilians is relaxed in comparison with what it is in the normal domestic situation. The standard in war is more like the relaxed standard found in some unusual domestic situations, such as utility employees working to avoid an imminent gas explosion. Given the urgency of the situation, the employees are allowed to impose a greater risk to bystanders than would normally be expected. In the context of war, the standard are always relaxed. Like those working to avoid the gas explosions, combatants in battle are permitted to put civilians under greater risk than normally would be allowed in civilian life.
THE IDEA OF DOUBLE INTENTION FURTHER DEVELOPED
There are criticisms to be made and developments to be offered in regard to each of these three points.
Regarding Walzer’s first point, it is certainly right that the extent of decrease in civilian risk (and the corresponding extent of increase in combatant risk) required by the idea of double intention varies from case to case, depending on the circumstances. One thing this means, as Walzer notes, is that we cannot say that civilians have a right (corresponding to the combatant’s duty) to no more than “such and such a degree” of risk, such as “a one-in-ten chance of death.” This is a wrong way to look at it because the extent of the risk that may be imposed on civilians is not fixed, but varies. The extent to which civilian risk should be decreased is not unchanging, but relative to the features of the particular situation.
But despite the variability in the answer from case to case regarding how much civilian risk should be decreased, some general claims can be made about how the answer is to be determined in particular cases. In each case, there are different kinds of factors that must be considered, that must be weighed or balanced against each other, to determine the answer. But one general claim Walzer makes about these factors stands in need of correction. He asserts that concern to achieve the military objective may place a fixed upper limit on how far combatants need go to reduce civilian risk. The principle of discrimination cannot require actions that would seriously jeopardize the success of a mission. An alternative means of attempting to achieve a military objective that would make achieving that objective sufficiently unlikely (that would “doom the military venture”) need not be chosen, no matter how effective it is at reducing civilian risk. Such alternatives may simply be ruled out of consideration.
Certainly the likelihood that an alternative would achieve the military objective is a factor to be considered in the choice of an alternative. But Walzer’s claim about this factor misrepresents how it should be considered. Every alternative means of attempting to achieve a military objective carries with it an expected likelihood of achieving the objective (should that alternative be chosen), and this factor must be considered in relation to other factors in choosing an alternative. The problem is that Walzer gives this factor absolute weight in comparison with other factors in the case of alternatives for which the likelihood of achieving the objective is at a certain low point. But this is not how things work. First, there is no nonarbitrary line to be drawn between alternatives where the likelihood of achieving the objective is very low and alternatives where the likelihood is higher than this, as Walzer’s idea would require. There is no magical cut-off point. Second, even for alternatives where the likelihood is very low, it is still possible to imagine that a very great decrease in civilian risk would make that alternative one that could be chosen. This is part of the logic of balancing civilian risk against combatant risk due to the connection between combatant risk and likelihood of success in achieving the objective. This factor of likelihood varies inversely with the factor of combatant risk: a higher combatant risk generally implies a lower likelihood that the objective would be achieved.
This inverse relationship between combatant risk and likelihood of achieving an objective can readily be seen in Walzer’s historical example illustrating his discussion of double intention. He cites an episode from a World War I memoir in which allied soldiers, were throwing grenades into cellars in a French village, seeking to clear it of German troops. Before throwing, they shouted down into the cellars to give any civilians that might be hiding there a chance to come out. Walzer’s point is that while the soldiers decreased the civilian risk by shouting first, they at the same time increased the risk to themselves. But, by increasing the risk to themselves, the soldiers decreased the likelihood that the objective would be achieved. For, if they had become casualties, and their shouting increased the chances they would, the objective would have been less likely to have been achieved. In general, a greater risk to combatants implies a lesser likelihood of achieving the objective.
What factors are relevant, from the perspective of the idea of double intention, in determining whether some alternative means of achieving a military objective is permissible? Well, first, as discussed earlier, that alternative must be a member of the set of lesser-civilian-risk alternatives. But, when looking at alternatives within that set, what other factors should be considered? One is, of course, the extent of the risk that action imposes on civilians, but we have seen two others. In choosing whether an alternative means to a military objective is permissible, combatants should balance or weigh against each other the following factors:
(a) the extent of the risk the alternative poses to civilians;
(b) the extent of the risk the alternative poses to combatants; and
(c) the likelihood that the alternative would achieve the military objective.
A balancing or weighing of these factors in each particular case will yield a value for (a) that will divide members of the set of lesser-civilian-risk alternatives into permissible and the impermissible alternatives.
The basic idea of double intention, as presented by Walzer, is that combatants should choose a means to a military objective that goes at least some distance toward decreasing (a) at the expense of increasing (b). But (c) must be involved in the calculations as well. Walzer seems to handle (c) by saying that if an alternative decreases (c) sufficiently, then that alternative is simply ruled out. But determining the permissibility of an alternative requires considering (c) in relation to (a) and (b) across the full range of values for (c). Even if (c) is small in the case of some alternative, that alternative cannot be ruled out without considering how the value of (a) for that alternative. The idea of double intention is that in choosing an alternative means to a military objective, we trade off some increase in (b) and decrease in (c) for some decrease in (a).
That factor (c) should be fully in play across its full range may be seen from the following example. Imagine again a military command post in a city, and imagine further that the only reliable way to destroy it is to bomb the city. If an alternative such as sending in a commando force were tried, the likelihood of failure would be high. As I have characterized it, Walzer’s position would be that the commando-force alternative may be rejected out of hand because it would place the achievement of the military objective in great jeopardy. But surely, given the large number of civilian lives that would be saved by choosing the commando-force alternative in preference to the bombing alternative, the commando-force alternative should at least be considered, the increase in (b) and the decrease in (c) being weighed against the great decrease in (a). We must have a way of rejecting an alternative that would reliably achieve an objective in favor of one that would not when the latter carries a great benefit in terms of saving civilian lives.
But the question remains, in comparing available alternatives, how much of an increase in (b) and a decrease in (c) should be traded off for how much of a decrease in (a). This leads us to Walzer’s second point, that civilians have a right to an exercise of “due care” on the part of combatants, on analogy with negligence in tort law. The way that combatants are to weigh or balance the various relevant factors (such as (a), (b), and (c) above) is determine if an alternative is permissible or impermissible is to exercise due care toward civilians. This is an appeal to a domestic analogy, modeling appropriate behavior in war on appropriate behavior in a legal system. Exercising due care toward our compatriots in our daily lives is also a matter of comparing alternatives in the light of factors such as those governing choices in war. Walzer mentions that his thinking on this point was aided by a work of Charles Fried, which explicates the idea of tort liability in terms of the notion of a “risk pool.” The idea is that “all persons by virtue of their interactions contribute, as it were, to a common pool of risks which they may impose upon each other, and on which they may draw when pursuing ends of the appropriate degree of seriousness.” It is permissible for me to impose risks on you, by driving my automobile, say, because in your social activities, you impose risks on me. It is a matter of social cooperation within a community.
But this use of the domestic analogy here may be more finely tuned, if we note the distinction between negligent and reckless behavior. There are at least two reasons to think that when combatants impose impermissible risk on civilians, their behavior is more like recklessness than negligence. First, Fried’s idea of a risk pool is not applicable to risks that combatants impose on enemy civilians. Combatants and enemy civilians, being from different societies, do not share a risk pool. This makes it hard to understand how combatants could be negligent, in Fried’s sense, towards enemy civilians. This connects with an earlier point raised in the discussion of Fleury’s criticisms of Walzer. Fleury rejected the idea of double intention in favor of double effect because, he claimed, combatants do not owe foreigners that great a degree of concern. We rejected this criticism on the grounds that the applicability of the principle of discrimination does not depend on the nationality of those put at risk in military actions. The morally relevant fact about civilians is that they are innocent, in some sense of that term, not that they are citizens of this or that nation. Now, given that a risk pool relates only members of the same society, viewing behavior that violates the principle of discrimination as negligent, in Fried’s sense, would incorrectly open up the idea of double intention to Fleury’s sort of criticism.
The second reason that recklessness is better than negligence as an analogue for behavior that violates the principle of discrimination is based on the fact that recklessness is regarded in the law as involving a form of intentionality. The reckless person has awareness while acting of the increased risk his or her behavior imposes on others (whether we regard that awareness as an indirect intention or as mere foresight). In contrast, the negligent are usually unaware of the risk their behavior imposes on others (though they should be aware of it). Combatants who violate the principle of discrimination, like the reckless and unlike the negligent, are generally aware that their actions impose risks on innocent parties. Thus, recklessness is a better domestic analogue than negligence. In fact, behavior that violates the principle of discrimination is very much like the standard example of recklessness, that is, the discharging of a pistol in a crowded room. Such behavior is reckless rather than negligent because the shooter is aware that her action imposes great risks on innocent individuals. This is an awareness that combatants generally have regarding risks their actions impose on civilians.
In the domestic case, behavior is reckless when the risk it imposes on others is unreasonable. So, we may say that the idea of double intention requires that combatants do not impose unreasonable risks on civilians. This leads to Walzer’s third point. The appropriate standard of reasonableness differs in the two spheres. Combatants are given more latitude in the risks they impose on civilians than are their domestic counterparts who, for example, discharge pistols in crowded rooms. The standard for how much civilian risk it is reasonable to impose is always relaxed in war, in comparison with normal domestic situations. Sometimes the standard is relaxed in domestic situations as well, as with efforts to avoid an imminent gas explosion, but they are always relaxed in war. So much is at stake in war that it is sometimes reasonable, not reckless, not a violation of the principle of discrimination, to fire pistols (or more destructive weapons) in environments where civilians are present. But sometimes not, and this is why there is a need for the second intention, an intention to reduce the risk to civilians, to make civilians safer that they might be in the context of the combatants’ pursuit of their military objective.
But how much more relaxed is the standard in war? How is reasonableness of risk-imposition determined in the military context? We can give a partial answer to this question by noting that the reason the standard is relaxed in the military context is that a lot is at stake in war. When the standard is relaxed in the domestic context, it is because there is a great deal at stake, such as the situation of an imminent gas explosion. In war, there is always a great deal as stake. But while the stakes in war are higher, they are not always equally high. The height of the stakes varies from case to case. Analogously, in an unusual domestic situation, the standard is more or less relaxed depending on the height of the stakes. The domestic standard would be more relaxed in the case of efforts to stop an imminent terrorist nuclear attack than it is in the case of efforts to stop an imminent gas explosion. In the military context, some skirmishes are key to victory in a battle and some are not; some battles are key to victory in the war and some are not.
Thus, we must add to our list of factors to be considered when determining reasonableness of civilian risk-imposition in a military context:
(d) the extent to which the achievement of the military objective would further the cause of victory in the war.
Note that (d) is that, unlike (a), (b), and (c), it is independent of all the other factors. One might be tempted to add a fifth factor: the importance of victory in war. But to do so would undermine the idea that what is permissible in jus in bello is permissible to both sides equally, independent of which side is fighting with the just cause, who is the aggressor and who the defender.
The addition of factor (d) does not make the calculations involved in balancing and weighing the factors any more precise-seeming. But we should take heart from the fact that reasonableness in the law regarding recklessness is no more precise-seeming in the abstract, involving as it does the same kind of weighing of factors. Yet the law is able to use the notion of reasonableness reliably nonetheless. Some individuals charged with recklessness are convicted and some are acquitted, and which are which is not an arbitrary matter. In addition, there is case law that tends to make the notion of reasonableness more precise and predictable over time. We may expect that such an informal case law in military traditions, and perhaps a more formal one in military law, would over time make the requirements of double intention more precise as well.
Now, we may try to put all of the pieces together into an account of the idea of double intention, in terms of another revision of the original statement (2):
(2’’’) Civilians must not be intentionally harmed, and positive efforts must be make to reduce the risk to them. Combatants should choose, in respect to each military objective, a means to achieve it that is a member of the set of lesser-civilian-risk alternatives for that objective. They may choose any member of that set which imposes a reasonable amount of risk on civilians, reasonableness being determined by a weighing and balancing, for each alternative, of the extent of the civilian risk it imposes against other relevant factors, such as: the extent of the combatant risk the alternative imposes; the likelihood that it would achieve the military objective; and the extent to which the achievement of the objective would further the cause of victory in the war.
This is not a standard that is to be applied just at the tactical level, in the choices made by individual combatants, such as the allied soldiers in the World War I story who called out before throwing the grenades. It is also to be applied at the operational and strategic levels, where broad alternative means to military objectives are chosen and where rules of engagement are formulated.
DOUBLE INTENTION AND ASYMMETRIC WARFARE
Finally, let me briefly discuss one contemporary application of the idea of double intention. Three phrases are batted about in discussion of war in our era, asymmetric warfare, the revolution in military affairs (RMA), and radical force protection (RFP). War is asymmetric when the military capabilities of the belligerents are greatly mismatched, especially when the mismatch is technological. RMA is a label for the great improvement in military technology brought about in recent decades by the computer chip, making possible the great military advantages that flow from such things as great computing power, accurate and inexpensive remote sensing, and greatly increased guidance systems (allowing for PGM—precision guided munitions). RFP, as its name suggests, is a policy that mandates fighting in a way that will completely avoid combatant casualties. To put it roughly, RFP is made possible by RMA and by the technological disparities between belligerents leading to asymmetric warfare.
Some, like Paul Kahn, have argued that RFP by itself is at odds with just war theory, due to the fact that the apotheosis of RFP, so-called riskless war, such as the NATO military action in Kosovo, contradicts the fundamental idea of jus in bello, what Michael Walzer calls the moral equality of soldiers. The idea is that combatants on one side are justified in trying to kill combatants on the other, only if those other combatants are a threat to their lives. As your killing in self-defense is justified, only if the person killed was posing a threat to your life, so killing in war is justified, only if the combatants killed were a threat to those who kill them. This is not the case in riskless war. NATO combatants killed Serbian combatants when Serbian combatants, because of the asymmetric character of the war, did not pose a threat to the lives of NATO combatants. But this objection to RFP depends on the issue of combatant risk, and the concern of this paper is civilian risk. So, the objection I want to raise is different.
Note, to begin, that RMA is of potentially great benefit to efforts to reduce civilian risk, and it has certainly been seen as such. PGM, for example, by allowing for greater precision in weapons’ delivery, can greatly reduce civilian damage in war. This is so not only because bombs can be more accurately delivered on military targets, but also because the greater accuracy allows achievement of the military objective with a smaller explosive, further reducing collateral damage. In this sense, RMA may be of great moral advantage.
But the RMA is a double-edged sword in its relation to the principle of discrimination. The basic problem is that the characteristics of the technologically advanced weapons themselves may come to be seen to carry the more of the military’s responsibility to adhere to moral standards, in particular, to the principle of discrimination, than they can in fact carry. There are at least three ways in which this may be the case. First, according to the idea of double intention, the characteristic of the weapons we use, by themselves, cannot substitute for the second intention to reduce civilian risk. The weapons we use are part of what determine the optimal alternative means to achieve a military objective, and the idea of double intention requires that we choose an alternative that carries less of a civilian risk than the optimal alternative. Thus, while PGM, for example, help us to lower civilian risk, the idea of double intention requires that we go beyond this to reduce the risk even further.
The second way in which RMA technology may come to be seen to carry a greater share of the moral load than it can in fact carry is when the weapons of RMA are used in a policy of RFP, as they were in Kosovo. An RFP policy was instituted in Kosovo by relying exclusively on air power rather than ground troops to bring about Belgrade’s capitulation on the demand to withdraw its forces from Kosovo. But to achieve this objective by air power alone required attacking not only purely military targets, but also so-called dual-use targets, that is, targets, such as the elements of basic social infrastructure, that have both a military and a civilian use. Destruction of the dual-use targets caused significant harm to civilians, including probably a great number of civilian deaths, due, for example, to the degradation of sanitation facilities and the loss of electrical power in hospitals. In fact, the idea of double intention required the choice of an alternative means to the objective of Belgrade’s capitulation than a policy that avoided military casualties at the cost of the civilian harm from the destruction of dual-use targets.
The third way in which the advanced weapons could not carry the full moral load arises from the fact that, without adequate intelligence, the most precise weapon in the world cannot be used so to limit civilian damage. But better intelligence requires having one’s people on the ground, and hence putting combatants at greater risk. As a result, efforts to gather better intelligence are at cross purposes with a policy of RFP. The idea of double intention requires gathering that better intelligence, and hence is itself at odds with a policy of RFP.
In all three of these ways, it can be seen that in the case of asymmetric warfare, when a nation’s advanced military technology is used against a nation that lacks it, the technology itself cannot do all the moral work, thereby freeing the nation to pursue a policy of RFP. In all three of these ways it is clear that advanced technology does not a repeal a basic moral truth of warfare, that (a) is in inverse relation to (b), that reducing civilian risk requires increasing combatant risk. Nations with advanced military technologies still are required by the principle of discrimination to put their combatants at increased risk to reduce risk to civilians. The principle of discrimination, which includes the idea of double intention, morally precludes the pursuit of a policy of RFP.
 Statement (2) represents only a partial account of the doctrine of double effect. For a fuller account, see Michael Walzer, Just and Unjust Wars (New York: Basic Books, 1977), p. 153.
 I owe this phrasing to Judith Litchenberg, “War, Innocence, and the Doctrine of Double Effect,” Philosophical Studies 74 (1994), pp. 355.
 In one of his public explanations for bombing Hiroshima, President Truman pointed out that the city contained important military facilities.
 Walzer, Just and Unjust Wars, pp. 151-159.
 Walzer, Just and Unjust Wars, p. 156.
 Walzer, Just and Unjust Wars, p. 156.
 Colonel J.G. Fluery, “Jus in Bello and Military Necessity,” Advanced Military Studies Course 1, (Department of National Defense (Canada), 1998), http://198.231.12/papers/amsc1/o12.html, accessed 12/16/03, p. 7.
 Fluery, “Jus in Bello and Military Necessity,” p. 6.
 Fluery, “Jus in Bello and Military Necessity,” p. 5. Fleury cites in this regard Sheldon Cohen’s Arms and Judgment: Law, Morality, and the Conduct of War in the Twentieth Century (Boulder, CO: Westview Press, 1989), p. 127.
 Walzer, Just and Unjust Wars, p. 158.
 Note that this formula does not cover the case in which the OA is the alternative that poses the least risk to civilians, which is certainly possible. In that case, of course the OA should be the alternative chosen.
 Walzer, Just and Unjust Wars, p. 155.
 Walzer, Just and Unjust Wars, p. 156.
 Walzer, Just and Unjust Wars, p. 156.
 Walzer, Just and Unjust Wars, p. 156.
 Walzer, Just and Unjust Wars, p. 156.
 Walzer, Just and Unjust Wars, p. 156.
 Walzer, Just and Unjust Wars, p. 157
 Walzer, Just and Unjust Wars, p 152.
 The role played in the choice of alternative by factor (3) shows that the answer to how much the risk to civilians should be decreased cannot be given merely in terms of a comparison between the value of a combatant’s life and the value of a civilian’s life.
 I ignore the principle of proportionality for the same reason I did in the earlier use of this example.
 Charles Fried, “Imposing Risks on Others,” chapter 12, in Charles Fried, An Anatomy of Values (Cambridge, MA: Harvard University Press, 1970), pp. 183-206, quotation from p. 189.
 One implication of this understanding of double intention is that the question of how much civilian risk should be decreased is relative not directly to the military objective in question, but to each alternative means of achieving that objective.
 See Paul Kahn, “The Paradox of riskless Warfare,” Philosophy and Public Policy Quarterly 22, no. 3 (Summer 2002), pp. 2-8. For a discussion of the moral equality of soldiers, see Walzer, Just and Unjust Wars, pp. 34-41.