Public reason in just war theory

Declarations of war are an example of how an action's justifiability depends on the way it is actively justified. This example fits within the ‘public reason’ tradition about ethical burdens of democratic argument. The purpose of this paper is not to defend a public reason account of just war theory; instead, it is to use the underlying intuitions that guide public reason to make sense of declarations of war. This section describes the difference between public and private reason-based accounts of international ethics, and ties the public reason model to questions about the role of international institutions. The latter argument is particularly important. For many modern commentators, public reason is difficult without an institutional backdrop that enables fair, impartial decisions to be reached. The next several sections focus on this specific problem, showing that declarations of war are rooted in a weak interpretation of fairness that meets the demands of public reason.

Many just war theorists often set the discussion of just war against realism. Realism traditionally denies that moral claims should figure prominently in discussions about war. Perhaps the most famous statement of this position is by Hans Morgenthau (Reference Morgenthau2006, chap. 1), whose textbook on international politics begins by explicitly stating that political leaders should not consider ethical questions.Footnote ⁵ Just war theorists and many moral philosophers strongly disagree, arguing that political elites have obligations to protect their own citizens as well as innocent victims of genocide and ethnic cleansing abroad, and to maintain the rules of war to ensure that soldiers’ rights are protected when they surrender.Footnote ⁶

The arguments just war theorists use are usually rooted in ‘private reason’. By this, I mean that just war theorists have sought to marshal intuitions that any reasonable individual has – about limits to the right of self-defense or the limits of necessity for example – to prescribe moral rules for considering war. The argument of this paper is not that the private reason model is misguided. Instead, when considering some specific elements of the decision to go to war, a focus on ‘public reason’ is better able to capture our intuitions about political justification. Focusing on public reason requires focusing on the interplay between how an action is justified and its justifiability.

A simple example of the difference between public reason and private reason is failing a student in class. A focus on private reason means focusing on the grounds that any person might reasonably have for failing a student: the student does not attend, does not turn in a final, cheats, or throws apples at other students. A focus on public reason, in contrast, draws attention to a different set of concerns; the way the failure is explained conditions the justifiability of failing the student. Can one fail a student without providing an explanation? Does one need to justify the failure with regard to the student's interests (training them to be more successful later in life), the university's interests (fairness toward other students), or larger social interests (training them to be members of a society)? If one justifies failing a student in light of the student's interests, does one need her consent to do so? In sum, a public reason model does not supplant the private reason account but complements it by pointing to connections between the process of justification and the justification of actions, highlighting broader issues.

Within the context of war, public reason raises many important questions. Is a war just if it has not been appropriately justified? Can one ‘rightly intend’ a humanitarian war on victims’ behalf if one does not have the victims’ consent? Can one fight a war to defend a community if the community does not support the war? Do agreements between belligerents on how to treat captured soldiers obtain moral force? Each of these questions relate to public reason because there is an interplay between the process of political justification – the explanations for an action or public agreements – and the justifiability of an action. If public reason has traction on ethical questions related to war, then declarations of war are a critical test case. The central question I ask is whether the justifiability of a war depends how the war has been politically justified. Here, the connections between declarations of war and public reason are at their starkest.

One important question about the value of using public reason-based concepts in international ethics is the role of institutions in international life. In political theory, questions about public reason and political justification usually emerge within the dense institutional context of the state. Two leading thinkers in the public reason tradition – John Rawls (Reference Rawls1997) and Jürgen Habermas (Reference Habermas1990) – ask how public (or communicative) reason affects the kinds of political or ethical arguments one can give within the public sphere. These ideas, especially for Rawls (Reference Rawls2001), are often limited to reason within communities with robust national institutions. Introducing public reason into debates about just war theory raises questions about international institutions. How does public reason extend beyond the state to a world without the firm institutions that characterize good government?

One solution to extending the insights of public reason is to concentrate on formal IOs that may serve as an institutional backdrop for public reason. One resource within just war theory may point toward these institutions. A traditional element of just war theory is that a legitimate authority must declare war.Footnote ⁷ For traditional just war theorists, this implies that a state's sovereign must decide to go to war, effectively barring civil war and private violence. Modern commentators and much of the public discussion on international law, however, emphasizes a different form of legitimacy. Whereas realism and just war theory may describe power politics before 1945, there has been a steady shift toward the creation of new centers of legitimate authority in IOs, such as the United Nations, the International Court of Justice, and regional security alliances. Today, multilateral institutions are the seat of significant legal authority, such as the Security Council in the case of Iraq, and the NATO in the case of Kosovo (Schroeder Reference Schroeder2004, 185–6; Dolan Reference Dolan2005, 86–101; van der Linden Reference van der Linden2007, 68). While neither organization is perfectly impartial, they are ‘legitimate’ repositories of the sovereign right to make war.

One powerful moral argument for the growth of IOs is that private reason may be inappropriate when considering issues of war and peace. Within a society, private reason is often insufficient to ensure a just decision unless clear procedures are followed to ensure that private reason is not tainted with private interest. This is captured in the maxim that one cannot be party, judge, jury, and executioner in one's own case. Without an independent agency to render impartial decisions, there is no fairness; and, without fairness, there is no justice (e.g., Calhoun Reference Calhoun2002). Without the expectation for impartiality, an international society capable of employing the language of justice as a brake on power politics is impossible.Footnote ⁸ Key demands for increasing representation and limiting veto power in the UN Security Council, as well as debates about ensuring that decisions reached by the International Court of Justice are impartial, reflect these underlying moral concerns.

The core problem with a focus on international institutions is that international institutions have not obtained the status of background institutions that enable fairness. States have not effectively transferred decision-making power, and even if they could, this would not ensure that decisions are fair and impartial.Footnote ⁹ This does not mean that IOs play no role in enhancing support for interventions. The UN, in particular, has the ability to marshal support and legitimacy for war (Hurd Reference Hurd2007; Thompson Reference Thompson2009), and on many non-security issues, international courts are able to make reasonably impartial claims. Yet, international institutions have not yet solved the problem of partiality. Therefore, the challenge – if public reason is to have a role in international politics – is to craft an argument for political justification without background institutions in which political justification operates impartially.

Strong and weak interpretations of justice

The challenge for just war theory, if it wants to assert the importance of impartiality, is to do so without relying on institutional solutions that require decision making by independent third parties. How do we develop a fair and impartial system of justice in international politics without fair and independent courts, like in domestic politics? The argument of this section is two-fold. First, we can distinguish between a strong and a weak interpretation of the principle of impartiality. The strong interpretation maintains that justice requires impartial (independent) decision making. The weak interpretation, advanced here, posits that justice, in international politics, implies that states are not permitted to reserve the right of moral judgment (as opposed to legal judgment) in their own case. Second, the weak norm that states should declare war meets the demands of a weak interpretation of justice as impartiality.

The sense of justice developed in this section is procedural. A traditional injustice is taking the right to judge in one's own case. This principle is a cornerstone of society. Letting an individual judge her own case gives her the power to make arbitrary and self-interested decisions. While I do not develop the argument, this intuition concerning fairness is historically at the heart of international law and is a crucial element of many accounts of moral and political justification (Franck Reference Franck1995; Cheng Reference Cheng2006, 279–289).

Advocates of world government often advance the importance of impartiality in its strongest form. The strong interpretation holds that actors should never make enforceable decisions in disputes where their rights or well-being are concerned.Footnote ¹⁰ Within the state, this intuitive interpretation finds solid footing; judges and legislators are supposed to recuse themselves if their own interests are at stake because judgments should be rendered by disinterested third parties. In international politics, this strong interpretation finds traction in the claims discussed in the previous section; an international body is necessary to ensure that the behavior of strong states is not arbitrary and capricious. If the strong interpretation were the only reasonable interpretation of the principle of justice, then just war theory is in trouble for the reasons discussed earlier. A strong interpretation of justice requires strong norms capable of overriding state interests.

In contrast, a weak interpretation of justice posits that communities should not claim the sole right to make a judgment in a dispute where its rights are at issue. There are multiple audiences who have a right to judge of a state's foreign policy: its public, the target state and its public, and the rest of the international community (including third party states, IOs, and other transnational actors). A state cannot rightly refuse to enable these audiences to render a judgment on the moral defensibility of a war; it is obligated to make a case that is public: sincere, inspectable, and not subject to change based on political or legal circumstance. This is weak because, unlike the strong interpretation, it does not require that these moral judgments determine the course of policy. However, it does require states to publicly explain their reasons for war; to make a formal case.

In the next section, I argue that states are morally bound to meet the weak interpretation of justice. Before doing so, I want to suggest that declarations of war meet certain requirements of the weak interpretation with reference to two cases.

Declaring war by giving the moral reasons meets this weak interpretation. A reasoned and conditional statement of intent makes a case to the target, its public and the international community. It provides a written record of a state's reasons for fighting and enshrines a clear moral case for causus belli. The refusal to declare war, by contrast, is a refusal to grant others the right to sit as judges over a state's decision. This refusal is akin to the barring the accused from confronting their accusers in a courtroom, removing the right of self-defense through argument. Cicero (1913, 37) explains, ‘there are two ways of settling a dispute; first, by discussion; second, by physical force; and since the former is characteristic of man, the latter of the brute, we must resort to force only in case we may not avail ourselves of discussion’.

There are two paradigmatic cases where this refusal creates problems. Before the War of 1812, the British imposed a blockade on continental Europe. The Orders in Council (1807) meant that US ships could not sail to Europe, drastically reducing American trade. Even worse for public opinion, British ships frequently impressed American sailors into often abusive British maritime service. At the same time, the Madison administration wrongly thought the British were to blame for a revivalist Native American movement. On June 1, Madison issued his War Message to Congress, claiming that war was necessary for redress these grievances, and on June 18, Congress issued a declaration of war.

If the Madison administration made the case that war was justified because of impressments to the British and neutrals, then it would have met the weak interpretation of justice. Instead, the Madison administration complained to the British of abuse, but never issued a threat of war. In fact, the declaration of war came as a surprise in the United Kingdom. The government ordered the rescinding of the Orders in Council on June 23, more than a month before it learned that the United States had declared war. Madison's war message was secret, the justification for war was weak, and the British thought they had responded to the just American demands. On August 3, when news reached London of the declaration of war, the British were toasting to the end of the Orders in Council and a new period of peace between the US and Britain (Latimer Reference Latimer2007, 34). When news of British concessions reached Washington, Madison did not end the war; the confused case for war, which cited unreasonable concerns alongside reasonable trade and naval concerns, left Madison free to soldier on (Hickey Reference Hickey1989, 42–3). Madison's refusal to make a clear case for war to the British – both moral grievances and their seriousness – is a refusal to explain to the British government that it is on trial – that war is the result of refusal to address grievances – and a refusal to enable the British to provide compensation.Footnote ¹¹

The decision by the United States to attack Iraq bears a resemblance to 1812. The US Authorization to Use Military Force (AUMF) includes three pages of demands: weapons of mass destruction (WMD), defiance of international law, repression of its population, the attempted assassination of George H.W. Bush, harboring terrorism, and, to finish, a demand for regime change. Like Madison's war message, the United States did not limit itself to establishing legitimate, redressable moral grievances; the AUMF did not provide a clear roadmap to avoid war. The most significant difference between 1812 and 2003 relates to Hussein's more prescient reaction to US demands. Whereas the British thought that making concessions on trade would avoid war, Hussein thought that that making concessions on weapons inspections would not stop the US-led invasion (Iraq Survey Group 2004). He was right. Tony Blair recently claimed that even if Iraq had no WMD, he would have supported the invasion of Iraq because of widespread hostile intent (Burns and Cowell Reference Burns and Cowell2010). Blair's admission confirms many judgments of US policymakers: demands for WMD made publicly were only one of the grievances cited against the Hussein regime and war would occur if other conditions, especially regime change, were not met.

On March 17, in a televised speech Bush publicly changed his demands: ‘Saddam Hussein and his sons must leave Iraq within 48 hours. Their refusal to do so will result in military conflict commenced at a time of our choosing’. Even this demand was insincere. In a meeting later that evening with the House and Senate leaders, he remarked, ‘If Hussein leaves, we'll go in anyway’ (Woodward Reference Woodward2004, 369). The result of this free-for-all process was a situation in which the moral demands that the United States made on Iraq resembled a shell game. The international community could not judge the moral grievances that the United States held regarding Iraq because the reasons were shifty, informal, and undeclared. Hussein rightly understood that there was no path to meet legitimate US demands and avoid war.

The obligation to declare war

The weak interpretation of justice posits that others have a right to judge our conduct. Alone, however, this does not mean that decisions to go to war that do not meet the weak interpretation of justice are necessarily unjust. This section makes the case that states have an obligation to meet the weak interpretation of justice. To do so, I first make a distinction between offensive and defensive wars to explain how justification works in two contexts. Then, I posit three reasons to declare war to meet the weak sense of justice: the target community and bystanders’ rights, and a discursive argument.

Some actions can be just even if they are not publicly justified in a fair and impartial way. Giving money to charity, skipping lunch with friends, going to a party, and other everyday life events are justifiable even if they are not justified in advance. Similarly, there are cases where actions are justified implicitly or after the fact. I can fail a student, for example, without writing an unasked for note explaining the cause of the failure, or I can fire an employee for failing to show up without sending her a written letter beforehand. The failure to explain these actions prospectively does not mean the actions themselves are unjust. Why do undeclared wars themselves become unjust when they are not justified prospectively?

To make sense of different ways in which justifications are given, I propose a distinction between defensive and offensive wars. International law and moral philosophy tends to distinguish between acts of aggression and reactions to aggression (May Reference May2008). At issue in the aggression-non-aggression distinction is who is criminally responsible for a war, and whether certain classes of war, such as preventive wars or preemptive strikes, count as aggression. In contrast, the offensive–defensive distinction I propose is intended to make sense of our intuitions about the need to publicly justify war. A state fights an offensive war when it makes the first step over the brink to war (e.g., starts shooting), and a state fights a defensive war when it reacts to another state's offensive moves. The purpose of the offensive–defense distinction is not to supplant the aggression-non-aggression distinction, but to develop an account more useful for thinking about public justification.

War belongs to a special set of actions whose moral permissiveness depends on advance justification. By way of contrast, some actions that benefit others while doing harm to no one rarely, if ever, require moral justification. When I give money to charity, I can do so anonymously and without explanation. Other actions harm individuals, but are usually implicitly justified. When I fail a student, I do not need to provide written reasons because implicitly the student knows they received a low score. Defensive wars, where a community's borders have been crossed and its military is defending itself, like France in 1914 or American sailors at Pearl Harbor, do not need to justify resisting because self-defense against naked aggression is a clear moral rule which, like a syllabus, provides a, implicit justification for action. Defensive wars, therefore, do not need to be actively justified.Footnote ¹²

Offensive wars, where a community crosses its borders to do harm to another community, belong to a set of actions that are only justifiable if they are publicly justified. Offensive wars include humanitarian interventions, counter-interventions, pre-emptive and preventive wars, along with wars of expansion or conquest. Many of these wars are illegal because the United Nations Security Council has not authorized them, and therefore they are not legally justifiable. This does not mean, however, that they are unjustifiable on other grounds. Nicholas Wheeler (Reference Wheeler2000), for example, convincingly shows that Tanzania's intervention in Uganda, Vietnam's intervention in Cambodia, and India's intervention in East Pakistan, were illegal (because they did not have Security Council authorization), but they are morally justifiable because they ended intense humanitarian suffering. In these cases, however, states still need to justify the wars they fight. When the United States intervenes in Iraq, Afghanistan, Kuwait, and Kosovo, assuredly it will contravene the rights and damage the well-being of innocents in other communities. There are three sets of reasons that, in this class of cases, giving reasons publicly and thereby allowing other communities to sit as judges over the case for war beforehand is crucial to the ability of a war to be considered just.

The first set of reasons concerns the right of the target community. In a legal context, the accused usually has the right to appear at trial along with a right to hear the evidence and confront the accuser. When a murderer is on trial, even if a decision to convict is factually right, because the murderer killed someone, our intuitions suggest that the decision to convict is wrong if the accused does not have access to or is not allowed to dispute evidence, or was tried in absentia. The conviction is unjust because the procedure used to obtain it is unjust. These same intuitions that guide legal cases, connecting right procedures to just decisions, also operate in the international context. Even if an aggressor believes it has just cause, the target's community has the right to hear and dispute the aggressor's claims. Otherwise, force replaces argument. This argument is more fully developed in the next section related to responsibility, so I do not concentrate on it here.

The process of public justification more substantially affects the interests of third parties. Bystanders – neutral third-party states – deserve a moral explanation for war. There are two sets of reasons. The first set of reasons related to harm that may flow across borders from an act of intervention. Every conflict, from humanitarian crises to great power war, tends to do harm to other communities. The failure of the United States intervention in Somalia, for example, undermined domestic support for intervening in future human rights conflicts, influencing the UN's decision to avoid intervention in Rwanda (Barnett Reference Barnett2002, 35–48). The 2003 Iraq War had ripples effects as oil prices dramatically rose after the intervention and created opportunities for political destabilization in neighboring states. The United States intervention in Afghanistan led to increased violence in areas of Pakistan where militants fled.

To begin a war in such a way that the worth of these affected communities is not respected is to fight an unjust war. One plausible reason is that when one does harm to bystanders, one needs their consent. Recent work on humanitarian intervention notes that if one intervenes on a victim's behalf, it should be done with their consent (Tesón Reference Tesón1997; Pattison Reference Pattison2007). Otherwise, one is behaving paternalistically, replacing one's own judgments about the costs and benefits of war for the victims who have to bear those costs and benefits. We do not, however, intuitively think that harm to bystanders always requires their consent, especially if the harm inflicted on bystanders is substantially lower than the benefits that accrue to the victims’ community. One reason for the lack of a need for consent is that intervening in a way that does harm to third parties is different in kind because it does not substitute one's own judgments about the costs and benefits of the war for those who the war is intended to help. Anti-paternalism is not sufficient to ground the obligation to declare in a way that requires consent.

Even though bystanders’ consent is not obligatory, engaging in discussions with bystanders (when possible) is important. Respecting other communities’ moral worth means affording them opportunities to make judgments concerning actions that affect them, particularly when they are affected bystanders. Within a society, we show respect for others by reasoning with them, thereby engaging their distinctly human capacities, and not relying on force alone to gain their compliance (Larmore Reference Larmore1996, 136–41). When politicians make decisions concerning their constituents, for example, they show moral respect by explaining the reasons for their actions. Or, when a parent explains a decision to a nearly grown child, decisions are often explained to supply reasons that can be evaluated and disputed. By analogy, showing respect for other communities requires giving one's reasons, enabling other communities to express their views and render moral judgments. A state that claims to have cause for war shows moral respect for other communities by explaining why the war is morally necessary and the reasoning through which it is justified.

A second set of reasons related to bystanders’ rights is that a declaration of war may directly minimize harm. Undeclared wars, or wars in which moral reasoning is insincere, tend to harm the international community as a whole. Larry May (Reference May2005), by way of making an argument in a different context, points out that random violence (e.g., a crime of opportunity) affects the public interest because the violence proceeds against a person qua person. A random mugging, for example, harms the public because any member of the public could have happened to be in the same place at the same time, providing the same opportunity to be a victim. By contrast, private violence, where a person is engaged in violence against a person because of some particular reason, is not a threat to the community because it does not target a person because they are a person, but because of some special feature (e.g., a crime of passion). May's concerns about random violence make sense of our intuitions about wars in which moral reasoning is obviously insincere and self-serving. Nazi aggression seemed a threat to the international community, in part, because Poland, France, and others were not targets for some specific reason (because of a long-term grudge, for example), but because they were independent communities that stood in the way of Nazi conquest. They were targeted as independent communities rather than, say, breakers of a specific covenant, making the threat to them of the public interest for other independent communities. Publicly declaring the reasons for war turns public violence into private violence. By making a case that outlines specific grievances against a community, a state shows that there is a specific relationship of grievance that justifies the use of force, respecting bystanders’ rights not to be harmed.

Moreover, refusing to justify war likely results in partial understandings of the justifiability of war. When individuals judge their own cases, even when they sincerely do their best, they are likely to obtain partial judgments. One reason Aristotle gives for a mixed society is that different groups (the many, few, or one) are likely each to advance claims regarding just distributions that reflect their political and economic positions. Allowing one group to judge runs the risk of replacing a full sense of justice (all claims considered) with a partial interpretation (one group's claims dominate). Judgments concerning justice are fullest when they take into account the perspectives of different individuals within a society (Aristotle Politics, 1280a9–1281a9).

Aristotle's worry that groups will advance partial claims concerning justice seems appropriate to describing concerns over justice in the international system. Left to its own devices, any single community is likely to have a partial understanding of justice in its own case. Issues of war and peace are not without emotion, and motivated and unmotivated biases often lead communities to make mistakes in their moral reasoning. Declarations of war partially address this concern. Requiring states to make a case for just cause provides a mechanism for others to challenge the case. Refusing to make oneself sole judge means being willing to have one's claims challenged. Insofar as discussion, argument, and public opinion provide potential checks on partial portrayals of justice, it reduces the risks of partial justice.Footnote ¹³

These three reasons to declare war explain part of our skepticism of the Madison and Bush administrations’ conduct. Both administrations acted as if they were the exclusive judge of the rightness of their claims; they never gave their targets open and public trials on the sincere causes for war. Neither showed moral respect for their target or the international community by publicly making an open case. Therefore, the judgment in both cases seemed self-serving and partial. The Bush administration and the Congress are not morally liable for violating international law. No member of the Security Council was impartial; France and Russia, who led the opposition to Security Council authorization, had substantial stakes in the outcome of the debate. And, if only Russia (and not France) stood in the way of action in Iraq, I doubt the illegality of the intervention would have been as important of an issue for most commentators on the war. Instead, they were wrong not to make a publicly reasoned and sincere formal case.

Debating last resort

Why should states pursue war only as a last resort? The importance of the last resort standard has been the subject of much recent debate. On the one hand, requiring states to pursue war as a last resort seems laudable because it means that non-violent means have been tried before violent ones. If war is the first resort instead of the last, then states have not tested opportunities for peace. On the other hand, requiring war as a last resort can have substantial costs. Claiming that every avenue must be tried, including those that are unlikely to work, permits aggressors time to finish murdering innocents.

One way to try to strike a balance between the costs of delay and the need to try peace is to emphasize the costs of delay over the importance of peace. Jean Bethke Elshtain (Reference Elshtain2003, 60), for example, defines a last resort as ‘a resort to armed force taken after deliberation rather than as an immediate reaction’. Michael Walzer (Reference Walzer2004, 88) suggests that the reason for the rule is to ensure deliberation: sending soldiers into harm's way incurs so many costs that ‘political leaders must cross this threshold only with great reluctance and trepidation’. This version of last resorts captures an important intuition – one should recognize the costs of war – but inaccurately links it to claims about last resorts. If a political leader crosses the threshold to war without exhausting all peaceful and diplomatic means, she can still do so with ‘great reluctance and trepidation’. John F. Kennedy, for example, may have ordered the bombing of Cuba in 1962 while shaking at the consequences of his decision (meeting the reason behind Walzer's standard), but still be liable for not attempting a quarantine or another round of diplomacy.

A second way to strike the balance is to emphasize the need to search for peace. James Turner Johnson (Reference Johnson2004, 38) argues that a last resort has been reached ‘when no other means will achieve the justified ends sought’. By this, Johnson does not mean that every peaceful alternative needs to be tried, including those likely to be ineffectual. Instead, every peaceful alternative likely to work must be pursued. Johnson's definition comes closer to an adequate balancing act but still suffers from critical ambiguities. How likely must the success of alternative courses of action be to count as plausible? Does it matter if there are no costs of delay? At what point do gathering threats require immediate action, rather than allowing for another round of diplomacy that has a low chance of succeeding? At its root, this balancing act is ambiguous (Crawford Reference Crawford2003; Lango Reference Lango2005, 260–3).

Private reason and responsibility

A more nuanced approach to understanding the last resort standard concentrates on the responsibility for war. When one fights an undeclared war – even with just cause – one is responsible for the war and its inevitable side-effects. The basic intuition is that one is responsible for an action if one has control over it and foreknowledge about its effects. When a state declares war, it is providing the target state with foreknowledge that not addressing legitimate grievances will result in war, as well as providing ways the target state can address those grievances, thereby giving the target partial control over the outbreak of war. When states do not declare war – when they launch a surprise attack or do not provide a reasonable list of grievances that can be addressed – then the target state is not responsible for the war because it has no control. In other words, failing to declare means war is not a ‘last resort’ because one has not given the threat of war a real chance to succeed.

One traditional approach to identifying moral responsibility is to focus on control over actions (Dennett Reference Dennett1984; Fischer and Ravizza Reference Fischer and Ravizza1998). If one does not control an action – if there is no choice – then one is not responsible because one could not have acted otherwise. The ‘control’ principle makes intuitive sense in handling problems of war and peace. A state is not responsible for a war if someone invades them; they are responsible for a war if they invade others. Germany, not Poland, is responsible for the invasion of Poland in 1939 because Germany, not Poland, made the decision.

One difficulty with the control principle relates to problems of moral luck. In some circumstances, the moral evaluation of a decision depends on circumstances that are not under agents’ control (Nagel Reference Nagel1979, chap. 3; Williams Reference Williams1981). Attempted murder, for example, is treated differently than successful murder, even though the difference between success and failure is not controlled by the agent. This is a dilemma because we intuitively believe that we should treat attempted and successful murder differently, but a would-be murderer has no control over whether the murder is successful, violating the control principle.

The specific type of moral luck related to declarations is ‘circumstantial luck’ (see Nagel Reference Nagel1979, chap. 3). In cases of circumstantial luck, one is held responsible for actions taken in circumstances that are not of one's choosing, and those circumstances help determine the moral evaluation of an action. The moral evaluation of a soldier following orders, for example, depends on the nature of the orders that are beyond her control. Circumstantial luck complicates attribution of responsibility. For example, the question ‘Who was responsible for the war in Afghanistan?’ is a complicated question. According to the control principle, the Taliban are not candidates for responsibility: they not choose a US intervention in Afghanistan and likely would have preferred that the United States not attack them. The United States, in contrast, had control over the decision to intervene and therefore was responsible for the war. Moral luck, however, influences the way we think about control. The United States had control over the decision to fight, but no control over the circumstances that prompted the war. Circumstances beyond the Bush administration's control – the September 11th terrorist attacks and the Taliban's refusal to cooperate in finding terrorists – determine how we treat the United States even though it has no control over those circumstances.

By itself, this poses no special difficulties for the way traditional just war theory frames jus ad bellum questions. The United States was responsible for the war (because the war was the product of US decision making) and the war was not blameworthy due to cruel ‘luck’ (because of just cause due to Taliban actions). Therefore, the United States simply fought a just war.

Does this ‘private reason’ account adequately capture our understanding of responsibility for the US intervention in Afghanistan? The United States – if it is responsible for the war – is also responsible for the collateral damage and civilian deaths that we always know are the products of war. In one powerful way, the United States did not feel itself responsible for the war or its effects. The September 11 attacks – not US decision making – created a circumstance in which justice cried out for intervention. By refusing to address US demands, the Taliban created a world in which the only route to just cause addressed was through violence. I return to this case after discussing public reason and responsibility.

Public reasons and shifting responsibilities

Was the United States responsible for the war in Afghanistan? A private reason approach – which focuses on the decisions made by states and the circumstances in which states find themselves – assigns responsibility to the United States. This section contends that focusing on public reason reaches a different and more reasonable conclusion. Public demands can shift responsibility shifting control to a different agent. This section concentrates on the example of the war in Afghanistan, showing that the Taliban's refusal to address US demands shifts responsibility for the war away from the United States and toward the Taliban, and that a declaration of war is necessary for this shift in responsibility.

The key argument relates to the nature of luck. Luck defines conditions beyond the actor's control. There is an important moral difference, however, between circumstances that are outside of a person's control and those outside of every person's control. One example of the latter type – canonical to questions about moral luck – relates to control over the future. The moral valence assigned to some actions depends on the success of those actions, and success can depend on random chance that is beyond everyone's present control.Footnote ¹⁴ The difference between attempted and successful murder, for example, is treated differently in criminal law, although whether the bullet is deflected may depend on chance. Most real-world military situations do not depend on random chance, but on others’ actions. The circumstances in which the United States found itself after September 11 did not arise from uncontrollable natural factors, but from deliberate decisions taken by al-Qaeda and the Taliban. The phrase moral luck, in other words, does not capture the extent to which others’ intentional decisions created the circumstances that led to the war. The September 11 attacks and the Taliban's refusal to hand over those responsible was not the product of a natural lottery in which random chance determined outcomes, but an unnatural lottery in which the Taliban and al Qaeda put the United States in a circumstance that required action (see Card Reference Card1996).

The question related to public reason is the balance of responsibility between the actor who controls a decision (the United States) and the actor who is responsible for the conditions that lead to the decision (the Taliban). Public reason provides tools to shift responsibility away from the agent who commits an action, toward the agent who is responsible for the circumstances surrounding an action. In other words, control over outcomes can be handed off through dialogue. To understand how responsibility shifts, one needs to draw a distinction between two types of agents: a control bearer (who makes the decision over an action) and a circumstance creator (who creates the situation in which a normally illegitimate action is morally appropriate). In Afghanistan, for example, the United States is the control bearer and the Taliban is the circumstance creator.Footnote ¹⁵

How does responsibility shift to the circumstance creator? If a circumstance creator knows that the consequence of creating a circumstance is some effect, then the circumstance creator shares responsibility for creating that effect. In other words, if I know that I put you in a situation where you will do x, I am partly responsible for x. For a circumstance creator to have control over an action, two conditions need to apply. First, the circumstance creator needs foreknowledge about the effects of continuing to create the circumstance. Foreknowledge is crucial because, without foreknowledge, one cannot realize and thereby control the likely effects of an action (Dennett Reference Dennett1984, 54–5). For the Taliban to be responsible for the war, for example, they need to have foreknowledge that refusing to cooperate with the United States will have the effect of causing a war. Second, to be responsible, the circumstance creator needs to know the courses of action that will avoid the harmful outcome. If the circumstance creator does not know how to avoid the harmful outcome, then that outcome is beyond their control. In other words, it is not enough for the Taliban to know that it must do something; they must believe that something within their area of control can be done that will reduce the risk of invasion. Any formal or informal device that connects the creation of a circumstance to the action can partly shift responsibility from the control bearer to the circumstance bearer.

In the context of war, to say that a war is a last resort is akin to saying, ‘we are not entirely to blame for the war’. By providing non-violent means to avoid war, one becomes less responsible for the foreseeable loss of innocent life likely occur in the war's commission. Circumstances over which we have no control have provided just cause, and the refusal of others to address those circumstances make the action justifiable. War, in other words, is a last resort in the sense that one provides others non-violent options to address just cause; if those non-violence options are ignored, then war becomes the last resort to achieve just cause. Declarations of war enable a shift in responsibility by requiring states to make clear and explicit demands on each other. They provide foreknowledge that war may result because of specific grievances, and they provide specific avenues to address those problems.

George W. Bush provided this type of declaration of war on the Taliban in his September 20th speech to Congress. Although not a ‘formal’ written declaration, it did make the demands clear, specific, and promised that war would not occur if the demands were met:

Bush specified that refusing the demands would lead to war: ‘They will hand over the terrorists, or they will share in their fate’. By making these clear demands, Bush shifted responsibility to the Taliban by giving them control over the circumstances that justified the war.

Conversely, if a state does not declare war, then it is fully responsible for the war even if it has just cause. A declaration of war distributes control over the war by providing clear courses of action for others to make a difference; refusal to threaten war by launching a surprise attack prevents states from having the kind of foreknowledge that enables control, and not specifying clear routes to avoiding war prevents states from assuming responsibility.

The 2003 Iraq War draws the appropriate contrast. Before the war, George W. Bush's public rhetoric tried to shift responsibility in the same way he had with the Taliban. In his 2003 State of the Union address, he tried to show he had a forced hand: ‘We seek peace. We strive for peace. And sometimes peace must be defended. A future lived at the mercy of terrible threats is no peace at all. If war is forced upon us, we will fight in a just cause and by just means…’Footnote ¹⁷ Here, Bush is making the responsibility argument discussed above. Saddam Hussein, not the United States, is responsible for the war because Hussein created a circumstance that made the war just.

Unlike Afghanistan, however, the shiftiness of the Bush administration's case prevented Hussein from having effective control over the war. In contrast to the clarity of the demands on al Qaeda, Bush consistently changed his public demands related to Iraq. As noted earlier, in public and private statements, the Bush administration continually changed the public justification for the war, changing the focus from WMD to human rights to regime change to terrorism in a slipshod fashion, and in the final days, demanded that Hussein must leave Iraq. And, even if Hussein met every demand, Bush indicated in a secret meeting with the congressional leadership that he would still attack Iraq. Therefore, the Bush administration is responsible for the war because it was not a last resort; Hussein had no control.

Surprise attacks

One might worry that requiring states to declare war puts them at needless risk. Publicly justifying a war in advance might risk telegraphing our punches, alerting our enemies, and puts more lives at risk. The Japanese attacks on Pearl Harbor or Port Arthur would fail if there was advance warning of an attack. Concerning the latter, two historians note that ‘surprise at Port Arthur was so vital that any thought of the need to declare war never entered his [Togo's] head… To have signaled his intentions in advance by a declaration of war would have been, in Togo's estimation, an act of criminal folly’ (Warner and Warner Reference Warner and Warner1974, 19). Japanese success depended on catching enemy fleets unprepared in a harbor; therefore, it is no surprise that Japan did not declare war until after the attacks began. A critic might suggest that a state that declares war in these cases is ignoring its responsibility to its public to ensure their security and protect their soldiers.

There are several reasons why this objection is not persuasive. Many intuitively think that surprise attacks are wrong for the reasons suggested earlier: adversaries and innocents have a right to be notified and negotiations under the threat of war should be given an opportunity to succeed. In the debate within the Kennedy administration during the Cuban missile crisis, Robert Kennedy and George Ball suggested that launching a surprise airstrike against Cuba was akin to the Japanese attack against Pearl Harbor, favoring alternative strategies (including a declaration of war or an ultimatum) to ensure that US conduct was in keeping with its traditions. George Ball wrote ‘we tried Japanese as war criminals because of the sneak attack on Pearl Harbor’ and that a surprise attack would be ‘pursuing a course of action that would cut directly athwart everything we have stood for during our national history, and condemn us as hypocrites in the opinion of the world’ (in May and Zelikow Reference May and Zelikow1997, 121). Moreover, most surprise attacks are parts of unjust wars. Few endorse Japanese motives during the Second World War or the Russo-Japanese War; the failure to declare war is one more issue in a litany of wartime abuses.

More importantly, declaring war does not eliminate most elements of tactical surprise. One can issue a declaration of war against Libya, Iraq, or Afghanistan, and achieve tactical surprise by timing strikes at unexpected times, from unexpected directions, or through the use of special operations forces. Even if there are exceptional cases, a state can come awfully close to declaring. In 1967, Israel did not formally give a declaration of war, but orally explained the long-term Israeli policy treated a close of the Straits as causus belli. An imperfect situation often leads to an imperfect policy, and criticism of Israel should take into account the force of circumstances. Some situations may be even less perfect where the nature of technology may require secrecy, and most importantly, time, leading to less clear obligations to declare. Understanding that a general obligation to declare exists does not mean states must compromise their war plans, but that we can criticize states that hide behind war plans to avoid reason-giving.

Secret diplomacy

The second important criticism is that secret diplomacy may be more effective than public declarations. In particular, a public declaration should not always be treated as a last resort because, in certain cases, secret diplomacy may settle problems more effectively. Once a state begins to publicly make a case for war, emotions run high, politicians become committed, and the war machinery is set in motion.Footnote ¹⁸

There are several reasons, however, that this objection misses the mark. Before modern wars begin, especially in democratic states, the public normally learns about the impending war in advance. Mobilizing the military is difficult without public knowledge. Therefore, whether or not a war is declared, the public will learn about a war and politicians will likely make a case for the war. The crucial question is whether the case is rightly made, not whether it is made. More importantly, secret negotiations and public diplomacy are not mutually exclusive. If secret diplomacy has a chance of working, it can be pursued before the declaration of war is made or in tandem with it. During the Cuban missile crisis, the United States made public demands on the Soviet Union while working in secret to avoid the war through direct negotiations.

A related question concerns bargaining in war. Many suggest that negotiations are part of a bargaining process (Schelling Reference Schelling1966; Fearon Reference Fearon1995), and states often make more demands than they are willing to settle for. To ask states to limit themselves to just demands is unrealistic because, in part, no one's initial position is limited to the outcome that they will settle for; it is akin to requiring car buyers to give their reserve price before negotiations could begin.

The crucial question is when must states give their reserve price: the just demand that will avoid the war. Pre-war negotiations are often a two-step process. In the first stage, states negotiate without the threat of war, often for long periods of time. In a second stage, the dispute usually begins a process of escalation that may lead to war. During the first round, where demands are made without threat of war, states are not required to give their reserve price. Unjust demands can be included in the bargaining process before the period is reached where force is considered an immediate option. Only in the latter phase is the refusal to stick to just demands a problem. In short, if the intention of states – during rounds of negotiations that precede declarations of war – is to use unjust demands to obtain just demands, then there is nothing objectionable about the unjust demands on this account.Footnote ¹⁹

Giving the reserve price by declaring war in the second stage – when escalation is planned – is crucially important. Before a war starts, states should limit themselves to those demands that give just cause (through a declaration of war). To do otherwise risks not affording the target the opportunity to make peace by complying with reasonable demands. The reason is familiar to deterrence theorists. A state may choose to escalate a conflict if it believes that the gains that could be made outweigh the costs. If a state chooses to not give its reserve price – if it bargains too hard – then a state may choose to escalate because the ‘true’ minimum bargaining position of other states is concealed, making the costs of compliance with demands higher.Footnote ²⁰