Essay: The Just War Tradition
Robert Skidelsky
Prospect | Wednesday, December 01, 2004

In recent years there has been a revival of war as a policy of choice. Since the collapse of communism, the US and its allies have attacked Iraq (twice), Yugoslavia and Afghanistan. With "hot war" released from its cold war constraints, it is important to consider the conditions under which resort to war is justifiable, and what methods of fighting wars are right. This is the domain of "just war" theory. There is also the related issue of how just war doctrine may be fruitfully applied by the UN, the main custodian of international law.
The return of war as a policy of choice overturns the post-second world war assumption, enshrined in the UN charter, which allowed only wars of self-defence and proscribed intervention in the domestic affairs of sovereign states. This position - a reaction to the two world wars of the 20th century - was reinforced by the unique destructiveness of nuclear weapons. The threat of MAD (mutually assured destruction) prevented war between the two cold war superpowers, and made possible the semi-pacifism of Europe and Japan, sheltering under the nuclear umbrella. The utility as well as the morality of war was questioned. The conventional view was that superior firepower could not prevail against ideologically inspired guerrillas. After Vietnam, the US became reluctant to accept casualties.
None of this meant that warfare disappeared from the planet. Hot wars punctuated the cold war era. Some of these were extremely bloody. But their locus was in the unstable, postcolonial peripheries of the world system. In the imperial age, the great powers fought each other and kept peace in their colonies. After the second world war, the great powers lived in peace with each other, while some of their former possessions were engulfed by violence, largely as a result of decolonisation.The UN's role was confined to the sporadic provision of unarmed peacekeepers, there on the sufferance of the warring factions.
Renunciation of war as a method of settling disputes between the "civilised" powers remains. But an attempt is now underway to pacify the disordered parts of the world, springing from a mixture of fear, self-interest and moralism. Yesterday's peaceniks are transformed into today's warriors.
Three changes have liberated war from its cold war constraints. The first and most obvious is the collapse of the Soviet Union, which left the US as sole superpower. The two Iraq wars, the Kosovo conflict and the Afghan war would probably have been too risky in cold war days. Second, globalisation has weakened the doctrinal hold of national sovereignty as a barrier to coercive intervention in the domestic affairs of states. Finally, vast military superiority has removed the bodybag constraint. In the first three post-Soviet Union wars, the US and its allies suffered almost no casualties (although the current problems in Iraq may be changing this).
In this more permissive context, two specific motives for waging war have emerged. The first is international terrorism. The events of 9/11 seemed to make obsolete the quest for security through deterrence. From the ruins of the World Trade Centre sprang Bush's doctrine of preventative war - an ambitious programme of removing latent threats to US security by promoting regime change. The Iraq war of 2003 was the first fruit of this new doctrine.
The second motive for making war is the much higher profile given to the protection of human rights. Humanitarian concern has been vastly stimulated by an increased flow of information about suffering, particularly via television. In 1998, Kofi Annan said: "State frontiers should no longer be seen as watertight protection for war criminals or mass murderers."
These two motives are to some extent interdependent. The US, in particular, links the fight against international terrorism to the battle for human rights and democracy.
But the freedom of the west to make war is far from absolute. It cannot contemplate war against strong non-western powers like China, Russia or India. The world balance of power may have collapsed, but some regional balances remain intact: North Korea cannot be assailed with impunity, "rogue state" though it is, because of China. The west's freedom to make war is also constrained by public opinion. The glamour has long gone out of war, killed off in the trenches in 1915. Western societies are inherently pacific; to be stirred out of their pacifism the cause has to be compelling. Moreover, there are many, particularly on the left, who argue that the return of war as a policy choice is tantamount to the revival of imperialism. The doctrine of state sovereignty retains a powerful appeal.
What has been regained is the freedom to fight limited wars in some areas, where technological supremacy can secure swift victories at little cost to the victors. This opportunity to make war, coupled with the emergence of a set of motives for doing so, raises the question of when it is right to do so.
The theologian James Turner Johnson calls the just war tradition a "moral tradition of justifiable and limited war." Of Christian provenance, it may be distinguished from two other Christian approaches to war: pacifism and holy war. Pacifism must be rejected, since war may be a way to a juster peace. But holy war - the Christian equivalent to the Muslim jihad - is equally wrong, since perfect justice is unattainable in this world.
Historically, the intellectual stimulus to curb unbridled force has been strongest in periods of imperial conquest or extreme lawlessness. The just war tradition, which originates in St Augustine's 5th-century City of God, was a compromise between the Christian church and the Roman state. In the middle ages, Aquinas specified the conditions of a just war; Michael Howard enumerated these in a 1992 essay as: "a war waged by a legitimate authority, for a cause in itself just, to make reparation for an injury or to restore what had been wrongly seized, and with the intention of advancing good and avoiding evil." The modern tradition starts in the 16th century with the Spanish discussion of the moral and prudential justification of the Spanish conquest of Latin America.
In the Westphalian era, which started in the 17th century, just war theory fell into abeyance. The sovereignty of states replaced the sovereignty of God: the state alone had the right to determine the extent of its legal and moral obligations. The problem of how to restrain sovereigns from attacking each other was partially solved de facto by the emergence of a balance of power. Discussion of the justice of war came to be largely confined to its conduct - on issues such as the distinction between combatants and non-combatants and the need for force to be proportionate to aim. The legacy of this discussion is today enshrined in the Geneva conventions, specifying treatment of prisoners, casualties of war and civilians in war zones. Rules of war, fashioned for battles between professional armies, in turn broke down in the 20th century, as military technology and mass armies combined to make war "total." The object of war shifted from the defeat of enemy forces to breaking the resistance of the civilian population.
Today, a rich and sophisticated discussion over centuries is usually presented as a list of propositions covering three topics - jus ad bellum, jus in bello and jus ad pacem. The first aims to determine whether a war is justly started, the second is concerned with the just conduct of the war, and the third - and most shadowy - explores the conditions of a just peace. The propositions support each other, like bricks in an arch. Just war theory is not a scholastic exercise, but an invitation to moral argument.
The requirements for a just war are traditionally reduced to six: just cause, legitimate authority, last resort, proportionality between offence and response, reasonable chance of success and right intention.
Just cause governs everything. An unjust war cannot be justly fought; it ought not to be fought at all. The two justifiable aims of war are self-defence and protection of the innocent. Either could make war moral. However, to accommodate the Westphalian system of sovereign states, self-defence became virtually the sole criterion of justice, with protection of the innocent relegated to protection of non-combatants in war. In affirming the justice of war to protect innocent lives, just war theory sits comfortably enough with the growing demand for a right to intervene to prevent humanitarian disasters.
Self-defence seems unproblematic. However, it is too elastic to be useful without qualification. Few would deny that it can include pre-emption if the threat of attack is clear and imminent. The problem arises if the threat of attack is seen as arising from the character of the regime, as in the concept of the "rogue state." When true security is said to require universal democracy, as Tony Blair claimed in his 1999 Chicago speech, self-defence can sound suspiciously like attack. The very existence of a dictatorship can be seen as a threat to security justifying preventative war. This extension of the concept of self-defence was evident in the 2003 Iraq war.
Against this tendency to inflate the meaning of self-defence, just war theory erects two main barriers. The first is that for a war of self-defence to be just, it has to be "clearly an act of redress of rights actually violated or defence against unjust demands backed by the threat of force" (JB Elshtain). The standard of proof implied by such language is much higher than that required for claiming that a threat is latent or potential. The badness of a state is not a just ground for waging war against it. It is what the state does. This reduces the margin for error or lying. (The fact that a threat turns out not to have existed does not make the war in itself unjust, provided the mistake was honest and not just manufactured to justify war.)
The second barrier, added by Aquinas, is legitimate authority. In his conception, legitimate authority was split between rulers and church. In the Westphalian period, rulers alone became the judges of whether wars were just. But in the UN charter there is a revival of the older idea that a war has to be sanctioned by a higher authority. Signatories to the UN charter recognise the UN itself as the legitimate war-authorising authority. In making war against Iraq in 2003, the British government at least felt constrained to argue that it was acting under a UN resolution authorising war.
The other criteria also seek to curb recourse to war. The last resort requirement expands the scope for negotiation. From the strategic point of view it often makes sense to go to war before all opportunities for negotiation have been exhausted, since prolonging negotiation might be interpreted as a sign of weakness. War was started in Kosovo (1999) and Iraq (2003) before negotiation had reached its end point.
Proportionate response is a strong defence against the holy war mentality. The proportionality criterion requires us to measure offence and response on some common scale. This is impossible if the things to be defended are values, since body counts and values are incommensurable. The strength of the just war tradition is that it limits what is to be protected to physical things such as lives, property, livelihood - the things most obviously at risk from aggression. This is consistent with a pluralist view of the world. It is also consistent with humanitarian intervention to prevent genocide, mass murder, mass starvation and torture. Blairite wars to establish freedom and democracy are not consistent with the just war tradition, as they tend towards a holy war mentality.
The reasonable chance of success stipulation also sits naturally with proportionate response. The more unlimited one's ambitions for war, the remoter one's chance of achieving them. The reasonable chance of success criterion is relevant not just to the winning of a war but to the pacification of a country following a war. Twenty months after President Bush declared victory in Iraq, Iraq is far from pacified.
Right intention is a further support for just cause. One may be mistaken in one's view of threats, but one must not intend war for any purpose other than self-defence or protection of innocent lives. In particular, one cannot use a good outcome to justify, in retrospect, a wrong intention. The fact that a war of aggression may produce good results - for example, a higher standard of living or a democratic government in the defeated country - does not justify it. The right intention criterion is based on the historical experience that wars of aggression, even if successful in their immediate aims, often create more problems than they solve.
The two main criteria of just conduct of war are discrimination and proportionality: civilians (non-combatants) must never be deliberately attacked, and force must be proportionate to ends. The first, as has been mentioned, becomes possible again with the precise technology which limits collateral damage. Some of the older questions - when is a civilian to be legitimately regarded as a non-uniformed soldier? - have receded, at least for a time, in the face of the awesome, but very precise, military means available to the technologically advanced.
The second criterion guards against overkill: the force used must be the minimum to achieve the goal desired. If war is made to protect the innocent then the military means should be proportional to that end. This criterion is clearly difficult to apply, since it depends on the military capacity of the warmaker. A badly trained army is likely to kill more people (soldiers and civilians) than is a highly professional one.
A just peace must leave the situation better than it was before the war started. This might suggest a Carthaginian peace (named after the physical destruction of Carthage and the liquidation of its people, carried out by the Romans after the third Punic war). The logic is that the physical elimination of the enemy guarantees a non-recurrence of the war. A similar aim underlay the Morgenthau plan for the pastoralisation of Germany, which was (briefly) accepted by the three Allied war leaders in 1944. However, a Carthaginian peace is incompatible with protection of the innocent. The just war tradition favours a peace of reconciliation, not just on prudential grounds, but because it is the moral consequence of a war started for just cause.
The main usefulness of the just war tradition is that it provides a moral standard for judging the legitimacy of a recourse to violence. At a time when war is once again a policy of choice for the strong, it is important to revive the tradition of limits, if only to prevent the strong abusing the weak. Just war theory makes up for the lack of a balance of power.
It may be argued that moral restraint is built into the conduct of civilised countries. But moral restraints on the use of violence can be loosened, especially if war is seen as a moral crusade. Just war theory is an antidote to the "neocon" division of the world into good and evil, with its demand for unlimited freedom to secure regime change.
State sovereignty is fragmenting but world government is still far off. The just war tradition, which accords a real, but only relative, moral value to sovereign states, better captures the challenges of a globalising, but not yet single, world.
What about the UN? The UN system was not set up to deal with the problems posed by rogue and failed states. Rogue states are latent threats to the security of others, producing a demand for preventative war. Failed states produce humanitarian disasters. The result is either that these problems are dealt with outside the UN framework, as in the Iraq war in 2003, or they are not tackled at all, as in Rwanda in 1994.
The inability of the UN to handle these two issues derives from the purposes for which it was formed. The UN charter, signed in 1945, was largely a reaction to German aggression in the second world war. In article 2, members pledge themselves to refrain from "the threat or use of force against the territorial integrity or political independence of any state" and not to intervene in matters "which are essentially within the domestic jurisdiction of any state."
Under chapter VII of the charter, the security council can authorise the use of force on determining that there exists a "threat to the peace, breach of the peace, or act of aggression" (article 39) and that force is the only way of dealing with it (article 42). The quoted words give quite a wide latitude, but the situation that the founding fathers sought to interdict was war or threatened war between member states. The charter transfers "rightful authority" to use force from the member state to the security council itself except for the "inherent right of individual or collective self-defence if an armed attack occurs against a member…" During the cold war, the right of veto given to the five permanent members (the US, Soviet Union, Britain, China and France) was regularly used by the two superpowers to protect their clients. This made chapter VII largely inoperative.
The protection of member states against acts of aggression was thus the main purpose of the UN. The question of the domestic behaviour of those states was a secondary issue. However, the atrocious behaviour of Nazi Germany did put human rights on the agenda. The universal declaration of human rights approved by the general assembly in 1948 set out minimum human rights - freedom from discrimination on the basis of race or sex, rights to life, liberty and security of person, rights to be free from slavery, torture and inhuman treatment. Subsequent treaties and covenants have amplified these. A convention on the prevention and punishment of genocide came into force in 1951, and a convention on torture in 1984 has been ratified by 134 states.
The important point, though, was that crimes of this sort are not a ground for chapter VII action. Human rights were to be protected by economic and diplomatic pressure, and their abuse punished by courts, not military action, unless it constituted a "threat to international peace." This remains the situation, with an international criminal court now set up to try individuals for crimes against humanity. Thus of the two grounds for war recognised by the just war tradition - self-defence and protection of the innocent - the UN charter recognises the first but not the second.
The question today is whether a collective security system set up to protect states against would-be conquerors can become an instrument for limiting state sovereignty to make the world better. If it cannot, the UN will become an irrelevance, with military actions undertaken without reference to it - as happened to the League of Nations. The answer hinges on the extent to which member states are willing to recognise circumstances other than overt acts of aggression as legitimate grounds for war.
US rhetoric to the contrary, the "war on terrorism" raises no new issues for the UN - as was shown when the security council authorised the attack on Afghanistan. As early as 1954, the international law commission (a UN body set up to promote the development of international law) recognised a clear threat to the "peace and security of mankind" from states that allowed their territory to be used as a base for terrorist attacks on other states. During the cold war, states which harboured terrorists were able to get away with it, but there is no reason why they should today.
Under chapter VII, the security council can impose sanctions on states which knowingly sponsor terrorist groups. Terrorism, though, is more likely to be internal: this is a problem for domestic (and international) police work, not for chapter VII action.
A more contentious issue concerns the right of pre-emptive or anticipatory attack. Pre-emption is recognised by customary law - and the just war tradition - as part of the inherent right of self-defence. However, by convention, the threat of armed attack must be imminent, requiring evidence not only of the possession of weapons but also of an intention to use them. Self-defence cannot justly be stretched to cover preventative war, aimed, for example, at overthrowing an unsavoury state which has a nuclear weapons programme. The problem is that in such a case evidence of intent and capacity is nearly always lacking, and anyway open to argument. The war against Iraq in 2003 was a preventative war dressed up as a pre-emptive one. As such, it falls foul of the just war criterion of self-defence, and was also illegal in terms of the UN charter.
True enough, the security council's power to order military action is not limited by any requirement that the threat to peace be imminent. But it is unlikely that it would ever authorise a preventative war, for the same reason that the latter falls outside the just war tradition - the evidence requirement is too severe. Also, the precedent would be too disturbing.
The demand for preventative war could occur most obviously to stop the spread of weapons of mass destruction. The 1968 treaty on the non-proliferation of nuclear weapons (NPT) bound all states without nuclear weapons not to acquire them, and all the nuclear ones to eliminate them. This bargain has not been kept. The nuclear states have not disarmed; Israel, India and Pakistan did not ratify the treaty, and proceeded to acquire nuclear weapons (Israel has not admitted this); in 2003, North Korea withdrew from the NPT; other signatories, notably South Africa, Iraq, Iran and Libya developed secret nuclear programmes. The UN system cannot prevent determined proliferators from acquiring nuclear weapons. The biological and chemical weapons conventions are virtually dead letters.
However, if the political will were there, it would not be too difficult to convert the present incomplete patchwork of voluntary renunciations into a binding and effectively policed regime, covering the whole world. Latin America and Africa are already nuclear-free zones. Diplomatic and economic pressure should suffice to get North Korea and Iran to renounce their nuclear ambitions, as South Africa and Libya have already done. Similar freezes could be applied to biological and chemical weapons. Any reported breach in the agreements would automatically trigger UN sanctions. This proposal has the disadvantage of leaving the present possessors with a monopoly of the most deadly weapons. But a pacific hegemony of the nuclear powers is surely better than the US fighting a series of preventative wars. And if a secure non-proliferation regime can be established, existing stockpiles of WMD can be gradually eliminated.
Such a regime would probably be acceptable everywhere except in the middle east. Arab nations will not give up their nuclear ambitions as long as Israel has the bomb, and Israel will not give up the bomb as long as it feels threatened by its neighbours. Therefore there can be no nuclear-free zone in the middle east until the outstanding Arab-Israeli issues are settled. This will not happen soon. Until it does, the middle east will remain the world's powder keg.
In the Westphalian era, protection of the innocent was the responsibility of states. The UN charter was crafted in this tradition; the conventions which proscribe genocide and torture provide only legal redress against perpetrators. The Rwandan genocide of 1994, followed by other episodes, shattered the assumptions of this doctrine. As Kofi Annan put it in 2000: "If humanitarian intervention is an unacceptable assault on sovereignty, how should we respond to a Rwanda, to a Srebrenica - to gross and systematic violations of human rights that offend every precept of our common humanity?"
The international commission on intervention and state sovereignty, set up by the Canadian government after the Kosovo conflict, issued a report in 2001 called "The Responsibility to Protect." This sought to transfer the ultimate duty for protecting the innocent from states to the international community. Following the just war logic described above, it outlined an interlocking set of six criteria for military intervention for humanitarian purposes: just cause, right intention, last resort, proportional means, reasonable prospects of success, and right authority.
The just cause threshold was set deliberately high. For military intervention to be justified, at least one of two things must be occurring or "imminently likely": a) "large-scale loss of life, actual or apprehended, with genocidal intent or not, which is the product either of deliberate state action, or state neglect or inability to act, or a failed state situation," or b) "large-scale 'ethnic cleansing,' actual or apprehended, forced expulsion, acts of terror or rape." Two other aspects of the report should be noted. First, "protection" implies more than "intervention." It requires prevention, reaction and rebuilding. Secondly, "There is no more appropriate body than the UN security council to authorise military intervention for human protection purposes. The task is not to find alternatives to the security council as a source of authority, but to make the security council work better than it has."
As we have seen, abuse of human rights is not a ground for military intervention under the charter. The logical answer would be to amend the charter to allow for it. But this is not practicable: an amendment would need to be ratified by states which might themselves become targets for humanitarian intervention. The only way forward is to proceed on a case law basis, in which interventions such as in Kosovo become precedents for subsequent action.
A second problem, not fully recognised by the Canadian report, is that prevention of genocide, mass murder or mass starvation runs into the same problem of evidence as other claims to wage preventative war. In practice, prevention means stopping a disaster which is already happening. But this at least is better than letting it run its natural course.
What the report does recognise is that protection of human rights may require a long-term commitment to the country in which they have been violated. It is not just a case of postwar reconstruction, but the construction of a viable state which can protect human rights in the long term. The UN has a mechanism, the trusteeship council, which in principle could be used for this purpose. The council has been suspended since 1994, when Palau, the last territory under its administration, became self-governing, but it has not yet been abolished and various proposals to revive it have been floated over the past ten years. It was the successor to the League of Nations mandate system, which was set up in 1919 to administer the colonial territories of the defeated central powers. According to the UN charter, the trustee undertakes a variety of responsibilities, which include "political, economic, social and educational advancement" and "respect for human rights and for fundamental freedoms." The trusteeship may be held by a single state, a group of states, or by the UN itself, and it may be terminated in accordance with the "freely expressed wishes of the peoples concerned." The UN took on trustee-like roles in 1999 in Kosovo and East Timor. (Although falling outside the humanitarian frame, a prime candidate for a UN trusteeship today is Palestine, in partial restoration of the old League of Nations mandate.)
A consensus is developing that it is best to handle humanitarian interventions as part of regional security systems. This vests responsibility for maintaining peace in a region in its main regional powers, acting on a mandate from the security council. This has the advantage of removing any suspicion of a new colonialism, and also provides a sensible division of labour between the US and other countries.
There is no reason, for example, for the US to be involved in Africa. The African Union has recently agreed a common defence policy which allows intervention in states whose authorities are carrying out genocide or mass murder. On the initiative of Nigeria and South Africa, it is to set up the African standby force: a grouping of five regional brigades of 3,000 troops each, based on national battalions which are to train together. These plans mirror the EU's efforts to create a rapid reaction force. Unfortunately, the African states lack the money to train or transport the putative intervention force. This logistical and financial support could be provided by the EU.
From these messy beginnings, precedents are being established for humanitarian intervention by specially trained regional forces, followed by the installation of a temporary UN civil administration.
What conclusions can be drawn from this survey? The most important is that force remains necessary for peace and security. The end of the cold war has given the US and its allies a limited discretion to use force. Just war theory offers the chance to subject that use of force to rules. A great part of the UN charter is consistent with the just war tradition. But it was set up to protect states against external aggression. A Westphalian document needs to be turned into a just war charter.
There is no way in which the current US demand for the right to make preventative war can be brought within the framework of the charter, or indeed of the just war tradition. As Kofi Annan said in September 2003: "This logic represents a fundamental challenge to the principles on which, however imperfectly, world peace and stability have rested for the last 58 years." However, the need for preventative war could be obviated by the rigorous enforcement of a non-proliferation regime. Recognition of the need for such a regime may be the best result of the Iraq war. Meanwhile, the duty to protect the innocent, even against their own governments, when the scale of atrocities shocks the world's conscience, is close to becoming accepted doctrine. Little by little, and too timidly, the aims and methods of such humanitarian interventions are being worked out, as part of the larger project of providing for good government, economic development and poverty elimination in some of the poorest countries in the world.