The 20th century was the most murderous in recorded history. The total number of deaths caused by or associated with its wars has been estimated at 187 million, the equivalent of more than 10 per cent of the world’s population in 1913. Taken as having begun in 1914, it was a century of almost unbroken war, with few and brief periods without organised armed conflict somewhere. It was dominated by world wars: that is to say, by wars between territorial states or alliances of states. The period from 1914 to 1945 can be regarded as a single ‘thirty years’ war’ interrupted only by a pause in the 1920s – between the final withdrawal of the Japanese from the Soviet Far East in 1922 and the attack on Manchuria in 1931. This was followed, almost immediately, by some forty years of Cold War, which conformed to Hobbes’s definition of war as consisting ‘not in battle only or the act of fighting, but in a tract of time wherein the will to contend by battle is sufficiently known’. It is a matter for debate how far the actions in which US Armed Forces have been involved since the end of the Cold War in various parts of the globe constitute a continuation of the era of world war. There can be no doubt, however, that the 1990s were filled with formal and informal military conflict in Europe, Africa and Western and Central Asia. The world as a whole has not been at peace since 1914, and is not at peace now.

Nevertheless, the century cannot be treated as a single block, either chronologically or geographically. Chronologically, it falls into three periods: the era of world war centred on Germany (1914 to 1945), the era of confrontation between the two superpowers (1945 to 1989), and the era since the end of the classic international power system. I shall call these periods I, II and III. Geographically, the impact of military operations has been highly unequal. With one exception (the Chaco War of 1932-35), there were no significant inter-state wars (as distinct from civil wars) in the Western hemisphere (the Americas) in the 20th century. Enemy military operations have barely touched these territories: hence the shock of the bombing of the World Trade Center and the Pentagon on 11 September. Since 1945 inter-state wars have also disappeared from Europe, which had until then been the main battlefield region. Although in period III war returned to South-East Europe, it seems very unlikely to recur in the rest of the continent. On the other hand, during period II inter-state wars, not necessarily unconnected with the global confrontation, remained endemic in the Middle East and South Asia, and major wars directly springing from the global confrontation took place in East and South-East Asia (Korea, Indochina). At the same time, areas such as sub-Saharan Africa, which had been comparatively unaffected by war in period I (apart from Ethiopia, belatedly subject to colonial conquest by Italy in 1935-36), came to be theatres of armed conflict during period II, and witnessed major scenes of carnage and suffering in period III.

Two other characteristics of war in the 20th century stand out, the first less obviously than the second. At the start of the 21st century we find ourselves in a world where armed operations are no longer essentially in the hands of governments or their authorised agents, and where the contending parties have no common characteristics, status or objectives, except the willingness to use violence. Inter-state wars dominated the image of war so much in periods I and II that civil wars or other armed conflicts within the territories of existing states or empires were somewhat obscured. Even the civil wars in the territories of the Russian Empire after the October Revolution, and those which took place after the collapse of the Chinese Empire, could be fitted into the framework of international conflicts, insofar as they were inseparable from them. On the other hand, Latin America may not have seen armies crossing state frontiers in the 20th century, but it has been the scene of major civil conflicts: in Mexico after 1911, for instance, in Colombia since 1948, and in various Central American countries during period II. It is not generally recognised that the number of international wars has declined fairly continuously since the mid-1960s, when internal conflicts became more common than those fought between states. The number of conflicts within state frontiers continued to rise steeply until it levelled off in the 1990s.

More familiar is the erosion of the distinction between combatants and non-combatants. The two world wars of the first half of the century involved the entire populations of belligerent countries; both combatants and non-combatants suffered. In the course of the century, however, the burden of war shifted increasingly from armed forces to civilians, who were not only its victims, but increasingly the object of military or military-political operations. The contrast between the First World War and the Second is dramatic: only 5 per cent of those who died in World War One were civilians; in World War Two the figure increased to 66 per cent. It is generally supposed that 80 to 90 per cent of those affected by war today are civilians. The proportion has increased since the end of the Cold War because most military operations since then have been conducted not by conscript armies, but by quite small bodies of regular or irregular troops, in many cases operating high-technology weapons and protected against the risk of incurring casualties. While it is true that high-tech weaponry has made it possible in some cases to re-establish a distinction between military and civilian targets, and therefore between combatants and non-combatants, there is no reason to doubt that the main victims of war will continue to be civilians.

What is more, the suffering of civilians is not proportionate to the scale or intensity of military operations. In strictly military terms, the two-week war between India and Pakistan over the independence of Bangladesh in 1971 was a modest affair, but it produced ten million refugees. The fighting between armed units in Africa during the 1990s can hardly have involved more than a few thousand, mostly ill-armed combatants, yet it produced, at its peak, almost seven million refugees – a far greater number than at any time during the Cold War, when the continent had been the scene of proxy wars between the superpowers.

This phenomenon isn’t confined to poor and remote areas. In some ways the effect of war on civilian life is magnified by globalisation and the world’s increasing reliance on a constant, unbroken flow of communications, technical services, deliveries and supplies. Even a comparatively brief interruption of this flow – for instance, the few days’ closure of US airspace after 11 September – can have considerable, perhaps lasting, effects on the global economy.

It would be easier to write about the subject of war and peace in the 20th century if the difference between the two remained as clear-cut as it was supposed to be at the beginning of the century, in the days when the Hague Conventions of 1899 and 1907 codified the rules of war. Conflicts were supposed to take place primarily between sovereign states or, if they occurred within the territory of one particular state, between parties sufficiently organised to be accorded belligerent status by other sovereign states. War was supposed to be sharply distinguished from peace, by a declaration of war at one end and a treaty of peace at the other. Military operations were supposed to distinguish clearly between combatants – marked as such by the uniforms they wore, or by other signs of belonging to an organised armed force – and non-combatant civilians. War was supposed to be between combatants. Non-combatants should, so far as possible, be protected in wartime. It was always understood that these conventions did not cover all civil and international armed conflicts, and notably not those arising out of the imperial expansion of Western states in regions not under the jurisdiction of internationally recognised sovereign states, even though some (but by no means all) of these conflicts were known as ‘wars’. Nor did they cover large rebellions against established states, such as the so-called Indian Mutiny; nor the recurrent armed activity in regions beyond the effective control of the states or imperial authorities nominally ruling them, such as the raiding and blood-feuding in the mountains of Afghanistan or Morocco. Nevertheless, the Hague Conventions still served as guidelines in the First World War. In the course of the 20th century, this relative clarity was replaced by confusion.

First, the line between inter-state conflicts and conflicts within states – that is, between international and civil wars – became hazy, because the 20th century was characteristically a century not only of wars, but also of revolutions and the break-up of empires. Revolutions or liberation struggles within a state had implications for the international situation, particularly during the Cold War. Conversely, after the Russian Revolution, intervention by states in the internal affairs of other states of which they disapproved became common, at least where it seemed comparatively risk-free. This remains the case.

Second, the clear distinction between war and peace became obscure. Except here and there, the Second World War neither began with declarations of war nor ended with treaties of peace. It was followed by a period so hard to classify as either war or peace in the old sense that the neologism ‘Cold War’ had to be invented to describe it. The sheer obscurity of the position since the Cold War is illustrated by the current state of affairs in the Middle East. Neither ‘peace’ nor ‘war’ exactly describes the situation in Iraq since the formal end of the Gulf War – the country is still bombed almost daily by foreign powers – or the relations between Palestinians and Israelis, or those between Israel and its neighbours Lebanon and Syria. All this is an unfortunate legacy of the 20th-century world wars, but also of war’s increasingly powerful machinery of mass propaganda, and of a period of confrontation between incompatible and passion-laden ideologies which brought into wars a crusading element comparable to that seen in religious conflicts of the past. These conflicts, unlike the traditional wars of the international power system, were increasingly waged for non-negotiable ends such as ‘unconditional surrender’. Since both wars and victories were seen as total, any limitation on a belligerent’s capacity to win that might be imposed by the accepted conventions of 18th and 19th-century warfare – even formal declarations of war – was rejected. So was any limitation on the victors’ power to assert their will. Experience had shown that agreements reached in peace treaties could easily be broken.

In recent years the situation has been further complicated by the tendency in public rhetoric for the term ‘war’ to be used to refer to the deployment of organised force against various national or international activities regarded as anti-social – ‘the war against the Mafia’, for example, or ‘the war against drug cartels’. Not only is the fight to control, or even to eliminate, such organisations or networks, including small-scale terrorist groups, quite different from the major operations of war: it also confuses the actions of two types of armed force. One – let’s call them ‘soldiers’ – is directed against other armed forces with the object of defeating them. The other – let’s call them ‘police’ – sets out to maintain or re-establish the required degree of law and public order within an existing political entity, typically a state. Victory, which has no necessary moral connotation, is the object of one force; the bringing to justice of offenders against the law, which does have a moral connotation, is the object of the other. Such a distinction is easier to draw in theory than in practice, however. Homicide by a soldier in battle is not, in itself, a breach of the law. But what if a member of the IRA regards himself as a belligerent, even though official UK law regards him as a murderer? Were the operations in Northern Ireland a war, as the IRA held, or an attempt in the face of law-breakers to maintain orderly government in one province of the UK? Since not only a formidable local police force but a national army was mobilised against the IRA for thirty years or so, we may conclude that it was a war, but one systematically run like a police operation, in a way that minimised casualties and the disruption of life in the province. In the end, there was a negotiated settlement; one which, typically, has not so far brought peace, but merely an extended absence of fighting. Such are the complexities and confusions of the relations between peace and war at the start of the new century. They are well illustrated by the military and other operations in which the US and its allies are at present engaged.

There is now, as there was throughout the 20th century, a complete absence of any effective global authority capable of controlling or settling armed disputes. Globalisation has advanced in almost every respect – economically, technologically, culturally, even linguistically – except one: politically and militarily, territorial states remain the only effective authorities. There are officially about two hundred states, but in practice only a handful count, of which the US is overwhelmingly the most powerful. However, no state or empire has ever been large, rich or powerful enough to maintain hegemony over the political world, let alone to establish political and military supremacy over the globe. The world is too big, complicated and plural. There is no likelihood that the US, or any other conceivable single-state power, could establish lasting control, even if it wanted to.

A single superpower cannot compensate for the absence of global authorities, especially given the lack of conventions – relating to international disarmament, for instance, or weapons control – strong enough to be voluntarily accepted as binding by major states. Some such authorities exist, notably the UN, various technical and financial bodies such as the IMF, the World Bank and the WTO, and some international tribunals. But none has any effective power other than that granted to them by agreements between states, or thanks to the backing of powerful states, or voluntarily accepted by states. Regrettable as this may be, it isn’t likely to change in the foreseeable future.

Since only states wield real power, the risk is that international institutions will be ineffective or lack universal legitimacy when they try to deal with offences such as ‘war crimes’. Even when world courts are established by general agreement (for example, the International Criminal Court set up by the UN Rome Statute of 17 July 1998), their judgments will not necessarily be accepted as legitimate and binding, so long as powerful states are in a position to disregard them. A consortium of powerful states may be strong enough to ensure that some offenders from weaker states are brought before these tribunals, perhaps curbing the cruelty of armed conflict in certain areas. This is an example, however, of the traditional exercise of power and influence within an international state system, not of the exercise of international law.*

There is, however, a major difference between the 21st and the 20th century: the idea that war takes place in a world divided into territorial areas under the authority of effective governments which possess a monopoly of the means of public power and coercion has ceased to apply. It was never applicable to countries experiencing revolution, or to the fragments of disintegrated empires, but until recently most new revolutionary or post-colonial regimes – China between 1911 and 1949 is the main exception – emerged fairly quickly as more or less organised and functioning successor regimes and states.

Over the past thirty years or so, however, the territorial state has, for various reasons, lost its traditional monopoly of armed force, much of its former stability and power, and, increasingly, the fundamental sense of legitimacy, or at least of accepted permanence, which allows governments to impose burdens such as taxes and conscription on willing citizens. The material equipment for warfare is now widely available to private bodies, as are the means of financing non-state warfare. In this way, the balance between state and non-state organisations has changed.

Armed conflicts within states have become more serious and can continue for decades without any serious prospect of victory or settlement: Kashmir, Angola, Sri Lanka, Chechnya, Colombia. In extreme cases, as in parts of Africa, the state may have virtually ceased to exist; or may, as in Colombia, no longer exercise power over part of its territory. Even in strong and stable states it has been difficult to eliminate small unofficial armed groups, such as the IRA in Britain and ETA in Spain. The novelty of this situation is indicated by the fact that the most powerful state on the planet, having suffered a terrorist attack, feels obliged to launch a formal operation against a small, international, non-governmental organisation or network lacking both a territory and a recognisable army.

How do these changes affect the balance of war and peace in the coming century? I would rather not make predictions about the wars that are likely to take place or their possible outcomes. However, both the structure of armed conflict and the methods of settlement have been changed profoundly by the transformation of the world system of sovereign states.

The dissolution of the USSR means that the Great Power system which governed international relations for almost two centuries and, with obvious exceptions, exercised some control over conflicts between states, no longer exists. Its disappearance has removed a major restraint on inter-state warfare and the armed intervention of states in the affairs of other states – foreign territorial borders were largely uncrossed by armed forces during the Cold War. The international system was potentially unstable even then, however, as a result of the multiplication of small, sometimes quite weak states, which were nevertheless officially ‘sovereign’ members of the UN. The disintegration of the USSR and the European Communist regimes plainly increased this instability. Separatist tendencies of varying strength in hitherto stable nation-states such as Britain, Spain, Belgium and Italy might well increase it further. At the same time, the number of private actors on the world scene has multiplied. Under these circumstances, it is not surprising that cross-border wars and armed interventions have increased since the end of the Cold War.

What mechanisms are there for controlling and settling such conflicts? The record is not promising. None of the armed conflicts of the 1990s ended with a stable settlement. The survival of Cold War institutions, assumptions and rhetoric has kept old suspicions alive, exacerbating the post-Communist disintegration of South-East Europe and making the settlement of the region once known as Yugoslavia more difficult.

These Cold War assumptions, both ideological and power-political, will have to be dispensed with if we are to develop some means of controlling armed conflict. It is also evident that the US has failed, and will inevitably fail, to impose a new world order (of any kind) by unilateral force, however much power relations are skewed in its favour at present, and even if it is backed by an (inevitably shortlived) alliance. The international system will remain multilateral and its regulation will depend on the ability of several major units to agree with one another, even though one of these states enjoys military predominance. How far international military action taken by the US is dependent on the negotiated agreement of other states is already clear. It is also clear that the political settlement of wars, even those in which the US is involved, will be by negotiation and not by unilateral imposition. The era of wars ending in unconditional surrender will not return in the foreseeable future.

The role of existing international bodies, notably the UN, must also be rethought. Always present, and usually called upon, it has no defined role in the settlement of disputes. Its strategy and operation are always at the mercy of shifting power politics. The absence of an international intermediary genuinely considered neutral, and capable of taking action without prior authorisation by the Security Council, has been the most obvious gap in the system of dispute management.

Since the end of the Cold War the management of peace and war has been improvised. At best, as in the Balkans, armed conflicts have been stopped by outside armed intervention, and the status quo at the end of hostilities maintained by the armies of third parties. This sort of long-term intervention has been applied for many years by individual strong states in their sphere of influence (Syria in Lebanon, for instance). As a form of collective action, however, it has been used only by the US and its allies (sometimes under UN auspices, sometimes not). The result has so far been unsatisfactory for all parties. It commits the interveners to maintain troops indefinitely, and at disproportionate cost, in areas in which they have no particular interest and from which they derive no benefit. It makes them dependent on the passivity of the occupied population, which cannot be guaranteed – if there is armed resistance, small forces of armed ‘peacekeepers’ have to be replaced by much larger forces. Poor and weak countries may resent this kind of intervention as a reminder of the days of colonies and protectorates, especially when much of the local economy becomes parasitic on the needs of the occupying forces. Whether a general model for the future control of armed conflict can emerge from such interventions remains unclear.

The balance of war and peace in the 21st century will depend not on devising more effective mechanisms for negotiation and settlement but on internal stability and the avoidance of military conflict. With a few exceptions, the rivalries and frictions between existing states that led to armed conflict in the past are less likely to do so today. There are, for instance, comparatively few burning disputes between governments about international borders. On the other hand, internal conflicts can easily become violent: the main danger of war lies in the involvement of outside states or military actors in these conflicts.

States with thriving, stable economies and a relatively equitable distribution of goods among their inhabitants are likely to be less shaky – socially and politically – than poor, highly inegalitarian and economically unstable ones. A dramatic increase in economic and social inequality within, as well as between, countries will reduce the chances of peace. The avoidance or control of internal armed violence depends even more immediately, however, on the powers and effective performance of national governments and their legitimacy in the eyes of the majority of their inhabitants. No government today can take for granted the existence of an unarmed civilian population or the degree of public order long familiar in large parts of Europe. No government today is in a position to overlook or eliminate internal armed minorities. Yet the world is increasingly divided into states capable of administering their territories and citizens effectively – even when faced, as the UK was, by decades of armed action by an internal enemy – and into a growing number of territories bounded by officially recognised international frontiers, with national governments ranging from the weak and corrupt to the non-existent. These zones produce bloody internal struggles and international conflicts, such as those we have seen in Central Africa. There is, however, no immediate prospect for lasting improvement in such regions, and a further weakening of central government in unstable countries, or a further Balkanisation of the world map, would undoubtedly increase the dangers of armed conflict.

A tentative forecast: war in the 21st century is not likely to be as murderous as it was in the 20th. But armed violence, creating disproportionate suffering and loss, will remain omnipresent and endemic – occasionally epidemic – in a large part of the world. The prospect of a century of peace is remote.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences