Untitled Document


Global systems of the 20 th century were designed to address inter-state tensions and civil wars. War between nation-states and civil war have a given logic…21 st century violence does not fit the 20 th century mould…Violence and conflict have not been banished…But because of the success in reducing inter-state war, the remaining forms of violence do not fit neatly either into “war” or “peace”, or into “political” or “criminal” violence.

( World Bank 2011 )

The idea that twenty-first century organised violence is different from the wars of the twentieth century has been widely debated in both the scholarly and the policy literature. Various terms have been used to conceptualise contemporary conflict – wars among the people, wars of the third kind, hybrid wars, privatized wars, post-modern wars as well as ‘new wars’ ( Duffield 2001 ; Eppler 2002 ; Hables Gray 1997 ; Hoffman 2007 ; Holsti 1996 ; Kaldor 2012 ; Munkler 2005 ; Smith 2005 ; Snow 1996 ; Van Creveld 1991 ). But it is the term ‘new’ that seems to have stuck and become the main butt of the critics.

This article 1 defends the concept of ‘new wars’. Engaging with and countering the various criticisms that have been brought forward against the term ‘new’, it makes the argument that the ‘new’ in ‘new wars’ has to be understood as a research strategy and a guide for policy. Because the ‘old’ is enshrined in the concept of the ‘new’ the term enables us to grapple with the overall logic that is inherent in contemporary violent conflicts and that makes them different in kind from ‘old wars’. It is a logic that goes beyond specific components of contemporary conflicts – identity politics or economic predation, for example. Rather, it provides an integrative framework for analysis.

This essay addresses four main thrusts of criticism: whether new wars are ‘new’; whether new wars are ‘war’; whether existing data confirms or negates the findings about the nature of new wars; and whether new wars can be described as post-Clausewitzean. Before doing so, it is worth issuing a note of caution. One of the problems with many of the critics is that they lump together the different versions of the argument and treat criticism of one particular aspect contained in one particular version as a criticism of the whole argument. Such claims include the identification of new wars with civil wars, the claim that they are only fought by non-state actors and only motivated by economic gain, or that they are deadlier than earlier wars ( Berdal 2003 ; de Graaf 2003 ; Kalyvas 2001 ; Mellow 2010 ). In particular, many of the critics employ reductionist arguments whereby new wars are associated with a particular aspect of contemporary wars, for example, crime or privatisation or brutality, and fail to take into account the overall conceptual framework that relates actors, goals, methods and forms of finance. This essay will try to avoid this trap and focus on my own version of New Wars ( Kaldor 1999 ). Before discussing the critiques, I will start with a summary of this particular ‘new wars’ argument.

The logic of new wars

New Wars are the wars of the era of globalisation. Typically, they take place in areas where authoritarian states have been greatly weakened as a consequence of opening up to the rest of the world. In such contexts, the distinction between state and non-state, public and private, external and internal, economic and political, and even war and peace are breaking down. Moreover the break down of these binary distinctions is both a cause and a consequence of violence.

New wars have a logic that is different from the logic of what I call ‘old wars’ – the idea of war that predominated in the nineteenth and twentieth centuries. In the original version of the argument, I derived this logic from the differences between old and new wars in actors, goals, methods and forms of finance. These are:

  • Actors: Old wars were fought by the regular armed forces of states. New wars are fought by varying combinations of networks of state and non-state actors – regular armed forces, private security contractors, mercenaries, jihadists, warlords, paramilitaries, etc.
  • Goals: Old wars were fought for geo-political interests or for ideology (democracy or socialism). New wars are fought in the name of identity (ethnic, religious or tribal). Identity politics has a different logic from geo-politics or ideology. The aim is to gain access to the state for particular groups (that may be both local and transnational) rather than to carry out particular policies or programmes in the broader public interest. The rise of identity politics is associated with new communications technologies, with migration both from country to town and across the world, and the erosion of more inclusive (often state-based) political ideologies like socialism or post-colonial nationalism. Perhaps most importantly, identity politics is constructed through war. Thus political mobilisation around identity is the aim of war rather than an instrument of war, as was the case in ‘old wars’.
  • Methods: In old wars, battle was the decisive encounter. The method of waging war consisted of capturing territory through military means. In new wars, battles are rare and territory is captured through political means, through control of the population. A typical technique is population displacement – the forcible removal of those with a different identity or different opinions. Violence is largely directed against civilians as a way of controlling territory rather than against enemy forces.
  • Forms of Finance: Old wars were largely financed by states (taxation or by outside patrons). In weak states, tax revenue is falling and new forms of predatory private finance include loot and pillage, ‘taxation’ of humanitarian aid, Diaspora support, kidnapping, or smuggling in oil, diamonds, drugs, people, etc. It is sometimes argued that new wars are motivated by economic gain, but it is difficult to distinguish between those who use the cover of political violence for economic reasons and those who engage in predatory economic activities to finance their political cause. Whereas old war economies were typically centralising, autarchic and mobilised the population, new wars are part of an open globalised decentralised economy in which participation is low and revenue depends on continued violence.

The implication of these differences is that, whereas old wars tended to extremes as each side tried to win, new wars tend to spread and to persist or recur as each side gains in political or economic ways from violence itself rather than ‘winning’ (see Keen 2012 ). Whereas old wars were associated with state building, new wars are the opposite; they tend to contribute to the dismantling of the state.

It is this logic of persistence and spread that I have come to understand as the key difference with old wars – something I elaborate in the last section, where I discuss whether new wars are post-Clausewitzean. Clausewitz was par excellence the theorist of old wars – for him, war was a contest of wills. In my version of new wars, war is rather a violent enterprise framed in political terms. It is important to stress that both old and new wars, in my formulation, are ideal types. They are ideas of war rather than empirical descriptions of war. The test of how well they fit empirical reality depends on whether they provide a guide to useful policy. As I discuss in the following sections, it is this point that is most often missed by the critics of the new wars thesis. 2

Are new wars ‘New’?

The most common criticism of the ‘new wars’ argument is that new wars are not new. It is argued that the Cold War clouded our ability to analyse ‘small wars’ or ‘low-intensity wars’, that many of the characteristics of new wars associated with weak states can be found in the early modern period and that phenomena like banditry, mass rape, forced population displacement, or atrocities against civilians all have a long history.

Of course this is true. Many of the features of new wars can be found in earlier wars. Of course the dominance of the East-West conflict obscured other types of conflict. But there is an important reason, which is neglected by the preoccupation with empirical claims, for insisting on the adjective ‘new’.

Critics of the ‘new wars’ thesis often concede that what is useful about the analysis of ‘new wars’ is the policy implication of the argument. But this is precisely the point. The term ‘new’ is a way to exclude ‘old’ assumptions about the nature of war and to provide the basis for a novel research methodology. The aim of describing the conflicts of the 1990s as ‘new’ is to change the way scholars investigate these conflicts and thus to change the way policy-makers and policy-shapers perceive these conflicts. Dominant understandings of these conflicts that underpin policy are of two kinds. On the one hand, there is a tendency to impose a stereotyped version of war, drawn from the experience of the last two centuries in Europe, in which war consists of a conflict between two warring parties, generally states or proto-states with legitimate interests, what I call ‘Old Wars’. This term refers to a stylised form of war rather than to all earlier wars. In such wars, the solution is either negotiation or victory by one side and outside intervention takes the form of either traditional peace-keeping – in which the peace-keepers are supposed to guarantee a negotiated agreement and the ruling principles are consent, neutrality and impartiality – or traditional war-fighting on one side or the other, as in Korea or the Gulf War. On the other hand, where policy-makers recognise the shortcomings of the stereotypical understanding, there is a tendency to treat these wars as anarchy, barbarism, ancient rivalries, where the best policy response is containment, i.e. protecting the borders of the West from this malady. The use of the term ‘new’ is a way of demonstrating that neither of these approaches are appropriate, that these are wars with their own logic but a logic that is different from ‘old wars’ and which therefore dictates a very different research strategy and a different policy response. In other words, the ‘new wars’ thesis is both about the changing character of organised violence and about developing a way of understanding, interpreting and explaining the interrelated characteristics of such violence.

As Jacob Mundy ( 2011 ) puts it, in one of the more thoughtful contributions to the debate:

‘Whether we choose to reject, embrace or reformulate concepts such as…. new wars, our justifications should not be based on claims of alleged coherence with particular representations of history. Rather such concepts should be judged in terms of their ability to address the very phenomena they seek to ameliorate’.

Even so, it can be argued that there are some genuinely new elements of contemporary conflicts. Indeed, it would be odd if there were not. The main new elements have to do with globalisation and technology.

First of all, the increase in the destructiveness and accuracy of all forms of military technology has made symmetrical war – war between similarly armed opponents – increasingly destructive and therefore difficult to win. The first Gulf war between Iraq and Iran was perhaps the most recent example of symmetrical war – a war, much like the First World War, that lasted for years and killed millions of young men, for almost no political result. Hence, tactics in the new wars necessarily have to deal with this reality.

Secondly, new forms of communications (information technology, television and radio, cheap air travel) have had a range of implications. Even though most contemporary conflicts are very local, global connections are much more extensive, including criminal networks, Diaspora links, as well as the presence of international agencies, NGOs, and journalists. The ability to mobilise around both exclusivist causes and human rights causes has been speeded up by new communications. Communications are also increasingly a tool of war, making it easier, for example, to spread fear and panic than in earlier periods – hence, spectacular acts of terrorism. This does not mean, as Berdal ( 2011 ) suggests, that the argument implies that all contemporary wars involve global connections or that those connections are necessarily regressive. Rather, it is an element in theorising the logic of new wars.

Thirdly, even though it may be the case that, as globalisation theorists argue, globalisation has not led to the demise of the state but rather its transformation, it is important to delineate the different ways in which states are changing. Perhaps the most important aspect of state transformation is the changing role of the state in relation to organised violence. On the one hand, the monopoly of violence is eroded from above, as some states are increasingly embedded in a set of international rules and institutions. On the other hand, the monopoly of violence is eroded from below as other states become weaker under the impact of globalisation. There is, it can be argued, a big difference between the sort of privatised wars that characterised the pre-modern period and the ‘new wars’ which come after the modern period and are about disintegration.

These new elements are not the reason for the adjective ‘new’, however, even though they may help to explain the evolution of new wars. The point of the adjective ‘new’ does not have to do with any particular feature of contemporary conflicts nor how well it resembles our assumptions about reality, but rather it has to do with the model of war and how the model I spell out is different from the prevailing models that underpin both policy and scholarship. It is a model that entails a specific political, economic and military logic.

Many of the critics miss the point about the logic of new wars. For example, both Berdal ( 2011 ) and Malesovic ( 2010 ) make the point that identity politics are also about ideas – the idea of Greater Croatia, for example, says Berdal. In a trivial sense, that is true just as ideological conflicts can also be reduced to identity – a communist or a fascist identity as opposed to an ethnic or tribal identity, for example. But the point of making this distinction is to illuminate different political logics, the way in which identity politics is associated with different practices, different methods of warfare and different ways of relating to authority. Identity politics is about the right to power in the name of a specific group; ideological politics is about winning power in order to carry out a particular ideological programme. Typically, in new war contexts, for example, access to the state is about access to resources rather than about changing state behaviour; in such situations, competition for power tends to be based on identity rather than on programmatic debate, even if the latter is more of an ideal than a reality. This helps to explain military tactics – population displacement as a method of exerting political control – or the persistence of new wars, as fear is a necessary long-term ingredient of identity politics. Berdal and Malesevic seem to be implying that the term ‘identity politics’ suggests that politics is a mask , which is instrumentalised for economic reasons; of course new wars are about politics – that is why they are wars – and of course identity is constructed, but so are all other forms of ideology. The point is that the distinction that I make between identity politics and ideology (democracy or socialism) and geo-political interest implies a different set of political practices and a different methodology of war.

Some critics of the ‘new wars’ argument say the term is too fuzzy – a ‘hodgepodge’, say Henderson and Singer ( 2002 ). Indeed, similar terms – like hybrid warfare, multivariant warfare, or complex warfighting – are explicitly about being a mixture. Thus, for example, multi-variant warfare refers to a ‘spectrum of conflict marked by unrestrained Mad Max ways in which symmetric and asymmetric wars merge and in which Microsoft coexists with machetes and stealth technology met by suicide bombers’ ( Evans 2003 ; Hoffman 2007 ). The problem with existing categorisations of conflict, however, is that they do not easily fit contemporary reality, a point that will be elaborated in the data section, and consequently the policy prescriptions that emerge out of them are confused and distorted. It is to be hoped that the current debate will further refine the concept and lead to new categories that may displace the term ‘new’.

A typical example of this type of criticism is the article by Sven Chojnacki. Chojnacki ( 2006 ) argues that the term ‘new wars’ is too vague and also ‘methodologically problematic because the criteria for identifying “new” wars are highly arbitrary, difficult to reproduce inter-subjectively, and difficult to reconcile with conflict theory’ (italics added). Chojnacki then goes on to establish his own categories based on actors – inter-state, extra-state, intra-state, and sub-state – which entirely misses the point of new wars, in which the actors are both state and non-state, internal and external. It misses the point that the term ‘new wars’ is a critique of prevailing conflict theory.

Some critics concede that something like new wars exists. But that does not mean that ‘old wars’ have gone away. Particularly after the wars in Iraq and Afghanistan, some scholars and policy makers warn of assuming that future wars will look like Iraq and Afghanistan. It is to be hoped that future wars will not be like Iraq and Afghanistan because these wars have been exacerbated by outside military interventions. But nor are future wars likely to look like the wars of the twentieth century. Of course, a return to old wars cannot be ruled out. It is possible to imagine continued competitive arming by states, growing interstate tensions, and a tendency to forget the suffering of previous generations. But failure to deal with the ‘new wars’ of the present might make that possibility more plausible. The reconstruction of militarised states through external wars might come to be viewed as a way of re-establishing the monopoly of violence at national levels. As John Keegan puts it: ‘The great work of disarming tribes, sects, warlords and criminals – a principal achievement of monarchs in the 17th century and empires in the 19th – threatens to need doing all over again’ (Quoted in Mueller 2004: 172 ). In the present economic crisis, where states are cutting defence budgets, there is a tendency to protect what is seen as the core defence task – preparation for ‘old war’ – and to squeeze the emerging capacity to contribute to global peace enforcement efforts.

Are new wars ‘War’?

Some writers argue that contemporary violence is mainly privatised and/or criminal and cannot therefore be properly described as war. A good example of this kind of thinking is John Mueller’s interesting book The Remnants of War. He claims that war is becoming obsolescent and what is left are thugs who are the ‘residual combatants’ ( Mueller 2004 ). In other words, he defines war as ‘old war’. A similar argument is made by Martin Shaw ( 2003 ), who talks about ‘degenerate wars’.

According to Mueller ( 2004: 115 ), ‘most of what passes for warfare to-day is centrally characterised by the opportunistic and improvisatory clash of thugs, not by the programmed and/or primordial clash of civilisations –although many of the perpetrators do cagily apply ethnic, national or ideological rhetoric to justify their activities because to stress the thrill and profit of predation would be politically incorrect’.

There is a lot of sense in this line of argument. New wars can be described as mixtures of war (organised violence for political ends), crime (organised violence for private ends) and human rights violations (violence against civilians). The advantage of not using the term ‘war’ is that all forms of contemporary violence can be regarded as wholly illegitimate, requiring a policing rather than a political/military response. Moreover, much contemporary violence – like the drugs wars in Mexico or gang warfare in major cities – appears to have a similar logic to new wars, but has to be classified as criminal. The same sort of argument has been used in relation to terrorism. There has been widespread criticism of the term ‘war on terror’ because it implies a military response to terrorist violence when policing and intelligence methods, it is argued, would be more effective ( Howard 2002 ).

On the other hand, the political element does have to be taken seriously; it is part of the solution. Articulating a cosmopolitan politics as an alternative to exclusivist identity is the only way to establish legitimate institutions that can provide the kind of effective governance and security that Mueller is proposing as a solution. War does imply organised violence in the service of political ends. This is the way it legitimises criminal activity. Suicide bombers in their farewell videos describe themselves as soldiers not as murderers. Even if it is the case, and it often is, that those who frame the violence in ethnic, religious or ideological terms are purely instrumental, these political narratives are internalised through the process of engaging in or suffering from violence. Indeed, this is the point of the violence; it is only possible to win elections or to mobilise political support through the politics of fear. This is a point made strongly by Kalyvas in his Logic of Violence in Civil Wars. He quotes Thucydides on ‘the violent fanaticism which came into play once the struggle had broken out ….society had become divided into two ideologically hostile camps, and each side viewed the other with suspicion’ ( Kalyvas 2006: 78 ). Overcoming fear and hostility does not necessarily come about through compromise, even if that is possible, because compromise can entrench exclusivist positions; rather it requires a different kind of politics, the construction of a shared discourse that has to underpin any legal response.

A related terminological issue concerns the word ‘conflict’. There is a legal difference between ‘war’ and ‘armed conflict’, which has to do with whether or not war has been formally declared. Most data sets assume a threshold below which violence cannot be counted as war – say a thousand battle deaths per year, as in the Correlates of War database (Correlates of War Project). Without wishing to be overly semantic, the term conflict does seem to imply a contestation around a legitimate grievance that can be resolved either by victory of one side or through compromise; the term used in the Uppsala University Conflict Dataset is ‘contested incompatibility’ ( UCDP 1988 ). Actually, conflict is endemic in all societies and necessary for change and adaptation. Democracy is a peaceful mechanism for managing conflict. Violence, as Michel Wievorka ( 2009 ) contends, tends to be the opposite of conflict; it closes down debates and ‘encourages ruptures’. In ‘new wars’ the ‘sides’ need an ‘incompatibility’ in order to justify their existence.

The Debate about data

The ‘new wars’ argument is largely based on qualitative rather than quantitative data. It came out of empirical studies of the wars in the former Yugoslavia and the South Caucasus as well as Sub Saharan Africa ( Kaldor and Vashee 1997 ). This knowledge has since been augmented by research on Iraq and Afghanistan, but there were two quantitative claims that I used to back up the arguments that battles are becoming rare and most violence is directed against civilians. One concerned the dramatic increase in the ratio of civilian to military casualties and the other concerned the rise in the numbers of displaced people per conflict. Other data that could be relevant relate to the recurrence and/or persistence of contemporary conflicts as well as the tendency to spread.

In fact, the quantitative data, despite claims to the contrary, does seem to confirm the claims about the nature of new wars even though this data has to be used cautiously because it largely derives from ‘old’ assumptions about conflict. The debate about data covers three broad areas: the numbers and duration of wars; the numbers of casualties; and the levels of forced displacement.

The numbers and duration of wars

There are three main sources for data on numbers of wars. These are:

All three data sets are based on ‘old war’ assumptions. For violence to be counted as a war, there has to be a state involved at least on one side and there have to be a certain number of battle deaths. Moreover, they all distinguish between intra-state and inter-state war, and some have added sub-state or non-state categories. Yet central to the ‘new wars’ argument is the difficulty of distinguishing between what is state or non-state and what is external or internal. So, none of these numbers are really able to capture the nature of new wars.

In particular, the emphasis on battle deaths has the counter-intuitive effect of leaving out major episodes of violence. As Milton Leitenberg ( 2006 ) puts it: ‘There were few “battle deaths” in Cambodia between 1975 and 1978, comparatively few in Somalia in 1990 and 1991, or in Rwanda in 1994: but it would simply be bizarre if two million dead in Cambodia, 350,000 in Somalia and 8000 or more in Rwanda were omitted from compilations’.

Nevertheless, the findings from the three databases do have some relevance to the new wars thesis. They all tend to concur in the following conclusions:

  • The virtual disappearance of wars between states;
  • The decline of all high intensity wars, involving more than a thousand battle deaths;
  • The decline in the deadliness of war measured in terms of battle deaths;
  • The increase in the duration and/or recurrence of wars; and
  • The risk factor of proximity to other wars.

In other words, there does seem to be a decline in ‘old wars’, which is largely what this data measures. There is also a decline in the numbers killed in battles, which is consistent with the argument about the decline of battle. And there does seem to be evidence for the argument that new wars are difficult to end and they tend to spread if we assume that the data does catch some ‘new war’ elements.

The UCDP has made the most effort to adjust to the new realities and has added data on episodes of one-sided violence and on non-state violent conflicts. Both of these numbers seem to be increasing and this again is consistent with the argument that new wars could be treated as cases of mutual one-sided violence and that low-level, low intensity persistent conflicts may be more typical nowadays.

Those who have criticised the new wars argument using this sort of data have tended to set up straw men to attack. Thus it is argued that new wars are civil wars and the decline in civil wars suggests that new wars are not increasing. But new wars are not the same as civil wars and no one has claimed that new wars are increasing or decreasing; the argument was always about the changing character of war. Bizarrely, critics have also suggested that the decline of battle severity is a critique of new wars when on the contrary it confirms the new wars argument ( Melunder, Oberg and Hall 2009 )


The problem with calculations about the ratio of civilian to military casualties is three fold. First, figures on civilian casualties are notoriously inaccurate. There are a variety of methods for calculating these numbers: reliance on media and other reports of individual deaths, epidemiological surveys, opinion surveys and, where available, official death certificates. The results vary widely. Thus, casualties in the Bosnia war vary from 260,000 (the number given by the Bosnian Information Ministry and widely used by international agencies at the time), of which 60,000 were military, to 40,000 in the World Disasters Report ( Roberts 2010 ). Similarly, civilian casualties in the Iraq war have been the subject of huge debate; the numbers vary widely, from around 100,000 civilian casualties from violence (as of a 2011 estimate by Iraq Body Count , which relies on media reports and official documents) to over a million (based on an opinion survey in 2007, which asked Iraqis in all 18 governorates whether any member of their family had been killed) ( ORB International ).

Secondly, it is very difficult to distinguish combatants from civilians. The only figures for which there are accurate statistics are military casualties because these are formally recorded by their governments. Hence, we know that, as of September 2012, there were some 4804 military casualties in Iraq, of which 4486 were American, and some 3202 military casualties in Afghanistan, of which some 2136 were American ( Iraq Coalition Casualty Count ). But, since many combatants in new wars are police, militia, private contractors, mercenaries, para-militaries or criminals of various kinds, the figures for other military and civilian casualties are very difficult to identify. A good example are the figures produced by the Sarajevo Research and Documentation Centre. They collected death certificates for people killed in the 1992–5 war and estimated that some 97,207 people were killed, of which 39,684 or 41% were civilian and 62,626 or 59% were soldiers. However, the number for soldiers included all men of military age. Since we know that it was mainly men of military age that were killed in ethnic cleansing operations and the majority of displaced people were women – and we also know that participation in the violence was very low, about 6.5% of the population – it is simply not credible that all those men were soldiers. It would presuppose that nearly all the 8000 men and boys killed in Srebrenica were soldiers, for example.

Thirdly it is very difficult to distinguish whether civilians were killed as a side effect of battle, as a result of deliberate violence (political or criminal), or as a result of the indirect effects of war – privation and disease. The Human Security Report suggests that deaths as an indirect effect of war have declined in contemporary wars. This is because wars are often highly localised and low-level and general improvements in healthcare or in immunisation continue during the wars. The main method of calculating these indirect effects is through calculating the excess deaths that took place over and above what might have been expected from previous trends. The Human Security Report, for example, criticises the IRC report on casualties in the war in the Democratic Republic of Congo, which estimates that 5.4 million people died during the war who would not have died ‘had there been no war’; more than 90% were estimated to have died from war-exacerbated disease and malnutrition. The HSR argues that their estimate was based on an estimated infant mortality rate prior to the conflict that was too low, that their surveys were biased in favour of areas with a small population and a high death toll and that the true figure is probably much lower.

So what can be said about the data on casualties? First of all, the data suggests an overall decline in all war-related deaths. One of the misapplied criticisms that have been made of the new wars thesis is that new wars scholars claim that atrocities in new wars are worse than in previous wars. The only claim that the new wars thesis makes is most violence in new wars consists of violence against civilians rather than combat – it would be mad to claim that violence against civilians is worse than the modernist state-based atrocities like the holocaust or the Soviet purges. Secondly, there has been a dramatic decline in battle deaths. If we compare all war-related deaths to battle deaths rather than civilian to military casualties, then it is possible to assert that the ratio has increased on a scale commensurate with the ‘new wars’ original claim ( Lacin and Gleditsch 2005 ). Thirdly, casualties among regular soldiers are a very small proportion of total deaths in wars, both because there are fewer regular soldiers taking part in wars and because of the decline in battle.

Finally, what is shocking about this whole debate is the fact that we have good and accurate statistics for the deaths of men in state-based uniforms, but information about the vast majority of victims is totally inadequate.

Forced displacement

No one disputes that the overall total displaced population has increased. Indeed according to UNHCR, the figures for forcibly displaced people in 2010 were at their highest in fifteen years at 43.7 million, including 15.4 million refugees, some 27.5 million internally displaced persons and 837,500 individuals whose asylum applications had not been processed. But critics suggest that these numbers should be qualified in two respects. First, data collection has greatly improved, especially in relation to internally displaced persons. In particular, the main source of IDP data is the Norwegian Refugee Council’s Internal Displacement Monitoring Centre, which has only been collecting data since 1998 ( IDMC ). Before that date, the main source was UNHCR’s estimates of those IDPs of concern to UNHCR, a much lower figure. Secondly, refugee and IDP data tends to be cumulative, since many people do not return to their homes.

Nevertheless, recent conflicts – especially in Iraq, Somalia and Pakistan – do seem to confirm the contention that forcible displacement is a central methodology of new wars. In Iraq, for example, some 4 million people were displaced at the height of the war in 2006–2008; roughly half were refugees and half were internally displaced. Indeed, it can be argued that one reason for lower levels of deaths in war is that it is easier to spread fear and panic using new communications, so that more people leave their homes than formerly. At the same time, there does seem to be a trend towards increasing displacement per conflict. Using the American Refugee Council data, Myron Weiner ( 1996 ) calculated that the number of refugees and internally displaced persons per conflict increased from 327,000 per conflict in 1969 to 1,316,000 in 1992 (1992 was, of course, a peak year for conflict). Using the Uppsala Conflict Database and figures from UNHC and the IDMC, an upward trend in refugees and internally based persons can be observed per conflict. Figure 1 is broader, showing the rise in annual numbers of internally displaced persons in countries experiencing not only armed conflict, but what the UCDP describe as substate conflict and one-sided violence. 3

Rise in annual numbers of internally displaced persons in countries experiencing armed conflict, substate conflict, and one-sided violence.

Figure 1: Rise in annual numbers of internally displaced persons in countries experiencing armed conflict, substate conflict, and one-sided violence.

One conclusion from this discussion is the need to refine the displacement data, which could well offer a better indicator of human insecurity than some of the other numbers that are used.

The Debate about Clausewitz

The final set of criticisms against the ‘new wars’ thesis has to do with the claim that new wars are post-Clausewitzean ( Strachan and Herberg-Rothe 2007 ; Schuurman 2010 ). The reasons that are normally put forward for claiming that new wars are post-Clausewitzean have to do with the Trinitarian conception of war, the primacy of politics and the role of reason. Both John Keegan ( 2004 ) and Martin Van Creveld ( 1991 ) have suggested that the Trinitarian concept of war, with its tripartite distinction of the state, the army and the people, is no longer relevant. Other authors suggest that war is no longer an instrument of politics and, indeed, that the ‘divorce of war from politics’ is characteristic of both pre-Clausewitzean and post-Clausewitzean wars (Snow quoted in Angstrom 2003: 8 ). Along with these arguments, critics have also questioned the rationality of war. Van Creveld, for example, argues that it is ‘preposterous…to think that just because some people wield power, they act like calculating machines that are unswayed by passions. In fact, they are no more rational than the rest of us’(1991: 10).

These arguments are rather trivial and, depending on how Clausewitz is interpreted, they can all be refuted. Huw Strachan ( 2007 ) points out that the trinity refers to ‘tendencies’ or motivations rather than empirical categories. The point of the concept is to explain how a complex social organisation, made up of many different individuals with many different motivations, can become, in his words, the ‘personalised state’ – a ‘side’ in or party to war. ‘War’ says Clausewitz, ‘is, therefore, not only chameleon-like in character, because it changes colour in some degree in each particular case, but it is, also, as a whole, in relation to the predominant tendencies which are in it, a wonderful trinity, composed of the original violence of its elements, hatred and animosity, which may be looked upon as blind instinct; the play of probabilities and chance, which make it a free activity of the soul; and of the subordinate nature of a political instrument, by which it belongs to pure reason’ ( 1968: 24 ). These different ‘tendencies’ – reason, chance and emotion – are mainly associated with the state, the generals and the people, respectively, but the word ‘mainly’ or ‘more’ suggests that they are not exclusively associated with these different components or levels of warfare.

Clausewitz argues that war is what unites the trinity. The trinity was ‘wondrous’ because it made possible the coming together of the people and the modern state. Obviously, the distinction between the state, the military, and the people is blurred in most new wars. New wars are fought by networks of state and non-state actors and often it is difficult to distinguish between combatants and civilians. So, if we think of the trinity in terms of the institutions of the state, the army and the people, then it cannot apply. But if we think of the trinity as a concept for explaining how disparate social and ethical tendencies are united in war, then it is clearly very relevant.

A second issue is the primacy of politics. Among translators of Clausewitz, there is a debate about whether the German word politik should be translated as policy or politics. It can be argued that it applies to both if we roughly define policy as external, in terms of relations with other states, and politics as the domestic process of mediating different interests and views.

New Wars are also fought for political ends and, indeed, war itself can be viewed as a form of politics. The political narrative of the warring parties is what holds together dispersed loose networks of paramilitary groups, regular forces, criminals, mercenaries and fanatics, representing a wide array of tendencies – economic and/or criminal self-interest, love of adventure, personal or family vendettas, or even just a fascination with violence. It is what provides a license for these varying tendencies. Moreover, these political narratives are often constructed through war. Just as Clausewitz described how patriotism is kindled through war, so these identities are forged through fear and hatred, through the polarisation of us and them. In other words, war itself is a form of political mobilisation, a way of bringing together, of fusing the disparate elements that are organised for war.

Understood in this way, war is an instrument of politics rather than policy. It is about domestic politics even if it is a politics that crosses borders rather than the external policy of states. If, for Clausewitz, the aim of war is external policy and political mobilisation, this means, in new wars, it is the other way round. Mobilisation around a political narrative is the aim of the war and external policy or policy vis-à-vis the proclaimed enemy is the justification.

So if new wars are an instrument of politics, what is the role of reason? ‘New wars’ are rational in the sense of instrumental rationality. But is rationality the same as reason? The enlightenment version of reason was different from instrumental rationality. As used by Hegel, who was a contemporary in Berlin of Clausewitz, it had something to do with the way the state was identified with universal values, the agency that was responsible for the public as opposed to the private interest. The state brought together diverse groups and classes for the purpose of progress – democracy and economic development. Clausewitz puts considerable emphasis on the role of the cabinet in formulating policy and argues that the Commander-in-Chief should be a member of the cabinet. The cabinet, which in Clausewitz’s time was a group of ministers advising the monarch, was thought to play a role in bringing together different interests and motivations and providing unifying, publicly justifiable arguments for both war and the conduct of war. Of course, members of the cabinet had their own private motivations, as do generals (glory, enrichment, jealousy, etc), but it is incumbent on them to come to some agreement, to provide the public face of the war and to direct the war, and this has to be based on arguments that are universally acceptable (universal, here, referring to those who are citizens of the state). In his description of the evolution of warfare and the state, which echoes Hegel’s stadial theory of history, he argues that only in the modern period can the state be regarded as ‘an intelligent being acting in accordance with simple logical rules’ ( Clausewitz 1968: 342 ) and that this is associated with the rise of cabinet government where the ‘cabinet had become a complete unity, acting for the state in all its external relations’ ( Clausewitz 1968: 344 ).

The political narratives of new wars are based on particularist interests; they are exclusive rather than universalist. They deliberately violate the rules and norms of war. They are rational in the sense of being instrumental. But they are not reasonable. Reason has something to do with universally accepted norms that underpin national and international law.

However there is another argument about why new wars are post-Clausewitzean. This has to do with the fundamental tenets of Clausewitzean thought – his notion of ideal war. This is derived from his definition of war. ‘War’ he says ‘is nothing but a duel on an extended scale. If we would conceive as a unit the countless number of duels which make up a war, we shall do so best by supposing to ourselves two wrestlers. Each strives by physical force to compel the other to submit to his will: each endeavours to throw his adversary, and thus render him incapable of further resistance. War therefore is an act of violence intended to compel our opponent to fulfil our will’ ( Clausewitz 1968: 5 ; italics in the original). Violence, he says, is the means. The ultimate object is the ‘compulsory submission of the enemy to our will’ and, in order to achieve this, the enemy must be disarmed.

He then goes on to explain why this must lead to the extreme use of violence. ‘Now philanthropists may easily imagine there is a skilful method of disarming and overcoming an enemy without causing great bloodshed….However plausible this may appear, still it is an error, which must be extirpated; for in such dangerous things as war, the errors which proceed from a spirit of benevolence are the worst. As the use of physical power to the utmost extent by no means excludes the co-operation of intelligence, it follows that he who uses forces unsparingly, without reference to the bloodshed involved, must obtain a superiority if his adversary uses less vigour in its application. The former then dictates the law to the latter, and both proceed to extremities to which the only limitations are those imposed by the amount of counteracting force on each side ’ ( Clausewitz 1968: 6 ; italics added). In other words, the inner nature of war – Absolute War – follows logically from the definition as each side is pushed to make fresh efforts to defeat the other, a proposition that Clausewitz elaborates in Chapter 1, through what he calls the three reciprocal actions according to which violence is ‘pushed to its utmost bounds’ ( 1968: 7 ). For Clausewitz, combat is the decisive moment of war.

Real war may depart from ideal war for a variety of reasons, but as long as war fits his definition, it contains the logic of extremes and, in Chapter 2 of my book, I describe how that logic applied to ‘Old Wars’. It is this logic of extremes that I believe no longer applies in ‘new wars’. I have therefore reformulated the definition of war. I have defined war as ‘an act of violence involving two or more organised groups framed in political terms’. According to the logic of this definition, war could either be a ‘contest of wills’ as is implied by Clausewitz’s definition or it could be a ‘mutual enterprise’. A contest of wills implies that the enemy must be crushed and therefore war tends to extremes. A mutual enterprise implies that both sides need the other in order to carry on the enterprise of war and therefore war tends to be long and inconclusive.

‘New wars’ tend to be mutual enterprises rather than a contest of wills. The warring parties are interested in the enterprise of war rather than winning or losing, for both political and economic reasons. The inner tendency of such wars is not war without limits, but war without end. Wars, defined in this way, create shared self-perpetuating interest in war to reproduce political identity and to further economic interests.

As in the Clausewitzean schema, real wars are likely to be different from the ideal description of war. The hostility that is kindled by war among the population may provoke disorganised violence or there may be real policy aims that can be achieved. There may be outside intervention aimed at suppressing the mutual enterprise or the wars may produce unexpectedly an animosity to violence among the population, undermining the premise of political mobilisation on which such wars are based.

This redefinition of war constitutes a different interpretation of war, a theory of war, whose test is how well it offers a guide to practice. Since it is an ideal type, examples can be used to support the theory, but it is, in principle, unprovable. The question is whether it is useful. Take the example of the ‘War on Terror’. Antonio Echevarria defines the ‘War on Terror’ in classic Clausewitzean terms: ‘Both antagonists seek the political destruction of the other and, at this point, neither appears open to negotiated settlement’ ( 2007: 211 ). Understood in this way, each act of terrorism calls forth a military response, which, in turns, produces a more extreme counterreaction. The problem is that there can be no decisive blow. The terrorists cannot be destroyed by military means because they cannot be distinguished from the population. Nor can the terrorists destroy the military forces of the United States. But if we understand the ‘War on Terror’ as a mutual enterprise – whatever the individual antagonists believe – in which the American Administration shores up its image as the protector of the American people and the defender of democracy, those with a vested interest in a high military budget are rewarded and extremist Islamists are able to substantiate the idea of a Global Jihad and to mobilise young Muslims behind the cause, then action and counterreaction merely contribute to ‘long war’, which benefits both sides. Understood in Clausewitzean terms, the proposed course of action is total defeat of the terrorists by military means. Understood in post-Clausewitzean terms, the proposed course of action is very different; it has to do with both with the application of law and the mobilisation of public opinion not on one side or the other, but against the mutual enterprise.

The contrast between new and old wars, put forward here, is thus a contrast between ideal types of war rather than a contrast between actual historical experiences. Of course, the wars of the twentieth century, at least in Europe, were close to the old war ideal and the wars of the twenty first century are closer to my depiction of new wars. Contemporary wars may not actually conform to this description any more than earlier wars conformed to the old war description. Perhaps another way to describe the difference is between realist interpretations of war as conflicts between groups, usually states, that act on behalf of the group as a whole and interpretations of war in which the behaviour of political leaders is viewed as the expression of a complex set of political and perhaps bureaucratic struggles pursuing their particular interest or the interests of their faction or factions, rather than those of the whole. It can be argued that in the Westphalian era of sovereign nation-states, a realist interpretation had more relevance than it does today.

This conceptual distinction is not quite the same as the way I originally described ‘new wars’ in terms of the involvement of non-state actors, the role of identity politics, the blurring of the distinction between war (political violence) and crime (violence for private interests) as well as the fact that, in new wars, battles are rare and violence is mainly directed against civilians ( Kaldor 2007 ). But it is not inconsistent with that earlier description; it merely involves a higher level of abstraction.


The debate about new wars has helped to refine and reformulate the argument. The debate about Clausewitz has facilitated a more conceptual interpretation of new wars, while the debate about data has led to the identification of new sources of evidence that have helped to substantiate the main proposition.

The one thing the critics tend to agree is that the new war thesis has been important in opening up new scholarly analysis and new policy perspectives, which, as I have stressed, was the point of the argument ( Newman 2004 ; Henderson and Singer 2002 ). The debate has taken this further. It has contributed to the burgeoning field of conflict studies. And it has had an influence on the intensive policy debates that are taking place especially within the military, ministries of defence and international organisations – the debates about counter-insurgency in the Pentagon, for example, or about human security in the European Union and indeed about non-traditional approaches to security in general.

What is still lacking in the debate is the demand for a cosmopolitan political response. In the end, policing, the rule of law, justice mechanisms and institution-building depend on the spread of norms at local, national and global levels. And norms are constructed both through scholarship and public debate. If we are to reconceptualise political violence as ‘new war’ or crime and the use of force as cosmopolitan law enforcement rather than war-fighting, then we have to be able to challenge the claims of those who conceptualise political violence as ‘old war’, and this can only be done through critical publicly-engaged analysis.


1 I am grateful to Denisa Kostovicova and Sabine Selchow for comments on an earlier draft and to Tom Kirk for help with the literature search.

2 An exception is Ken Booth’s ( 2001 ) thoughtful essay that accepts the point about the logic of new wars, but is critical of what he sees as top-down, overly militarised policy implications. I have not addressed this argument in this essay, but it is a concern in much of my work on human security.

3 I am grateful to Anouk Rigterink for assistance with these numbers.


Angstrom, J 2003 Introduction. In: Angstrom, J et al The Nature of Modern War: Clausewitz and his Critics Revisited . Stockholm: Department of War Studies, Swedish National Defence College.

Berdal, M 2003 How ‘new’ are ‘new wars’? Global economic change and the study of civil war. Global Governance : 477–502.

Berdal, M 2011 The ‘new wars’ thesis revisited. In: Strachan, H and Scheipers, S The Changing Character of War . Oxford: Oxford University Press.

Booth, K 2001 New wars for old. Civil Wars 4,2: 163–170

Chojnacki, S 2006 Anything new or more of the same? Wars and military intervention in the international system 1946–2003. Global Society 20,1: 25–46. DOI: http://dx.doi.org/10.1080/13600820500405442 .

Correlates of War Project . Available at http://www.correlatesofwar.org [Last accessed 5 February 2013].

De Graaff, B 2003 The wars in former Yugoslavia in the 1990s: Bringing the state back in. In: Angstrom, J and Duyvesteyn, I The Nature of Modern War: Clausewitz and his Critics Revisited . Stockholm: Department of War Studies, Swedish National Defence College.

Duffield, M 2001 Global governance and the new wars: The merging of development and security . London: Zed Books

Echevarria, A 2007 Clausewitz and contemporary war . Oxford: Oxford University Press.

Eppler, E 2002 Vom Gewaktmärkte zum Gewaltmarkt? Frankfurt: Suhrkmamp.

Evans, M 2003 From Kadesh to Kandahar; Military theory and the future of war. Naval War College Review Summer 2003.

Hables Gray, C 1997 Post-modern war: The new politics of conflict . London: Routledge.

Henderson, E A and Singer, D 2002 ‘New wars’ and rumours of ‘new wars’. International Interactions 28, 2: 165–190. DOI: http://dx.doi.org/10.1080/03050620212098 .

Hoffman, F 2007 Conflict in the 21 st century; The rise of hybrid wars . Arlington: Potomac Institute for Policy Studies.

Holsti, K J 1996 The state, war and the state of war . Cambridge: Cambridge University Press.

Howard, M 2002 What’s in a name? Foreign Affairs , January/February 2002.

Human Security Report . Available at www.hsrgroup.org .

Internal Displacement Monitoring Centre , Norwegian Refugee Council. Available at: http://www.internal-displacement.org [Last accessed 5 February 2013].

Iraq Body Count . Available at http://www.iraqbodycount.org/ [Last accessed 5 February 2013].

Iraq Coalition Casualty Count . Available at: http://icasualties.org/ [Last accessed 5 February 2013].

Kalyvas, S N 2001 ‘New’ and ‘old’ civil wars: A valid distinction? World Politics 54, October: 99–118.

Kalyvas, S N 2006 The logic of violence in civil wars . Cambridge: Cambridge University Press.

Kaldor, M 1999 New and old wars: Organised violence in a global era . 1 st edition. Cambridge: Polity Press.

Kaldor, M 2007 New and old wars: Organised violence in a global era . 2 nd edition. Cambridge: Polity Press.

Kaldor, M 2012 New and old wars: Organised violence in a global era . 3 rd ed. Cambridge: Polity Press.

Kaldor, M and Vashhee, B, eds. 1997 New Wars . London: Pinter Press.

Keegan, J A 2004 History of warfare . London: Pimlico.

Keen, D 2012 Useful enemies: When waging war is more important than winning them . New Haven: Yale University Press.

Lacina, B and Gleditsch, N P 2005 Monitoring trends in global combat: A new dataset of battle deaths. European Journal of Population 21: 145–166. DOI: http://dx.doi.org/10.1007/s10680-005-6851-6 .

Leitenberg, M 2006 Deaths in war and conflicts in the 20 th century. Cornell University Peace Studies Program Occasional Paper 29, 3 rd edition.

Malesovic, S 2010 The sociology of war and violence . Cambridge: Cambridge University Press.

Mellow, P A 2010 Review article: In search of new wars: The debate about the transformation of war. European Journal of International Relations , 16: 297. DOI: http://dx.doi.org/10.1177/1354066109350053 .

Melunder, E, Oberg, M and www.hsrgroup.org 2009 Are ‘New Wars’ More Atrocious? Battle Severity, Civilians Killed and Forced Migration Before and After the End of the Cold War. European Journal of International Relations, 15:505. DOI: http://dx.doi.org/10.1177/1354066109338243 .

Mueller, J 2004 The remnants of war . Ithica: Cornell University Press.

Mundy, J 2011 Deconstructing civil wars: Beyond the new wars debate. Security Dialogue 42: 279–295. DOI: http://dx.doi.org/10.1177/0967010611405378 .

Munkler, H 2005 The new wars . Cambridge: Polity Press.

Newman, E 2004 The ‘new wars’ debate: A historical perspective is needed. Security Dialogue 35, 2: 173–189. DOI: http://dx.doi.org/10.1177/0967010604044975 .

ORB International Available at www.opinion.co.uk .

Peace and Conflict , Centre for International Development and Conflict Management, University of Maryland. Available at http://www.cidcm.umd.edu/pc/ [Last accessed 5 February 2013].

Roberts, A 2010 Lives and statistics: Are 90% of war victims civilians?. Survival 52, 3: 115–136.

Schuurman, B 2010 Clausewitz and the ‘new wars’ scholars. Parameters Spring 2010: 89–100.

Shaw, M 2003 War and genocide . Cambridge: Polity Press.

SIPRI Available at www.sipri.org .

Smith, R 2005 The Utility of Force . London: Alfred A. Knopf.

Snow, D 1996 Uncivil wars: International security and the new internal conflicts . Boulder: Lynne Rienner.

Strachan, H 2007 Carl Von Clausewitz’s ‘On war’: A book that shook the world . London: Atlantic Books.

Stracan, H and Herbheg-Rothe, A 2007 Clausewitz and the twenty first century . Oxford: Oxford University Press.

UCDP Department of Peace and Conflict Research, Uppsala University. Available at http://www.pcr.uu.se/research/ucdp/definitions/definition_of_armed_conflict/ [Last accessed 5 February 2013].

Van Creveld, M 1991 The transformation of war . New York: The Free Press.

Von Clausewitz, C 1968 [German edition: 1832] On war. London: Penguin Books.

Weiner, M 1996 ‘Bad neighbours, bad neighbourhoods: an inquiry into the causes of refugee flows’ International Security 21, 1: 5–42.

Wievorka, M Violence: A new approach. London: Sage Publications.

World Bank 2011 World Development Report 2011: Conflict, Security and development. Available at http://go.worldbank.org/UCTHLNS530 [Last accessed 5 February 2013].