Conditional Legitimacy, Reinterpreted Monopolies:

Anna Leander

Globalisation and the Evolving State Monopoly on Legitimate Violence

Paper prepared for the ISA Panel: Legitimacy and Violence: Globalization and the Displacement of the State New Orleans, 24-27 March 2002

Anna Leander, Copenhagen Peace Research Institute, Fredericiagade 18, DK- 1310 Copenhagen aleander@copri.dk

Abstract:

The article argues that globalisation is altering the nature and meaning of the state monopoly on legitimate violence. It is accentuating the tensions around the meaning of 'legitimacy'. The relativism implied in the idea that states can define which use of violence is 'legitimate' (and which is not) is increasingly contested both by the international society of states and in a world society of transnational actors. At the same time a profound redefinition of what it means to have a 'monopoly' of violence is going on. Increasing the private ownership and allocation of the means of coercion are blurring the responsibility of states beyond their own borders and, for some states, even within them. As a consequence the differentiation among states is growing, private actors are central to war and peace, and the system of national states might be undergoing a fundamental change.

Outline:

Introduction.................................................................................................... 3

The SMLV and International Relations................................................................ 4

The Continued Centrality of the SMLV....................................................... 4

The importance of context and shifting meanings....................................... 8

The Increasingly Contested and Conditional Legitimacy of the State Monopoly on Violence 10

The significance of tensions in legitimacy................................................ 11

The dislocation of politics and contestation of legitimacy........................... 11

'ccentuating the relativist conundrum in international society..................... 14

The State Monopoly on Legitimate Violence Reinterpreted .................................. 17

Multiple Monopolies.............................................................................. 17

Privatising Ownership and Allocation of the Means of Coercion.................. 18

Challenges to the Decision-Making Authority of States.............................. 21

Conclusion: Increased Differentiation Among States and Obliged Reference to Private Actors 23

References..................................................................................................... 25

1. Introduction

'Globalisation' holds an increasingly central place in International Relations (IR) discussions. Many people have begun to treat it as an emerging 'perspective' or even 'paradigm'^[1] through which international relations can be analysed. Yet, the link between globalisation and the traditional core of IR, namely the study of war and peace, has paradoxically occupied a relatively marginal place in the debate. Although, the 'retreat of the state' and the challenges to the Westphalian system loom large in these discussions,^[2] the link between these changes and war and peace is rarely put at the centre of the discussion.

In this paper, I want to challenge this way of thinking and draw a direct link between globalisation and war and peace. Like much of the 'globalisation literature' I will do this by thinking about the state as a set of institutions which fill different functions and ask myself to what extent these functions can still be filled and especially in what way they are being filled. But instead of asking myself what is happening to the way and the capacity of the state to manage the economy, provide social welfare, cater for the environment or pursue a cultural policy, I look at what is happening to the capacity of the state to control organised violence and the way it does it. This question is central to IR. Indeed, the bulk of IR is grounded on an (explicit or implicit) adherence to a Weberian definition of the state as an institution which successfully monopolises the use of legitimate organised violence. If this definition turns out to be evolving in fundamental ways, this points to the need to rethink the fundamental issues and approaches.

And indeed, the argument I want to make is that globalisation is altering the nature and meaning of the state monopoly on legitimate violence (SMLV). It is accentuating the tensions around the meaning of 'legitimacy'. The relativism implied in the idea that states can define which use of violence is 'legitimate' (and which is not) is increasingly contested both by the international society of states and in a world society of transnational actors. At the same time a profound redefinition of what it means to have a 'monopoly' of violence is going on. Increasing private ownership and allocation of the means of coercion are blurring the responsibility of states beyond their own borders and, for some states, even within them. This changed meaning and nature of the SMLV, is having profoundly diverging effects on the capacity and ways in which various states 'fill their function' of regulating violence. And as the paper concludes, it is increasing the differentiation between states. The overall conclusion then, is that globalisation is undermining two cornerstones of thinking about war and peace: that war and peace are essentially a matter of relations between states and that these states are like units. The paper argues that the private-public boundary is increasingly blurred and that states are not (even in principle) like units.

2. The SMLV and International Relations

Before moving on to the discussion about globalisation and the state's capacity and way to regulate violence, it seems important to give some indications about what it means to treat this subject. In the first place I want to underline the centrality and importance of thinking about the role of the state in regulating violence. It is not only because IR thinking broadly has been anchored in the idea that the state is defined by its monopoly on legitimate violence that the issue is important. Rather, it is because the SMLV has been seen as the solution to the fundamental problem of how to handle violence (nationally and internationally). And I want to stress that this is a perennial problem which is unlikely to disappear and that therefore thinking about what is happening to the SMLV is not merely a matter of criticising IR theory. It is important in itself. Second, I want to point out that the discussion that follows (of course and obviously) is not the first of its kind. Rather, much is drawn from existing ideas and arguments made by scholars who have asked themselves questions similar to mine. However, very unfortunately and partly because of the way that the arguments have been formulated and advanced, the discussion has tended to get stranded in two dead ends: one where scholars are pushed to place the issue of whether the present developments are historically unprecedented or not at the heart of the discussion and a second which focuses on whether they entail 'the end of the state' or not. I want to make very clear that these debates are dead ends. They miss the point. Because they are Manichean (it is either black or white), they have no way of discussing shades of grey. Yet this is precisely what the present changes are all about: it is about a changing nature and meaning of the SMLV, not about eliminating it or reformulating it from scratch. It is neither about the novelty of the changes nor about the disappearance of the state. It is about shifts (to which there are analogies in the past) in nature of the state.

1. The Continued Centrality of the SMLV

Any discussion about the SMLV at present is bound to run into two different kinds of objections which have in common that they amount to saying (in more subtle language of course) why bother discussing it? It is therefore useful to start with answering this question. I will argue below that the answer is that the SMLV (as an ideal type) remains central to (national and international) political thinking for the very good reason that we do not have any good replacements for it in spite of the well known tensions and ambiguities which surround it. And the fact of the matter is that we are far from ready to do away with it theoretically, normatively or politically and that therefore it is absolutely crucial to understand what is happening to it.

The first kind of objection facing anyone who 'bothers' with the SMLV is that it is no longer important. It might have been an important and even fundamental aspect of political organisation in the past. But in a globalised world things are different. The economy, finance, the environment, or issue centred politics are far more important for politics. The regulation of violence has become a marginal aspect of social and political organisation; it has been turned into 'an archaic and rudimentary instrument of action in a civilizing world (in the sense of Norbert Elias)'^[3] The army has become a 'zombie institution'surviving itself by way of inertia.^[4] In so many words, we are made aware that the state, sovereignty and the related claims to a SMLV are rendered obsolete by the altered and globalised nature of politics where which have lead to a 'de-politicisation of the state'or a 'disintermediation of politics'.^[5] Somehow (but how?), we are told that the SMLV has lost its significance. We no longer need to worry about it or discuss it.

The second kind of objection follows a different, and more interesting, path. It recalls the oppressive nature of states and modern states in particular to drive home the idea that worrying about what is happening to the SMLV might be superfluous. States, it is pointed out, are major sources of violence and oppression. State violence whether exercised via the theoretically perfectly 'legitimate' implementation of legal norms or the violation of these norms through torture, death squads and extra-judicial executions has been responsible for a very large share (if not the largest share) of organised violence in our century.^[6] Now, since we all know this, should we not rejoice if the role of the state in managing violence is finally diminished? And even if we don't rejoice what are we actually lamenting? The weakening of one of the main sources of oppression and suffering in modernity? Is discussing and digging into what is happening with the SMLV not simply an exercise reflecting the (unreflected and uniformed) projection of an idealised, democratic, western state onto a world where 'real states' are anything but ideal and democratic? Does it not simply amount to justifying the oppressing practices and violence of the really rather nasty structures we term states?

Both types of objections are weighty and make valid points about the SMLV in general and its role at present in particular. But before rushing off to embrace the idea that we might be better off without the constant referencing to the SMLV, it might be useful to give some consideration to why the SMLV as an ideal type for thinking about the regulation of violence has held such a central place in political thinking for such a long time. And this consideration in turn, I think, makes clear that there are also good reasons for not discarding the idea too rapidly. 't the very least (and this is what matters for this paper), there is a strong case to be made for taking interest in what is happening to it and, as I will argue, the most promising way of doing this is precisely to explore what is happening to the tensions and ambiguities that have always surrounded the idea.

There are two (opposed and very weighty) reasons for thinking that the SMLV is important.^[7] The first one is a managerial order view, which grows out from a tradition of thought which IR scholars would tend to identify as 'realist'.^[8] We need the SMLV to manage violence in society, to preserve the minimal order which is absolutely essential for social and political life. On this account violence is an inevitable part of social and political life (because of 'human nature', psychological drives to aggression, the inevitable plurality of views or something else) which will result in conflict and hence potentially in the use of force.^[9] And in order to check 'the rumblings of uncontrolled violence'where anyone can use violence against anyone else for any reason it is important that there be a superior authority which can 'successfully'impose a monopoly on the legitimate use of violence. We need the state to keep the lid on violence.

This position is usually charged with underestimating or obfuscating the suffering that violence by states entails. In part it does so by leaving wide open the question of whether or not the state is legitimate and on what basis that legitimacy is defined. This issue is assumed away and de facto legitimacy is granted from without, via the recognition of statehood (usually without much further discussion).^[10] This makes for a profound relativist dilemma: what is it that makes the state use of violence legitimate? And how far can one go in simply being silent on the use and implications of that violence?^[11] Moreover, the argument downplays the significance of state violence by 'banalising'it. It builds on an idea that the use of violence is simply an instrument among others, in continuity with other instruments of government. And this makes violence (in one form or another) simply appear as a natural part of the governance process.^[12]

These charges are usually answered with one version or another of Hobbes' argument that:

..the estate of Man can never be without some incommodity or other; and that the greatest, that in any form of Government can possibly happen to the people in generall, is scarce sensible, in respect of the miseries, and horrible calamities, that accompany a Civill warre; or that dissolute condition of masterlesse men, without subjection to Lawes, and a coërcive Power to tye their hands from rapine, and revenge...^[13]

This answer is clearly not satisfying. It just repeats the assumptions: government violence is necessary to check the 'rumblings of uncontrolled violence' which is more threatening than state violence. However, even if we admit that this is not satisfactory, does it mean that we are ready to throw the SMLV as an ideal type for checking violence in society and in politics over board?

I would tend to think that we are not. Even if we do not assume that violence is an inevitable part of politics, it clearly would be pollyanna like to assume that it could not be and has never been. ' brief excursion into contemporary no-go-areas (be they in big 'Western' cities or in the developing world) or historical equivalents such as the Violencia in Columbia are sobering in this respect.^[14] And provide plenty of ammunition for those unwilling to let go of the idea that a SMLV is still important.^[15] That is, the question of how to deal with violence and 'keep a lid on it' cannot simply be swept aside. And since the SMLV is the modern answer we have, it would seem excessively hurried to sweep it aside on the grounds that state violence can be--and has--been terrible. Rather the logical road to take seems to be to take a closer look at the ambiguities surrounding the legitimacy aspect of the state monopoly on violence and how these ambiguities are handled.

The second view is a 'liberal' view where the SMLV is important not for putting a lid on inevitable violence but to keep (avoidable) violence out of the polity. On this account violence is by no means an inescapable part of the political landscape. On the contrary, it is something which threatens to destroy that landscape. It is not the foundation of power. It is a threat to power. Violence is employed when power is weakened or eroding and it is likely to lead to further weakening and erosion.^[16] This said a SMLV still has an important role to play.^[17] It is important to protect the polity, and the power on which it rests, from inside and outside threat. This is a precondition for protecting the rights that the functioning of the polity rests on that it also confers upon its members. The SMLV is the ultimate guarantee that these will not be overhauled by outside powers or by inside opposition if for some reason either should appear.

Also this liberal argumentation is profoundly contested. Its assumptions are charged with being far too blue eyed about the nature of politics and its relationship to violence. It is also seen as far too optimistic on the role of the state in creating and protecting rights. Critics rightly point out that since the rights that states purportedly defend are not necessarily covering the right problems, nor are they necessarily covering all groups in society which are in need of protection. They might for instance do little or nothing about defending economic and social rights, or exclude chunks of the population (ethnic or religious minorities, or women) from the protection of rights they actually need. In that sense the legal system might well be a major source of oppression.^[18] It is even possible to pin point a 'paradox of violence' whereby the most violent and repressive regimes are those where violence is the least visible and the least discussed.^[19] Moreover, both history and the present day politics are ripe with cases where states do engage in massive violations even of the imperfect rights which they themselves have in theory accorded to their citizens including electoral laws, legislation on property rights, but also the right to a trial, to free movement or the right not to be tortured.

These arguments are by and large valid and justified. However, for the present discussion what matters is whether or not they justify discarding the SMLV as an unimportant and/or uninteresting subject of investigation. And, as argued above for the realist position I am not convinced that they actually do. Even if we admit that there are many problems when it comes to seeing the SMLV as the ultimate guarantee of the rights on which the polity is based and which it confers upon its members, the question remains if we can do without it and if we actually have a better alternative. And on this account, the experience of people stripped of their citizenship is an interesting reference point. Indeed, as refugees at all times have discovered rights only exist in so far as there is a superior authority to which one can refer to claim these rights (or to protest their non implementation).^[20] This might make us think of the potential for creating cosmopolitan arrangements for ensuing rights. But again, for these to be effective we ultimately need some equivalent or substitute for the state to guarantee and protect rights.^[21]

Consequently, it seems to me that there is a strong case to be made for thinking about the SMLV, not only in terms of a critique of IR theory. The fundamental issues which have placed the SMLV at the centre of (national and international) political thinking have not disappeared. On the contrary, both the issue of how to control violence and the issue of how to protect rights in the polity continue to be central political issues. And neither the fact that other things are also important nor the fact that rights are imperfect and/or imperfectly implemented can make them disappear. Rather, it should be taken as indication that it is important to remain aware that the idea of a SMLV (as an ideal typical answer to the problem of violence) is inherently imperfect. In practice and theoretically it is filled with tensions and ambiguities. This in turn is more of an indication of the importance of discussing and exploring these tensions, than of the need to rid oneself of the basic issue.

2. The importance of context and shifting meanings

For the present paper, the central question is what -- if anything - 'globalisation^[22] 'has meant for the nature and role of the SMLV as an answer to the basic problem of violence in politics and in society. And even if the most common reaction in IR has been to discard the question, there is a great deal of interesting work on the topic. However, many of the important points made in that work have been lost because of the tendency to drive the discussion into two dead ends. I hence want to argue for the importance of moving the discussion away from these dead ends to a debate which makes it possible to describe and discuss shades of grey.

The probably most common attitude in the theoretical IR community towards 'globalisation' has been to discard it as basically irrelevant for the question of violence and for the role of the state in managing violence. If it is discussed in the context of war and peace it is mostly in terms of its indirect effects. It is seen as altering the causes, means and context of war. But not the centrality, the nature or the meaning of the SMLV. Thus, Alexander Wendt justifies his state centrismBand his neglect of transnational phenomena--with the argument that the control of violence is the precondition for all other social activities and 'states are still the primary medium through which the effects of other actors on the regulation of violence are channelled into the world system'.^[23] That is, the 'great divide'^[24] between inside and outside is seen as essentially (depending on what is meant by 'primary medium') untouched. At least when it comes to the regulation of violence.

This is not to say that no interesting work has been done on the topic. On the contrary most people busying themselves with actual wars and the way they develop have been pushed to say a great deal about it. Hence, there is considerable discussion surrounding both the legitimacy of the state monopoly of legitimate violence and surrounding the evolving nature of that monopoly which the discussion below will draw extensively on and develop. From a variety of perspectives it is pointed out that the legitimacy of the state to use violence is increasingly contested. From 'above' by the international society of states by the development of what is sometimes referred to as the 'new military humanism'^[25]. But also from below by a variety of movements contesting the boundaries of existing polities and hence the legitimacy of the states ruling them.^[26]

In parallel, a rather different kind of literature makes the point that the nature of the state monopoly on violence is evolving in fundamental ways. There is an incontestable trend towards increasing the role of private actors both when it comes to providing the means of violence (through the privatisation) but arguably also when it comes to allocating these means.^[27] Indeed, not only is the production of military equipment and arms increasingly placed in private hands but so is the provision of soldiers via the creation and growing acceptance of so called private military companies.^[28] In addition to this there is an increasing acceptance that market allocation, based on the economic logic of who pays, might be allowed to replace or at least supplement state allocation, based on the political logic of who is a deserving ally, of the means of violence. This vast literature and the discussions surrounding it provides obvious and interesting food for thought for anyone with an interest in what is happening to the SMLV.

Yet, the discussion which this literature deserves has been stifled by an unfortunate tendency to drive it into two equally unpalatable dead-ends. One of these is what I would call the old-new dead end which locks the debate into a dispute over whether or not the present developments are truly new or not. The reason for which the debate gets into this corner is easy to grasp. The logic is to say that if people argue that 'globalisation'(which they claim is new) is altering the regulation of violence, then they must presumably be saying that the changes they are describing are also new. This presumption is seemingly confirmed by the wording of Kaldor's contrast between 'New Wars and Old Wars'.^[29] The result is that the discussion turns into a dispute over whether or not what we are seeing is historically unprecedented or not. The outcome is exceptionally predictable (and unexciting): most of the 'new' phenomena already existed in some form in earlier periods: There have been mercenaries, organised crime and paramilitaries as long as there has been warfare. Identity politics and ethnic exclusion are long standing goals of war. Warfare has always had a side to it which was dispersed and fragmented, hitting above all the civilian population. And so on.

However, there are several problems with this reasoning. Firstly, there is no need (even for Mary Kaldor) to claim that the developments are historically unprecedented. Globalisation (which is new) might well provoke changes to forms of regulation of violence that have historical analogies and in that sense are not new. But by focussing exclusively on these forms and their historical precedents, the discussion misses what is indeed new, and that is their link to a global context. That is, the debate fails to contextualise the forms of regulation observed. Secondly, this form of discussion simply misses the very banal and basic point that the reason we are interested in these changes at all, is that they matter for our precent conceptions and normative understandings of the relationship between violence and politics. And for this, it might not matter much that mercenaries constituted the bulk of the armed forces in renaissance Italy or in absolutist France. Neither context was marked by any great concern for individual rights or for limiting the impact of violence in politics and society.

The second dead end into which the discussion tends to wander is the discussion about whether the state is disappearing or not. 'gain there are understandable reasons for why so many debaters are caught in this trap. If indeed it is the case that we define the state by its monopoly of legitimate violence, then, might we not also conclude that those who argue that the SMLV is being challenged are actually arguing that the state is disappearing? The stage is set for a Don Quichotte exercise whereby one can seemingly win the argument by showing that the state is still around and influential in regulating violence.

But besides not being particularly productive or interesting, this exercise has two major problems. First it assumes that the relationship between the SMLV and the state is a fixed one; that the SMLV is quintessential to statehood. But this gives the Weberian definition of statehood too much centrality. 's persuasively argued and shown by Reus-Smit the nature of statehood evolves over time.^[30] And there is therefore every reason to believe that the relationship between statehood and violence might also vary (in time and space). Second, the exercise assumes that a fixed meaning is attached to the SMLV. But this might not be the case. 's soon as we scratch the surface of the ideal typical SMLV, we become aware that it can be given a great variety of meanings. The limits to which the 'private' can encroach on the public monopoly and the basis of legitimacy are both exceedingly fuzzy and unclear. But any exploration of this fuzziness, and of changes within it, are excluded from the outset by a debate depicting the world in purely Manichean terms.

To sum up, the SMLV is still a central part of our thinking about violence and politics. It continues to be central to our understanding of how to manage violence as well as how to guarantee rights. Hence there are good reasons to think about if and how it is influenced by globalisation, in what direction and what the implications are. There is a considerable body of thought on this issue. However, it has been difficult to get at the core issue because of the way much of the debate has been framed. It has decontextualised change, assumed a fixed relationship between the SMLV and statehood, and assigned SMLV a fixed meaning. This has made it difficult to describe the evolving nature of the SMLV and to assess the implications of the changes that are taking place.

In order to move forward in the debate, it therefore seems useful to take the ambiguities and variability of the SMLV as the point of departure and to consider how globalisation affects these. I propose to do this by focussing on how globalisation influences the ambiguities surrounding the legitimacy and the monopoly on violence granted to and claimed by the state. These tensions and ambiguities are staple goods in the IR literature. With regard to legitimacy the central issue is how to deal with the relativism implied in the idea that the state defines what is legitimate and what to do when the boundaries and/or nature of the state is contested. With regard to the monopoly the central issue is what degree of state control of violence is implied by the notion of monopoly. And, as I will show below, on both accounts globalisation has had an important impact. It has made legitimacy more conditional and it has increased the weight of private actors in the control of violence.

3. The Increasingly Contested and Conditional Legitimacy of the State Monopoly on Violence

The first set of changes implied by globalisation I want to discuss are those related to legitimacy. I will argue that globalisation has had the effect of making legitimacy more central by making it more contested and conditional. Contested because the dislocation (or deterritorialization) of politics has made the correspondence between the jurisdiction of the state and the polity more contested. Hence undermining the fundamental idea that the state can claim legitimacy in relation to a polity. It is more conditional because by the same token there is increasing pressure on other states to take the substance of legitimacy seriously and hence to deny the legitimacy of states that do not respect some set of rules. Both of these changes correspond to a deepening of tensions which have always been part of the idea that the state monopoly on violence is legitimate. 's I will begin by showing, precisely these tensions make it difficult to take the road of simply letting the issue of legitimacydrop out of the picture.

1. The significance of tensions in legitimacy

The question of how important 'legitimacy' is for the control of violence and for the definition of the state is a vexed one. Although it figures prominently in the Weberian state definition, many authors prefer to simply let it drop out. Thus, Tilly prefers to refer to 'controlling the principal means of coercion within a given territory'^[31] and Giddens to the 'direct control of the means of internal and external violence'.^[32] The reason for the hesitance to refer to legitimacy is that there are problematic normative implications of doing so. The glaringly obvious question that this reference raises is what legitimacy, defined by whom.

The conventional way of dealing with the question of legitimacy in the IR literature, and an answer which is usually seen as linked to the Westphalian peace, is that each state (or polity) is allowed to decide for itself what is legitimate and the basis of sovereignty is that it should be free to do so without outside interference. This is a way of limiting conflict in view of the inevitable plurality of views in the world. However, the implication is that the substance of 'legitimacy' drops out of the picture of international relations. ' state is legitimate because it is recognised as a state by the international society of states. The substance of legitimacy is dismissed as a purely internal matter and the reference to it hence becomes vacuous.

However, this move creates two kinds of tensions which have always made it difficult to drop the issue altogether. The first, tension arises when several authorities are making claims to statehood on the same territory. Classically this is the case in regime changes (revolutions, coups, or restorations) where competing authorities claim statehood on the same territory and in secessions when a part of a state claims its independence from the central state. A second tension inherent in the move is that it demands a relativism which is not always lived up to. The international society of states has repeatedly denied the legitimacy of some states. And indeed, for some IR scholars, sovereignty has always been conditional.^[33] The development of international law where the individual (rather than the state) is the subject enshrines this conditionality.

Through both these tensions the question of legitimacy sneaks in through the backdoor. They make the substance of 'legitimacy' (and not only the capacity to monopolise the means of coercion) central for the recognition of statehood. In the former case 'legitimacy' is determining for which contested authority is recognised and in the latter denial, recognition and contestation is conditional upon legitimate behaviour. The point I want to argue in this section is that both tensions have been deepened in the process of globalisation. Globalisation has increased the contestation of state legitimacy by making the boundaries and foundations of the polity increasingly blurred. It has also increased the relativist conundrum by increasing the pressure on international society to take the substance of legitimacy more seriously.

2. The dislocation of politics and contestation of legitimacy

Second, even if there is no immediate personal reason to feel concern, people may well enlarge the polity with which they feel concern and participate in. The globalisation of communication and the media certainly plays an important role in this process. The media brings the politics of a much larger polity in peoples daily lives. It facilitates the spread of ideas and values. This might work in favour of rallying around specific values or ideas. Thus, one might argue with Habermas that since the defeat of fascism after the second world war there is an attachment to human rights and democracy and a belief that these are principles which demand universal respect,^[39] or join Halliday in pointing out that 'enlightenment rationalism may look a bit tired from the metropolitan vantage point: it looks very different if you are being tortured or beaten in jail, or if your male relatives are forcing you to submit to their will, or denying you an education'.^[40] However, inversely it might fuel opposition on behalf of those who do not want to take part in or be influenced by a transnationally defined political space; possibly because they think of themselves as disadvantaged by the rules of the game in this space.^[41] Thus, also those who strive to carve out space for their own identities who increasingly define their politics in relation to a transnational sphere. And this is what has made a large number of authors draw links between the rise of 'identity politics'and globalisation.^[42]

Third, the discrepancy between the state and the boundaries of the polity is accentuated by the enlargement of the political agenda to issues which are inherently transnational. Indeed, there has been a tendency to expand the sphere of the political and 'previously de-politicized areas of decision-making now find themselves politicized'.^[43] Thus, issues such as ecology, science, food safety or the gender relations have been placed solidly on the political agenda. And to the extent that the polity is defined in relationship to these issues it tends to become a polity which has no a priori reason to coincide with state boundaries.

And finally, to avoid giving an overly voluntaristic picture of what drives the dislocation of politics it is important to underline that it is not only a matter of changing self-definitions, universal values or expanding political agendas. Rather, the expansion is just as often imposed by the de facto linking up of social spaces through what one might term structural changes. Indeed, there is a discrepancy between the polity concerned by a (political) decision (or development) and the location of the authority making that decision (or setting the development in motion).^[44] For example, the decision about what to do with Bulgaria's nuclear plants regards not only the Bulgarian population. ' change in US interests rates influences financial markets across the world. ' change in the US steel industry affects everyone linked to it directly or indirectly. And the development of a vocal issue centred movement (e.g. environmentalists, feminists, or 'TT'C) influences not only those who created it but everyone concerned as the movements alter the image of the issue, the politics surrounding it and the regulation of it.^[45] The result is that often there is no opting out.^[46] ' state (or its citizens) cannot simply declare that it does not want to be affected by e.g. a nuclear disaster, developments in international financial markets or reconceptualizations of what the is a legitimate use of violence.

This is all the more true since the authorities which make the decisions are increasingly private and hence escape the logic of state control. Globalisation is intimately linked to (and driven by) a diffusion of authority. Indeed, the pressure to open up and privatise economies across the world has greatly increased the importance of the decisions made by private business, banks, rating agencies, accountants or financial market operators.^[47] And in turn these actors often exercise a private pendant to the multilateral conditionality imposed by international organisations such as the IMF or the World Bank.^[48] In addition to this, there has been a spectacular 'NGOization'of politics since the late 1980s. The end of the cold war and decreasing willingness of outside (state) allies to finance and get involved in conflicts is part of the explanation for this as states have increasingly withdrawn both financially and politically and instead channelled increasing amounts of their aid and involvement through NGOs which have become correspondingly more influential.^[49] The fact that NGOs carry much of the expanded political agenda is another.^[50] However, this growing clout of private actors is particularly damaging to the legitimacy claims of states. They can mostly not be identified and rarely held responsible for their choices. Their impact is that of an impersonal effect on structures and the conditions of choice which is profound but difficult to pin down.^[51] Private actors thus have what Gill refers to a constitutional powers, that is the possibility of setting the boundaries and the rules of the game within which politics is taking place^[52]. The question is indeed 'who elected the bankers'^[53], the media, and the NGOs.

What this amounts to then is a situation where the legitimacy of the state in general is increasingly up for grabs. A growing number of actors can--and do--contest both the borders of the polity and the content of the state policies within them. They play an important part in shaping national policy processes and also in the creation of policy processes which are de facto transnational. The consequence is that a simple image of legitimacy as being defined by the state is very problematic. In Beck's wording: 'where the dominant political image of modernity was Leviathan, the moral standing of 'national' powers and superpowers will, for the future, be captured in the picture of Lemuel Gulliver, waking from an unthinking sleep to find himself tethered by innumerable tiny bonds'.^[54]

This state of affairs is ripe with implications also for the legitimate use and control of force. For one it increasingly makes the question of the legitimacy of any state's control and use of violence transnational. It is no longer the preserve of the national polity and/or the international society of states. Rather, as argued by Shaw 'wars [and one might say more broadly the use and control of violence] lose their spatial location, and, through their telegeneic (re-)presentation, become political crises in which questions of justice and intervention must also be publicly discussed and decided in the far-off centres of global civil society'.^[55] For two, and relatedly, a further implication is that the legitimacy of the state use of violence can be challenged, contested or confirmed by a correspondingly wide range of actors and movements in a correspondingly transnationalised space. Thus, Kurdish terrorism on the territory of the German republic becomes a way of contesting the legitimacy of the Turkish state and its use of violence, lobbying by 'mnesty International in the UN a way of contesting the legitimacy of the Myanmar governments use of violence against the Karen, and blocking Shell stations in Germany a way of contesting the Nigerian government's violent treatment of the Ibo. The result is a blurring of the internal and external as well as the private and public contestation of the monopolisation and control over violence. And relatedly a pressure to blur the lines separating crime fighting from war fighting.^[56] But above all, there is an imposed necessity to think legitimacy in broadly transnational terms.

3. 'ccentuating the relativist conundrum in international society

Globalisation has not only made legitimacy more contested, it has also accentuated the difficulty of international society to close its eyes on the relativism to which it (at least according to much of IR theory) adheres. 's already pointed out, this has always been difficult in practice. There have been limits to what degrees of oppression and violence the international society of states has been willing to accept as 'legitimate'.^[57] There are causes which make justify intervention and war even in classical thinking (including notably genocide). Of course the temptation to protest and intervene and even deny statehood has been all the stronger as the polity concerned has been of immediate concern. And as I will argue the deterritorialisation of politics (just discussed) has increased the pressure on states to place the substance of legitimacy on the international political agenda. This pressure has been all the more effective as it has come in pair with the argument that the issue is not between intervening and not intervening in some other state's sovereign affairs; it is a choice between forms of intervention. The consequence has been that legitimacy is not only increasingly transnational but also increasingly conditional on the approval of other states.

Indeed, the dislocation of politics just discussed has not left states unaffected. On the contrary, they have been under pressure to react to it and to integrate also legitimacy questions on their own (international) political agenda. 'rguably much of the transnational politisation runs through states.^[58] In part the pressure to move in this direction come from the constitution of 'epistemic communities'of experts working on similar issues from a similar perspective often across state boundaries.^[59] In part, the pressure comes from advocacy groups of various forms, including private business and NGOs. And finally, it comes through citizens who do not necessarily lobby directly, but still have firm beliefs about what is legitimate or not for other states to do. They find it important that children should not be used as soldiers in Columbia or Falungong members tortured in China. In different ways, these groups place pressure on states to place the substance of the legitimacy of other states on the centre of their international agenda.

Concretely translated this means that there is pressure on states to in various ways intervene with the definition of what is legitimate in other states. 't a very fundamental level this is visible in the expansion of international law, and in particular the privatisation of human rights law, which is setting increasingly strict boundaries around what is actually legitimate state behaviour and which also gives an important formal basis for condemning certain states and asking for policy changes within them. But it is also visible in the use of sanctions such as the present EU Sanctions against Zimbabwe or political conditionality imposed on loans and aid granted directly or through multilateral arrangements such as the IMF and the World Bank. Finally, there is growing pressure on states to control the activities of their own nationals and see that they do not encourage or benefit from 'illegitimate' practices in other countries. Hence, there is a strong movement for pressuring firms to become 'good citizens' and shoulder their 'social responsibilities'.^[60]

The pressure on states to hence make the legitimacy of other states an important determinant of their foreign policies has been all the more important as it is argued to be something which states anyway do interfere with. In part this point is made through the argument of historical responsibility. Many states, their conflicts and ways of claiming legitimate control over violence is shaped by international factors and concretely by the cold war. In Buzan's wording the international system is increasingly penetrating.^[61] The situation in Mozambique, Angola, Columbia or Afghanistan are hardly understandable without the past interferences of other states. This leads on to the deeper point, that the choice is not between meddling or not with the (internal question of legitimacy); but it is a matter of choosing how to meddle. Indeed, at the most basic because external recognition is so important for statehood, there is no way around the fact that granting statehood is already meddling: it grants at least the beginnings of legitimacy. Moreover, the increasing linkages of social space in its various form makes it even more illusory to claim non-interference. Hence non-action becomes an choice of non-intervention. Just as there is no opting out from the international system, there is no opting out of being involved.^[62]

For the present discussion this matters because it drives home the point that legitimacy is increasingly at the centre also for the international society of states. Granting, contesting and denying legitimacy has been made even more important by globalisation as it increases pressures on states to make it an explicit part of their policy-making. It bears emphasising, that this is all the more significant as external recognition has become quintessential to statehood. Indeed, Tilly's monumental overview of state-making concludes on a 'drift from internal to external state building'that is the increased importance of access and handling of external (as opposed to internal) capital and means of coercion.^[63] Similarly, most studies of states in Africa (and more generally the developing world) concur on the importance of the international granting of statehood rather than its internal constitution.^[64]

Stressing that legitimacy is increasingly an international and transnational affair is not to imply that everyone is equal. Not all abuses of state violence provoke international reaction and condemnation. In fact many would think that far from enough do.^[65] And reactions, when they are there, vary greatly. But this does not alter the point argued in this section: that legitimacy is increasingly conditional to the approval of other states. The point is not that this approval follows some well defined rules equally applied to everyone. Nor is it that international politics has become more 'moral'(even Blair certainly tried to make it so). Rather, the point is that globalisation has increased the pressure on states to take a position on what is 'legitimate'use and control over violence in other states and often to follow up these positions with practical measures. That is, a state qua state can not count on its internal use of violence automatically being recognised as legitimate (as pointed out above in reality it probably never could). It has to consider the legitimacy of its monopoly on violence in international terms and count on the fact that the recognition of its legitimacy might very well be conditional upon following specific rules.

To sum up then, this section has argued that globalisation has altered the understanding of the legitimacy of the state monopoly on violence fundamentally. It has drawn the substance of legitimacy to the centre of the agenda by accentuating two conventional tensions which make it hard to ignore the issue. It has increased the contestation of the legitimacy of the state monopoly on violence by a variety of private actors, and it has made the legitimacy conditional to the approval of other states.

4. The State Monopoly on Legitimate Violence Reinterpreted

The second set of changes to the SMLV I want to discuss has to do with the monopoly of the state on legitimate violence. Just as with legitimacy, globalisation accentuated the lack of clarity that surrounds the meaning of monopoly. Indeed, it has always been somewhat fuzzy how much private involvement a 'monopoly'allows for and how far the monopoly claim actually extends. In this section I will argue that globalisation has accentuated these tensions. Both when it comes to the control over the means of violence and to their allocation, increasing space has been made for the private sector. In fact, as I will argue in the last section, the reinterpretation has been so far reaching that it impinges on the decision making authority of states, particularly in the international sphere. And for some states it is even questionable whether it is merely a matter of reinterpreting or whether it is possible to talk in any meaningful way about a monopoly of the state on legitimate violence.

1. Multiple Monopolies

The first thing to be clear on is that there has always been a considerable degree of ambiguity about what a state 'monopoly on legitimate violence'referred to. It is an ideal type to solve some of the basic problems of dealing with violence. But what does it actually mean to have a monopoly and how is that meaning given practical substance? Even the most cursory look at the ways in which states have interpreted monopoly and have tried to control violence, makes it amply clear that statehood and claims to monopolise the use of legitimate violence has meant very different things historically. Both the degree to which states have accepted private involvement in ownership and allocation have varied greatly and the degree to which states have extended their monopoly claim to cover also the realm extending beyond their own borders.

First of all, there have been varying degrees of acceptance of the presence and weight of the private in the state monopoly. In fact, to discuss the question it helps to disentangle the notion of monopoly a bit and make three distinctions depending on what aspect of the control over violence we are talking about. 1) There is the question of controlling what violence is used for, or the decision making authority. 2) There is the question of ownership over the means of coercion and finally 3) there is the issue of how the means should be allocated. Of these three aspects of control, the decision making authority is clearly the central one. It is the decision making authority of the state which matters for it to make an effective claim to state control over violence. If it can not decide for what purpose and against whom to deploy violence, it cannot possibly claim to control (let alone monopolise) the means of violence in any meaningful way. On the other two aspects of the control of violence there has been considerable flexibility. State decision making authority can (has and does) coexist with various degrees of private sector involvement in the control over the allocation and in the ownership of the means of coercion. For example, a state may lease troops to an ally (and hence rely on private allocation) or itself use privateers for its armed forces (and hence rely on privately owned coercion). It was only in the course of the 19^th that states began to give the role of private actors the kind restrictive interpretation which has become standard.^[66] And even in the 20^th century private allocation and ownership has continued to play a considerable role in the regulation of violence.

Second, it is important to emphasise that there is an ambiguity about which sphere is being covered by the state's claim to monopolise violence. Indeed, in the course of the 19^th Century a profound transformation of sovereignty took place as states extended the claims of monopoly control over violence to cover also the international sphere and not only their own jurisdiction. Prior to that, private violence in the international sphere was considered as precisely that: private. That is states did not have to take responsibility for it. They instead tended to use and draw benefits from the private violence when it suited them and shirk responsibility when it did not. However, eventually this practice had a number of 'unintended consequences'(getting states entangled in conflicts with each other as well as with private authorities) which eventually made the 'collective of state rulers'change their practices. Consequently, from the late 19^th Century onwards:

Traditional states were transformed into a system of national states that held one another accountable for any individual violence emanating from their respective territories. Sovereignty was redefined such that the state not only claimed ultimate authority within its jurisdiction, defined in geographic terms, but accepted responsibility for transborder violence emanating from its territory.^[67]

This shows that when trying to capture the changes in the monopoly of the state on legitimate violence, it becomes fundamental to clarify which aspect of this monopoly is being discussed and what monopoly claim. Moreover, it is yet another reminder of the importance of not falling prey to the Manichean temptation of rushing off to proclaim the 'end of the state monopoly on violence'. Rather, what clearly needs to be considered is how significant the various changes are for what part of the monopoly and then of course what the implication of this is for the 'system of national states'.

2. Privatising Ownership and 'llocation of the Means of Coercion

Therefore I want to start by discussing the impact of globalisation on the ownership and the allocation of the means of coercion internationally. I want to argue that globalisation has contributed to shape the clear (and related) moves to privatise in both spheres in particular by putting pressure on government to cut their defense budgets but also by retuning market allocation to the honourable status it had largely lost in the course of the twentieth century.

The private sector of course never entirely lost its role in the production and control over the means of coercion. Even at the height of the cold war private firms played an important role in arms production and the market (albeit restrained and regulated) continued to play an important role in the allocation of the means of coercion. Moreover, mercenaries, that is private military personnel fighting for a pay did not disappear. Companies selling fighting services (e.g. the Nepalese Ghurka with roots in UK imperial policies) as well as mercenaries have existed all along.^[68] This said, the over past decades we have witnessed a marked increase in the weight of private actors. There has been a trend to privatise the state owned defense industries in most countries as well as to press private companies to become less reliant on state subsidies and more market oriented. They are no longer allowed to rely only (or even mainly) on state contracts but are on the contrary encouraged to look for profitability on market terms.^[69] Even the involvement of the state increasingly follows a commercial logic and state institutions are turned into commercial ventures. Thus, in the UK e.g., the Ministry of Defence procurement executive has been replaced with an independent, commercially ruled, Defence Procurement Agency.^[70] The most discussed and dramatic change from this perspective is no doubt the rise (and acceptance) of Private Military Companies (PMCs) which sell military services on a commercial basis to both public and private clients.^[71] 'n indication of the magnitude is that a compilation of available information on mercenary activity in Africa from the 1950s onwards shows 15 entries for the 40 years spanning 1950-1989, and 65 for the period 1990-98.^[72]

'Globalisation' alone cannot account for this overall privatisation trend. Regime changes in Central and Eastern Europe clearly play a role in the privatisation of the ownership and control of the means of coercion. Similarly, it is not uncommon to point to the importance of the 'revolution of military affairs'(RM') and the related changing nature of warfare as an important determinant of privatisation. It is argued that the RM' has increased needs for private sector specialists and consultants as well as the role of off-the-shelf technology.^[73] And this might well be true, even if I my inclination would be to think that the relationship goes both ways: the private sector is also dependent on public sector subsidies and contracts.^[74]

This said, 'globalisation' is an important part of the explanation. It has fuelled both the demand and the supply of private military services and contributed to justify the reliance on markets for the allocation. On the supply side, it has had the dual effect of pressuring governments to reduce defence expenditures and to facilitate the creation of private alternatives. Indeed, globalisation has pushed governments to follow a strategy of 'embedded financial orthodoxy'.^[75] Although (as the recently passed mammoth size US defence budget, reminds us) this pressure is neither absolute, nor equal for all, nor impossible to circumvent, it certainly has pushed most governments to look for cheaper (often private) alternatives. At the same time globalisation has made the provision of these private alternatives easier. In particular, the deregulation and integration of financial markets plays an important role in much the same way as it does for internationally oriented firms in any sector. It greatly facilitates international business by making it easier to organise transfers of services, by making it more profitable because of the possibilities it open to minimise taxation and other regulatory costs, and finally it makes it easier to keep activities non-transparent (through the reliance on tax-havens^[76]). Because of the political sensitivity of the issue, the last point is of particular significance for firms involved in the trade of means of coercion.

Second globalisation has played a significant role in fuelling the demand for private services. Indeed, central to this demand is the 'gap'between the security needs of various actors and the capacity of states to meet these needs.^[77] In particular the spread of 'internal'wars and ethnic conflict has pushed a both public and private actors to rely increasingly on private services. And here 'globalisation'comes in at least two ways. The first, is that it has been significant in prompting the erosion of state structures which epitomises the internal war situations. The financial pressures on states via financial market actors and/or via good governance conditionality has been intended to break down the clientelistic practices of weak states. The drama is that it has worked well, but that the result has not necessarily been more effective (democratic and non-clientelistic) state structures but a privatisation, often accompanied by violent contestation, of state authority.^[78] The second way it comes is through the shortage of foreign currency that is also a consequence of financial market pressure. This has made aid distribution and the operations of foreign firms a very important source of revenue. Thus the politicisation of these activities and the pressure on them ranging from 'taxation'to outright extortion and kidnappings has increased greatly.^[79] This has pushed both aid organisations and firms to rely on private protection or no operate in conflict areas.^[80] And if we ask the industry itself it is precisely in protecting private firms and aid organisations they see the best future prospects.^[81]

Finally globalisation has played a role in shifting the attitude towards private allocation. And it has done so essentially by consolidating what some have called a 'market civilisation'or alternatively a consensus around the virtues of neo-liberal policies.^[82] Indeed, the idea that the market and hence money should be determining for the allocation of the means of coercion is by no means self-evident. Indeed, control over the allocation is often considered essential, as it decides who has access on what terms to what coercive capabilities. And this in turn is of central importance for preserving the decision making authority of the state over how these means should be used. 's illustrated by the efforts to limit weapons proliferation or to protect military technology this remains an important concern and it is important not to overstate the shifts that have taken place. However, it is also true that a in pair with the 'neo-liberal revolution'it has become increasingly accepted to argue that market allocation might after all be more effective and justified and that certainly it might be preferable to the ineffectiveness of the (rent-seeking and corrupt) state.^[83] In the words of an observer of the present developments: 'if any policy would claim popularity, at least among the world elites, it would certainly be privatisation.'^[84] This is reflected both in the pressure on governments and states to accept/promote market allocation and in the (absence of) efforts to regulate the market.

In sum, the increased role of private actors in the allocation and ownership of the means of production is not solely accounted for by globalisation. This said, globalisation has been an important development fuelling it. It has facilitated the expansion of an international private market and the development of the firms acting within it. It has been part of the reasons for the increased demand for private services and it has contributed to making private allocation more acceptable.

3. Challenges to the Decision-Making Authority of States?

Even if globalisation has contributed to a trend whereby the weight of private actors in the allocation and ownership of the means of violence has increased, this still leaves fully open the question about the extent to which this is actually affecting the decision making authority of states and hence the core of the state monopoly on legitimate violence. Indeed, it is only if this decision making authority seeps away from the state and is taken over by other authorities that we can talk about an erosion of the state monopoly on legitimate violence.^[85] 'gain however, it is important to look at shades of grey and variety. And as I will argue in this section, even if there are considerable signs that state decision making authority is being impinged upon, it is not clear that all monopoly claims are seriously affected. However, it does emerge that in the international sphere, it is no longer clear that states cannot shirk the responsibility for non-state violence by referring to it as a private act. And also that for many states privatisation has worked to empower non-state actors hence undermining the monopoly claims of the state even within the own territorial boundaries.

First, the privatisation of allocation and of ownership of the means of coercion is having an impact on the control of states over how violence is used in conflict situations. Indeed, the privatisation of ownership and allocation is making it increasingly hard for states to control that their decision making authority can actually be translated into actual military operations. There is always the risk that the private firms will not fulfill their contracts, or not do so fully.^[86] Or that they will shift sides in the middle of a conflict or simply run away when the situation becomes to unpleasant.^[87] Or, again finally that they may turn against their employers and work for their overthrow by a ruler more sensitive to their own concern.^[88] That is, Machiavelli's prime concern with how to deal with the 'whores of war'are reappearing at full and posing very similar problems to the ones that he spent so much time on.

Second, privatisation of ownership and allocation is changing the capacity of states to decide who is entitled to use what kind of force. Indeed, by definition private firms are in the business to make a profit. For some firms this might require keeping a good reputation for selling only to respectable clients.^[89] This said, there is no monitoring system in place to check firms. The notion of respectable client is highly circumspect in internal war situations. And it is therefore hardly surprising to find numerous examples of firms selling services to non-state organisations including rebel groups, extractive firms, or outright organised crime (involved e.g. in drug trading, human trafficking, or trade in illegally extracted diamonds).^[90] Moreover, what kind of force can be used also becomes more difficult to check. The firms have an interest in selling what they control.^[91] Since PMCs also tend to sell consultancy and training services, they are well placed to contribute to the definition of the security needs of a state.^[92] Lastly, the very fact that private firms increase and improve the level of armaments in some states lead other states to upgrade and hence initiates a spiral of upgrading armaments, at times of conflict fuelled by the fact that private firms sell to both sides.^[93]

Third, privatisation is affecting the way that decision making authority is exercised in the state and possibly more widely the structure of state institutions. In states are important providers of private military services the consequence of privatisation tends to be one of de-politicising the issue by virtue of moving the question of 'what kind of use of force is being used where by what nationals'out of the public arena of debate. Private firms (unlike governments) do not need political approval. And indeed, the wish to circumvent political debates about intervention in far off places and the related fear of seeing body bags coming back is seen as an important part of why states have been so willing to use private forms of intervention.^[94] Similarly, in states which buy the private military services the balance between political actors is shifted. The government can rid itself of the costly and (in view of the propensity of the officers to make or participate in coups) politically risky need to pay for a national army.^[95] Moreover, the possibility of relying on outside means of coercion (and finance) reduces the need for the state to engage in the kind of institution and state building which Tilly argued forced governments to engage with their societies in a process which led to the 'forging of mutual constraints between rulers and ruled'that was so fundamental for the civilianisation of modern Western states.^[96]

Finally, there are clear signs that the issue of who is held accountable for violence is becoming increasingly blurred.^[97] For one, governments are unclear on what responsibility they have for the actions of their nationals as epitomised by the general confusion surrounding the status of mercenaries. They are denied the various protections granted to soldiers in the Geneva Convention, and therefore do not have prisoners of war status, but it also silent onh the question of what the alternative is.^[98] To complicate the matter further, mercenaries are often involved as 'corporate entities'and it is very hard to hold governments responsible for the actions of firms. Not least because the nationality of these is hard to determine: the headquarters will often be placed in a tax-haven location even when the firm has strong and obvious links with a state. This lack of clarity about accountability is essential: the system of national states is characterised by the fact that states actually do take the responsibility for acts of violence beyond their borders. Yet, because of the increasing privatisation it is no longer clear that this is the case. Indeed, states can and do now deny responsibility for violence in the international sphere by pointing out that it is private.

So to sum up, globalisation has contributed to a decisive increase in the weight of private ownership of the means of coercion as well as in the role played by private (market) allocation of these means. This privatisation has itself impinged on the decision making authority of states. It has altered state control over how violence is used, by whom and in what form it is used, which state institutions make decisions about its use and finally who is ultimately accountable for the way violence is used.

The question is of course what this entails for the state monopoly on legitimate violence. I have argued that it definitely amounts to a redrawing of the public-private boundary. I have also pointed out that it has seriously undermines 'monopoly'of states in the international spheres. And I have indicated that for some states also it has meant increasing difficulties to defend the monopoly claim even at home. The question is how much practices can change and yet remain consistent with the a state monopoly on legitimate violence. From Thompson's perspective, they have not changed enough to be inconsistent since states still take responsibility for international violence.^[99] However, as argued above, the increasing weight of mercenaries particularly through the growth of private military companies, makes this assertion contestable. But is this shift sufficient to declare that we are now in a 'post-sovereign system' where wars have become 'post-modern'^[100]? The question is clearly important and calls for much more careful research and consideration.

5. Conclusion: Increased Differentiation Among States and Obliged Reference to Private Actors

So far I have discussed the way globalisation has influenced the meaning of the SMLV in general. And since the SMLV exists as an ideal type and fundamental institution in the international system, this way of proceeding is justified and important. Moreover, I would certainly claim that the shifts that have been discussed and described above are of a general nature and that they do influence all states in the international system. However, as has already been repeatedly emphasised in the preceding discussion, they influence states in very different ways. And in guise of conclusion I want to make a point out of this difference.

First, the increased centrality, contestation and conditionality of legitimacy has a profoundly diverging meaning for different states. I have argued that states, in virtue of being states, cannot settle what is a legitimate use of force and what is not. Other states and transnational groups participate in the decision. Yet, they do so on very unequal terms. While some states and groups have become what one might term legitimacy makers, others are transformed into legitimacy takers. While the first can define for themselves and others what the substance of legitimacy is, the others are pushed either to defend their own conception or (more likely) to adjust in one way or another. Thus inclusion of the substance of legitimacy, and the military humanism related to it, which 'is a free choice for some descends as cruel fate upon others'.^[101] Clearly one can still read this development in terms of the emergence of 'cosmopolitan' checks on the use of violence^[102] or as the emergence of a 'global state' which is being established through (an unfinished) revolution.^[103] However, there can be no doubt that this cosmopolitan community or embryonic state (as most communities and states) is a highly unequal one.

Similarly, the increased clout of private actors, although affecting all states, is accentuating the differentiation among them. For some states, its impact is limited to the changes it entails for the regulation of violence in the international sphere. Privatisation might make it more strenuous to determine what means of coercion are used by whom. It is likely to make it more difficult to ascertain who is to be held accountable for the use of violence internationally both because it shifts the weight of decision making institutions and because the accountability of private actors is inherently problematic. However, there is no real challenge to the state monopoly on violence within the territorial boundaries of the state. For other states however, privatisation has meant that their monopoly on the use of legitimate violence has significantly eroded or even disappeared. Indeed, some states are no more than one of many actors competing to regulate violence. 't the most extreme, violence becomes a resource in itself because it allows looting of civilians and/or siphoning off resources from humanitarian organisations. Wars turn into 'complex emergencies'where states are but one of many competitors in a 'development security complex'.^[104] One might describe this role of the state as 'mimicking war lords'and call it an 'innovative strategy'.^[105] But the bottom line is that it is a strategy where private authorities are on an equal footing with the state when it comes to controlling violence on the national territory.

It seems to me that this dual differentiation of states through their (non)say in defining legitimacy and their (non)control over violence is already a central part of the academic and policy-making discussions. In neither is a state simply a state. 'Rogue states'(behaving in illegitimate ways) and 'failed states'(unable to control violence in their territories) figure prominently in both. Yet, there is a curious gap between the readiness to add adjectives to states and the unwillingness to explore the implications of doing so for the fundamental institutions of international relations, including the state monopoly on legitimate violence. In this paper, I cannot claim to have explored these implications fully. But I hope that the paper has been clear and persuasive in making the points that it is important to do so. Moreover, even a preliminary investigation, such as this one, can conclude on three things: that legitimate control over violence is increasingly transnationally constituted, that private actors have an important (and growing) role in regulating violence, and this has profound (and diverging) implications for the relationship between the SMLV and statehood and possibly more broadly for the organisation of the international system of national states.

References

Adams, James (1998) The Next World War: Computers are Weapons and the Front Line Is Everywhere, New York: Simon and Schuster.

Adams, Thomas K. (1999) 'The New Mercenaries and the Privatization of Conflict', Parameters 29, 2, pp. 103-116.

Andreas, Peter and Richard Price (2001) 'From War Fighting to Crime Fighting: Transforming the 'merican National Security State', International Studies Review 3, 3, pp. 31-52.

Angoustures, 'line and Valérie Pascal (1993) 'Diasporas et financement des conflits', in François Jean and Jean-Christophe Rufin, eds, Économie des guerres civiles, Paris: Hachette, pp. 495-543.

Appadurai, 'rjun (1998) 'Dead Certainty: Ethnic Violence in the Era of Globalization', Public Culture 10, 2, pp. 225-247.

Arendt, Hannah (1969) On Violence, New York: ' Harvest Book Harcourt, Brace &World, Inc.

--- (1979/1951) The Origins of Totalitarianism, San Diego et al.: ' Harvest Book. Harcourt Brace & Company.

Arnold, Guy (1999) Mercenaries. The Scourge of the Third World, London: Macmillan.

Avant, Deoborah (2000) 'From Mercenaries to Citzen 'rmies: Explaining Change in the Practice of War', International Organization 54, 1, pp. 41-72.

Ayoob, Mohammed (1992) 'The Security Predicament of the Third World State: Reflections on State-Making in a Comparative Perspective', in Brian L. Job, ed., The Insecurity Dilemma: National Security of Third World States, Boulder & London: Lynne Rienner Publisher, pp. 63-80.

Badie, Bertrand, Chrisitian Coulon, Bernard Coubertafond, Paul Dumont, et al., eds (1987) Contestations en Pays Islamiques, Paris: Centre des Hautes Etudes sur l'Afrique et l'Asie Modernes.

Bauman, Zygmunt (1989) Modernity and the Holocaust, Ithaca: Cornell University Press.

--- (1998) Globalization: The Human Consequences, Oxford: Polity Press.

Bayart, Jean François, Stephen Ellis and Béatrice Hibou (1997) La criminalisation de l'Etat en 'frique, Paris: Editions Complexe.

Beck, Ulrich (1997) The Reinvention of Politics: Rethinking Modernity in the Global Social Order, Cambridge: Polity Press.

Beck, Ulrich G. (1998) 'Wie wird Demokratie im Zeitalter der Globalisierung möglich?', in Ulrich Beck, ed., Politik der Globalisierung, Frankfurt a/M: Suhrkamp, pp. 7-67.

--- (1999) 'Über den postnationalen Krieg', Blätter für deutsche und internationale Politik, 8, pp. 984-990.

--- (2000a) Die Zukunft von 'rbeit und Demokratie, Frankfurt: Suhrkamp.

--- (2000b) What is Globalization?, Oxford: Blackwell.

Bozarslan, Hamit (1993) 'Kurdistan: économie de guerre, économie dans la guerre', in François Jean and Jean-Christophe Rufin, eds, Économie des guerres civiles, Paris: Hachette, pp. 105-146.

Brauer, Jürgen (1998) ''n Economic Perspective on Mercernaries: Military Companies and the Privatisation of Force', Cambridge Review of International 'ffairs XIII, 1, pp. 130-146.

Buzan, Barry (1995) 'The Security Problematique in the "New World Order"', in Ronnie D. Lipschutz, ed., On Security, New York: Columbia University Press, pp. 196-211.

Buzan, Barry and Ole Waever (2002, forthcoming) Regions and Powers: The Structure of International Security, Cambridge: Cambridge University Press.

Buzan, Barry, Ole Wæver and Jaap de Wilde (1998) Security. ' New Framework for 'nalysis, Boulder, London: Lynne Rienner.

Caporaso, James ed. (2000) 'Special Issue on: Continuity and Change in the Westphalian Order', International Studies Review 2, 2.

Castells, Manuel (1996a) The Power of Identity, Oxford: Blackwell Publishers.

--- (1996b) The Rise of the Network Society, Oxford: Blackwell Publishers.

Cerny, Philip (1994) 'The Infrastructure of the Infrastrcuture? Toward "Embedded Financial Orthodoxhy" in the International Political Economy', in Ronen Palan and Barry Gills, eds, Transcending the State Global Divide, Boulder, CO: Lynne Rienner, pp. 223-249.

Clapham, Christopher (1996) Africa and the International System. the Politics of State Survival, Cambridge: Cambridge University Press.

Clark, Ian (1999) Globalization and International Relations Theory, Oxford: Oxford University Press.

Cullen, Patick (2000) 'Keeping the New Dog of War on a Thight Leash: 'sssessing the 'ccountability for Private Military', Conflict Trends, June, pp. 36-39.

Duffield, Mark (1994) 'The Political Economy of Internal War: 'sset Transfer, Complex Emergencies and International 'id', in Joanna Macrae, 'nthony Zwi and with Mark Duffield and Hugo Slim, eds, War and Hunger. Rethinking International Responses to Complex Emergencies, London and New Jersey: Zed Books, pp. 50-69.

--- (1998) 'Post-modern Conflict: Warlords, Post-adjustment States and Private Protection', Civil Wars 1, 1, pp. 65-102.

--- (2001) 'The Emerging Development-Security Complex', Manuscript.

Edmonds, Martin (1998) 'Defense Privatisation: From State Enterprise to Commercialism', Cambridge Review of International 'ffairs XIII, 1, pp. 114-129.

Elias, Norbert (1982) 'Civilization and Violence: On the State Monopoly of Physical Violence and its Infringements', Telos, 16, pp. 133-154.

--- (1998) The Civilizing Process, Chicago.

Evans, Peter (1997) 'The Eclipse of the State? Reflections on Stateness in an Era of Globalization', World Politics 50, October, pp. 62-87.

Finnemore, Martha and Kathryn Sikkink (1998) 'International Norm Dynamics and Political Change', International Organization 52, 4, 'utumn, pp. 887-917.

Fredland, Eric and 'drian Kendry (1998) 'The Privatisation of Military Force: Economic Virtues, Vices and Government Responsibility', Cambridge Review of International 'ffairs XIII, 1, pp. 147-164.

Friedman, Irving S. (1983) 'Private Bank Conditionality: Comparison with the IMF and the World Bank', in John Williamson, ed., IMF Conditionality, Washington D.C.: Institute for International Economics, pp. 109-124.

Giddens, 'nthony (1985) The Nation State and Violence, Berkely: University of California Press.

Gill, Stephen (1995) 'Market Civilization and Disciplinary Neoliberalism', Millennium 24, 3, pp. 399-423.

--- (1997) ''n Emu or an Ostich? EMU and Neo-Liberal Economic Integration; Limits and 'lternatives', in Petri Minkkinen and Heikki Patomäki, eds, The Politics of Economic and Monetary Union, Helsinki: The Finnish Institute of international 'ffairs, pp. 205-228.

Guéhenno, Jean-Marie (1998-99) 'The Impact of Globalisation on Strategy', Survival 40, 4, pp. 5-20.

Guzzini, Stefano (2001) 'The Enduring Dilemmas of Realism in International Relations', COPRI Working Paper, 43.

Göle, Nilufer (1998) Modern Mahrem: medeniyet ve örtünme, Istanbul: Metis Yayinlari.

Habermas, Jürgen (1998) Die postnationale Konstellation. Politische Essays, Frankfurt a.M.: Suhrkamp.

Halliday, Fred (1999) Revolution and World Politics : The Rise and Fall of the Sixth Great Power, Durham, NC.: Duke University Press.

--- (2000) The World at 2000: Perils and Promises, New York: Palgrave.

Hassner, Pierre (1995a) 'Le XXe siècle, la guerre et la paix', in La violence et la paix. De la bombe atomique au nettoyage ethnique, Paris: Editions Esprit, pp. 371-391.

--- (1995b) 'Vers un universalisme pluriel?', in La violence et la paix. De la bombe atomique au nettoyage ethnique, Paris: Editions Esprit, pp. 355-371.

--- (1998) 'Refugees: a Special Case for Cosmopolitan Citizenship?', in Daniele 'rchibugi, David Held and Martin Köhler, eds, Re-Imagining Political Community. Studies in Cosmopolitan Democracy, Cambridge: Polity Press, pp. 273-286.

Held, David (1991) 'Democracy, the Nation State, and the Global system', in David Held, ed., Political Theory Today, Stanford: Stanford University Press.

--- (1993) 'Democracy: From City States to Cosmopolitan Order?', in David Held, ed., Prospects for Democracy: Polity Press.

--- (1995) Democracy and the Global Order: From the Modern State to Cosmopolitian Governance, Cambridge: Polity Press.

Held, David, 'nthony G. McGrew, David Goldblatt and Jonathan Perraton (1999) Global Transformations: Politics, Economics and Culture, Stanford, C': Stanford University Press.

Hobbes, Thomas (1651 (1985)) Leviathan, London: Penguin Books.

Hobsbawm, Eric J. (1963) 'The 'natomy of Violence', New Society, 28, pp. 16-18.

Holsti, Kalevi J. (1996) The State, War, and the State of War, Cambridge: Cambridge University Press.

Howe, Herbert M. (2001) 'mbiguous Order: Military Forces in African States, Boulder CO.: Lynne Rienner.

Haas, Peter (1992) 'Knowledge, Power and International Policy Coordination', International Organization, a special issue 46, 1.

Ignatieff, Michael (2000) Virtual War. Kosovo and Beyond, London: Chatto&Windus.

Jervis, Robert (1976) Perception and Misperception in International Politics, Princeton: Princeton University Press.

Kaldor, Mary (1999) New and Old Wars: Organized Violence in a Global Era, Oxford: Polity Press.

--- (2000) 'Cosmopolitanism and organised violence', First Press www.theglobalsite.ac.uk, 2.

Keck, Margaret E. and Kathryn Sikkink (1998) 'ctivists Beyond Boarders: 'dvocacy Networks in International Politics, Ithaca, NY: Cornell University Press.

Kinvall, Catarina (2001) 'Globalization and Religious Nationalism. Self, Identity and the Strife for Ontological Security', Paper presented at the International Soceity of Political Psychology, Mexico July 2001.

Kurdle, Robert T. and Lorrain Eden (2001) 'The New 'ttack on Tax Havens: Understanding Politics and Policy', Paper presented at IS' Chicago 21-4.02, 2001.

Leander, 'nna (2000) 'Strange Looks on Developing Countries: ' Neglected Kaleidoscope of Questions', in Thomas C. Lawton, James N. Rosenau and 'my C. Verdun, eds, Strange Power. Shaping the Parameters of International Relations and International Political Economy, 'ldershot et. al.: 'shgate, pp. 343-365.

--- (2001a) 'Dependency Today: Finance, Firms, Mafias and the State', Third World Quarterly 22, 1, pp. 115-128.

--- (2001b) 'Global Ungovernance: Mercenaries and the Control over Violence', Paper presented at the Seminar on Private Military Companies: Centre for Development Research, Copenhagen, 7 Dec.

--- (2001c) 'The Globalisation Debate: Dead Ends and Tensions to Explore', Journal of International Relations and Development 4, 3, pp. 274-285.

--- (2001d) 'Globalisation, Transnational Polities and the Dislocation of Politics', COPRI Working Papers, 12.

--- (2002) 'Global Ungovernance: Mercenaries and the Control over Violence', COPRI Working Paper, 4.

--- (2002 forthcoming) 'Wars and State-building: Revisiting, Reversing and Varying the Process', in Stefano Guzzini and Dietrich Jung, eds, Copenhagen Peace Research: Conceptual Innovations and Contemporary Security 'nalysis.

Leiteritz, Ralf (2001) 'Sovereignty, developing countries and international financial institutions: a Reply to David Williams', Review of International Studies 27, 3, pp. 435-440.

Lipschutz, Ronnie D. (2000) 'fter 'uthority. War, Peace and Global Politics in the 21st Century, 'lbany: State Univeristy of New York Press.

Michaud, Yves (1978) Violence et Politique, Paris: Gallimard.

--- (1996) La violence apprivoisée, Paris: Hachette.

--- (1999) La violence, Paris: Puf.

Minc, 'lain (1993) Le Nouveau Moyen 'ge, Paris: Gallimard.

Mittelman, James (2002) 'Globalization: 'n 'scendant Paradigm', International Studies Perspectives 3, 1, pp. 1-14.

Musah, 'bdel-Fatah and Kayode J. Fayemi, eds (2000) Mercenaries: 'n African Security Dilemma, London: Pluto Press.

Neumann, Iver B. (1997) 'John Vinvent and the English School of Interntional Relations', in Iver B. Neumann and Ole Waever, eds, The Future of International Relations. Masters in the Making?, London and New York: Routledge, pp. 38-65.

O'Brien, Robert, 'nne Marie Goetz, Jan 'art Scholte and Marc Williams (2000) Contesting Global Governance: Multilateral Economic Institutions and Global Social Movements, Cambridge: Cambridge University Press.

Palan, Ronen (1998) 'Trying to Have Your Cake and Eating It: How and Why the State System Has Created Offshore', International Studies Quarterly 42, pp. 625-644.

Patomäki, Heikki (2001) Democratising Globalisation: The Leverage of the Tobin Tax, London: Zed Books.

Pauly, Louis W. (1997) Who Elected the Bankers? Surveillance and Control in the World Economy, Ithaca and London: Cornell University Press.

Porter, Tony (1999) 'The Transnational 'genda for Financial Regulation in Developing Countries', in Leslie Elliott 'rmijo, ed., Financial Globalization and Democracy in Emerging Markets, Basingstoke and New York: Palgrave, pp. 91-117.

Reno, William (1998) Warlord Politics and African States, Boulder, London: Lynne Rienner.

Reus-Smit, Christian (1999) The Moral Purpose of the State. Culture, Social Identity, and Institutional Rationality in International Relations, Princeton: Princeton University Press.

Rufin, Jean-Christophe (1993) 'Les économies de guerre dans les conflits internes', in François Jean and Jean-Christophe Rufin, eds, Économie des guerres civiles, Paris: Hachette, pp. 19-59.

Scholte, Jan 'art (2000) Globalization: ' Critical Introduction, London and New York: Macmillan and St. Martin's Press.

Shaw, Martin (1996) Civil Society and the Media in Global Crises, London: Pinter.

--- (2000) Theory of the Global State: Globality as Unfinished Revolution, Cambridge: Cambridge University Press.

Singer, P. W. (2001/2) 'Corporate Warriors: The Rise of the Privatized Military Industry and Its Ramifications for International Security', International Security 26, 3, pp. 186-220.

Spearin, Christopher (2001) 'Private Security Companies and Humanitarians: ' Corporate Solution to Securing Humanitarian Spaces?', International Peacekeeping 8, 1, pp. 20-43.

Spicer, Tim Lt Col (1998) 'Interview with Lt Col Time Spicer', Cambridge Review of International 'ffairs XIII, 1, pp. 165-173.

Strange, Susan (1990) 'The Name of the Game', in Nicos Rizopoulos, ed., Sea Changes. 'merican Foreign Policy in a World Transformed, New York and London: Council on Foreign Relations Press, pp. 238-274.

--- (1995) 'The Defective State', Daedalus 24, 2, pp. 55-74.

--- (1996) The Retreat of the State: The Diffusion of Power in the World Economy, Cambridge: Cambridge University Press.

--- (1999) 'Corporate Managers in World Politics', in Michel Girard, ed., Individualism and World Politics, London: Macmillan, pp. 145-160.

Thompson, Janice (1994) Mercenaries, Pirates, and Sovereigns: State-building and Extraterritorial Violence in Early Modern Europe, Princeton: Princeton University Press.

Tibi, Bassam (1991) Die Krise des modernen Islams, Frankfurt: Suhrkamp.

Tilly, Charles, ed. (1975) The Formation of National States in Western Europe, Princeton: Princeton University Press.

--- (1985) 'War Making and State Making as Organized Crime', in Peter Evans, Dietrich Rueschemeyer and Theda Skocpol, eds, Bringing the State Back In, Cambridge: Cambridge University Press, pp. 169-191.

--- (1990) Coercion, Capital and European States, 'D 990-1992, Cambridge M' and Oxford UK: Blackwell.

UN, Mr. Enrique Bernales Ballesteros and Special Rapporteur (2001) Report on the question of the use of mercenaries as a means of violating human rights and impeding the exercise of the right of peoples to self-determination, pursuant to Commission resolution 2000/3, New York: UN, E/CN.4/2001/19, available at http://www.unhchr.ch/Huridocda/Huridoca.nsf/0/b6e7abef3d39af97c1256a140059c839/$FILE/G0110170.pdf.

van Creveld, Martin (1991) The Transformation of War, New York: The Free Press.

Walker, R. B J. (1993) 'Violence, modernity, silence: from Max Weber to international relations', in David Campbell and Michael Dillon, eds, The Political Subject of Violence, Manchester and New York: Manchester University Press, pp. 137-160.

Weissman, Fabrice (1993) 'Mozambique: La guerre du ventre', in François Jean and Jean-Christophe Rufin, eds, Économie des guerres civiles, Paris: Hachette, pp. 299-341.

Wendt, 'lexander (1999) Social Theory of International Politics, Cambridge: Cambridge University Press.

Williams, David (2000) ''id and Sovereignty: Quasi-states and the International Financial Institutions', Review of International Studies 26, 4, pp. 557-573.

Zarate, Juan Carlos (1998) 'The Emergence of a New Dog of War: Private International Security Companies, International Law and the New World Disorder', Stanford Journal of International Law 34, 1, pp. 75-162.

Zürn, Michael, Gregor Walter, Sabine Dreher and Marianne Beisheim (2000) 'Globalisierung und die Handlungsfähigkeit des Nationalstaats', Zeitschirft für Internationale Beziehungen 7, 2, pp. 297-331.

^[1] (Mittelman 2002; Buzan and Waever 2002, forthcoming).

^[2] (Strange 1996; Caporaso 2000).

^[3] (Elias 1982; Michaud 1996: 7; Elias 1998).

^[4] (Beck 1997).

^[5] (Guéhenno 1998-99; Beck 2000b).

^[6] For a wonderful account of the role of legal norms in state violence see (Bauman 1989). For the argument in general (Michaud 1999: 29-34).

^[7] The two traditions are derived from (Arendt 1969).

^[8] But it is far from clear what this tradition actually consists of (Guzzini 2001) and hence many of those sharing this view on violence might well not be realists and some who don't actually may so this labelling should be taken with a pinch of salt.

^[9] (Michaud 1978, 1999).

^[10] It is the recognition of this which makes it possible for Thompson to bracket the legitimacy question entirely from her study of the establishment of a state control over violence (Thompson 1994).

^[11] (Walker 1993).

^[12] (Michaud 1978).

^[13] (Hobbes 1651 (1985): 238).

^[14] (Hobsbawm 1963; Minc 1993; Strange 1995).

^[15] (Ayoob 1992; Holsti 1996).

^[16] This argument is developed in detail in (Arendt 1969).

^[17] It might be useful to confirm that liberals would not refer to the monopoly on violence as legitimate.

^[18] (Walker 1993).

^[19] (Michaud 1978).

^[20] (Arendt 1979/1951).

^[21](Hassner 1998).

^[22] In this paper I use globalisation as refering to the creation of transnational social space, which tends to imply a Atime space compression' and result in a Adeterritorialisation of social space'. For a more detailed discussion of the reasons to prefer this definition to its many competitors see (Scholte 2000: 46-50; Leander 2001c).

^[23] (Wendt 1999: 9).

^[24] (Clark 1999).

^[25] (Beck 1999; Ignatieff 2000).

^[26] (Castells 1996a).

^[27] This very useful and clarifying distinction is drawn from the excellent work of (Thompson 1994?).

^[28] (Edmonds 1998; Fredland and Kendry 1998; Leander 2001b).

^[29] (Kaldor 1999).

^[30] (Reus-Smit 1999).

^[31] (Tilly 1975: 638).

^[32] (Giddens 1985: 121).

^[33] (Neumann 1997).

^[34] (Halliday 1999).

^[35] I have developed these arguments in greater detail in (Leander 2001a, 2001d).

^[36] (Held, McGrew et al. 1999 : 321-326; Beck 2000b: 72-77).

^[37] Political parties in countries of emigration do their best to organise the emigrant vote. The religious Refah/Virtue party in Turkey e.g. has organised transport back to Turkey for voting for its supporters on a large scale.

^[38] (Angoustures and Pascal 1993; Bozarslan 1993; Weissman 1993).

^[39] (Habermas 1998: 71-9).

^[40] (Halliday 2000: 153).

^[41] This is why Beck thinks that the Ameta politics of politics' whereby the rules of the game of politics are contested is becoming more salient (Beck 2000a).

^[42] For example, (Badie, Coulon et al. 1987; Tibi 1991; Castells 1996a; Appadurai 1998; Göle 1998; Lipschutz 2000; Kinvall 2001).

^[43] (Beck 2000b: 99). + Pizzorno on absolute pols.

^[44] This insight is underlying much of the work pursued by Held, (Held 1991, 1993, 1995).

^[45] For elaboration on this point see e.g. (Finnemore and Sikkink 1998).

^[46] (Leander 2000).

^[47] (Patomäki 2001: chaps 1-3).

^[48] (Friedman 1983; Porter 1999).

^[49] (Rufin 1993; Duffield 1994; Clapham 1996).

^[50] (Keck and Sikkink 1998; O'Brien, Goetz et al. 2000).

^[51] An insight which has been used to argue that conditionality is in fact a lesser evil because more transparent and democratic. (See the debate Williams 2000; Leiteritz 2001).

^[52] (Gill 1995, 1997).

^[53] (Pauly 1997).

^[54] (Beck 2000: 72).

^[55] (Shaw 1996).

^[56] (Andreas and Price 2001).

^[57] (Hassner 1995b).

^[58] (Zürn, Walter et al. 2000).

^[59] (Haas 1992).

^[60] For the opposed views from Porto Allegro and New York both confirming the point see: ACampaigners set to focus on world's biggest corporations', Financial Times (February 4 2002); and ACompanies pledge better 'corporate citizenship' Financial Times (February 4 2002).

^[61] Buzan even argues that Ait is not clear how states develop under these conditions [a very strong and penetrating international system], or even whether they can' (Buzan 1995: 195).

^[62] (Beck 1998).

^[63] (Tilly 1990).

^[64] (Clapham 1996; Bayart, Ellis et al. 1997; Reno 1998).

^[65] Characteristic of this position is Hassner's lament/question ACan we be on our little bourgeois island protected by tariff or police barriers, without even looking at the suffering of the planet at the TV, refusing the refugees and letting everyone massacre themselves around us? Such a perspective seems unacceptable, but, at the same time, I cannot demonstrate how we will pass from the present prevailing non-engagement to a >contagion' which, eventually, would bring us back to our responsibilities for the world.' (Hassner 1995a: 381).

^[66] (van Creveld 1991; Avant 2000).

^[67] (Thompson 1994: 19)

^[68] (Arnold 1999).

^[69] (Edmonds 1998).

^[70] (Fredland and Kendry 1998).

^[71] (Brauer 1998; Leander 2002).

^[72] (Musah and Fayemi 2000). This indication should only be taken as indicative of the shift not as reliable indicator. Since it is inherently difficult to find information on this issue.

^[73] (Adams 1998).

^[74] After describing the military origins of the internet and having drawn up the main steps of the technological revolution Castells argues that Ait is indeed by this interface between macro-research programmes and large markets developed by the state, on the one hand, and decentralised innovation stimulated by a culture of technological creativity and role models of fast personal success, on the other hand, that new information technologies came to blossom' (Castells 1996b: 60).

^[75] (Cerny 1994).

^[76] For good overviews of the services offered by tax havens and the discussions surrounding them see (Palan 1998; Kurdle and Eden 2001).

^[77] (Singer 2001/2).

^[78] (Clapham 1996).

^[79] (Duffield 1994, 2001).

^[80] As a consequence, DSL (Defence Systems Limited), lists among its clients: the International Rescue Committee, CARE, Caritas, USAID, GOAL and World Vision. And in the UN system the UNHCR, UNICEF, UNDP, WFP amongst others declare to have used private security services (Spearin 2001).

^[81] (Spicer 1998).

^[82] For excellent discussions of this process see (Strange 1990; Gill 1995; Strange 1999).

^[83] (Evans 1997).

^[84] (Singer 2001/2: 197).

^[85] This explains the centrality for the debate on privatisation about the extent to which privatisation undermines the decision making authority of the state and the energy put into defending the idea that after all the PMCs are, in reality an extension of state policies, when in fact many of them (and particular the resource based ones) have the same kind of ambiguous relationship to states as did the colonial companies of the 18^th Century.

^[86] Brown &Root is reportedly failed to deliver or severely overcharged on 4 of its 7 obligations to the US Army.

^[87] In 1994 the Gurkha e.g. fled Sierra Leone after their commander had been killed (and reputedly cannibalized).

^[88] It seems that EO might have been active in the 1996 ousting of the president of Sierra Leone. This and the two preceding examples stem from (Singer 2001/2: 205-6).

^[89] The larger PMCs have repeatedly protested that their activities might contribute to fuelling conflicts and empowering non-state actors.

^[90] (UN and Rapporteur 2001).

^[91] Arguably the centrality of private actors in determining the allocation was cemented already by the transition from a conventional armament logic to a logic where proliferation became the central concern (Hassner 1995a)..

^[92] For the importance of perceptions and threat definitions (Asecuritization') see (Jervis 1976; Buzan, Wæver et al. 1998). A concrete illustration is perhaps the capacity of MPRI to get ever more expensive contracts for training the Saudi Army (Arnold 1999).

^[93] (Singer 2001/2; Leander 2002).

^[94] For example, the American firm MPRI (Military Professional Resources Inc.) e.g. was involved in the sophisticated Croat AOperation Storm' whereby Serb held Krajina was recovered in 1995, a turning point in the war, and then in rearming and training the Bosnian armed forces. Both interventions made it possible to decisively tip the military balance without taking the national and international political debate which open interventions would have provoked (Adams 1999: 110).

^[95] (Howe 2001).

^[96] (Tilly 1985: 186; 1990; Leander 2002 forthcoming).

^[97] (Zarate 1998; Cullen 2000).

^[98] As revealed among other things by the discussions about what to do with mercenaries in Afghanistan.

^[99] (Thompson 1994: 153).

^[100] (Duffield 1998).

^[101] (Bauman 1998).

^[102] (Kaldor 2000).

^[103] (Shaw 2000).

^[104] (Duffield 1994, 2001).

^[105] This is a central point made by (Reno 1998).