Hostname: page-component-7b7567c8c-2rsb6 Total loading time: 0 Render date: 2026-11-13T00:31:00.359Z Has data issue: false hasContentIssue false
  • English
  • Français

The same as it never was? Uncertainty and the changing contours of international law

Published online by Cambridge University Press:  28 October 2011

OLIVER KESSLER
Rights & Permissions [Opens in a new window]

Abstract

International law has changed significantly since the end of the Cold War. As long as the international was thought to be populated by sovereign states predominantly, international law was conceived of as a means for peaceful dispute settlement. That is: the reference to state sovereignty not only divided public from private international law, but structured most of public international law itself; from the very definition of and associated rights and duties to the attribution of responsibility. With the emergence of the post-national constellation, a reduction of law to questions of states' practices is increasingly problematic. At the same time, the post-national constellation denotes more than just a structural shift in the world polity. It challenges established dogmas rooted in an individualistic philosophy of science and thereby calls for a different understanding of how the world is (made) known. What uncertainty has to offer is the provision of a different vocabulary detached from the state through which we can reconsider some changes in international law.

Information

Type
Research Article
Information
Review of International Studies , Volume 37 , Issue 5 , December 2011, pp. 2163 - 2182
Copyright
Copyright © British International Studies Association 2011

Introduction

Although it might have not been taken seriously in its own right, the notion of uncertainty always played a significant role in International Relations (IR) theorising.Footnote 1 Already Hans Morgenthau referred to the uncertainty of the balance of power to emphasise the fragility of global stability and the dangers of over-confidence in one's own convictions.Footnote 2 John Herz referred to uncertainty when he formulated the security dilemma,Footnote 3and in the context of regime theory, the function of institutions is usually associated with their capacity to reduce uncertainty.Footnote 4 While uncertainty has somehow always been around, it was, consciously or not, always embedded in a semantic field of anarchy, rationality and national interests where it is common to treat uncertainty as a category of interstate relations and thereby make it accessible for rational approaches to IR.Footnote 5 In this context, uncertainty was essentially confined to an economic (and some would probably say economistic) decision theory, including approaches of information economics and game theory.Footnote 6 Here, uncertainty was always defined in terms of ‘risk’: in these models, it is always possible to capture the existing uncertainty by some probability distribution over outcomes, types or whatever, and thereby apply the same instrumental rationality and insurance mathematics as we know it from ordinary ‘risks’.Footnote 7 This conceptualisation of uncertainty was and still is supported, legitimised, and reproduced by the overall dominance of a positivist philosophy of science in IR that directs us to ask certain questions, see certain problems, and come up with a specific solution.Footnote 8

Yet this positivist conceptualisation of uncertainty appears to be inadequate today for a rather simple reason: In game – and contract theory, games are ontologically prior to interactions: actors (including their interests and action space), common probability distributions over types (in Bayesian games), and in particular the basic categories of the game are fixed before the game itself is played (some would probably say that a model has to be well defined). This implies thatchange can only take place within fixed categories.Footnote 9 A change in the payoff matrix might lead to a different equilibrium, but the basic categories of the game are static and given – dynamic modelling notwithstanding. Given the widely observed systemic changes today which are altering the constitutive rules of world politics, such a notion of uncertainty does not get us very far indeed. We have witnessed a move to privatise, transnationalise, and de-politicise security threats that are answered by new illiberal practices. We have witnessed the advent of new actors, new political spaces, and new practices that hardly fits in the traditional conceptual framework of national vs. international politics or public vs. private actors.Footnote 10 Additionally, it is meanwhile somewhat of a consensus that an understanding of systemic changes requires abandoning established dogmas rooted in an individualistic philosophy of science and thereby calling for a different understanding in IR of how the world is (made) known.

In other words, to capture these changes of the constitutive rules of world politics, it is necessary to leave the traditional and rather positivist notion of uncertainty behind and to develop a different conceptual framework that actually helps us to understand those changes. This raises two questions that these contribution seeks to address: first, how can such a concept of uncertainty look like? And second, what implications has such an uncertainty approach for understanding the politics of international law. The first question is answered by distinguishing two notions of uncertainty: ontological uncertainty associates contingency with things, games, and decisions that treat situations of uncertainty similarly to risk decisions. Epistemological uncertainty is juxtaposed to risk and is directed towards analysing processes of world (sense-)making. In this context, uncertainty is continuously reduced to risk through the application of categories, discourses, and disciplinary knowledge whereby fields, discourses or systems are constituted, reproduced, or changed. At the same time, the very process of uncertainty reduction produces new uncertainties. It creates new non-knowledge, new problems, antinomies, or contingencies.Footnote 11 From this perspective, world society is characterised by a continuous process of uncertainty absorption and production that, over time, shifts the horizon of possibilities, the range of possible contingencies, the seeable and the sayable. From this perspective, the second question is answered in terms of a redefined process of uncertainty absorption and production in international law: Risk provides a new vocabulary to assess authority, to allocate responsibility and accountability that redefines the politics of space in international law to a politics of time. This might well be the reason why many legal institutions that ascribe responsibility and accountability on the basis of ‘norms’ are either redefined or collapse with the consequence that legal norms legitimise and do not constrain violence.

To substantiate this argument, this contribution is divided in three sections. The first section discusses the recent literature on risk and uncertainty as used in security studies. The objective of this discussion is, however, not to identify some variables or common denominator in order to unify them, but to get a better sense of what this literature actually tries to accomplish. In my understanding, it seeks to capture current social, spatial, and temporal changes in world society, including a different understanding of what ‘social order’ is. This implies a redefined concept of politics where the observation of political processes has less to do with states per se, but with the management of uncertainty and contingency.Footnote 12 Politics is located at the very processes of sense and thus world making. The second part addresses the first question and explores this ‘politics of uncertainty’ by discussing different conceptualisations of risk and uncertainty. The third part turns to the second question and shows how such an understanding of uncertainty, risk, and norms allows us to reconstruct current changes in international law. I suggest that risk provides an alternative vocabulary to norms as the very mode of operative closure of international law to assess authority, jurisdiction, and responsibility. If the notion of legalisation would not have such an unlucky meaning in IR, one could say that the politics of international law can be understood as legalisation, that is, the processes whereby things, problems, issues, and facts are made ‘legal’.

Three notions of risk

Especially with the so-called ‘War on Terror’ and the current ‘world economic crisis’, risk is increasingly used to challenge existing narratives and to provide a different assessment of practices and dynamics of world politics.Footnote 13 This interest in risk is predominantly fuelled by either one of three different approaches: Beck's notion of risk society,Footnote 14 Foucault's notion of risk as new ‘dispositif’,Footnote 15 or Niklas Luhmann's sociology of riskFootnote 16. This section briefly discusses these approaches to reconstruct their family resemblance.Footnote 17 Although they are fundamentally different, they are carried by the conviction that we currently witness a transformative change and hence a reconfiguration of the social, temporal, and spatial conditions of world politics.

Risk society

To assess current changes in world politics, we are used to talk about post-industrial society, post-Fordism, the transformation of the Keynesian welfare into the Schumpeterian workfare state, we are used to characterise different types of statehood, or to talk about the ‘transformation of the state’. For Beck, these various descriptions of ‘post’ this and ‘post’ that are just signs of a new emerging social order: the transformation of industrial society to risk society.Footnote 18 Industrial society of the post-war era was built on the basis of a societal belief in progress, rationality, and control of social relations. The Keynesian welfare state promised economic security, economic policy sought to fine-tune the economy, and the breathtaking technological advancements once promised a new world of freedom and security.

However, these high hopes of modernity lay shattered before us. The unintended consequences of modern society like pollution, genetically modified food or global financial risks, feed back negatively and now threaten the very foundations of modern societies: nuclear energy produced Chernobyl and Fukushima, globalisation produced new terrorist threats and financial practices just brought the world economy on the brink of collapse. Modern societies are occupied with the assessment, management and prevention of the risks they themselves have produced.Footnote 19 Societies have to realise for the first time, that their existence is fragile and marked by contingency. They have to become reflexive. As a result, modern beliefs in rationality, progress, and science are unmasked as false beliefs (if not even ideology) and bring to the fore the paradox of modern life, as Ulrich Beck has put it: ‘the more emphatically the existence of world risk society is denied, the more easily it can become a reality. The ignorance of the globalization of risk increases the globalization of risk.’Footnote 20

The characteristics and consequences of these risks like future terrorist attacks, ecological collapse, or financial crises are best characterised by what they are not: they are not catastrophes. As in particular his latest book on World Risk Society emphasises, risks are the anticipation of catastrophes. Once a risk materialises, it ceases to be a risk. Risk is future oriented and inherently puts emphasis on visualisation, imaginations and expectations of the future that feed back onto the present and require and inform decisions.Footnote 21 Yet the very attempt to control the future transforms the spatial, temporal, and social conditions of world society. Socially, world risk society wipes out the class defined concepts of inequality that could be addressed through the redistribution of wealth. Rather, the confines of industrial society are transformed and left behind where risk society puts focus on the distribution and individualisation of risks.Footnote 22 Spatially, risks don't follow the logic of territoriality. Whether we deal the fight against terrorism, the environment or financial crises, they are not reducible to ‘individual’ or state-level interests, beliefs, or counter-measures, but require a common solution. For Beck, global risks thus constitute a cosmopolitan moment when states realise that they need alliances for both external and internal security. National measures might not be a bad thing, but they won't help much once a risk turns into a catastrophe. Temporally, governments cannot simply sit and wait for the next catastrophe to come and then engage in reactive measures to deal with the consequences only. Rather, governments are increasingly required to address these risks proactively by introducing preventive measures. Preventive measures, however, are the attempt by governments, scientists, and companies to convey security and calculability where there is none. Risks escape the logic of numbers and science, they are incalculable and cannot be compensated for, as Beck warns: ‘The irony of risk is that rationality, that is, the experience of the past, encourages anticipation of the wrong kind of risk, the one we believe we can calculate and control, whereas the disaster arises from what we do not know and cannot calculate’.Footnote 23

Beck's notion of risk is clearly different from the positivist understanding of risk. As he emphasises in World Risk Society, and I think also thereby taking on part of the critique by Niklas Luhmann, the driving force of world risk society is neither an accumulation nor the growth of knowledge, but different kinds of non-knowledge. Or as Beck coins it, ‘reflexive non-knowledge’, that is knowledge about one's non-knowledge serves as the medium of reflexive modernity’.Footnote 24 Non-knowledge can take on different forms and guises ranging from the not-wanted-to-be-known to the ‘unknown unknowns’.Footnote 25 This focus on non-knowledge leads Beck to identify societal struggle over the very definition of risks that becomes indistinguishable from the reality of risk itself. Different forces and groups try to capture risks by defining the magnitude of possible harm, how risks are produced, what they actually are, and what kind of evidence is legitimate to proof their existence today. These fights over the definition of risk constitute a specific mode of knowledge and non-knowledge that defines the way risks are made known and how non-knowledge is governed. The central fight in risk society is not about means of production and modes of production, but about the knowns and unknowns of global risks. Whatever the specific definition might be, it certainlycreates a specific power relation and constitutes certain inequalities by including certain aspects and excluding others.

However, to the extent to which Beck focuses on the interplay between knowledge, non-knowledge and the definition and anticipation of risks, it is interesting that not the risks themselves, but uncertainty constitutes these regimes of non-knowledge. The assessment, management and prevention of modern risks does exactly not follow a logic of modern risk management, but a logic of uncertainty: states and other actors need to convey the image of control where we are faced with unstructured uncertainty. It is uncertainty that calls for preventive measures, that requires giving global risks a face, a design, or geographical space while the very idea of control, calculus, and intention is actually passé.Footnote 26 At the same time, however, one has to admit that Beck does not develop the notion of uncertainty fully. It is unclear what ontological status global risks actually have. On the one hand, risks are defined, anticipated, and designed. On the other hand, they are endowed with causal powers to constitute a cosmopolitan moment and redefine the basic contours of society.

Risk as dispositif

The writings of Michel Foucault, and especially his two lectures at the Collège de France in 1978 and 1979, have become highly influential for current IR theorising.Footnote 27 There are many interpretations and opinions out there as to how the lectures relate to his other ‘periods’, and I do not try to claim unity where Foucault himself resisted it. Any attempt to emphasise unity over difference runs the danger of engaging in performative contradictions at this point. Nevertheless, a short reconstruction of some of his themes and interests provides some helpful background knowledge for what the concept of governmentality ‘does’.

In his The Order of Things, Foucault was interested in how discourses create and occupy fields, how these fields become possible and reproduced.Footnote 28 How is it possible that we actually assume language ‘refers’ to something, that we can understand language in terms of representations and signs. This interest in the operations of discourse continued later on. In Discipline and Punish,Footnote 29 the prison is not only a modern site to punish by ‘exclusion through inclusion’, but also a site of knowledge. It creates prisoners, it allows for the formation of a truth regime where we can know about their behaviour, their health, their sexuality. Modern institutions like the prison, the clinic, and the asylum that are usually seen as progress to previous forms of punishment and care, are sites of power that constitute knowledge, practices, and rules that govern conduct. They are not embodiments of enlightenment ideas. These sites allow sentences to be true, they structure the horizon of possible statements, their interconnectedness and reproduction. As Foucault notes:

Let us not … ask why certain people want to dominate, what they seek, what is their overall strategy. Let us ask, instead how things work at the level of those continuous and uninterrupted processes which subject our bodies, govern our gestures, dictate our behaviours, etc. In other words, rather than ask ourselves how the sovereign appears to us in his lofty isolation, we should try to discover how it is that subjects are gradually, progressively, really and materially constituted though a multiplicity of organisms, forces, energies, materials, desires, thoughts, etc. We should try to grasp subjection in its material instance as a constitution of subjects.Footnote 30

What counts as truth, as statement, as reference to some empirical fact, is the effect of these techniques and practices. Discourses are not simply talk or mere language, but they set the boundaries of writing, saying, thinking, and knowing in specific historical contexts. His concept of arché actually nicely captures how statements don't follow the hierarchical rules of logic or the binary codes of structuralism. Statements stand in relation to other statements, they overlap, they produce continuities and discontinuities along historically specific rules of ordering.Footnote 31 A point that was continued in his geneaological writings, but it was augmented by an interest in how techniques and practices create new fields and thus power and systems of exclusions.

Already at this rather crude and general level can one recognise two central themes that inform the governmentality literature: the interplay of discourses, knowledge, power and creation of subjectivity; and a focus on techniques, rationalities, instruments, and practices instead of the state or some fixed institution per se. Governmentality stands somewhat in between these two and shows how these techniques and practices relate to fields that are the product of the set of possible discourses themselves.Footnote 32 What certainly differs in these lectures from his previous writings is a new emphasis on non-discursive elements and thus a focus on the boundary between discursive and non-discoursive practices.

The lectures are certainly rich and touch on many themes. One of them is a description of the processes that led to the emergence of the population as a new ‘site of techniques’. How knowledge about populations became possible, how the population reconfigured the reality of sovereigns, led to the emergence of new practices, techniques and rationalities. He shows apparent achievements like freedom, rights, and autonomy are produced and controlled by the art of government. The state is thus not this Weberian administration, but set of practices, strategies, and technologies of the self whereby each is governed. In this context, the term ‘dispositif’ refers to the ways in which a ‘heterogeneous ensemble consisting of discourses, institutions, architectural forms, regulatory decisions, laws, administrative measures, scientific statements, philosophical, moral and philanthropic propositions’, that is, discursive and non-discursive practices, create objects and events and make them governable.Footnote 33 A dispositif is a site of struggle and prevails in a given period of time with a set of techniques, strategies and ‘apparisations’ (or etatisations for that matter).

To understand risk as a dispositif puts the focus exactly on how risk constitutes a heterogeneous ensemble that govern individuals. Risk allows the grouping of possible terrorists under a specific category were, again, knowledge is produced, sentences made true, techniques of intervention applied. Risk enacts actors and is therefore not just a description of our society, but it is ‘ordering our world through managing social problems and surveying populations’.Footnote 34 It creates new heterotopias such as Guantanamo, new contingencies and new promises of future control of these very contingencies.Footnote 35 Foucauldians have in particular pointed to neoliberalism as a specific form of governmentality. Setting aside the rather unfortunate use of neoliberalism for a moment, it can be characterised as the entry of the market in politics: techniques on the basis of market principles, instrumental rationality, benchmarking, statistics, and risk based categorisations. Risk fosters exactly this neoliberal governmentality, as Leander and van Munster nicely summarised: ‘Neo-liberal governmentality situates human activity within an economic calculus. It relocates activities from the public realm to the market by means of direct privatization, public-private partnerships, outsources, the creation of new markets, the making of internal markets and the establishment of conditions for competitive success.’Footnote 36 Similarly, Aradau and van Munster showed how risk allows for a reconfiguration of the present and the future in security practices.Footnote 37 They showed how the future is subject to control, how it is calculated, insured, and how this unknown future is now subject to a specific governmentality that controls an unknown future through the employment of precautionary measures.Footnote 38 Precautionary measures reconfigure reactive in proactive security policies that focus on surveillance of populations and circulations (money, mobility, and information).Footnote 39

Yet here again, it is the unknown, the uncertain and not only risk in itself that provides the rationale that stabilises and connects statements. As they argue: ‘precautionary risk introduces within the computation of the future its very limit, the finity of uncertainty and potential damage. It is therefore exactly the opposite of prudence. If the latter recommended what “precautions” to take under conditions of knowledge, the former demands that we act under scientific and causal uncertainty.’Footnote 40 It is this uncertainty beyond risk that provides a different rationale where now the neoliberal governmentality is enacted.Footnote 41 Precautionary principles don't redefine risk, but actually operate at the border of risk and uncertainty, that controls uncertainty and impose rationality on it. While analyses inspired by Beck and Foucault highlight precautionary measures, anticipation, and the attempted control of an unknown future, it should be highlighted that they nevertheless differ in their assessment. Beck emphasises the incalculability of risks and claims risks cannot be tamed and controlled. Foucauldian analyses show how the relation between the present and the future is controlled and subject to governmentality through risk and economic rationality.

Risk as mode of observation

Niklas Luhmann's autopoetic systems theory provides a different notion of risk embedded in his general systems theory.Footnote 42 Generally speaking, systems are similar to discursively constituted fields in Foucault. We naturally talk about the economic system, the legal system, the health system or sports, etc. For Luhmann, these systems are not closed boxes, nor are they ontological entities, but they are continuously reproduced through communications.Footnote 43 Without economic communication, the economy would cease to exist. Without the continuous production of legal texts, decisions, opinions, and cases, it would be impossible to identify a legal system that is separated from other social spheres. They exist in the very moment in which communications connect and produce new communications. Thus systems limit the range of what is sayable, thinkable, knowable and at the same time limit the range of possible further communications.Footnote 44 There are always only a range of possible arguments in economic negotiations, legal hearings, and political debates. These range of themes and possible contributions are historically specific and shift over time. The themes, the semantics, the argumentative structures, and in for that matter social relations, techniques, practices, and subjectivities were different in the Middle Ages than they are today.

Yet without the reproduction of arguments, of texts, of actions, the economic, legal or political discourse would cease and become indistinguishable from their environment. The way they are reproduced, the range of social relations, the temporality and spatiality they give rise to, are systems specific. The ‘gaze’ of a system, the horizon of possible communications, is not reducible to extra systemic forces like spirit, reason, or ideas but lies within the system itself. Neither experience nor reason, but only the discourse itself provides the conditions of discourse formation, of transformation and correlation. Each system has its own constitutive rules, rationality, ideas of what society is while society is characterised by the clash of these different societal rationalities, rules, and discourses.

Through the continuous reproduction of communication, boundaries in the form of semantic distinctions emerge that stabilise and structure further communications: the distinction between facts and norms is of crucial importance within the legal system, but hardly plays any role in political debates. The distinction between demand and supply is crucial within the economic system, but plays no role in the sport system. Systems theorists are now particularly interested in how semantic distinctions form such boundaries, how they emerge, become naturalised and taken for granted; how they create their own reality, how they reconfigure social relations, time, and space. They are interested in how the connectivity of communications produces facts, events, or things. How these things acquire meaning and how their meaning changes with the change of concepts and discourses.

Insofar as systems operate by the continuing enactment of distinctions, risk has no existence outside of communication. Neither is risk a psychological attribute nor some ontological sequence. There ‘is’ no ‘risk’ that could execute force on the nation state or that somehow demands states to act in a certain way. Rather, risk is a mode of observation based on specific distinctions. As every ‘thing’ or unity result from temporary condensation of meaning using and reproducing distinctions, the meaning of words and concept is not located ‘inside’ but ‘between’ concepts. This includes the very term of risk itself. The notion of risk in itself is meaningless until it is separated from counter-concepts. One cannot simply observe risk without at the same time differentiating it from something else. To know what risk ‘is’ or what it ‘does’, one cannot simply look at it but needs to reconstruct the hidden and (in)visible semantic distinctions. And it is this setting apart that gives meaning to risk.

In this sense, whatever distinction one chooses to start with, risk is not a thing, but an ascription and a specific mode of observation that creates and performs its own reality.Footnote 45 From this perspective, risk is not simply a form of knowledge or some material force or necessity, but equally characterised by our non-knowledge. Risk marks the boundary between knowledge and non-knowledge and the way in which this boundary might be made known. At the same time, this boundary between knowledge and non-knowledge and the knowledge about our non-knowledge are of a different kind than simply the knowledge of some fact. The question can only be what the use of risk does and not what risk ‘is’ in an ontological sense. The notion and its use inevitably rest and reproduce societal meaning structures concerning the social, time and reality that are inscribed in the connectivity of communications.

This opens the way to see risk embedded in and reproduced by a multiplicity of overlapping and conflicting discourses. What the risk of ‘pollution’ means is different in each functional system such as law, health, economy, religion, etc. Each system, based on own rules of significations, sense-making and logic reconstructs and endows risk with a different meaning. The societal observation of some risk, for example of a terrorist attack, then presents itself a flux of contradictions and a multiplicity of logics simultaneously at play.

Exactly this multiplicity or polycontextuality of risk instantly refers back to a constitutive uncertainty. As every discourse reconstructs some assumed ‘objective’ risks in its own terms and logic, uncertainty refers to the specific way in which communication is connected. It is not simply some residual category, but the constantly reproduction of uncertainty absorption and production provides the space of the possible. Uncertainty is always reduced to risk by social systems (that is, by processes of categorisation, world disclosure, interpretation and observation, etc.) thereby producing new internal contingencies and uncertainties in the form of other excluded alternatives, other possible worlds and perspectives that always set one's own position contingent. Uncertainty is the noise allowing the construction of order that depends on forms of social differentiation.

As way of summary, these three approaches differ substantially in the way they conceptualise risk. Luhmann and Beck share their interest in the structure imposing role of non-knowledge. Foucault and Luhmann are quite similar in their understanding of rules of formation, correlation, and transformation of discourses (or systems). Both want to unearth naturalised distinctions and lay open their contingency. Beck and Foucault both emphasise the anticipation of risks, the specific rationale of governmental techniques they give rise to, even though they differ in their assessment. I certainly do not want to claim that either one is better or more convincing than the other. However, it is quite telling that in all three approaches the notion of uncertainty plays a crucial albeit somewhat unrecognised role. Given their different conceptualisation of risk, they differ in their conceptualisation of uncertainty. Yet as different as they might be, all three certainly differ from the notion of uncertainty we usually encounter in traditional IR theory. This is, I assume, the very reason why they attract so much attention today. The notion of uncertainty is used to observe structures of sense-making, the processes whereby risks are framed, defined, managed and given meaning and on the basis of that subject to certain techniques and practices. All three in a sense share a critique of a technocratic understanding of risk that buys into unwarranted believes in technological ‘progress’. Given that all three approaches emphasise uncertainty and use the distinction of risk and uncertainty to generate specific dynamics and insights, the next section attempts to shape the contours of a notion of uncertainty that could then help us to assess the current changes in international law from such a perspective.

Risk, uncertainty, and the false promise of Keynes

The last section provided a reconstruction of the current risk literature and highlighted the importance of uncertainty. Given that the notion of uncertainty in the writing of Beck, Foucault and Luhmann differs from what we usually encounter in IR theorising, we have to ask what are these differences and what are the consequences? This section pursues this question by engaging in more detail with the distinction of risk and uncertainty and risk versus uncertainty.

Let us start with the boundary of the ‘traditional’ concept of uncertainty where uncertainty is embedded in a semantic field of games and instrumental rationality. Modern game and contract theory themselves are, on the other hand, based on expected utility calculus. The expected utility calculus took its modern shape with the publication of Theory of Games and Economic Behavior by Oskar Morgenstern and John von Neumann and was later refined by Arrow, Marschak and Savage.Footnote 46In its current form, the expected utility calculus is based on the four axioms of completeness, transitivity, independence, and continuity that ensure that an uncertain decision can be described in terms of comprehensive and mutually exclusive sets of possible states of the world. A decision under uncertainty is understood to be quite similar to the drawing of a ballot from a box.

Of course, the expected utility calculus was subject of a very intensive debate and all sorts of alternatives like regret theory or various forms of non-linear utility function were suggested, which kept decision theory very well on its toes for the last forty years. I won't discuss these different proposals in detail at this point as they all share the basic metaphor: to decide under uncertainty is to choose different pay off functions with probability distributions over possible states of the world. A decision under uncertainty is nothing more than a good bet. The notions of risk and uncertainty are thus the same. This is the concept of uncertainty we are familiar with from roulette or poker (and some boring classes in statistics): of course nobody knows what the next card will be or what hand the other player holds, nobody knows what number will be next – but the game structure is given and the odds can be calculated.

The boundaries of this approach are given by probability theory. Of course, to start a lengthy discussion on different probability theories might not be fruitful at this point, and it might be sufficient to point out that economic rationality is based upon probability theory and that the probability they focus on is the common interpretation of probability as a ‘long run’ frequency. As Morgenstern and von Neumann explain: ‘Probability has often been visualized as a subjective concept more or less in the nature of an estimation. Since we propose to use it in constructing an individual, numerical estimation of utility, the above view of probability would not serve our purpose. The simplest procedure is, therefore, to insist upon the alternative, perfectly well founded interpretation of probability as frequency in the long run.’Footnote 47

Richard von Mises is maybe one of the most well known proponents of the frequency interpretation. He explains further, the frequency theory is only admissible with well-defined ‘collectives’ (the presumption of well-defined games or settings in economic models does not come out of the blue). Probability theory from this point ‘is the study of transformations of admissible numbers, particularly the study of the change of distributions implied by such transformations’.Footnote 48 This implies that probability theory is understood as a statistical enterprise: the transformation of numbers, the ascription of numbers to mass phenomena and repeatable events.Footnote 49 Already on the level of probability theory, this fosters an empiricist understanding of science – including its well known ‘methods’. Or, as von Mises again assures us, ‘[i]n my opinion – the calculus of probability has nothing to do with metaphysics, at any rate not more than geometry or mechanics has.’Footnote 50

Yet it is important to remember that this approach presupposes that all the categories (ball, box, colours, etc.) are fixed and are prior to the decision, test, or ‘observation’. All epistemological questions are resolved and if they are raised (imperfect and incomplete information), then they are inferior to ontological questions. To start a discussion here for example on whether the three on the dice is really a three – and whether it is the same three for you as well as for me exactly misses the point. In other words: by definition, there can't be genuine uncertainty, that is, uncertainty that cannot be described in terms of probability distributions and possible states of the world – that is risk. Uncertainty can thus always be represented by numbers. Uncertainty that is ‘beyond’ number theory is excluded from analysis.

This helps us to understand why the call for genuine uncertainty is often phrased in terms of a critique on positivism, or why positivists often react quite uneasy when they are confronted (again) with their neglect of genuine uncertainty. To accept genuine uncertainty would require from them to give up their strongest beliefs. This is not an easy thing to do when this is all one actually has. One could almost feel sorry for them. On the other hand, if we are interested in more interesting questions and take the epistemological dimension seriously, we need to leave these narrow confines of rational choice behind and observe those processes whereby uncertainty is reduced to risk. At this point, there is often enough a reference to Knight and Keynes.Footnote 51 True, Keynes's notion of uncertainty derives from his logical theory of probability that is open for situations where probability distributions or a numerical representation of probability is not possible.Footnote 52 Situations of uncertainty are for Keynes qualitatively different from past experiences. In uncertain situations, i.e. situations that are not yet ‘reduced’ to well-defined settings, the relevant knowledge and non-knowledge is different from the quantitative, instrumental rationality.

Yet, I do think that Keynes's and Knight's notions of uncertainty are more part of the problem than part of the solution: while both are interested in a notion of uncertainty that is different from risk (and that I think it derives actually from Kant's philosophy, in particular his transcendential dialectics), both are interested in how uncertainty is reduced by conventions or functional expertise respectively. For both, however, conventions or expertise are empirical facts and don't imply an intersubjective epistemology: Keynes locates his discussion on uncertainty and conventions in his price theory. If we want to observe conventions, we can look at prices. Even if it is a convention how we name things, we still have to empirically examine them when we want to know them.

I think a more interesting notion of uncertainty can be found in Austrian Economics, in particular when we look at its founder Carl Menger and the later Hayek. Menger's notion of uncertainty is part of his subjective value theory that informs his interest in institutions. Menger wondered how institutions could emerge despite the fact that there is no general will, that they are not the result of an overall plan. His theory of spontaneous or unintended order is explained in book three of his Untersuchungen which influenced his later writings on the emergence of money as an institution.Footnote 53 What is surprising, is how early he recognised a very contemporary argument: already in the nineteenth century, Menger argued that economic institutions emerged by conventional agreement and change therefore organically. The organic nature of economic institutions transcends the capacity of individuals to ‘plan’ or ‘construct’ institutions which would require a mechanic order. Prices would therefore not describe equilibrium conditions as Walras or Smith would assume.Footnote 54 The entire concept of equilibrium as a steady state, a fixed point, is strange to Menger. For him, the economy is an ongoing process of transactions and communications. To explain the resulting order, the concept of equilibrium or ‘prices as representing equivalents’ is simply not necessary.

With recourse to Austrian Economics, we can say that, and quite in contrast to conceptions proposed by Knight or Keynes, that uncertainty cannot be reduced, bracketed or evaded by homogenous grouping over which probabilistic assessments are then made possible.Footnote 55 The subjectivity of experience and expectation formation always reorganise complexity in such a way that the reduction by conventions and norms is in constant need of reevaluation and reassessments. No individual can escape or step out from her human condition to take up an objective position. No one could claim to be in the possession of complete knowledge. While for rational approaches and for Keynes and Knight the observer stands outside his own model, theory, conceptual framework, for Hayek, for example, the observer is always part of the social phenomena he tries to understand.

To acknowledge the continuous reproduction of uncertainty leads us to a key distinction between the ontological (risk as uncertainty) and the epistemological concept (risk vs. uncertainty) of uncertainty: both entertain different ideas about contingency. For positivist approaches, the throw of a dice that determines the materialisation of a possible state of the world is treated as a natural phenomenon. It is unimportant whether it is actually a human being or a computer that ‘throws’ the dice. For an Austrian Economics inspired notion of uncertainty, contingency is always a social concept: the realm of the possible is structured by the net of actors, instruments, techniques and language. Over time, new actors, instruments, concepts etc alters the way actors observe themselves and others, what expectations they form and how things ‘emerge’ within the horizon of possible experiences.

From this perspective, processes of sense-making can be recast as transforming unstructured uncertainty into structured risks. This includes disciplinary knowledge: it makes a huge difference whether economists, lawyers, or catholic priests have the authority to define some situation and propose possible solutions. But their categorisation has practical consequences that are highly political: each discipline sees certain things and not others, it includes certain aspects and structurally excludes others, each discipline authorises some form of expertise and delegitimises other narratives. Hence, the absorption of uncertainty through categorisation produces new uncertainty in terms of these other excluded perspectives, the inscribed biases, and other possible worlds.

In this sense, the key focus in this risk literature is actually on the continuing absorption and production of uncertainty through ‘means of risk management’, that is, categorisation and sense-making. It might be too much to call ‘risk’ a misnomer, but I do think the reason why risk is so attractive as a theoretical concept is that it allows us to think differently about uncertainty, contingency, and the making of social order, that is, the reconfiguration of social relations, and of time and space in world politics.

To summarise the argument: the interest in uncertainty rests on at least three different pillars: the abandonment of positivist philosophy of science, a different concept of contingency, and a focus on the practical consequences of categorisations, discursive practices, and communications: our models, ideas, and concepts are not simply passive mirrors of empirical facts, but bring them about. From this perspective, as the writings of Beck, Foucault, and Luhmann actually do share, risk names a complex web of contingencies and thereby regulates the relationship between the present and the future: the contingent future sets the present contingent as it demands decisions, to take on a specific view or enact some discourse and not another. At the same time, the future is always visualised and imagined. As soon as the future becomes the present, a new future and horizon of possible alternatives is produced. Current changes in this relationship alter prevalent social relations and the role of time and space in world politics. The next section pursues the question of how such a perspective allows us to understand the politics of international law differently?

Risk and international law

The previous section outlined a notion of uncertainty that locates political processes right at processes of world making. How things are made economic or legal are not innocent processes, but shot through with epistemological and ontological assumptions and biases that have practical and political implications. When we now wonder about the ‘politics’ of international law, I think uncertainty directs us to the question how things are ‘made’ legal. That is: what happens when a fact, catastrophe, or some event is analysed, defined, framed in legal terms – and maybe not in economic or religious terms.

International law as we know it from textbooks is conceived of being based on state sovereignty where it provides a vocabulary to transform political disputes in legal claims. As long as the international was thought to be populated by sovereign states predominantly, international law was conceived of as a means for peaceful dispute settlement. Sometimes it worked, sometimes it didn't. By taking sovereignty as vantage point, the state was at the same time inside and outside of international law.Footnote 56 States were the author and the addressee of legal norms which automatically raised the question whether international law is really law or how it could be ‘enforced’ and backed up by sanctions. This predominant role of the state has also led IR theorists to the misguided idea that legal rules were epiphenomenal and consequently hardly worth the pain to be analysed in detail.

At first sight, this view seems to be supported by international law itself: the predominant self-description of international lawyers tells the story of emancipation from natural law through positive lawFootnote 57 where ultimately the state stabilised important legal distinctions like domestic vs. international and public vs. private law. The reference to state sovereignty not only divided public from private international law, but structured most of public international law itself: from the very definition of and associated rights and duties to the attribution of responsibility. Even the very distinction of law of peace, law to war and law of war refers to state sovereignty as organising principle.Footnote 58

The distinction of sovereignty on the one hand and the ‘normative authority’ of international law created the condition of possibility and the condition of impossibility of international law at the same time. It is worthwhile to remember Martti Koskenniemi's argument in his From Apology to Utopia where he identified the constitutive paradox of modern international law and showed us the horizon of possible arguments and the operational logic of international law.Footnote 59 He argues that legal arguments are based upon an uneasy tension between concreteness and normativity. As he explains in a later article:

An argument about concreteness is an argument about the closeness of a particular rule, principle, or doctrine, to state practice. But the closer to state practice an argument is, the less normative and the more political it seems. The more it seems as just another apology for existing power. An argument about normativity, on the other hand, is an argument which intends to demonstrate the rule's distance from state will and practice. The more normative a rule, the more political it seems because the less it is possible to argue it by reference to social context. It seems utopian and – like theories of natural justice – manipulable at will.Footnote 60

This paradox (that rules to be concrete and normative at the same time), structures the horizon of possible arguments in international law. At the same time, it creates and destroys law and has subsequently to be hidden by further distinctions like positivism/naturalism, autonomy/community, process/rule etc. This paradox and mode of self-description is however based on the notion of norms. The notion of norms is at the heart of when it comes to law, as it is common to define law as a system of norms. Indeed, it is often assumed that the unity of law is represented by a hierarchy of norms where an ultimate Grundnorm in Kelsen or the ultimate rule of recognition in H. L. A. Hart would secure the autonomy of law. The legal gaze and thus the question of the biases, finds its answer in the notion of norms whereby questions of legality or illegality of some fact, process, or conduct is determined. Law operates on the basis of norms, and not on prices, truth, justice, or love.

Of course, the notion of uncertainty as developed in part two resists the easy temptation to fall back into a positivist language game. The argument is not that international law is a system of norms, but that its operations take on the form of norms. Legal arguments, practices, communications refer to norms. As Friedrich Kratochwil has pointed out, norms have no existence outside of discourse, practical reasoning, or communication.Footnote 61 The question is thus not whether we face a hierarchy of norms, but how legal arguments are formed, connected, or enacted by modes of practical reasoning (like analogies). Yet, the construction, publication, and communication of legal arguments reproduce ‘law’ and it is the continuous operation that guarantees international law's existence.

Through the language of norms, lawyers can reduce uncertainty, make a situation comprehensible and then formulate legal arguments. The legal system is continuously reproduced by the recursive link of legal arguments. Yet these legal arguments can take only a particular form. A legal argument then provides a range of possible counterarguments which then might be evaluated in court. The language of norms constitutes a set of social relations, has its own temporality, space, and creates its own problems. Through norms, the politics of international law is also structured in a specific sense: as said above, the ‘function’ of international law is often said to provide a vocabulary for peaceful dispute resolution. It transforms political disputes in legal claims. Norms as ‘operational logic’ suggest that law has a civilising power on states. Ironically, this also implies that it can never reach that goal as it otherwise would eventually cease to exist.

When we shift the focus from sovereignty and actors to the reproduction of contingency and construction of the horizon of future possibilities, the question is not about either law or politics. Rather, we witness a change in the contingency production. As the kind of contingency legal rules address change, well established categories lose bite and we currently witness the development of new concepts. One of these new concepts is risk.

When risks regulate the relation between the present and the future, risks operate quite similar to norms: norms are counterfactually valid and stabilise expectations: although they certainly do not determine behaviour, they make some range of behaviour expectable and other forms hard to justify. Norms divide behaviour, actions, and communication in conformity/deviation and thereby relate the present to the future. When violence breaks out, a contract is breached, a right violated, the first question lawyers ask is what norm is to be applied, how that norm relates to other norms, and what legal consequences in terms of responsibility, accountability, or compensation arise. The language of norms structures unstructured complexity (a undefined situation for example) and thereby determines what ‘is the case’, who the relevant actors are and what rights and obligations have to be taken into consideration. From this perspective, the notions of norms and risks appear to be coextensive. However, there is a crucial difference: Although the concepts of risk and norms are functionally equivalent in a temporal sense, the notion of risk is more future-oriented as the notion of norms. Although norms stabilise expectations and may serve as a yardstick for expectable behaviour, the question of whether or not a norm has been violated requires an event to have already happened, while a risk is always related to a contingent future. One cannot violate risk like one can violate a norm.

That said, the introduction of the vocabulary of ‘risk’ in international law introduces a new temporality that threatens the stabilising functions of norms. As in particular the three risk approaches discussed in the first part emphasise: in particular the employment of precautionary measures redefines the temporality of international law: it reconfigures the politics of space into a politics of time: targeted killing, targeted sanctions, the revolution in military affairs, the surveillance and screening of movements and data are attempts to control time. Risk points to an acceleration in time which transcends the vocabulary of norms. It requires the control of possible threats, it provides a different rationality to allocate responsibility, authority, and accountability in international law. Risk redefines basic legal distinctions like public/private, war/crime, civil/military, inside/outside, or here/there. As a consequence, traditional confines of international law that are based on the temporality of norms are either circumvented, they collapse or readjust. With this different rationality also shifts the very function international law is said to fulfill: it is not a means for peaceful conflict resolution (alas uncertainty reduction), but a manager of contingency, a means to control contingency where international law does not constrain, but actually legitimises violence. Risk in international law creates nothing less than a different reality lawyers have to face: what is known, what is unknown, hidden, silenced, excluded, and how something is known or unknown. Risk subscribes to different concept of authority. Authority is now rooted in knowing what comes next or how the unknown future looks like, not in knowing one's legal history or the sound interpretation of legal cases. In other words, the vocabulary of risk may lead to a redefinition of the self-observation of international law.

Conclusion

This contribution pursued the question of how an uncertainty approach may allow us to reassess current changes in international law. It first provided a brief discussion of Beck's notion of world risk society, Foucault inspired understanding of risk as a dispositive, and Luhmann's sociology of risk to show that their arguments are based on a notion of uncertainty that is different from what we usually encounter in IR. The second part discussed in more detail different notions of uncertainty and suggested that a quick recourse to Keynes's famous quote doesn't do the trick. A broader perspective that takes his philosophy of probability into account suggests that this notion of uncertainty is deeply embedded in early analytic philosophy and does not entail an intersubjective epistemology. That means that the observer stands outside the model and can observe empirical processes objectively. That hardly captures what in particular Foucault and Luhmann are interested in. Rather, I suggested that a recourse to Austrian Economics where uncertainty described an inevitable human condition that includes the observer. This opens the way to understand uncertainty in terms of disciplinary knowledge or the autopoiesis of systems. What uncertainty points at, is thus a redefined production of contingency and the emergence of new contingencies that alter the connectivity of communications.Footnote 62 That is: disciplines or systems create their own reality. This allows us to observe political processes at the level of sense-making: the practical consequences of the biases when things are made economic or legal. Hence while politics is often enough used in terms of cui bono or is per definition hot-wired to the state, this literature observes politics at the level of sense-making through disciplines, discourses, fields, or systems. The third part then provided some of the current changes in international law. It suggested that risk provides a different temporality of legal operations that outgrow the contours and confines of norms. As a consequence, we see a redefinition of what international law is actually said to be for, what is it supposed to accomplish and do. Ultimately, this leads to different reasons and arguments for allocating authority and accountability in law.Footnote 63 These changing contours of the production of the horizon of possibilities and the way newness can emerge alters the way in which the uncertain possibility is absorbed into ‘manageable’ and actual risks. From this perspective, risk is not simply a calculative technique to control contingency. Risk excludes and includes certain knowledge. It determines what is not known and how it is not known.

In the end, the identified systemic shift raises a further conceptual difficulty that future research may need to address: insofar as ‘risk’ reconfigures the logic of inside/outside – that is, that they cannot be put into the boxes of national or international, public or private – we need a different conceptual apparatus that addresses these practices. At the same time, our political vocabulary by which we try to ascribe meaning to practices emerged with and is irremediably linked to the nation state. We simply do not have a vocabulary that allows us to go beyond the nation state to trace the current changes under way. Beck, Foucault, and Luhmann try to provide entry points from which the change of the intersubjective meaning structures can be reconstructed without falling back on a vocabulary that reinforces the individual actor and loses the intersubjective from sight. To make an additional step towards such a new vocabulary is probably the ‘promise’ of an uncertainty approach to international law.

References

1 See for example Adler, Emanuel and Haas, Peter, ‘Epistemic Communities, International Cooperation and World Order: Creating a Reflective Research Program’, International Organization, 46:2 (1992), p. 369CrossRefGoogle Scholar ; Rathburn, Brian, ‘Uncertain about Uncertainty: Understanding the multiple meaning of a crucial concept in International Relations Theory’, International Studies Quarterly, 51:3 (2007), pp. 533–557CrossRefGoogle Scholar .

2 Morgenthau, Hans, Politics among Nations: The Struggle for Power and Peace (New York: Knopf, 1948), p. 223 ffGoogle Scholar .

3 As one passage reads: ‘Groups or individuals … must be, and usually are, concerned about their security from being attacked, subjected, dominated, or annihilated by other groups and individuals. Striving to attain security from such attack, they are driven to acquire more and more power in order to escape the impact of the power of others. This, in turn, renders the others more insecure and compels them to prepare for the worst. Since none can ever feel entirely secure in such a world of competing unity, power competition ensues, and the vicious circle of security and power accumulation is on.’ Herz, J., ‘Idealist Internationalism and the Security Dilemma’, World Politics, 2:2 (1950), pp. 157–180CrossRefGoogle Scholar . For a deeper discussion on uncertainty and security dilemma, see Montogmery, E. B., ‘Breaking out of the Security Dilemma, Realism, Reassurance and the Problem of Uncertainty’, International Security, 31:2 (2006), p. 155 ffGoogle Scholar . See Jervis, Robert, ‘Cooperating under the Security Dilemma’, World Politics, 30:2 (1978), pp. 167–214CrossRefGoogle Scholar .

4 Emanuel Adler and Peter Haas, ‘Epistemic Communities’, p. 367.

5 Synder, G. H., ‘The Security Dilemma in Alliance Politics’, World Politics, 26:4 (1984), pp. 461–495CrossRefGoogle Scholar ; Glaser, C. L., ‘The Security Dilemma Revisited’, World Politics, 50:1 (1997), pp. 171–201CrossRefGoogle Scholar ; Taliaferro, J. W., ‘Security Seeking under Anarchy: Defensive Realism Revisited’, International Security, 25:3 (2000–1), pp. 128–161CrossRefGoogle Scholar . See also, Jervis, Robert, ‘Variation, Change and Transitions in International Politics’, Review of International Studies, 27 (2001), pp. 281–295CrossRefGoogle Scholar .

6 Of course one could argue that standard game theory hardly represents state-of-the art theorising. But even in newer contributions based on contract theory (or mechanism design), the semantic relation between uncertainty and risk is still constitutive for further analysis. See Koremenos, B., Lipson, C. and Snidal, D., ‘The Rational Design of International Institutions’, International Organization, 55:4 (2001), pp. 761–799CrossRefGoogle Scholar .

7 For a really fascinating discussion, see Harsanyi, John C., ‘Games with Incomplete Information Played by “Bayesian Players I-III.” Part I. The Basic Model’, Management Science, 14:3 (1968), pp. 159–182CrossRefGoogle Scholar .

8 For a further discussion see Kessler, O., Die Internationale Politische Okonomie des Risikos (Wiesbaden: VS-Verlag, 2008), chap. 4Google Scholar .

9 For a longer discussion see Kessler, Oliver and Daase, Christopher, ‘From Insecurity to Uncertainty: risk and the paradox of security policy’, Alternatives, 33:2 (2008), pp. 211–232CrossRefGoogle Scholar .

10 See in particular the contributions of Tanja E. Aalberts and Wouter Werner as well as of Anna Leander to this Special Section.

11 For other attempts to analyse uncertainty in IR see Best, Jaqueline, ‘Ambiguity, Uncertainty and Risk: Rethinking Indeterminacy’, International Political Sociology, 2:4 (2008), pp. 355–374CrossRefGoogle Scholar ; Rathbun, Brian, ‘Uncertain about Uncertainty: Understanding the Multiple Meanings of a crucial concept in International Relations Theory’,International Studies Quarterly, 51:3 (2007), pp. 533–557CrossRefGoogle Scholar . See also Kessler, ‘Uncertainty’, pp. 501–23.

12 Of course, the systemic shift identified in part two has more implications and affects diverse fields from aesthetics, ethics, design, art, competition, and collaboration between military companies that this contributions unfortunately cannot talk about. For a discussion see for example Lacy, Mark, ‘Designer Security: Control Society and MoMA's SAFE: Design Takes On Risk’, Security Dialogue, 39:2–3 (2008), pp. 333–357CrossRefGoogle Scholar . For an analysis of the ‘ethics of risk’ see Peter Burgess, Ethics of Risk, unpublished manuscript, and Peter Burgess, Anthony Amicelle, Khalid Koser, Keith Krause, Oscar Salemink, Edien Bartels, Rocco Bellanova, Alfio Cerami, Erik Eggum, Gunhild Hoogensen, Sonja Kittelsen, Kim Knibbe and Martijn de Koning, Promoting Human Security: Ethical, Normative and Educational Frameworks in Western Europe (Paris: UNESCO, 2007).

13 This section is a further development of Kessler, Oliver, ‘Risk’, in Burgess, Peter(eds), Handbook of New Security Studies (London: Routledge), Chapter 3Google Scholar .

14 Beck, Ulrich, ‘The Terrorist Threat: World Risk Society Revisited’, Theory, Culture & Society, 19:4 (2002), pp. 39–55CrossRefGoogle Scholar ; Beck, Ulrich, ‘The Silence of Words: On Terror and War’, Security Dialogue, 34:3 (2003), pp. 255–267CrossRefGoogle Scholar ; Beck, Ulrich, Die Weltrisikogesellschaft (Frankfurt/Main: Suhrkamp, 2007)Google Scholar .

15 Aradau, C. and Munster, R. van, ‘Governing terrorism through risk: Taking precautions, (un)knowing the future’, European Journal of International Relations, 13:1 (2007), pp. 89–115CrossRefGoogle Scholar ; Lobo-Guerrero, L., ‘Biopolitics of specialist risk: Kidnap and ransom insurance’, Security Dialogue, 38:3 (2007), pp. 315–334CrossRefGoogle Scholar ; Lobo-Guerrero, L., ‘Pirates, Stewards, and the securitisation of global circulation’, International Political Sociology, 2:3 (2008), pp. 219–235CrossRefGoogle Scholar ; Neal, A., ‘Cutting off the King's head: Foucault's Society Must Be Defended and the Problem of Sovereignty’, Alternatives: Global, Local, Political, 29:4 (2004), pp. 373–398CrossRefGoogle Scholar .

16 Daase, C. and Kessler, Oliver, ‘Known and Unknowns and the Political Construction of Danger’, Security Dialogue, 38:4 (2007), pp. 411–434CrossRefGoogle Scholar ; Kessler, Oliver and Werner, W., ‘Extra juridical killing as risk management: the breakdown of legal rationalities in the fight against terrorism’, Security Dialogue, 39:2 (2008), pp. 299–322Google Scholar ; Petersen, K. L., ‘When Risk Meets Security’, Alternatives: Global, Local, Political, 33:2 (2008), pp. 173–190CrossRefGoogle Scholar ; Petersen, K. L., ‘Risk, responsibility and roles redefined: Is counter-terrorism a corporate responsibility?’, Cambridge Review of International Affairs, 21:3 (2008), pp. 403–420CrossRefGoogle Scholar .

17 In this section, I will not discuss the various applications and avenues of these approaches, but rather focus on the theoretical arguments instead. A full discussion of the impact of these literatures on IR would require a different article and distract from the key argument concerning the role of uncertainty.

18 Beck, Ulrich, Risk Society: Towards a New Modernity (London: Sage, 1992)Google Scholar .

19 Beck, Ulrich, ‘Living in a Risk Society’, Economy and Society, 35:3 (2006), p. 332Google Scholar .

20 Ibid., p. 330.

21 Beck, , Die Weltrisikogesellschaft, pp. 29, 62Google Scholar .

22 Beck, , Risk Society, p. 14Google Scholar

23 Beck, ‘Living in a Risk Society’, p. 330.

24 Beck, , Risk Society, p. 230Google Scholar .

25 Ibid., p. 231.

26 Ibid., pp. 83, 85.

27 Unfortunately, this is not the place to engage with the promises and pitfalls of governmentality itself. See Neumann, I. B. and Sending, O., ‘The International as Governmentality’, Millennium – Journal of International Studies, 35:3, pp. 677–701CrossRefGoogle Scholar ; Leander, A. and Munster, R. van, ‘Private Security Contractors in the Debate about Darfour: Reflecting and Reinforcing Neo-Liberal Governmentality’, International Relations, 21:2 (2007), pp. 201–216CrossRefGoogle Scholar .

28 Foucault, Michel, The Order of Things: An Archaeology of the Human Sciences (London: Routledge 2001)Google Scholar , in original Les Mots et Les Choses (Paris: Gallimard, 1966)Google Scholar .

29 Foucault, Michel, Discipline and Punish: The Birth of the Prison (New York: Vintage, 1994)Google Scholar .

30 Foucault, Michel, ‘Knowledge/Power’, in Gordon, Colin and Foucault, Michel (eds), Power/Knowledge: selected interview and other writings 1972–1977 (Panthon Books, 1980), p. 97Google Scholar .

31 ‘By the word, I do not mean the mass of text which have been collected at a given period, or chanced to have survived oblivion from this period. I mean the set of rules which at a given period and for a definite society defined the limits and forms of expressability, the limits and the forms of conversation, the limits and the forms of memory and the limits and the forms of reactivation.’ Foucault, Michel, The Archaeology of Knowledge. (London: Pantheon, 1972), pp. 14–15Google Scholar .

32 See Foucault, Michel, Security, Territory, Population (Basingtoke: Palgrave, 2007)Google Scholar .

33 Foucault, Michel, ‘In the Flesh’, in Gordon, Colin (ed.), Power/Knowledge: Selected Interviews and Other Writings (New York: Pantheon, 1977a), pp. 194–228, 194Google Scholar .

34 Aradau/ van Munster, ‘Governing terrorism through risk’, p. 97.

35 Neal, ‘Cutting off the King's head’, pp. 373–98.

36 See van Munster and Leander, ‘Private Security Contractors’, p. 204.

37 C. Aradau and R. van Munster, ‘Governing terrorism through risk’, pp. 89–115; Aradau, C., Lobo-Guerrero, L. and Munster, R. van (eds), ‘Security, Technologies of Risk, and the Political’, Security Dialogue, 30:2–3 (2008), pp. 147–154CrossRefGoogle Scholar .

38 Ibid., p. 103; Lobo-Guerrero, ‘Pirates, Stewards, and the securitisation of global circulation’, pp. 219–35.

39 Ibid., p. 102. Also Daase and Kessler, ‘Known and Unknowns and the Political Construction of Danger’, pp. 411–34.

40 Ibid., p. 101.

41 Ibid., p. 102.

42 Niklas Luhmann's understanding of risk is best encapsulated in Luhmann, Niklas, Die Soziologie des Risikos (Berlin: de Gruyter, 1991)Google Scholar . For an introduction to his systems theory see Luhmann, Niklas, Social Systems (Princeton: Princeton University Press, 1996)Google Scholar . Unfortunately, I cannot go into detail about the promises of the notion of autopoiesis for overcoming many dead ends of traditional systems theory.

43 For an equivalent formulation see Derrida, Jacques, ‘Signature, event, context’, in Limited Inc (Vienna: Passagen, 2001)Google Scholar .

44 Foucault, Michel, ‘Réspone à une question’, Esprit, 371 (May 1968), pp. 850–874Google Scholar .

45 Also MacKenzie, D., An Engine, Not a Camera: How Financial Models Shape Markets (Cambridge, MA: MIT Press, 2006)CrossRefGoogle Scholar .

46 Marschak, Jacob, ‘Rational Behavior, Uncertain Prospects and Measurable Utility’, Econometrica, 18:2 (April 1950), Seite 111–141CrossRefGoogle Scholar ; Savage, Leonard J., ‘Une axiomatisation du comportement raisonnable face à l'incertitude’, Econométrie (Colloque International XL. Paris: CNRS)Google Scholar ; Arrow, Kenneth J., ‘Le rôle des valeurs boursières pour la répartition la meilleure des risques’, Econométrie (Paris:CNRS, 1953), pp. 41–47Google Scholar ; Übersetzt, , ‘The role of securities in the optimal allocation in the optimal allocation of risk-bearing’, Review of Economic Studies, 31 (1953), pp. 91–96Google Scholar ; Samuelson, Paul. A., ‘Probability, Utility and the Independence Axiom’, Econometrica, 20:4 (Oct. 1952), Seite 670–678CrossRefGoogle Scholar ; Malinvaud, Edmond, ‘Note on von Neumann-Morgenstern's Strong Independence Axiom’, Econometrica, 20:4 (Oct. 1952), Seite 679CrossRefGoogle Scholar .

47 von Neumann, John and Morgenstern, Oskar, Theory of Games (Princeton: Princeton University Press, 1944), p. 19Google Scholar .

48 Richard von Mises, ‘On the Foundations of Probability and Statistics’, Address delivered on 11 September 1940 at a meeting of the Institute of Mathematical Statistics in Hanover 1940, p. 193.

49 Mises, Richard von, Wahrscheinlichkeit, Statistik und Wahrheit (Wien: Springer, 1928)CrossRefGoogle Scholar .

50 Richard von Mises, ‘On the Foundations of Probability and Statistics’, p. 191.

51 Keynes, J. M., ‘The General Theory of Employment’, Quarterly Journal of Economics, 51:2 (1937), pp. 209–223CrossRefGoogle Scholar ; Knight, F., Risk, Uncertainty and Profit (Chicago: Chicago University Press, 1921)Google Scholar .

52 For a further discussion of the underlying probability theory, see Kessler, Oliver, ‘Uncertainty, Rationality and the Study of Social Institutions’, Review of Social Economy, 66:4 (2008), pp. 501–522CrossRefGoogle Scholar .

53 Menger, Carl, Untersuchungen über die Methode der Socialwissenschaften, und der politischen Oekonomie insbesondere (Leipzig, Duncker & Humblot, 1893)Google Scholar .

54 Menger, Carl, Grundsätze der Volkswirthschaftslehre (Wien: Baumüller, 1871)Google Scholar .

55 Knight, Frank Heynemann, Risk, Uncertainty and Profit, (New York: Harper and Row, 1921)Google Scholar ; Keynes, John Maynard, ‘The General Theory of Employment’, Quarterly Journal of Economics, 51:2 (1937), pp. 209–223CrossRefGoogle Scholar .

56 Koskenniemi, Martti, From Apology to Utopia: The Structure of International Legal Argument (Helsinki: Finnish Lawyers' Pub. Co, 1989)Google Scholar .

57 Koskenniemi, Martti, The Gentle Civilizer of Nations: The Rise and Fall of International Law, 1870–1960 (Cambridge and New York: Cambridge University Press, 2002)Google Scholar ; Kratochwil, Friedrich, ‘Sovereignty as Dominium’, in Lyons, G. and Mastanduno, M. (eds), Beyond Westphalia: State Sovereignty and Intervention (Baltimore: John Hopkins Press, 1995), pp. 21–42Google Scholar .

58 This prominent conceptual position was shared by realists and liberal alike. Even natural law and the Radbruch formula cannot avoid this reference.

59 Koskenniemi, Martti, From Apology to Utopia, reprint (Cambridge: Cambridge Univesity Press, 2005)Google Scholar .

60 The later article is Koskenniemi, Martti, The Politics of International Law, European Journal of International Law, 1:1, p. 8Google Scholar .

61 Kratochwil, Friedrich, Rules Norms and Decisions (Cambridge: Cambridge University Press, 1989), chap. 7CrossRefGoogle Scholar .

62 In Daase/Kessler, ‘Known and Unknowns and the Political Construction of Danger’, pp. 411–34; and Kessler, Oliver and Daase, C., ‘From insecurity to uncertainty: risk and the security paradox’, Alternatives, 33:2 (2008), pp. 211–232CrossRefGoogle Scholar . We tried to develop two typologies of the changing contours and interplay of knowledge and non-knowledge to give rise to different security practices in the War on Terror.

63 For a discussion see of course the other contributions to this Special Section. See also Oliver Kessler and Wouter Werner, ‘Targeted Killing’.