1

Plugging the GAP Working with Buzan: the Ilisu Dam as a security issue

by

Jeroen Warner1

jeroen.warner@wur.nl

Occasional Paper No 67 SOAS Water Issues Study Group

School of Oriental and African Studies/King’s College London University of London

Jan 2004

The paper was originally drafted in April 2000 for a seminar at McGill University, Montreal.

Summary This paper summarises and develops current theorising on the politicisation and securitisation of transboundary waters. Inspired by the ideas of Buzan, Waever and de Wilde (1998) it sheds light on the inter- and transnational politics of water projects and shows how linkage politics are explained and shown to be relevant in the water sector. The case of the Ilisu Dam project in south-eastern Turkey is used to illustrate how actors can operate in the domain of shared waters and how, through performing speech acts and creating material facts, they seek to influence the sovereignty and integrity of water resources and their use and allocation in complex circumstances. Key words: water, politicisation of water, securitisation of water, linkage in hydropolitics, trans-boundary waters, Ilisu Dam, Euphrates-Tigris basin. Introduction Eleven years after the fall of the Berlin Wall, we still have no consensus over what 'security' entails. While the concept has branched out into many directions, both horizontally (issue-areas) and vertically (levels of security), there is no unifying model of security in general use. More's the pity as the Babel of security talk not only prevents a clear academic debate, the discourse of security has rather important political implications. The present essay will argue that the security framework devised by Barry Buzan, Ole Waever and Jaap de Wilde can be useful in analysing international relations issues, as it introduces a grid in which non-traditional domains and levels of security can find a place. The concept of a 'securitising move' helps understand how 'security' can be subject to political manipulation. Enhanced by ideas from linkage politics, it can be used to make sense of different actor's strategies. With Buzan, I surmise that an instrumentality may underlie this: language may be used for the ulterior purpose of dominating a security domain, which in turn may help change the balance of power within the

1 Project co-ordinator (Multi-Stakeholder Platforms for Integrated Catchment Management), Irrigation and Water Engineering Group, Wageningen University, email: jeroen.warner@wur.nl , PhD candidate, Flood Hazard Research Centre, Middlesex University.

2

decision-making regime. At the substantive level, this article is especially concerned with the issue-area of water management. As a vital resource for which there is no substitute, water is fundamental to our collective survival and many have sounded alarm bells over potential 'water wars'. Moreover water is generally felt to be in crisis (e.g. Gleick 1993). Yet while the idea of 'water security' is clearly on the rise - the next Stockholm Water Symposium, scheduled for mid-August 2000, even has this as its central theme (see www.siwi.org), water security has not even been theorised. The present paper will kick off with a brief history of the bewildering expansion of the use of 'security' into a range of non-state, non-military areas. The Buzan model is then introduced in Section 1.2 as a way of structuring this rather chaotic field. In Section 2, a number of shortcomings are identified that need to be addressed to enhance the usefulness of the system in nontraditional security issues. I will argue that to arrive at a truly systematic understanding of security in its range of uses, and hopefully detect a pattern, we need to look into the way non-IR disciplines such as psychology and anthropology understand 'security'. One way of teasing out this pattern is presented in '2.1. It is noted that psychological theory reminds us that security is only one of two basic motives for action: security and risk, and that risk itself is two-sided as well ('2.2). Having explored the roots of security and expanded the model, Ole Waever's concept of 'securitisation' is introduced as a discursive strategy in '2.3, which in fact brings in an element of 'instrumentalist constructivism'. This strategy is expanded in '2.4. Section 3 outlines the case study, which deals with the controversy over the Ilisu Dam, which is part of the Turkish multi-dam Great Anatolia Project. The interest of this controversy is that it involves state, private and NGO actors, at the national and international level, and therefore likely to bring in different types of security strategies at different levels. The case is re-interpreted in light of the theory in Section 4, followed by a Conclusion. 1. Security revisited: beyond Westphalia Writing on security has traditionally been dominated by International Relations scholars, who traditionally wrote on the art of war and diplomacy. According to IR conventional wisdom, the present organisation of international society was created with the Peace of Westphalia of 1648. Formally independent, endowed with absolute sovereignty- the supreme, independent and final authority - clearly separated by borders and the obligation of non-intervention in domestic affairs of other states, states are commonly portrayed as unitary, impenetrable billiard balls. States are assumed to be acting to maximise its own 'national' interest and always on the lookout for power and stability. The state's key interest, however, is national security. A state, it is maintained, will always attempt to maximise means to safeguard its security in direct ratio to perceived threats to state survival. The Westphalian paradigm clearly gives priority to military security and diplomatic relations (high politics) over development issues including economic and environmental problems (low politics). High politics is followed closely by top decisionmakers; low politics is left to the relevant minister or delegated to civil servants, and only exceptionally capture high-level attention (Mouritzen 1996: 67). Countries that fail to solve their high politics differences will never cooperate on low politics (Lowi 1993). But the traditional division is not cast in stone - it is up to the state to decide what is high politics (Mouritzen, 1996: 73). Foreign affairs agendas become larger and more diverse, and the traditional hierarchy among issues are falling away. 'No longer can all issues be subordinated to military security.' (Keohane and Nye 1977). Most notably, 'economic security' (re)surfaced as a policy concern in the administration after Ezra Vogel questioned U.S. superiority (1979). As Japan's economic power grew to rival American hegemony, and trade wars with the European Union became commonplace in the mid-80s, students of International Political Economy started to focus on economic security, raising a formerly 'low-politics' issue into the domain of 'high politics'. There was not just a horizontal shift in 'security domains' going on, a renewed interest at other levels of security was also noticeable. At roughly the same time the concept of human security gained currency. The 1982 Palme (UN) Commission on Disarmament and Security issues was apparently the

3

first to include environmental security, and soon UNDP started including noneconomic indicators such as literacy, longevity and health in its Human Development Report in 1987, partly in response to the 1982 Palme Commission which is credited with introducing the concept of 'human security'. This shifted the level of analysis away from states towards communities and individuals, from warheads to hospital beds. Later the Bonn Declaration defined human security as '...an absence of the threat to human life, lifestyle or culture' (cited in Solomon 1999) suggests a cultural domain of security. Finally, environmental security drew attention to the subject of security: the environment, and people depending on it, though in the subsequent policy debate environmental security has generally come to mean threats to nation-states from conflict over environmental resources (see, for example, the thorough discussion of the literature by Gleditsch 1997). I will briefly touch on two types of criticism the rise of environmental security has elicited. A purist approach to security has warned against the inflationary use of security (e.g. Deudney 1990) and the extension of security to other domains than the military sector and the state. These are essentially two different problems. It is eminently possible to accept the broadening of categories to include non-military domains while still subscribing to a worldview in which the state is the key nexus between the local and global levels and be concerned with controlling risk and conflict. Others have recognised that military security, concerned with protecting territory and independence, is only one domain in which conflicts are meted out. The other type of unease came from a radical perspective pointing at the Hobbesian-Malthusian fallacy of equating scarcity with violent conflictm then presents strengthening the Westphalian state as the only solution. Critical authors have questioned the suitability of the state, notably the military, to provide environmental security, it being among the worst polluters (Dalby 1997) and warned against more pervasive intervention and the danger of a panoptic state (Klein 1997). Finally, Duffield (1999) has pointed at a worrying convergence of international development aid and security politics. Perhaps, as a close reading of his article suggests, the rise of 'environmental security' is part of a rather grander design, a concept of control? (Warner 2000). The next section will briefly venture into this possibility. Turbulence and insecurity Nonmilitary forms of security slowly but surely attracted interest in academic circles but also in the NATO community. It is tempting to surmise cynically that the rise of environmental security was about a discredited security establishment inventing new challengesto legitimse its privileged role. Indeed, in the early 1990s President Bush assigned his former employer, the Central Intelligence Agency, to search for and analyse new security problems (Lipschutz 1995). However, there was something to be said for the observation that the world had turned out to be a radically different place once the reassuring veil of the Cold War had been lifted, necessitating a new conceptual framework. First, the security community rediscovered nonstate conflict. Most conflicts are not battled out between tes but within states: the number of civil wars outnumbers violent interstate exchanges World-wide, governments do not so much fight each other as confront domestic opposition, and these conflicts not seldom involve natural resources (de Wilde 1995). This was only one phenomenon of many worrying top decisionmakers in the post-Cold war era. The speed of events seemed to assume unmanageable proportions - at the start of the 1990s, 'turbulence' (Rosenau 1990) was the order of the day and the spectre of 'ungovernability' became a prime concern in policy-making circles. The 'erosion of the state' in the face of 'globalisation' and the end of 'Cold-War stability' started to keep policymakers awake at night. Notably in the US Robert Kaplan's essay in the Atlantic Monthly 'The Coming Anarchy' touched a raw nerve. Drug trafficking, money laundering, mass migration, inner-city lawlessness in US cities, warlords in West Africa - it all seemed to come together. In the environmental area, the spectre of 'water wars' (Starr 1991; Bulloch & Darwish 1993; Homer-Dixon 1999, de Viliers 1999) seemed to necessitate greater capacities for (global) intervention. In a combination of Malthusian worries about environmental degradation and Hobbesian faith in the 'strong state', some started to advocate a strengthening environmental diplomacy which has resulted in the establishment of 'environmental hubs' in 12 potential troublespots (Dockser-Marcus & Brauchli 1997, see also Warner 2000 and Warner, forthcoming).

4

In a more general sense insecurity seems to have become a much more pervasive phenomenon. In the post-modern age, Bauman (1999) argues, security in its widest sense (German Sicherheit stands for security, safety as well as certainty) is eroding. This individual insecurity apparently is a world-wide phenomenon: the Council on Global Governance (1995: 79) notes that 'Despite the growing safety of most of the world's states, people in many areas now feel more insecure than ever. The source of this is rarely the threat of attack from outside. Other equally important security challenges arise from threats to the earth's life support systems, extreme economic deprivation, the proliferation of conventional small arms, the terrorising of civilian population by domestic factions, and gross violation of human rights.' The transnational nature of many hazards again erodes the capacity of the state to provide security to its citizens. The trend to outsource and privatise security functions raises questions in the realm of equity and sustainability. More and more, civil society capacities itself are tapped in contributing to its own security provision - in the domain of water: flood proofing and flood warnings). This changes the traditional2 'security contract' in complex ways (See Warner & Turton, forthcoming on the 'hydrosecurity contract') Fig. 1 Security exchange - traditional ideal-type \ Actors Deliverable

State Citizens

Security Supply Demand Legitimacy Demand Supply In view of all these types of security at different levels, a need has been to systematically organise these concepts. In the early 1990s, different typologies have been advanced, if not always with the same rigour. Lonergan's list, for example (economic, food, health, personal, community and political security; Lonergan, 1996) seems to mix up levels and types: health is an individual as well as a group characteristic (public health); community is at the group level and seem to refer to identity. In addition to domains, different levels of aggregation are now advanced for security analysis. When we discuss security issues, this is usually understood as interstate politics or, more recently, state-society relations (Ayoob 1997). This article sketches a case in which a new actor has taken the security stage: the (international) private sector. The ongoing privatisation of utilities brings in international actors, which in turn may bring their host governments into the fray. This produces quite a novel security dynamic. 1.2. Buzan's security model In response to the rise of new security concerns, different typologies have been advanced in the early 1990s, if not always with the same rigour. Lonergan's list, for example (economic, food, health, personal, community and political security; Lonergan, 1996) mixes levels and types: health is primarily an individual characteristic; community is at the group level and refers to identity. A more systematic categorisation is Buzan et al's (1995, 1998), who postulate five domains of security are military, economic, environmental, socio-cultural, political, at different levels: the international state system, the macroregion, the state and substate levels.

2 Note that 'traditional' is a relative concept: in most states water only became a public utility in the course of the present century, see e.g. Blokland, Braadbaart & Schwartz 1999.

5

level/ type Military Economic Environmental Societal Political

intl. system macro-region unit (state) institution individual prime concern: integrity wealth health identity legitimacy, sovereignty

Fig. 2: Security Diagram, after Buzan et al 1998 . Buzan's five domains of security are military, economic, environmental, societal and political. These are roughly the same categories as those identified by Nef (1996), but ignores the military sector altogether and frames societal security as a 'cutltural' domain. His categorisation lacks Buzan et al's vertical axis, that of levels of aggregation, that holds out the prospect of an even more structured analysis. The five levels of aggregation are international, macroregional, state, group and individual level. Arranged on two axes, this yields a 25-area security grid. For want of a more convincing classification, I shall therefore adopt Buzan's model, while allowing for the possibility that changes will need to be made to make it more suitable for analysis of water issues. 2. Working with Buzan: filling some gaps Given its nicely systematic approach to a complex issue, it is a pity that the diagram and underlying theory has not seen wider use. One reason could be its rather eclectic approach to epistemology. Confusingly - at least for those IR theorists who like their epistemologies tidy - Barry Buzan has called himself a 'liberal Realist' (Litfin 1999). This dualism is reflected in the hybrid methodological approach in his work. On the one hand, his work stands in the tradition of state-centric Realist power play (Realpolitik). On the other hand, he toys with regime theory in his theory of regional security complexes (applied in Ohlsson 1995) which hypothesis a continuum between anarchy and a security community, distinguished through each other by the number of regimes in place. And thirdly, the theory of 'securitisation' in Buzan et al, which originates in Ole Waever's work, suggests a constructivist approach in which 'security' is what you make it. Security is not an objective absolute, but a value you can manipulate. The eclecticism of combining these three strands (Realism, pluralism and constructivism) will be taken as given here. It should be noted that the only case study presented, in Buzan et al 1998, on European integration and security, does not provide any insights in the research method, the development of which merits a separate article. This article will make a claim, however that Buzan's grid, as well as Waever's (1995) related strategic concept of 'securitisation' could be more useful in understanding security issues and applied to water issues if sufficiently amended and 'deepened'. In this sense, an attempt is made to plumb some theoretical 'gaps' left by the Buzan framework, or make implicit possibilities more explicit. First, the authors do not make it clear why people should seek security. Drawing on psychology and systems theory, common criteria for 'security' are sought, which may then be applied to the specificities of each of the domains. Second, it will be argued that the unipolar concern with security excludes an equally important motive for human action: desire, or 'self-realisation'. It will be suggested that a two-attractor model could be a useful development from the Buzan model. This logically leads to a further critique: the relative arbitrariness of the 'core values' in the

6

model. While they certainly have an ad hoc plausibility, a truly constructivist approach would maintain that the definition of a core value itself for each domain is subject to contestation as well. It can therefore be surmised that this 'core value' may be the object of political struggle. Fourth, while Buzan et al propose a conflict within and between types of security, they do not really indicate why a certain issue may end up within a certain domain. Especially in complex issues such as water management, labelling it an 'environmental issue' is beside the point: it touches on all the domains and levels. Where it ends up, it may be argued, is subject to the success of strategic action. I shall approach this by drawing on another central theme to Buzan et al's work: 'securitisation', a speech act legitimising extreme measures by calling on existential threats, and connect this to concepts culled from 'linkage politics'. It will be argued that security and risk (parsed as threat and opportunity) are equally subject to management-of-meaning strategies, i.e. 'speech acts'. Finally, I will suggest a minor expansion to the labelling or understanding of two of Buzan's security domains, the military and societal sector, to tailor the analysis to water issues. As this paper reports work in progress, it cannot hope to provide a comprehensive path in any way. 2.1 Criteria for security '(A) great portion of all our activities, both conscious and unconscious (Y) seem to be directed toward

maintaining ourselves as living systems endowed with identities.' (Rapoport 1974). What is security? In its most widespread understanding, security is about protection from threats (in Hasenclever, Meyer & Rittberger 1997). Yet, everyday, dozens of ants may be threatened by our footsteps, but we are unconcerned. Thus, the implicit assumption is that there is something worth protecting, otherwise the threat would not concern us. Traditionally, about the only thing worth protecting was the Westphalian nation-state and its core values, often recast as goals. But if we accept the reasonableness of expanding the concept of 'security' beyond this state-centric concept, we find that in different areas of political debate, different things (material as well as immaterial) are felt to be 'under threat' and 'worth protecting' even if they are not always literally termed 'security'. From my current research into three flood protection projects alone, values as diverse as 'hallowed ground' ecosystem diversity, project integrity and community cohesion have been deemed inviolable and vital (in Buzan et al's terminology: 'securitised'). If it were possible to find some degree of consistency in the application of 'security', this could be useful in developing a systematic analysis. Let's take a very brief tour d'horizon down the road of psychology, human geography, anthropology and systems theory. In developmental psychology, 'security' is associated with attachment. As a child develops, it attaches itself to the mother, then to its immediate environment. If the attachment is not secure, developmental problems can be expected in later life. A child that is taken away from its mother will show signs of extreme fear. The attachment to one's 'environment' stays with most people. Geographers recognise that people are attached to their space, their place. We identify ourselves with our space - it becomes part an extension of the self (identity). This is why displacement and enforced resettlement/migration can be so traumatic. People crave at-home-ness (Tuan 1980; Hewitt 1997). The house, belongings, neighbourhood are seen as an extension of the self (Bauman 1999). Outside this 'safe area' is the 'big bad world'. We fear what we don't know, either people or territory, therefore we opt for the familiar rather than the unfamiliar, and have a preference for tenure and protection, a savings account and insurance. Even in migrating groups, in- and outgroup boundaries need to be maintained to maintain identity. 'The concern with identity leads us to distinguish ourselves from our environment, to stress difference.' (Rapoport 1974: 200). 'An ethnic group is about boundary maintenance; ethnicity is a way to structure interaction which allows the persistence of differences' (Vayrynen, 1998). 'Difference' may lead to the perception of conflict with others and conflict with nature and the desire to coerce this Other into submission. In such a world, danger is always present at the border (Douglas 1992). Without borders, there would be no danger. Boundaries discipline ambiguity and

7

contain contingency (uncertainty) - and help establish difference: 'us' and 'them', East and West, North and South. Campbell maintains that 'the ability to represent things as alien, subversive, dirty, or sick, has been pivotal to the articulation of danger in the American experience (Y) The mere existence of an alternative mode of being, the presence of which exemplifies that different identities are possible and thus denaturalizes the claim of a particular identity to be the true identity, is sometimes enough to produce the understanding of a threat.' (Campbell 1992). Psychologists (e.g. Kramer & Roberts) have noted the power of projecting one's fears on adversaries and one's uncertainties on victims to enhance one's security. In international relations this takes the form of seeking an external enemy with a view to uniting the population: Here, too, outward-directed conflict procures internal integration and cohesion (Rapoport 1974). This aspect is especially apt in light of the case study. Finally, systems theory teaches us that a system seeks stability over time (defined as the faculty of a system to maintain its state or revert to it in spite of or after disturbances). According to Talcott Parsons, one of the main functions of legitimacy is pattern-maintenance: the system's capacity to maintain its stability through the inculcation of shared values and the existence of widespread cultural norms' (Rush 1992). In light of the above, two observations come to mind. First, it is notable, then, that these uses of 'security' are not all concerned with survival, but rather with the maintenance of some vital values, properties or assets. It seems useful to separate out the uses of security that have to do with 'survival' (existential threat of annihilation) and those that are about 'loss' (nonexistential threat to system properties) while noting that the latter may still be framed in terms of the former. The Realist discourse of international relations associates security with stability and integrity. While 'stability' usually denotes the stability of governments or the international balance of power, 'integrity' is commonly associated with territorial integrity and 'identity' with cultural identity. However, on closer inspection these qualities apply to almost all categories (domains). For example, to those living on a territory, a landscape has a valued identity (Tuan 1974) and a cultural value system can have stability3. Note that the word 'maintenance' connotes preservation, indicating a conservative bias. Tansey (under review) criticises invoking psychological needs for security as the only explanation for the existence of social institutions, as it presents society as inherently conservative. Indeed, one persistent criticism the Realist in International Relations is its ahistoric, equilibrium-centered world view (see Strange 1982 on regime theory). Perspectives allowing for development and change, either in its linear or its non- linear manifestation bring in another dimension of security: metastability, that is, the stability in the face of disturbance and change. As the actor's environment changes, there is little hope of the actor being able to revert to its former state - if it could, it may not be able to maintain itself in a new environment. So the actor will have to change too. In that case, the concern is with identity: to what degree can the actor change without losing its identity? An ecological systems approach shows that the survival of a system can be realised in the face of change. This quality, often termed 'metastability' is also understood in terms of as system integrity, where, integrity refers to the ability of a system to maintain its organisation in the face of changing environmental conditions i.e. the ability to handle stress (Arnesen 1997). This understanding of the world, informed by insights from ecology, brings in the notion of resilience. In order to enhance the system's resilience, the actor will have to diversity its options. For the survival of the system as whole, cultural and ecological diversity may be relevant policy prescriptions. Another term used in ecology is persistence: ability of a system to maintain itself under different equilibria (van Dorp 1999). Brief, in addition to survival, identity, integrity, stability and persistence in the face of change are clues to the type of values that actors hold dear in each of the security domain. A threat to these characteristics are likely to incite people to seek security and make security moves. However, what will be the core value is subject to subjective politics. This is the burden of the next Section

3 'Security is the certainty of reliability and stability (Stätigkeit) of fundamental values produced by the system. (Deutsch, q in Barandat 1997).

8

2.3. Securitisation: narratives and speech acts So far, a framework for analysis has been constructed that understands people as needs-driven and, I daresay, rational. Cultural anthropologists teach us, however, that this rationality is relative, perhaps even spurious. We do not perceive the world 'as it is' - rather, we devise stories about the world that makes sense of the messy reality we are presented with. This section will argue that this construction may have a political function. Why do we construct stories? To understand the world, people ascribe meaning even if there isn't any. Risk is about loss, but without meaning, we ourselves are lost. We construct a coherent world-view that lends a logic, a meaning to our existence. Narratives lend a 'deep structure' and, hence, coherence to ambiguous situations. However, the construction of narratives to create coherent worldviews, can also be done actively, for political or other purposes. In their interaction with the world, people create representations that become legitimate certain images (Berger & Luckmann 1991 [1966]: 110ff). As people attach different interpretations to the meaning of a problem and its proposed solution(s), they may seek to may shape their representation of the issue in line with their perceived interests and try to convince others that it is the proper view. Authors such as Smircich & Morgan 1982, on the management of meaning) in management studies and Austin (1975, on declarative and performative acts) have shown how you can perform quite powerful feats with language to make things happen. Buzan et al's (1998) approach sees the speech act of 'securitisation' as a deliberate political use of language to move an issue into the language of the absolute. 'Politicisation means to make an issue appear to be open, a matter of choice, something that is decided upon and that therefore entails responsibility, in contrast to issues that either could not be different (laws of nature) or should not be put under political control (..) By contrast, securitization is a speech act legitimising extreme measures by calling on existential threats, as 'so important that it should not be exposed to the normal haggling of politics'.(Buzan et al 1998). In so doing the speech act moves politics beyond the normal, competitive rules of the game. Here, then is the quick and dirty recipe for securitisation: follow the security form, the grammar of security, and construct a plot that includes existential threat, points

of no return, and a possible way out - the general grammar of security as such plus the particular dialects of the different sectors, such as talk identity in the societal sector, recognition and sovereignty in the political sector, sustainability in the environmental sector, and so on. (..). it is implicitly assumed that if we talk of this (...), we are by definition in the area of urgency: by saying 'defense' (or in Holland, 'dikes'), one has implicitly said security and priority. (Buzan et al 1998: 27).

Buzan et al's approach is constructivist in that it is not concerned with the absolute truth or falsity of about the life-world - just like 'even a successful technology may be based on (partially) false assumptions' (Pinch & Bijker 1987) so a successful strategy may be based on a manifestly false theory about the world. The concepts in which we understand the world help to make it what it is (Berger & Luckmann 1991 [1966]). Note, however, that their constructivism is of the 'weak' type: it does not claim that perceptions of threats exist entirely independent of the threatening phenomenon. When foreign tanks cross borders, perceived risk and the actual event are virtually the same and in the pre-industrial world, '(s)tarvation, blight and famine are perennial threats. It is a bad joke to take this [type of] analysis as hinting that the dangers are imaginary.' (Douglas 1990) But still the intention with which the tanks are sent or the origin of the blight may still be interpreted in different ways, and indeed with the help of Douglas' anthropological work (e.g. Douglas and Wildavsky 1992) it can be shown how attribution of blame for insecurity and risk (a phenomenon which she claims is on the ascent in Western societies) has political outcomes - its history has shown the short avenue between perceiving a national threat and blaming the Jews, the Russians, immigrants, global capitalism for it. In the context of an international water, it is easy to blame the action of an upstream country for crop failure or flooding downstream, while reasons may

9

be far more complex. Blame, as part of framing the problem as a danger, will also point at a certain, politically welcome solution: expel the Jews/immigrants, increase the defense budget, punish the upstream country, etc. Managing the representation (meaning) therefore seems a vital ingredient of security strategy. Note that process of legitimation of certain problem images can become institutionalised and routine (Berger & Luckmann 1991 [1966]). Likewise, securitisation (as shorthand for 'invoking an existential threat to legitimise exceptional measures') can become the norm: 'where states have for long endured threats of armed coercion or invasion, and in response have built up standing bureaucracies, procedures and military establishments to deal with them' (Buzan et al. 1995, 1998). In the Netherlands, the knee-jerk response to floods has become to let a domineering Rijkswaterstaat (the Dutch waterways agency) devise a new round of flood-protection structures; the spectre of 'international anarchy' routinely legitimises a strong hegemon and seems to explain the persistence of NATO in the international arena after the Cold War. By bringing in speech acts, then, the authors effectively bring in a type of instrumentality, by which perceptions of reality can be manipulated for political reasons. Especially the strong value component that is germane to issues known as 'intractable problems' such as ethical issues or NIMBY issues (Fischer 1995), may cause them to be framed in extreme language, in so doing linking the value at issue to basic security (survival, integrity, identity, persistence). Especially in public controversies, ambiguity and uncertainty are intolerable. By providing a 'deep structure', narratives help create order and meaning in an ambiguous situation (Bruner q. by van Eeten, 1997). My additional contribution to this promising idea is to surmise that actors seek either to dominate a security domain by presenting an issue as 'absolute' in its terms, or by strategically recasting the issue in terms of a different domain. This will be explored in the next Section. Speech acts and domain linkage Apparently not everyone can perform a speech act: the success of a speech act also depends on the social position of the enunciator. Waever (1995) limits performative acts to authority figures: 'security is articulated only from a specific place, in an institutional voice, by elites', or as Buzan et al have it, 'the enunciator must be in a position of authority' This does not limit the floor to public officers. One social group that suggests itself in its ability to 'speak authoritatively' is experts, a strategy successfuilly employed by NGOs. Litfin (1999) however feels Buzan et al's approach makes for too exclusive a securitising club. She feels this approach 'abdicates too much terrain to the security traditionalists'. Security speech acts, she claims, are performed 'on a daily basis by an increasingly diffuse group of scholars and practitioners.' The fact that they may or may not do this reflectively, i.e. 'conscious of how they construct their speech acts' does not detract from the outcome. But what about the recipient, who must validate the speech act as legitimate? Clearly not all speech acts are accepted, though - they will have to meet with 'felicity' to become a mobilising force (Tansey, under review). We have noted above that people crave control over their destinies, and dread uncertainty. Traditionally, knowledge gaps in risks were converted into certainties by religious or magical knowledge (Giddens 1990: 125) Today, sin and taboo are no longer met with felicity - but the magic wand of risk management often is (Douglas & Wildavsky 1994). A great measure of uncertainty and issue complexity therefore, is a likely precondition for the acceptance of a speech act (cf. van Eeten 1997). For example, in the context of alarming figures of ecocapacity, Tjallingii (1996) notes there are three responses: (a) the figures are not true, (b) the figures are not alarming, or (c) the figures justify declaring a state of emergency. The first response ('not true') is only possible where the issue is easily decided by scientific method; (b) gives occasion to 'undersecuritise' the issue for lack of conclusive scientific data, while option (c) may be invoked to justify undemocratic (eco-authoritarian) measures. I suggest that while the field is still open, actors can seek to (1) colonise different security domains in the context of an issue-area where none has been taken, (2) fight the dominant actor/value of that

10

domain or (3) cast doubt on the evidence underpinning the argument. It may prove especially helpful if we do not regard the speech acts as isolated elements; they are parts of a narrative, a 'fairytale' of how things have come to be the way they are (Van Eeten 1997). Especially in public controversies, ambiguity and uncertainty are intolerable. By providing a 'deep structure'. narratives help create order and meaning in an ambiguous situation (Bruner q. by van Eeten, 1997). Narratives frame the solution in terms of influencing or neutralising action - curbing opportunities for resistance (securitisation) or pushing for measures to check and correct untrammeled technocracy (desecurisation). Narratives frame the solution in terms of influencing or neutralising action - curbing (indammen) opportunities for resistance (securitisation) or pushing for measures to check and correct untrammeled technocracy (desecurisation). Buzan et al, then, have not taken the contested and constructed nature of security to its logical conclusion: that each area of the security grid is contested. Could it be that actors seek either to dominate a security domain by presenting an issue as 'absolute' in its terms, or by strategically recasting the issue in terms of a different domain? The complexity of water-related problems seems a perfect vehicle for linkage. Regime theory tells us that issue-areas (domains) are or can be linked, passively or actively. Hisschemöller and Olsthoorn cite the linkage of flood protection (saving lives) and environmental security (saving nature) in the Dutch Rhine flood management debate (Hisschemöller & Olsthoorn 1999). While it is likely that many very specific value conflicts can be confined to just one security domain, the conflict between the many uses and users of water seems to promote the strategic recasting of a water issues into the politically most advantageous domains. These links may be made by a veto actor who seeks to obstruct a policy, but also by a project initiator to co-opt resistance. Note that the capacity to link, just like the ability to 'securitise' depends on the power to mobilise opinion (Warner 1998). If that solution is not seen to address (perceived) genuine problems, it will not find ready acceptance. 2.3. Risk: threat and opportunity As we have seen, we opt for the safe rather than the possible. Therefore, people tend to look what they know - they seek certainty and security. But, as Buzan (1991) notes, a world of maximum security would be a very dull world indeed. If everyone was only concerned with survival and security, little initiative would be taken. In opposition to Freud, Abraham Maslow saw mankind as motivated not just by the need to keep afloat but to pursue personal growth. This is reflected in statements such as: 'That which is not growing starts to rot' (attributed to a Russian general, q. in Rapoport 1974). and 'Security is development, and without development there can be no security.' (Robert McNamara q. in Myers 1993). Apparently, two attractors are at work, which psychologists have termed eros and thanatos. these two strivings can help us understand the dual meaning of risk and uncertainty. Famously, the Chinese character for 'crisis' is composed of the character for 'threat' and 'opportunity'. Similarly, concepts of 'risk' and 'anxiety' can have positive as well as negative connotations. A risk can be experienced as a danger, as hazard, as well as as a thrill, an opportunity for achievement (Tuan 1980), see Fig. 3. Likewise, a security risk is not just a 'threat' (inspired by fear) but also an 'opportunity' (inspired by desire) and the Janus-faced concept of anxiety most neutrally denotes the awareness that something might happen. It is open to one's mental schema (interpretation) to attribute the event as positive or negative. We always interpret the unknown in terms of the known (Maslow), so the understanding of anxiety is likely to reflect one's personality.

11

Fig. 4: Two sides of the same coin: fear and curiosity curiosity (+) / anxiety \ fear (-) Note that the two 'attractors', embeddedness and emergence, are only polar opposites in that the former is 'active' or centrifugal while the other is 'inactive' or centripetal. The left-hand column lists active, dynamic, developmental, potentially tension-creating pursuits that, incidentally, may put the status quo at risk; the states of mind listed in the right-hand column are passive, static, attachment-seeking and tension-avoiding pursuits that may impede necessary change. 'Sustainable development' in this light is an attempt to have it both ways: growth as well as stability. A dual-attractor rather than single-attractor model is compatible with empirical findings in risk studies. We seek an equilibrium between the two states of security and risk rather than maximising opportunity. It is a well-known finding in risk studies that safety measures in traffic do not necessarily lead to fewer traffic accidents. Perceptions and behaviour get in the way: as traffic participants perceive greater safety, they tend to take greater risk. Similarly, accident-proof playgrounds lose their attractiveness to children (Adams, 1995). Conversely, inhabitants of flood-prone areas are much more likely to save themselves and reduce damage in the event of a flood than those 'protected' by dikes and barriers. Their heightened awareness has made them adapt to the flood risk [sources]. Apparently, there is a 'sensible equilibrium' between undue security and undue risk. Fig. 5: Life: a tug-of-war between two attractors Emergence Embeddedness openness to a stimulating world anxiety about the world self-realisation safety, familiarity, at-homeness growth safety differentiation unity freedom hope, joy anxiety, helplessness, impotence development defense mechanisms goal-directed pursuit satisfying needs exploration anxiety (rigidity) The power of opportunity If we now turn to legitimation of security measures, we find Buzan et al customarily present 'defensive' examples to support securitisation theory, such as controls on trade, emergency rule, disbanding parliament, curbs on the press and even dikes to protect an actor from harm (Buzan et al 1995; 1998). But we have seen in the previous Section that protection is not the only attractor for action. If all states were concerned with defense, none of them would seek empire, which can be read as self-actualisation at the state level. Colonisation is an offensive (often in more than one sense of the word) rather than defensive act and may involve, and as Swyngedouw has noted, colonisation of the

12

internal resource base as well as foreign territory. National development at the state level or political advancement at the individual level, may be deemed of such overriding importance to justify extraordinary sacrifice, of self or others. Notably, Ferguson (1996) has shown how framing an area as in need of development is a powerful way of moving inordinate amounts of resources serves as an 'anti-politics machine' - a striking parallel with securitisation theory indeed. While his example discusses agricultural projects in Lesotho, the example of dam-building at the expense of local livelihoods comes to mind in the context of water politics. Fig. 6: Security and opportunity politics: some opposites security demands opportunity demands secrecy, securitised information transparency, openness fast-track decision-making democratic competition uniformity, unity diversity, multiplicity rules and procedures informal participation planned economy, appointed contractors tendering, market competition The earlier examples, culled from Buzan et al, all implicitly presuppose a liberal order in which security measures are indeed extraordinary. If we take the counterfactual as a guide - an autocratic, secretive environment under martial law which puts in flood defences as a matter of course, couldn't an existential threat be invoked to legitimise competition, freedom of expression, participation and controlled flooding - brief: to legitimse and demand a much greater degree of uncertainty and risk? In fact this does intuitively not sound too remote from the World Bank and IMF's practice in the past two decades, forcing the adoption of competition and open elections by invoking the threat of economic implosion. To some degree, this is caught by Waever/Buzan et al's term 'desecuritisation (of economics)', but this only describes the release of an issue from the realm of security. It is also not the same as 'undersecuritisation', which Buzan et al also introduce in the precursor to their book (Buzan et al 1995) - that phrase denotes the wilful, strategic neglect of an issue other (silenced) actors judge extremely necessary. To the contrary, bringing an issue into the area of risk in its 'opportunity' sense again may be promoted with the same sense of urgency as a security issue would (see Fig. 5). Perhaps a new phrase, 'opportunitisation' - admittedly, as inelegant as 'securitisation' - should be invented to allow for this strategy. This, then, opens up the possibility of securitising, undersecuritising and opportunitising moves and countermoves in the same domain of security. A securitising move may be countered by a desecuritising move. Absolutes are never absolute for good - a securitised issue may be repoliticised. As Fischhof et al (1980) note once it has been decided that there is something to question, 'the elicitor may legitimate events that were previously viewed as unacceptable or cast doubts on events that were previously unquestioned.' Fig. 7: A double-attractor continuum: from securitisation to opportunitisation

Ø Domain: Attractor:

Politicized (‘talk’)

Depoliticised) strategy:

Threat (-) URGENCY securitisation Don’t talk., act NORMAL (re)politicisation,

desecuritisation First deliberate

TABOO undersecuritisation Don’t talk, don’t act

NORMAL (re)politicisation, de-opportunitisation

First deliberate

Opportunity (+) URGENCY "opportunitisation"? Don’t talk, act

13

2.4. Applying the Buzan model to a water framework In itself, water as a security issue is not very hard to grasp - droughts and floods are equally vital threats to people's survival. The role of strategic natural resources in international conflict has been recognised from time immemorial (Gleditsch, 1997), both as objectives and as weapons. Water has been an object of military defence - as a goal (possession of Shatt al-Arab in Iran-Iraq war), target (dams and hydropower installations were military targets in the 2nd World War, the Vietnam War and the 'second' Gulf War, instrument for attack or defence (controlled flooding). In river basins, rivers became the object of military strategy, what with upstream countries (e.g. Turkey) using water to secure power, downstream countries (Iraq, Egypt) using power to secure (the continued inflow of) water (Warner, 1993). The translation from a generic International Relations framework for understanding representations of security to water issues is not completely straightforward, though. From the very start, two minor amendments to the model will be introduced to make the model more suitable to water management. In Buzan's model, the military sector is primarily concerned with the defense of territory (Buzan et al 1995). I think it useful to amend the military sector to a more generic 'physical' domain to make room for both the territorial aspect of security, as well as its associated geographical properties, most notably water. Indeed, when discourse in flood 'defense' in the Netherlands traditionally defines 'security' in terms of the integrity and stability of its flood defense structures itself (dikes, dams), rather than its people. Such structures are both vulnerable to military attack and to erosion and weakening due to geophysical processes. Also, I will slightly redefine Buzan et al's 'societal' sector s 'socio-cultural' in view of the social and cultural values of water. This has the additional advantage of reinstating Nef's 'cultural' domain to its rightful place. 3. Case study: Mind the GAP? Framing the Ilisu Dam In the past 12 months, a dam in Turkey has made the headlines in Britain at least twice (The Guardian, 1 March 1999, The Independent on Sunday, 12 Dec. 1999) and most recently, the BBC Nine O'Clock News (22 Jan 2000). The reports were without exception alarmist - an excellent result for Friend of the Earth's anti-Ilisu campaign, but is it a victory for common sense? In the past decade, much (Starr & Stoll 1987, Bulloch & Darwish 1993, Homer-Dixon 1999, De Viliers 1999) has been written about coming 'water wars' in which states would be taking on each other to secure (access to) scarce resources. Mostly, the focus of these predictions is the Middle East, it being a notoriously parched and conflict-ridden region. And indeed, troops have been mobilised along the Turco-Syrian border with some frequency over water conflict. A closer look at the conflict over the waters of the Euphrates and Tigris rivers shows that disagreement over water distribution is only one of many factors in a complex problem, made even more complex by the progressive liberalisation of the water sector. Moreover, the scope of the conflict cannot really be grasped without taking the colonial legacy into account. Just like the Serbian trauma over their 1389 defeat (significantly, at the hands of the Turks) still informs today's passions over Kosovo, the Turkish have very long memories, the memory of the giant Ottoman empire dominating the political rhetoric in present-day Turkey. Playing at lingering Ottomans sentiments, it is good (domestic) politics to declare the 21st century the Turkish century (itself a speech act!), in which the Turkish Empire would span from the Adriatic to the Chinese Wall, as the late President Turgut Özal used to do (Zürcher 1998: 335). The present article starts from the premise that this claim amounts to more than just demagogy. The hydraulic imperative For Turkey, the age of enlightenment started in 1922. As happened elsewhere, the traumatic loss of empire is often followed by an intensive 'internal colonisation': the expansive energy is now directed at development and harnessing of the natural resources in its own territory. Peripheral regions are

14

(re)integrated, the administrative system centralised, class divisions ironed out. Water security and development add up to the magic formula to align mutually opposing forces in a project that depoliticises those contradictions (Swyngedouw 1999). This was true of Spain after the loss of Mexico and the Philippines, and it certainly seems a plausible explanation for the Turkish course of action after 1920. When the French and English laid down the current Turkish boundaries at San Remo - blithely ignoring natural boundaries and denying the Kurds their nation-state - they laid the basis for many current resource conflicts in the region. The Ottoman empire was miniaturized: the Fertile Crescent (Iraq, Syria, Lebanon, Palestine and Jordan) became a Franco-British mandate, Northern Mesopotamia fell to Britain and except for Thrace all European territories had to be ceded. All in all, some irredenta! Rather than look back in anger, the Turkish government decided to look forward. The 'hydraulic imperative' for the development of its hinterland proved a manageable project to weld together feudal and modernising (industrialising) forces into a historic, nationalist compromise. The towering figure of 'papa' Atatürk plays a crucial role in the new strategy. First, the early twenties saw the intensive homogeneisation of the Turkish populations including a massive exchange of ethnic and religious minorities as a consequence of the war with Greece (1921-1922), and large-scale 'ethnic cleansing' of Armenians (Beeley 1995) The Lausanne Treaty of 1923 gave birth to the First Turkish Republic an indivisible, unitary Turkish state in which the Kurds formally do not even exist. The Kurdish identity proved not so easily repressible, however, and remains a formidable obstacle to the homogenising thrust of the Republic of Turkey. The GAP project seems a new strategy to integrate the Kurdish minority into Turkey by economic means. A second pillar buttressing the internal colonisation drive is autarky. Turkey may be poor in oil and gas (Turkey imports both from Libya and Saudi Arabia) but the country is very well endowed with raw materials. Until long after the Second World War Turkey was a one-party state and until this day the army is the de facto ruler. Like many a post-colonial state, Turkey's development trajectory was state-led and authoritarian. Public money funded large infrastructural works and educational improvement to equalise the level of human development across the country's regions. A calculated foreign policy helped bring in external funds. NATO membership brought in external funds, notably through the American aid agency USAID. Internally, economic integration has not really taken off - even today, the average income in Ankara is still many times that in Anatolia. The Guneydogu Anadolu Projesi, more conveniently known internationally as the Greater Anatolia Project, is one instrument aiming to right that balance. Developing of the Euphrates and Tigris basins holds out an enormous potential of hydropower and irrigation farming, the former as a substitute for imported fossil energy, the latter as a 'breadbasket' for the region - all other Middle East countries are now massive net food importers. Southeast Anatolia is rich in fertile soils - in 1996 only some 120 000 hectares out of a potential 314 million were under irrigation (Mutlu 1996). The hydropower and irrigation projects (80 dams, 66 hydropower stations) are to develop a 2 million-hectare area, that is an area the collective size of the Benelux countries. In theory the GAP is intended to reform the socio-economic situation in the most underdeveloped Turkish region. The projects are supposed to break the power of entrenched interests, latifundists, fight widespread poverty and provide much-needed physical infrastructure to the region. So far, the dams ave mainly produced electricity the provision of water for irrigation has fallen far behind. Moreover, the works have tended to promote the development of the regions of the West, not the impoverished East. However, the balance looks different when seen in light in of the strategy vis-a-vis the Kurds, a mixture of ruthless pursuit and integration. What the Turkish authority may not have envisaged in its drive to eradicate difference is that the Kurds reacted by a similar homogenising drive (Rygiel 1998). The internal war against the separatist Kurds, waged since 1984 was stepped up after a 'Kurdish intifada' in the early nineties. The Turkish army employed a slash-and-burn-tactic to root out settlements suspected of collaborating with the militant PKK, uprooting some 500,000 Kurds and swelling the numbers of the city of Diyarbakir, and, further afield, Ankara and boosting Kurdish migration into Western Europe in the process. Last year's offer of a ceasefire on the part of ther PKK has not been taken up by the Turkish authorities.

15

Raising the standard of living would seems to stand a better chance of controlling the elusive mountain Kurds. Once tempted out of their hiding places, it is figured, they will be less likely to provide sanctuary to the PKK. Economic development should also attract Turks from other regions, encouraging ethnic integration in the Kurdish-inhabited regions. This consideration seems even more important than the actual economic profitability of the GAP. Funding the project has been a problem from Day One due to the skepticism of a key external player, the World Bank (IBRD). The Bank now would like to see water-intensive agriculture curbed in favour of industry and urban supply. As the most important donor to the region by far it has proved highly effective in shaping economic policies in recipient states. Its veto on regionally sensitive projects tends to kill off a controversial water project for a considerable time. But in turning off its flow of funds has proved not to kill off project if the initiator is determined enough to find funds elsewhere. Moreover, the Bank needs to keep some customers on board in order to keep moving money; its clout is, to a great degree, a function of its huge budget. Sensing the Bank would show itself sensitive to protestations on the part of co-riparians Syria and Iraq; the Turks apparently never even formally applied for Bank backing; the veto only reinforced Turkish determination to continue the megaproject, even if it took up to 10% of the total annual budget. A mismanaged economy - the Treasury's coffers are chronically bottomless - groaned under the development effort. The lack of multilateral co-operation made itself felt in ever more painful ways when in the early nineties projects started to fall behind schedule further and further. More and more, the GAP seemed to look like the famed 'white elephant': the costly development project that never materialises. The fact that, after a temporary lull, the final stage of the Greater Anatolia Project - 21 dams, 19 hydropower stations - is now on full steam again, is only due to a radical institutional trick: privatisation. As early as in 1987 the means to fund the Izmit dam had run out. Instigated by then president Ozal (a civil engineer himself, as was his successor Demirel) a private consortium was created, Izmit Su, to complete the works. Stockholders are the municipality of Izmit, the Japanese conglomerates Sumitomo and Mitsui, Thames Water of Britain and two local companies, Gama and Guris. Thames Water is to run the utility for 15 years before returning it to the municipality of Izmit. Privatisation turned out to be a timely move: amazingly, water has developed into a global growth market. French and British giants like Vivendi (formerly Générale des Eaux), Suez Lyonnaise, Northwest Water, Severn Trent and Thames Water now carve up the globe for rich pickings in liberalised water utilities and infrastructural projects. Ten years after the abortive start of the Izmit project (dam, storage lake, sewage works and water utility) was ready to go onstream. Tellingly the key cost factor of project development involved the fee of Turkish lawyers struggling to legally enable the project. There was simply no available legislation for such projects in this sector. While privatisation had been advocated by several Turkish governments since the 1950s but is hardly compatible with the prevailing dirigisme. The privatisation law opposed by the secular and religious right, was finally pushed through parliament in November 1994 by Tansu Ciller, over islamist objections, well-timed to coincide with an important Galatasuray-Barcelona football match keeping many MPs glued to the TV screen (Zürcher 1998). Most of the megadams have so far been realised on the Euphrates (in Turkish: Firat): so far, Turkey has laid relatively limited claim to its sister river, the Tigris (Dicle). The last GAP dams are to change that. In March 1997 the Turkish government granted the contracts for the 182 m long, 135 m high Ilisu dam, 64 km from the Syrian and Iraqi border, the biggest Turkish hydropower project so far to a Swiss consortium consisting of Sulzer Hydro and Asea Brown Boveri (ABB), which firm has a 25% global market share. Funding was to be arranged for by the Swiss federal bank UBS. The project is only due to come onstream in 2006, but its announcement did not fail to elicit strongly-worded protests. The Ilisu dam alone will force fifteen to twenty thousand Kurds from 52 villages and 15 towns to resettle. Compensation is linked to land rights, meaning that nothing has been arranged for the countless landless. The town of Hasankeyf, which is on the Turkish heritage list, has to disappear to make the Turkish dream a reality. Stunningly, the GAP has united Syria and Iraq (Gulf War adversaries) in an alliance of

16

convenience that will not remain silent. In 1975 the two countries were on the brink of a water war themselves, and officially have not been on speaking terms since Syria joined the anti-Saddam coalition in the Gulf War. But after a five-day meeting the states decided jointly dispatch threatening letter to companies involved in building the Birecik dam. Syria filed compensation claims and threatened boycotts until a trilateral agreement was signed. The Ilisu dam would, again, reduce the amount of freshwater allowed to pass the border, impairing the diluting capacity to purify the wastewater flowing from the region's major cities. Baghdad, on its part, fears its flow to be contaminated by agricultural chemicals and pesticides. It claims a breach of international right and riparian water rights, which does not seem a really ingenious move. True, with some imagination a breach of a 1946 Turco-Syrian treaty stipulating consultation between riparians could be invoked (Guardian 1-3-99), but international law only provides only cold comfort for water plaintiffs - there are no agreed principles governing international rivers. Iraq may insist on the international law doctrine of absolute territorial integrity, stipulating that no riparian is allowed to impair the quality and quantity of the water resources flowing within its territory - but Turkey can with equal vigour juxtapose the doctrine of unlimited territorial sovereignty (Harmon doctrine): each state can treat the water within its boundaries any which way it likes. Along with China and Burundi, Turkey is the only state to refuse signing the 1997 UN treaty on non-navigable watercourses claiming it grants the downstream states excessive rights. Turkish politicians conveniently tend to view the entire Euphrates-Tigris-basin as a single Turkish river, denying its international character (in Allan 1995). It is easy to see why: the majority of the catchment may be in Iraq, where it drains into the Persian Gulf through the Shatt al-Arab (disputed by Iran). But the river receives 95% of its precipatation within Turkish territory, and the artisian springs just across the Syrian border are fed by Turkish rain, infiltrated into the soil and working its way down to Syria. Turkey also claims that the dams will also benefit its downstream neighbours as droughts and premature flooding can be prevented as a result of better regulation. Better timing would lead to more productive farming as well. The snag is, of course, that Turkey is denying its neighbours any real say in the regulatory decisions. The downstream states therefore have been sufficiently realist to agree on a percentage distribution of whatever Turkey leaves them: 42% for Syria and 58% for Iraq. This was agreed after Turkey closed the Atatürk Dam for a month in 1990 to fill the storage lake. It should be noted that a degree of opportunism is not alien to Turkish tactics - while Turkey views The Euphrates-Tigris basin as all-Turkish, it takes the reverse position vis-a-vis the Orontes (or Asi) on which it is downstream (Shapland 1998). Turkey as water hegemon Turkey manages to make much political capital out of its favourable geographical location. It connects three macro-regions Europe, Central Asia and the Middle East, a geopolitical nexus that virtually ensures NATO backing. The country is a member of the association of Islamic states, keeps knocking on the EU's door and competes with Iran for hegemony in unstable Central-Asia. While Iran can only count Armenia as its ally, Turkey has natural cultural bonds with the five Turkish-speaking states of Central Asia. As the Americans would rather not see expansion of the Iranian sphere of influence, they back Turkish inroads in the region. Although Erik Zürcher (1998) reasons Central Asia is too embedded in the Russian sphere of influence for Turkey to make much chance of pricing it away, that will not stop Turkey from trying. As Central Asia is so hot and dry, it should not come as a surprise that Turkey has made a water offer to this region (Hillel 1995). Turkey is especially conscious of its enviable position as a water-rich state in a water-poor region. Even though the state has great difficulty providing water and sanitation for its own megacities, still the Turks apparently see significant political gain in water export. The most infamous initiative has been the twin Peace Pipeline which was to carry water eastward to Saudi Arabia and westward to Israel, which was universally rejected, but now that Turkey works closely with Israel in the military area, it is to transport freshwater in giant, Norwegian-made 'Medusa' bags by sea to Israel, both annoying Arab neighbours and Turkish fundamentalists. Recently Turkey has also offered to carry water to Israel through pipelines under the sea (Spits, 16 July 1999).

17

But most importantly, its geographical location at the headwaters of the Euphrates and Tigris, the main sources of freshwater of its most troublesome neighbours, Syria and Iraq, is extremely convenient to Turkey. An unmistakable side effect (some might say: the key objective) of the intensive damming of the two rivers is that it enables the Turkish to turn the tap on or off. Turkey indeed has several bitter historic grievances against both neighbours it may feel could do with some hydraulic backing up. Turkey has historic claims to the oil-rich Mosul province in Iraq. Iraq needs Turkey to control its own Kurdish problem (PUK and PDK), allowing Turkey the right of hot pursuit on Iraqi territory, but reportedly provides logistic support to the PKK as well. Syria bets on the same horse: it allowed the Kurdish militants to train in the Syrian-occupied Biqa'a valley in Lebanon, where the extreme left Turkish urban guerilla Devsol is also based; prior to his recent arrest, PKK leader Ocalan used to live in a very nice mansion in the Syrian capital of Damascus ever since the Evren coup of 1980. And then there is the never resolved downing of MiG fighter planes above Turkish territory. Domestic and regional security issues are continuously linked. For example, when in 1995 Syria arrested five PKK activists and expelled Ocalan, they seemed to have presumed - wrongly - that Turkey would release more Euphrates water. Regional alliances are volatile and sometimes fairly inscrutable. Now that Prime Minister Netanyahu has left the political stage, an accord with Syria is a possibility again, alienating Syria from Iran. Jordan, which supported Iraq during the Gulf war, changed its mind, perhaps to facilitate negotiations about water with Israel, yet it now speaks for Iraq in the hullabaloo over Ilisu. When in early 1996 Turkey intercepted five Iranian lorries with arms, which Turkey claims were destined for the PKK, another diplomatic row ensured, but now that Syria is talking to Israel over the Golan Heights, Syria is backing off from Iran. Jordan, which supported Iraq during the Gulf War and suffered the consequences, then distanced itself from Iraq for a while, has now speaking for Iraq in the Ilisu case (Guardian 1-3-99). On top of that there is the overlay of American dominance in the Middle East, inspired by a desire to secure access to oil reserves, and expressed in extensive economic and military aid to Israel, Turkey and Egypt. While all states are fairly weak, the Americans can operate in the region as a patron and/or policeman. But the Turkish political leaders seem hellbent on changing the regional 'balance-of-weakness' into what American geographers John Kolars and William Mitchell (1991) have termed a 'pax aquarum', a hydraulic hegemony. Frequently, a 40% reduction is predicted for Syria and up until 80% less for Iraq; although this latter figure would be a cumulative effect of Turkish and Syrian dam projects. There are reports that two smaller Syrian rivers have run dry as a result of the reduced influx. Yet it would seem that the 3000-odd litres per capita available to Euphrates and Tigris riparians should comfortably see them through a reduction - Jordan scrapes by at a tenth of that amount. Syria itself is rather wasteful with its water, and is creating dustbowls, compounded by salination and evaporation (Economist, 13 Nov 1999) Turkey itself incidentally calculates a much punier 1830 litres per capita, based on its 60 million population, and argues that it will need a lot more in the future: The state feels it is legitimate in exploiting its climatological and economic advantages: a crop should grow where it can be marketed most profitably to ensure the most efficient land and water use. But more important than the real impact is the potential to give the water tap a twist in either direction. If all present dams in the catchment were to be closed all at the same time the entire volume of the rivers could now be stored many times over. off A simple sum shows why the downstream riparians are so outraged. The Ilisu storage lake will have a total storage capacity of just under 10.5 billion cubic meters and an operating capacity of 7.5 billion m3. Normally, that would leave a buffer capacity of 3 billion cubic metres. As the average annual inflow of the Tigris is 15 billion m3 the reservoir will account for half the total annual flow, while the spare capacity would enable a malevolent Turkey to arrest the river influx for some additional months, such that not a drop of Tigris water flows into Syria and Iraq (Berlin Declaration, 1999). And this is only one out of dozens of megadams built in this catchment. Not just closing but also the sudden opening of the floodgates would be disastrous. Several historic battles have been instructive. In 689 B.C. Sennacherib the Assyrian dammed the Euphrates upstream from Bagdad, only to destroy I after sufficient water had assembled behind the dam. The

18

sudden floodwave flooded the Mesopotamian capital and won Sennacherib the day (Gleick 1993). Such information teaches the downstream riparians some realism and Turkey to be laconic over even the gravest threats. On the one hand Turkey likes to show its most chivalrous side in public: the state solemnly intones it will never use the water weapon and has promised in 1987 to let a minimum of 500 cumecs (m3/sec), that is, as a two-month average. So far, the country appears to have kept its promise, if at a pinch: last July the Euphrates discharged just 42 cumecs (The Economist, 13 November, 1999). Turkey tends to politely announce what is going to happen and to call technical meetings to discuss the details - since 1980 a Turco-Iraqi committee has convened; in 1983 Syria joined the talks. But the parties prove to be very reticent in making available reliable information and the decision itself is always non-negotiable. Turkey clearly likes to remind its southerly neighbours who calls the shots around here. In 1990 the Euphrates was closed off for a month to fill the storage lake at the new Atatürk dam. Skeptics feel that the closure was totally unnecessary: the water could have come from the older Karakaya, upstream. But Syria took the unsubtle hint and made a big noise. There is a vaguely comical ritual side to all this. Another indignant Syrian mobilisation reinforces the media-friendly image of the (near-) water war, keeping international interest in the region alive. Some time later the world then learns that the countries have sat down together and that a cooperative treaty is in the offing, which subsequently never materialises. This ritual dance of near-wars, near-treaties and much verbal abuse have been a repeated phenomenon ever since Turkey has built large dams - since the seventies, well before the GAP, when the Karakaya, Keban and Karun dams were built. Still none of the plaintiffs dares try anything. Whoever takes on Turkey can expect NATO to step in, to whom Turkey has been a valuable alley for decades. And even though Iraq warned (in 1988) to start an international legal case, the road to the International Court of Justice has not been used. For a case to be accepted all parties involved will need to recognise its authority, as did Hungary and Slovakia in their dispute over the Gabcikovo dam on the Danube. While that recognition is not forthcoming, a repetition of moves can be expected. The ethics GAP It is therefore unsurprising that Turkey again gets its share of flak each time a new GAP dam is started; it is more interesting that the privatisation in the water sector brings new actors into play. As a result of the European commercial involvement European actors are now subject to sharply worded threats. Non-governmental organisations have seized on the internationalisation of water projects to underline their more broadly focussed protest against megadams. Local protests amplified by NGOs led to the (temporary?) discontinuation of the Narmada dam in India, the Arun II dam in Nepal and the Bakun dam in Malaysia. Companies are getting jittery, too: This, after all, is an industry that could barely survive without World Bank or government credit (Bosshard 1999). As investment in Turkey carries considerable political and economic risk, the international companies are loath to carry all that risk themselves. Governments of countries where civil engineering is an important export sector have so far, turned out surprisingly eager to provide export. Such a credit is no luxury: Anatolia is still under emergency rule, and GAP projects are under constant military surveillance. Now that the media, tipped off by human rights and environmental 'watchdogs' are on their trail, some governments that have issued export guarantees may be feeling increasingly uncomfortable. In Great Britain, for example, an affair erupted over the Ilisu dam. British construction company Balfour Beatty, another important international player, has been approached by ABB (which has 25% of the global market) and Sulzer Escher Wyss (Sulzer was also responsible for the largest construction, the Atatürk dam, which was finalised in 1992) to lead the construction consortium which is to realize the dam, the storage lake and hydropower station - further companies involved are Impreglio (Italy), Skanska (Sweden), and the Turkish companies Nurol, Kiska and Tekfen. The British Department of Trade and Industry (DTI) was quite ready to issue a 200mn-pound export credit, but had failed to confer with the newly ethics-conscious Foreign and Commonwealth Office (FCO), which embarrassingly had to read the news in The Guardian. Now DTI's Export Credit Guarantee Department (ECGD) which governs such export credits, has no ethical or environmental code governing those guarantees, while Foreign Minister Robin Cook prides itself on its ethical

19

standards. FCO and DTI have been known to be at loggerheads over values - for example, Cook does not want to condone any official British trade mission to Libya while no light has been shed on the death of policewoman Yvonne Fletcher, shot dead in 1984 from the Libyan embassy in London. (The Guardian 2 July 1999). But even the Office was ready to defend the project as a fine example of its ethical policy, claiming it would contribute to Middle East peace (Guardian 1-3-99) until the flak became too vehement In the House of Commons the trade minister, Brian Wilson, sought to reassure worried Liberal Democrats that no final decision had been taken. This affair, while still minor, could be painful to the Labour government which seeks to set itself apart from its Conservative predecessor, which some four years ago was embarrassed by a similar big dam project in Malaysia. The Malaysian Pergau dam project, involving the Swiss ABB and British company Biwater, turned into a scandal when technical development aid turned out to be tied in with British arms deliveries (the Malaysian government did not have to do much more than retract its Buy British Last-policy) and that Biwater had donated great sums to the then Tory government. The critical press has charged that Labour's support for Balfour Beatty is not coincidental either. Like Biwater, Balfour Beatty has been in the limelight for all the wrong reasons: the collapse of the Heathrow Airport Tunnel and bribery of officials to obtain permit for a Lesotho dam (e Eye; CorporateWatch) Other governments have tried to forestall the torrent of criticism in various ways. The Bundesrat, to which the Swiss central bank is accountable, justified its export risk guarantee go-ahead for 470 million Swiss francs with a view to new Swiss jobs (1200 full-time man years) (Bosshard 1999), Turkish development, and Turkish promises to look into expected negative side effects including forced resettlements, conflict over water rights with the downstream riparians, threatened cultural heritage, and malaria vectors associated with nonmoving water in a storage lake. The Swiss government in 1998 attached to its export credit (also covering a project in Ankara) the condition that an independent monitoring mechanism would be established. Sulzer Hydro argues Ilisu's environmental benefits (George, 1999). Mainly for external consumption, Turkey itself, too, is keen to present its scheme as a human and social development project, claiming conformity with UNDP environmental sustainability guidelines. While in so doing, the bilateral donors seem to follow the new World Bank criteria after all, it does not look as if such conditions could be any more than window dressing to secure continued access to a lucrative water market. Despite repeated NGO requests, the government remained reluctant to make the Environmental Impact Assessment for the Ilisu dam public. On this count, the Government has also changed its mind. 4. Understanding the GAP in security terms With the help of Buzan et al's grid, a story of security and risk trade-offs can be told. In terms of the diagram, Turkey has sought to enhance its national security since the 1920s by laying great stress on cultural identity and integrity (the unitary state) as well as economic development (self-actualisation, opportunity-seeking), which in turn should provide the government with greater legitimacy (political security). As an instrument to achieve both economic growth and integration of the Kurdish minority, the Greater Anatolia Project is legitimised and elevated beyond the realm of debate - in Buzan et al's terms, 'securitised'. As the project provides sufficient storing capacity to deprive its downstream neighbours of water for a considerable time, the project has the extra bonus (for Turkey) of enhancing external military security and contributing to the long-term military-political goal of macroregional hegemony. By denying that the basin is international (in other words, unilaterally declaring sovereignty over the integral basin in the name of integrated management), Turkey seeks to resist internationalisation of the water issue, which the downstream parties predictably reject and complicate by linking national and international issues. Turkey pursues its objectives at a considerable price. Domestically, the project has procured sufficient hostility from the Kurds to warrant the conclusion that it has contributed to internal cultural and military insecurity. Internationally, the controversy over the project has proved bad international

20

public relations (reputation), not improving its chances of EU membership, and landed Turkey on the brink of war with Syria on several occasions. Also, the project has deprived Turkey of international funding; burdening a stressed economy with spiralling project costs. That latter problem has now been partly 'solved' by liberalising the water sector, if at the cost of a degree of (temporary) loss of state autonomy over water resources to international companies (political insecurity). For these companies, the projects provide do not just provide opportunity but for several of them it provides much-needed security in terms of long-term income in a competitive market. However, it also brings considerable political and economic risk - not just by investing in a controversial project in a country that is effectively still at war with itself, but also the potential loss of its hardware or people due to attack in a region where the state of emergency is still in force and project workers reportedly experience occasional gunfire and threats. They have sought to alleviate this risk by securing export credits from their governments, which in Britain, Department of Trade and Industry at first was happy to provide. In addition to an export opportunity, the UK government even saw political capital in it, promoting it as a peace-promoting project. However, it had not counted on the well-orchestrated protest on the part of a European NGO coalition, spearheaded in Britain by Friends of the Earth, the project looked a choice opportunity to mobilise its political clout in opposing the Ilisu dam as a symbol for unethical British investment. Turning what was seen by its initiators and donors as an economic issue into a human rights issue, it allied with the Kurds who, seeing their cultural heritage endangered (symbolised by Hasankeyf), showed themselves unwilling to lose the remaining symbols of cultural identity in exchange for economic opportunity. This latter development seems a good example of active domain linkage. It is interesting that the military and cultural rather than the economic or environmental cards were played. The Ilisu Dam is but one in a series of Turkish megaprojects which have been consistently fought. The Syrian and Iraqi threats and writs against Turkey and its international investors on the basis of expected economic damage and environmental degradation (pollution) have failed to make an impact. NGOs, too, will never be able to win the day on economic arguments. By successfully recasting the issue as a human rights issue, they could play at a concern which to many people is an absolute, existential value at the individual and group level. In Buzan et al's phrase, they 'securitise' the issue by calling on an existential danger. However, for NGOs, the battle against large dams seems half-won, given the tightened World Bank rules on funding them and the rise of the World Commission on Dams. FoE may have judged the general public is growing increasingly blasé over issues of environmental quality which have not grabbed headlines of late. A 'water war' however is in a different - military - league, and a much more emotive 'spin' than the arguments Turkey and the UK foreign office advanced for the project as promoting regional peace. Moreover, casting the issue as a human rights violation will strike a chord in the sections of the European Parliament that, as noted above, are still wary of Turkish EU membership. Whatever the rationale, the move has been relatively successful. The Turkish initiators, sensing the change in international mood, have sought to counter it by casting the project as essentially humanitarianand ecologically sound, that is, it sought to defeat the opposition in the same security domain. While the project's opponents will feel this 'spin' on the project amounts to little more than nominal lip service, the price to be paid was to be at least subejcted to international scrutiny and environmental accountability. So, in order to improve the political strategy, it seems the different domains have indeed been linked or relinked to domains where a more successful outcome was anticipated. 5. Conclusion As security is instrumental in the enhancement of other values (Krause & Nye q in Buzan 1991) it is depressing that, as Buzan complains, the concept of security is 'underdeveloped'. We have seen that the model developed by him and his associates still leaves room for development, and I have no hope of claiming to have completed that task. However, some strides have been made: a bipolar model of security and opportunity;

21

securitisation and area linkage. The model introduced here can be used as a tool to zoom in on cases that (sometimes quite suddenly) become foci of international attention (Brent Spar comes to mind), and see which domains stand out. As one (wo)man's opportunity is another one's threat, controlling the problem definition in the debate is crucial in successfully carrying through or resisting the project. This may help us anticipate security strategies of both 'perpetrators' and 'victims' of contested acts. Expanded along the lines of this paper, the Buzan model is a way to help analyse political action, to clarify and think through the strategies of different actors embroiled in a conflictuous issue. This article, in sum, has called for a greater awareness of the 'management of meaning' in representations, perceptions and professed values than security studies have traditionally shown, while always keeping an eye open to the interests underlying the representation. The Buzan security diagram will be a helpful schematic means to this end. To phrase an issue in security terms invokes an issue of threat-defense (Waever 1995); I have suggested a mirror-image of opportunity-offence. If we want to break either mold, we need to question the structures that give rise to them, and the degree to which we can afford to let important issues be decided outside the political realm. While the new global security agenda is still under construction, it may not be to late to start that debate.

22

Literature John Adams, Risk, SAGE, London, 1995. Tor Arnesen, 'A method for analysis of ecological communication in regional management complexes

(REMA), 1997, http://www.ostforsk.no/per/Tor.Arnesen/publikasjoner/REMA.htm J. Langford Austin, How to do things with words, The William James Lectures 1955, Clarendon

Press, Oxford, ed. 1975. Zygmunt Bauman, In search of Politics, Polity Press 1999. Brian W. Beeley, 'On the geography of development in Turkey' in Eric Watkins (ed.) The Middle East

Environment. Selected Papers of the 1995 Conference of the British Society of Middle Eastern Studies, St Malo Press, Cambridge, 1995, pp. 109-118.

Peter L. Berger and Thomas Luckmann, The Social Construction of Reality. A Treatise of the Sociology of Knowledge, Penguin, London etc., 1991 [1966].

Ian Black, 'Fury as Cook bans MPs' trip to Libya', The Guardian, 2 July 1999 Peter Bosshard, 'BD opposes destructive dam project in Turkey', Berne Declaration, http:

www2.access.ch/eva/BD/bdnews.htm#dams, 17 June 1999, 13:03:42. Paul Brown, 'Britain Backs Controversial Dam', The Guardian, 1 March 1999. Paul Brown, Lib-Dems call for Inquiry into DTI links with Turkish dam', The Guardian,. XXXX J.S. Bruner, Actual Minds, Possible Worlds, Cambridge 1986. John Bulloch & Adel Darwish, Water Wars, Gollancz, London, 1993. E.J Burger, Health as a Surrogate for the Environment, Daedalus, 1990, 119(4), 133-50. Barry Buzan, People, States and Fear, an Agenda for International Security studies in the Post-Cold

War Era, Harvester Wheatsheaf, Hemel Hampstead, 2nd ed., 1991. Barry Buzan, Ole Waever & Jaap de Wilde, 'Environmental, Economic and Societal Security', COPRI

Working papers 10-95, Centre for Peace and Conflict Research, Copenhagen, 1995 Barry Buzan, Ole Waever & Jaap de Wilde, Security: a new framework, Harvester Wheatsheaf 1998. David Campbell, Writing Security, Manchester University Press, 1992. Council on Global Governance, Our Global Neighbourhood. Simon Dalby, 'Contesting an essential concept' in Krause & Williams 1997, pp. 3-32. Daniel Deudney, 'The case against linking degradation and national security', Millennium, 19(3)

1990, 148. pp. 461-476. van Dorp et al, Landschapsecologie, Boom, Amsterdam 1999. Mary Douglas, 'Risk as a forensic resource', Daedalus, 119(4), 1999, 1-16. Mary Douglas, Mary & Aaron Wildavsky, Risk and Culture: An Essay on the Selection of

Technological and Environmental Dangers, Berkeley 1992. Mark Duffield, 'Development and Security', paper presented at DSA Security and Development

meeting, South Bank University, 26 March 1999. Michel van Eeten, 'Sprookjes in rivierenland. Beleidsverhalen over wateroverlast en dijkversterking',

Beleid en maatschappij, 1997(1), 32-43. The Economist, 'Sharing Mesopotamia's water', 13 November 1999, pp. 81-82. James Ferguson, The Anti-Politics Machine. 'Development', Depoliticization, and Bureaucratic

Power in Lesotho', University of Minnesota Press, Minneapolis 1996. Peter Feuilherade, 'Water pressure', The Middle East, December 1993, pp. 31-33. Baruch Fischhof et al, Acceptable Risk, Cambridge University Press, Cambridge 1989 [1981] William Frank Fischer, Theories of anxiety, New York: London: Harper & Row, 1970. Martin Frimmel (Greenpeace Osterreich), 'Ökowaffe Wasser',

http://www.ecology.at/archiv/publikat/ew/ew9919.htm Alan George, 'Ilisu Dam consortium rejects criticism', The Middle East, May 1999.

http://dialspace.dial.pipex.com/town/terrace/lf41/me/may99/mebf0503.htm 07/05/99 11:12:43 Anthony Giddens, The Consequences of Modernity, Polity, Oxford, 1990. Nils Petter Gleditsch, 'Armed Conflict and the Environment, A Critique of the Literature',

Proceedings, KNAW Workshop on Environmental Change and International Security, Amsterdam, 20 January 1997, pp. 65-81.

Peter Gleick, Water in Crisis, Pacific Institute, 1993 (new version out now)

23

Andreas Hasenclever, Peter Meyer & Volker Rittberger, Theories of International Regimes, Cambridge University Press, 1997.

Mark Haugaard, The Constitution of Power. A theoretical analysis of power, knowledge and structure, Manchester University Press, Manchester/New York, 1997.

Kenneth Hewitt, Regions of risk, Longman, Singapore, 1997. Daniel Hillel, Rivers of Eden: the Struggle for Water and the Quest for Peace in the Middle East,

Oxford UP, New York, 1995. Thomas F. Homer-Dixon, 'Water Wars', Forum, International Red Cross, 1(1) 1999 1999. John F. Kolars en William A. Mitchell, The Euphrates River and the Southeast Anatolia Development

project, Southern Illnois University Press, Carbondale 1991. Bradley S. Klein, 'Conclusion: Every month is 'Security Awareness Month', Krause, Keith & Michael

C. Williams, Critical Security Studies. Concepts and Cases, UCL Press, London, 1997,, pp. 359-368.

Karen Litfin, 'Constructing Environmental Security and Ecological Interdependence', Global Governance, 5 (1999), pp. 359-377.

S.Lonergan, 'Water Resources and Security: Examples from the Middle East, in M.J. de Kok, proceedings International Workshop 'Environmental Security and Sustainable Development', in Dutch National Research Programme on Global Air Pollution and Climate Change, Bilthoven, 1996b.

Miriam Lowi, Water and Power. The politics of a scarce resource in the Jordan river basin. Cambridge University Press, Cambridge 1993.

Hans Mouritzen, 'Twining Plants of International Co-operation. Reflections on the Peculiarities of `Security' IGOs' in Jaap de Wilde & Hakan Wiberg, Organized Anarchy in Europe. The role of States and International organizations, IB Taurus 1996.

Norman Myers, Ultimate Security, W W Norton, New York/London, 1993. Chris Morris, 'Standing against the rising Tigris', The Guardian, July 17 1999. Servet Mutlu, The Southeastern Anatolia Project (GAP) of Turkey; its context, objectives and

prospects, Orient 37 (1996) 1, pp.58-87. Trevor J. Pinch & Wiebe E. Bijker, 'The Social Construction of Facts and Artifacts: Or How the

Sociology of Science and the Sociology of Technology Might Benefit Each Other', in The Social Construction of Technological Systems. New Directions in the Sociology and History of Technology The MIT Press, Cambridge, Mass/London, 1987, pp 17-51).

Anatol Rapoport, Conflict in man-made environment, Penguin, Harmondsworth 1974. James Rosenau, Turbulence in World Politics, Princeton, 1990. Michael Rush, Politics and society. An introduction to Political Sociology, Harvester Wheatsheaf

1992. Greg Shapland, Rivers of discord. International water disputes in the Middle East, Hurst & Co,

London, 1996. Linda Smircich.& Gareth Morgan, 'Leadership: The Management of Meaning', The Journal of

Applied Behavioural Science, 1982, 18(3): 257-273. Erik Swyngedouw, Modernity and Hybridity: The Production of Nature: Water and Modernisation in

Spain. Paper presented for SOAS Water Issues Study Group, University College of London, 25 January 1999.

Sybrand P. Tjallingi, Ecological Tjalingii, Strategiess and structures in environmental planning. IBN Scientific Contributions 2, DLO Institutions for Forestry and Nature Research (IBN-DLO), Wageningen 1996.

Yi-fu Tuan, Landscapes of Fear, Blackwell, Oxford, 1980. Tarja Vayrynen, 'Securitised ethnic identities and communal conflicts: A Need for Problem-

Constructing Conflict Resolution?', http://www.gmu/edu/academic/pcs/varyn.html Marq de Viliers, Water wars. Is the world running out of water?. Wendersfeld & Nicolson, 1999. Esra Vogel, Japan as Number One. Lessons for America, Harvard University Press,

Cambrdige/London 1979. Ole Waever, 'Securitization and Desecuritization' in Ronnie D. Lipschutz, On Security, Columbia

University Press, New York 1995.

24

Jeroen Warner, 'Nieuwe Ronde in Strijd om de Nijl', Vrede en Veiligheid, 2000(1). pp 510-526. Jeroen Warner, 'Environmental Security: a concept of control?' in Philip Stott et al. (ed.), Political

Ecology, Arnold, London, 2000 (in press). Jeroen Warner, 'Post-Westphalian water security', paper presented at a conference on the 350th

anniversary of the Peace of Westphalia, University of Twente, July 1998. Jeroen Warner, 'Images of water security: a more integrated perspective': AWIRU online-paper No. 3,

2000, http://www.up.ac.za/academic/libarts/polsci/awiruJeroen Warner, 'Towards a new hydrosocial contract for the 21st century', forthcoming AWIRU paper. Erik Zürcher, Turkey. A modern history, IB Tauris, London, 1993, 2nd ed. 1998. -, 'Build-Operate Transfer Model Law in Turkey' http:\\www.bircanoglu.com/botlaw.html. 11 06 99

15:22.18. -, 'ERG-Zusage für die Kraftwerke Ankara und Ilisu in der Türkei',

http://www.admin.ch/bawi/d/aktuell/presse/pm981130.htm.30 apr 1999, 11:47:41 -, 'Turkije biedt Israël waterpijp aan', Spits, 16 July 1999.