Incentives and conditions for burden-sharing in NATO. Alliance value, status enhancement, government institutional structure, and public opinion: A case study on the Dutch motivations and conditions for the deployment of Task Force Uruzgan - Afghanistan 2

(1)

INCENTIVES AND CONDITIONS FOR

BURDEN-SHARING IN NATO

Alliance Value, Status Enhancement, Government

Institutional Structure, and Public Opinion

A case study on the Dutch motivations and conditions for the deployment of

Task Force Uruzgan - Afghanistan 2006

Master Thesis Political Science: International Politics 2018/2019

Leiden University – Faculty of Social Sciences

Name: Lyke Kruizinga Student number: s1927604

Thesis Supervisor: Dr. C.C. van de Wetering Second Reader: Dr. M.B. Longo

Date: 11 June 2019 Word count: 9683

(2)

Page - 1 - of 38

Photo frontpage: Kaart Afghanistan fysisch milieu. Retrieved from: http://afghanistan-verenigdenaties-kaart.blogspot.com/2011/03/kaart-afghanistan-fysisch-milieu.html

(3)

Page - 2 - of 38

Abstract

This thesis attempts to answer the research question: ‘Which factors that shape states’ burden-sharing behaviour can explain the decision of the Dutch government to deploy Task Force Uruzgan as part of the NATO-led International Security Assistance Force in Afghanistan in 2006?’. In order to answer this question, a single case study design and content analysis were used as research methods. There are four explanatory factors, derived from previous literature and outlined in the theoretical framework used in this thesis, that can potentially explain the incentives and conditions for the Dutch burden-sharing behaviour in the case of the deployment of Task Force Uruzgan in 2006. These factors offer both systemic and domestic level explanations: alliance value, status enhancement, government institutional structure and public opinion, and are analysed through a neoclassical realist framework. The analysis demonstrates that all explanatory factors can, to some extent, explain the incentives and conditions for Dutch burden-sharing behaviour in NATO. Nonetheless, the most convincing and verified explanatory factor is alliance value. The outcome of this study contributes to a better understanding of middle powers’ burden-sharing behaviour in security alliances and paves the way for more

(4)

Page - 3 - of 38

Abbreviations

AIV Advisory Council on International Affairs

GDP Gross Domestic Product

ISAF International Security Assistance Force

NATO North Atlantic Treaty Organization or the Alliance

OEF Operation Enduring Freedom

PRT Provincial Reconstruction Team

SG Secretary-General

TFU Task Force Uruzgan

UN United Nations

U.K. United Kingdom

(6)

Page - 5 - of 38

Part 1: Introduction

After the attacks of 11 September 2001, the international community collectively decided to counteract terrorism and radicalism in Afghanistan. On 7 October 2001, the United States initiated Operation Enduring Freedom (OEF), an international operation against Al Qaida and Taliban forces in Afghanistan (Rijksoverheid, 2019). Two months after OEF started, the International Security Assistance Force (ISAF) was deployed in Afghanistan by a mandate of the United Nations (UN) Security Council (2003). ISAF’s primary objective was to support the Afghan interim government, settled in Kabul in 2001, “in the establishment and maintenance of a safe and secure environment in Kabul and its surrounding areas with full engagement of Afghan National Security Forces” (NATO, 2007). In 2003, the North Atlantic Treaty Organization (NATO or the Alliance) took command on ISAF until the mission’s completion in 2014 and the UN extended the mission’s mandate to entire Afghanistan (UN Security Council, 2003; NATO, 2015). The expansion resulted in a gradual stage-by-stage process with ISAF taking responsibility in the north in October 2003 (Stage 1), the west in September 2005 (Stage 2), the south in July 2006 (Stage 3), and the east in October 2006 (Stage 4). In stage 3, the number of ISAF forces in Afghanistan increased significantly from approximately 10,000 to about 20,000 troops after (NATO, 2007; NATO, 2015). Moreover, NATO’s first out-of-area and first ground combat operation was at its height more than 130,000 forces strong from 51 NATO and partner nations. Hence, ISAF represents one of the largest multinational coalitions in history (Gallis & Morelli, 2008; NATO, 2015).

The mission in Afghanistan has spawned debate within NATO about solidarity and the fair sharing of financial, political and human sacrifice (Ringsmose, 2010). Nonetheless, NATO succeeded in deploying tens thousands of troops to a sincerely hostile country (Ringsmose, 2010; Suhrke, 2008). Apart from the United States (U.S.) as largest contributor, the majority of these troops were European. About 75 percent of all European and non-U.S. troops in

(7)

Page - 6 - of 38

Afghanistan were represented by seven European countries (Siegel, 2009). Alongside of large military contributors such as the United Kingdom (U.K.) (9500), Germany (4335), France (3750), Italy (3160), and Poland (2140), the Netherlands deployed its largest contribution to ISAF from mid-2006 until the start of 2010 (Siegel, 2009). That is, the deployment of Task Force Uruzgan (TFU) in the southern-province of Afghanistan (stage 3) on the request of NATO in 2005. However, the NATO request for contribution to stage 3 was refused by France, Germany, Spain, Italy and several other NATO countries due to the risk of potential armed conflict (Suhrke, 2008). Nonetheless, the Netherlands accepted the request and contributed almost 2,000 Dutch soldiers per rotation in the south of Afghanistan between 2006 and 2010. In total, the Afghan war caused 25 fatal Dutch casualties (de Graaf, 2010). Hence, the Netherlands provided substantial and risky military support to ISAF through deploying this task force and acting as a ‘lead nation’ of Regional Command South of ISAF in Uruzgan (Dimitriu

& de Graaf, 2009; Siegel, 2009). Moreover, in February 2010, the Dutch government collapsed due to disagreement among the ruling coalition over the extension of TFU (Weitz, 2011). The deployment of TFU to ISAF in 2006 and its continuance until 2010 generated numerous debates and discussions in the national political and public arena (Dimitriu & de Graaf, 2009).

Therefore, one can ask why the Netherlands decided to contribute a significant number of Dutch troops to ISAF in a highly unsafe area while other NATO allies refused? This thesis seeks to understand what incentives led the Netherlands to commit and sustain forces in Uruzgan in what has become one of its most extensive expeditionary and risky military contributions to an international operation in a highly violent area since the Korean War in the 50’s (Klep, 2011).

Extensive research has sought to explain burden-sharing behaviour in alliances, in the vast majority of cases either from a systemic or international level perspective or from a domestic level perspective (Oneal, 1990;Sandler, 1992; Kreps, 2010; Oma, 2012). Nonetheless, it was found that combining both perspectives provides a better explanation of burden-sharing

(8)

Page - 7 - of 38

behaviour in alliances than the systemic and domestic level explanations do when studied on their own (Bennett, Lepgold, & Unger, 1994; Auerswald, 2004). Few studies, however, addressed both levels of perspective in explaining burden-sharing behaviour in alliances. The studies that do, rather consider dominant states in alliances (Kupchan, 1988; Bennet et al., 1994; Auerswald, 2004). More importantly, the vast majority of studies does not consider rather ideational incentives for burden-sharing behaviour (Davidson, 2011; Haesebrouck 2016; Massie & Zyla, 2018). Hence, this study seeks to contribute to academic debate and theoretical understanding of burden-sharing behaviour of middle powers1 in alliances and aims to provide new insights from both a systemic and domestic level of analysis, with a focus on ideational motivations. This requires a more refined theory, maintaining neorealist premises, but allowing for intervening domestic variables explaining states’ burden-sharing behaviour in alliances:

neoclassical realism. Therefore, this research considers both levels of analysis, that is, well-established domestic level variables and rather understudied, ideational grounded systemic level variables that can explain burden-sharing behaviour in alliances. Hence, the following research question will be answered through a neoclassical realist theoretical framework:

Which factors that shape states’ burden-sharing behaviour in alliances can explain the decision of the Dutch government to deploy Task Force Uruzgan as part of the NATO-led

International Security Assistance Force in Afghanistan in 2006?

This thesis starts by reviewing the academic literature on states’ burden-sharing behaviour in alliances. This literature review is followed by an explanation of the theoretical framework - which takes inspiration from, and combines systemic and domestic level theories on burden-sharing in alliances (Auerswald, 2004; Davidson, 2011; Massie & Zyla, 2018). Following from

1_{Middle powers can be defined as “states that are neither great nor small in terms of international power,}

capacity and influence, and demonstrate a propensity to promote cohesion and stability in the world system” (Jordaan, 2003: 165).

(9)

Page - 8 - of 38

the theoretical framework, four theoretical expectations are proposed based on the systemic level variables alliance value and status enhancement and domestic level burden-sharing variables governmental institutional structure and public opinion. Thereafter, the methodology and research design section will elaborate on the design of the study (single case study), the case selection (the deployment of Task Force Uruzgan in 2006), the research method (content analysis), operationalization and data collection. This is followed by an analysis of the data showing what incentives and conditions underpin the deployment of Task Force Uruzgan in 2006. The analysis is divided into three sections: (1) alliance value, (2) status enhancement, and (3) government institutional structure and public opinion. The last part provides an overall conclusion, theoretical implications of the findings, limitations, and suggestions for further research.

(10)

Page - 9 - of 38

Part 2: Literature Review

Extensive literature is available to explain burden-sharing in security alliances. A general definition of burden-sharing is given by Cimbala & Foster (2005: 1): “the distribution of costs and risks among members of a group in the process of accomplishing a common goal”. With regard to burden-sharing within NATO as the security alliance in this thesis, Oma (2012: 563) argued that “factors such as politically risky missions overseas, the lack of a common threat perception, and more generally a larger and less homogeneous group of member-states than during the Cold War have arguably added to the challenge of securing a ‘fair’ sharing of the alliance’s burdens”. Hence, the research field of alliance behaviour and burden-sharing is just

as relevant as ever. Burden-sharing debates have recently become more intense and alliances and their operations are confronted with a changing nature (Oma, 2012).

In the vast majority of cases, scholars sought to explain the factors that shape states’ alliance contribution from a neorealist systemic level perspective. The systemic level perspective explains burden-sharing through external ‘carrots’ and pressures. Frequently used systemic level explanations are based on collective action theory, balance-of-threat neorealism, and alliance dependence (Olson, 1965; Oneal 1990b; Bennet et al, 1994; Oma, 2012; Massie & Zyla, 2018). Moreover, systemic level explanations, such as alliance value and status seeking, are used to examine states’ burden-sharing behaviour in alliances (Massie & Zyla, 2018). Although neorealism based systemic level explanations succeed in laying hold of political leaders’ incentives to contribute to an alliance, the understanding of specific decision outcomes demands a neoclassical realist approach that also considers domestic level explanations. The domestic level perspective incorporates internal choice mechanisms of a state in explaining burden-sharing (Bennet et al., 1994; Oma, 2012). The domestic level explanations that are generally discussed in literature are state autonomy and public opinion, bureaucratic politics, or elite consensus and institutional government structures (Kupchan, 1988; Bennet et al, 1994;

(11)

Page - 10 - of 38

Auerswald, 2004; Massie & Zyla; 2018). The following sections will discuss each of these systemic and domestic level theories by their considerations in the academic literature on burden-sharing in alliances. Thereafter, the theoretical framework of this thesis will be delineated, taking inspiration from and adapting the hereafter discussed literature on burden-sharing behaviour in alliances.

2.1 Systemic level explanations for burden-sharing behaviour

The concept of collective action is frequently used in explaining the behaviour of nations in alliances by examining military expenditures and states’ Gross Domestic Product (GDP) (Oneal, 1990; Oma, 2012). The theory of collective action finds its roots in Olson’s (1965) work The Logic of Collective Action. Olson’s (1965) theory is mostly applied to research on cooperation among states in the security realm and proposes “a surprising tendency for the ‘exploitation’ of the great by the small” regarding a public good (Olson, 1965: 3; Oneal, 1990b).

Bennet, et al (1994: 41) define Olson’s argument by stating that “with respect to a good that would benefit each member of a group regardless of how much it costs to supply the good (i.e., a collective good), the smaller the state, the less likely it is to contribute”. Hence, the concept of collective action concerns the relationship between the gain of the collective good (if it were supplied to all) and the size of the state. Therefore, in early works it is argued that small states expect dominant states with “the greatest means” to bear most of the costs of the public good as the dominant states would gain also the most, even if small states do not contribute (Olson; 1965; Oneal, 1990b; Bennet, et al, 1994). With regard to this argumentation, Olson and Zeckhauser (1966) argued that, while assuming NATO as a security alliance providing a public good, small states within in NATO tend to ‘free ride’. However, the definition of a public good as an output provided by an alliance, non-rival and non-excludable, is challenged by critics (Oneal, 1990; Sandler, 1992). It is argued that alliance-based security “fails to meet the non-excludability requirement” (Ringsmose, 2009: 79). To conclude, recent studies demonstrate

(12)

Page - 11 - of 38

that the theory of collective action and its ‘exploitation hypothesis’ is most successful in explaining dominant states’ disproportional burden-sharing behaviour and the relative static distribution of burdens. However, it fails to explain other outcomes such as why middle powers or small powers provide meaningful contributions to an alliance’s provision of collective security outside its borders (Thies, 2003, Ringsmose, 2009; Oma, 2012; Auerswald 2004; Haesebrouck 2016).

A second dominant systemic level explanation in the literature for states’ burden-sharing behaviour in alliances, is the neorealist-based theory of balance-of-threat. It finds its origin in Walt’s (1987) balance-of-threat theory that suggests that states balance against threats in the

anarchic international system, instead of power. According to Bennet et al (1994), this systemic level explanation concerns the relation between the degree of threat a state is facing and its proportional contribution to oppose that threat. The degree of threat is defined by Walt (1987: 5) as the distance from a potential aggressor, evidence of its intentions, and its offensive capabilities. Hence, the balance-of-threat theory argues that the higher the threat perceived by a state, the more likely that state is to contribute to an alliance (Oma, 2012). Similar to the collective action theory, the balance-of-threat theory is more suitable to explain the behaviour of dominant powers in alliances. As Ringsmose (2009: 77) argued “the smallest among allies have little or no inducement to balance external threats by internal means”.

Existing research on alliance dependence, a third systemic level perspective explanation for burden-sharing behaviour, found strong support for the relation between the dependency of an ally and their burden-sharing behaviour (Bennet et al, 1994; Ringsmose, 2009). According to Snyder (1984), the theory of alliance dependence concerns the balance between two inherent fears of states in alliances, that is abandonment and entrapment, and alliance autonomy. Abandonment appears when “an ally realigns with one’s adversary or fails to help against it” (Bennet, et al., 1994: 44). Entrapment appears when “one becomes entangled in a conflict

(13)

Page - 12 - of 38

central to an ally’s interests but peripheral to one’s own in the hope that preserving the alliance

will outweigh the risk and costs of future war” (Bennet, et al., 1994: 44). This balance entails a trade-off between security and autonomy, and is based on the member state’s dependence on an ally (Oma, 2012). On a more general note, Oma (2012: 565) defined the argument of the alliance dependence theory as “the greater a state’s dependence on an ally, the more likely it is to accommodate its protector’s request for support”.

2.2 Domestic level explanations for burden-sharing behaviour

The neoclassical realist approach, that also takes domestic level explanations into consideration instead of only focusing on systemic level explanations of states’ burden-sharing behaviour, fares well in explaining for example the size and form of contributions (Oma, 2012: 565). Various studies examined the factors of state autonomy and public opinion (Kupchan, 1988; Bennet et al., 1994; Auerswald; 2004). Oma (2012: 566) summarized these expectations, based on liberal theories, as “states can be expected to be more prone to contribute when there is popular support for doing so; and, second, that institutionally weak executives should be more reluctant to use force”. This argumentation concerns the state’s autonomy vis-à-vis society, in other words “the state’s autonomy with respect to the preferences of civil society, most often reflected by legislators” (Bennet et al., 1994: 45). Hence, specific contribution can be expected only when a state can autonomously act upon the needed resources for a specific contribution (Bennet et al., 1994).

With regard to the relative autonomy of a state, Massie and Zyla (2018) argue that democratic regimes are more prone to political constraints that may hinder the implementation of their democratic executives’ policy preferences. That is, democratic executives are accountable to

electoral constituencies and their support. The authors state that elite consensus determines whether democratic states and their executives can implement their policy preferences (Massie & Zyla, 2018). Massie and Zyla (2018) refer to elite consensus as defined by Kreps (2010: 198):

(14)

Page - 13 - of 38

“the absence of a coherent opposition to the government’s foreign policy by political parties”.

Hence, Massie and Zyla (2018: 8) state that “the relative autonomy democratic executives have in implementing their foreign policy preferences thus ultimately rests on the degree to which there is an elite consensus on the value of committing military forces”.

However, Bennet et al., (1994: 47) argue that “even when the state is autonomous or the public favors a contribution, the level and kind depend also on the role and influence of the chief executive relative to top bureaucratic actors”. Hence, the domestic level approach of bureaucratic politics includes the argumentation that governments consist of several actors that share power and have partly overlying interests (Oma, 2012). This bureaucratic politics approach makes three claims: (1) bureaucratic bargaining results in compromises that determine government actions, (2) officials’ ‘seats’ shape their view on policy issues, and (3) top officials’

relative bureaucratic power determines the outcome of bureaucratic bargaining (Bennet, et al., 1994: 47). Auerswald (2004) more broadly merged these abovementioned argumentations on state autonomy, bureaucratic politics and elite consensus in ‘government institutional structures’. According to his study, the strength of democratic states’ executives can be divided in the categories of weak, medium and strong. In that regard, the stronger the states’ executives are, the easier they are willing to contribute significantly to an alliance (Auerswald, 2004).

(15)

Page - 14 - of 38

2.3 Theoretical framework

The discussion of the literature reveals that a substantial amount of research on burden-sharing either focuses on domestic level explanations or systemic level explanations (Kaarbo, 1996; Oma, 2012; Hudson, 2013; Oktay, 2014). As stated before, systemic level explanations well capture political leaders’ incentives to contribute to an alliance. However, specific decision outcomes can only be fully understood by addressing domestic level explanations (Oma, 2012). A few studies attempted to integrate the systemic level and domestic level explanations in order to explain states’ burden-sharing behaviour in alliances. As already discussed individually in

the sections above, Bennet et al. (1994) examined five externally and internally driven variables that could explain states’ burden-sharing behaviour in the case of the Operation Desert Storm coalition. The first three variables explain “foreign policy through external inducements and constraint, and are respectively collective action, balance-of-threat, and alliance dependence” (Bennet, et al., 1994: 41). Furthermore, the authors applied two domestic level variables, which are state autonomy and public opinion and bureaucratic politics (Bennet et al., 1994). The authors suggested an integrated model that explains how systemic and domestic level variables could be combined to produce outcomes. Moreover, they conclude that for most of the considered states that contributed to Operation Desert Storm, alliance dependence better explained states’ burden-sharing behaviour than collective action and balance-of-threat

variables (Bennet et al., 1994).

A second and more recent attempt to combine systemic and domestic level variables by designing a decision-making model to explain democratic states’ behaviour in alliances, is conducted in the study of Auerswald (2004). At first, he examined four existing systemic and domestic level theories to explore the contribution to NATO’s Operation Allied Force in Kosovo in 1999 of five prominent allies: the U.S., the U.K., France, Germany and Italy. Respectively, the existing approaches the author used are collective action, balance-of-threat,

(16)

Page - 15 - of 38

public opinion and government institutional structures (Auerswald, 2004). As an alternative, Auerswald (2004) also applied an integrated decision-making model, that includes the examined existing explanations, to his case. The research suggests that the integrated model is more appropriate in explaining democratic states’ burden-sharing behaviour in alliances than the existing approaches do separately. Auerswald’s (2004) integrated model departed from the model designed by Bennet et al. (1994) by considering mostly similar systemic and domestic level variables defined as:

(1) Collective action – “the existence of collective goods generates particular types of individual and group behaviour” (p. 636); this concept concerns the extent to which alliance members provide collective goods and would benefit from doing so. The core characteristics of collective goods are (1) “they are provided to all if they are provided at all” and (2) “one’s consumption of said good does not exhaust the supply of that good” (p. 636).

(2) Balance-of-threat – “threatened states in pre-existing alliances will contribute to an intervention that lessens that threat” (p. 639); this concept concerns the notion of alliances as a balancing mechanism “as a response to threats faced by individual states” (p. 639). Hence, “states facing significant threats will contribute/intervene” (p. 639).

(3) Public opinion – “state leaders should be more willing to use force were the public to strongly support such action” (p. 640); this concept concerns a ‘bottom-up’ perspective – public preferences shape foreign policies – whereby public opinion represents states’ burden-sharing behaviour (p. 640).

(4) Government institutional structure – “the domestic circumstances confronting a chief executive influence that individual’s conflict decisions, and those circumstances vary

predictably across democracies based on the structure of their domestic institutions of governance” (p. 641); this concept concerns “the relationship between chief executives and

(17)

Page - 16 - of 38

legislatures” whereby the domestic institution of governance (e.g. prime minister, majority party or a president) determines the strength of the executives in relation to public and/or legislative opposition. Hence, “institutionally strong executives have more domestic policy leeway and should hesitate less before using force than would weak executives operating under tight governmental constraints” (p. 642).

As can be seen, Auerswald (2004) excluded the approach of alliance dependence that Bennet et al. (1994) examined as an explanatory variable by arguing that fear of abandonment by the U.S. was inapplicable for NATO allies as the Clinton administration aimed at upholding a strong alliance in that time. Furthermore, the author’s integrated model did not take balance-of-threat theory into consideration because “we are interested in a class of conflicts – interventions of choice – in which the survival of our democratic states of interests is not immediately threatened” (Auerswald, 2004: 644).Therefore, as ISAF was an intervention of choice with no risk of NATO abandonment and terrorism posing no existential threat to the territorial integrity of allies or the Alliance itself (Advisory Council on International Affairs (AIV), 2017: 14; Suhrke, 2008), this research follows Auerswald (2004) by not considering alliance dependence and balance-of-threat theory. Hence, Auerswald’s (2004) brought forward an integrated decision-making model on allies’ burden-sharing behaviour based on the core concepts of theories on collective action, public opinion, and government institutional structure.2

However, by including the theory of collective action, Auerswald’s (2004) model is mainly designed to explain burden-sharing behaviour of dominant allies. As argued before in the literature review, collective action theory often fails to explain other outcomes than dominant powers accepting the largest share of the burden (Thies, 2003; Ringsmose, 2009; Oma, 2012;

2_{Auerswald’s integrated model is not shown here as the theoretical framework of this research only uses} his conceptualization of two of the four variables used in the integrated model. For a visualisation of the model, see Auerswald (2004: 643).

(18)

Page - 17 - of 38

Massie & Zyla, 2018). Hence, the theory of collective action does not perfectly align with the case of this research, as the Netherlands is perceived as a middle power in this research (Jordaan, 2003; Laatikainen, 2006; Oosterveld & Torossian, 2018). Moreover, it is argued that Auerswald’s (2004) model “leaves little room for different state motivations for contributing”

(Oma, 2012: 568). Hence, Auerswald’s (2004) systemic level concepts and expectations will not be taken into consideration for this theoretical framework.

In addition to existing and more frequent applied existing systemic level theories such as the collective action theory, balance-of-threat theory and alliance dependence, Massie and Zyla (2018) discussed two alternative and less studied systemic level theories with an ideational grounding, being alliance value and status enhancement, and applied them to the case of Canada’s disproportionate military burden-sharing in Afghanistan. The theory of alliance value is especially applicable in the case of NATO’s collective defense principle, as this is based on

Article V of the Washington Treaty that states that allies are committed to support its allies “in case of aggression with such action as it deems necessary” (Massie and Zyla, 2018: 5). Therefore, alliance value theory seeks to explain states’ burden-sharing behaviour with respect

to the extent a state values its alliance and the solidarity between allies that is created when valuing that alliance. Argued by Davidson (2011: 15), alliance value theory differs from alliance dependence theory as “value does not necessarily entail dependence’. Moreover, Davidson (2011: 15) emphasized a reason of alliance value as “the influence states expect to have on alliance policy”. However, as Davidson is confronted with little empirical support and significant overlaps with other explanations of alliance contribution, this study defines alliance value as ‘the ally’s emphasis on solidarity within the alliance’.

In addition to alliance value, Massy and Zyla (2018) examined status enhancement as an explanation for burden-sharing behaviour. Several authors did acknowledge that the phenomenon of status enhancement aspired by states in alliances can partly explain

(19)

burden-Page - 18 - of 38

sharing behaviour (Cooper, Higgott, and Nossal 1991; Willer, 2009; Kreps, 2010). Status can be defined as “a state’s attribute of a social role vis-à-vis a comparison group, such as begin a leader of a coalition of states” (Dafoe, Renshon & Huth, 2014; Massie & Zyla, 2018: 6). Massie and Zyla (2018: 8) expect that “states seeking to elevate their status as leading military allies are likely to commit sizeable and meaningful forces to coalition operations relative to their past behavior, military capabilities, and other allies”. The authors argue that theories of alliance value and status enhancement complement each other in explaining sizeable and meaningful contributions to U.S.-led multinational missions. Whereas alliance value explains general contributions to alliance operations, status enhancement can explain states’ specific contributions in a wide range of support options such as combat or noncombat military support (Jockel & Massie, 2017).

For this research, the theoretical framework takes inspiration from, and adapts the hereabove discussed domestic level variables proposed by Auerswald (2004) and adds alternative ideationist grounded systemic level variables and corresponding expectations developed and studied by other scholars (Davidson, 2011; Massie & Zyla, 2018). This research will follow the conceptualization of the domestic level variables public opinion and government institutional structures developed by Auerswald (2004) and the systemic level variables alliance value and status enhancement as defined by Davidson (2011) and Massie & Zyla (2018). However, the purpose of this research is not aimed at developing a new integrated decision-making model following Bennet et al., (1994) and Auerswald (2004), as this requires a large examination of the alliance burden-sharing behaviour of multiple states with varying outcomes of contribution which will exceed the scope of a Master thesis. Moreover, the academic debate on burden-sharing behaviour is remarkably focused on dominant allies. Rather, this research seeks to contribute to academic debate and theoretical understanding of middle power states’ burden-sharing behaviour in alliances and aims to provide new insights from different levels of

(20)

Page - 19 - of 38

analysis. It does so through considering both levels of analysis, that is, well-established domestic level variables and rather new and understudied systemic level variables that can explain burden-sharing behaviour in alliances. To that end, this research is an attempt at ‘bridge-building’ between two schools of thought by following a neoclassical realist approach through

examining factors from both systemic and domestic level of analysis. Hence, based on the hereabove discussed literature and argumentation, the following theoretical expectations for middle power states’ burden-sharing behaviour in alliances are derived for the purpose of this thesis:

(1) Alliance value: middle powers that value their security alliance are more likely to

contribute to an alliance operation.

(2) Status enhancement: middle powers that seek to enhance their status in an alliance are

more likely to contribute to an alliance operation.

(3) Government institutional structure and public opinion: middle powers that have

domestically strong executives, or medium executives enjoying legislative and public support, are more likely to contribute to an alliance operation.

(21)

Page - 20 - of 38

Part 3: Methodology

The goal of this study is to provide contextualized insight into the incentives and conditions that led to the deployment of TFU in 2006 by addressing three theoretical expectations on states’ burden-sharing behavior in alliances. In order to achieve this goal, a single case study

design is chosen with an explanatory aim (Yin, 2003). As the aim of this study is an in-depth analysis of the complex phenomenon of burden-sharing in alliances, a case study is the best choice as it “allows the exploration and understanding of complex issues” (Zainal, 2017: 1; Yin, 2003). A single case study design “entails the detailed and intensive analysis of a single case” (Bryman, 2016: 66). The single case in this study can be defined as a single event: the deployment of Task Force Uruzgan.

3.1 Case selection

The case selection is primary based on its relevance to the research objective of the study (George & Bennet, 2005). Founded with preliminary knowledge of the case, the crucial (most-likely) case selection was applied for this study (George & Bennet, 2005). As the Netherlands is expected to value NATO as a middle power, seeks to enhance its status after the failed mission in Srebrenica3, and enjoyed at least legislative support preceding its decision to deploy TFU, the case of deployment of TFU is perceived as most likely to be consistent with the theoretical predictions (Levy, 2008). As this study aims for verification of propositions, a crucial (most-likely) case is most appropriate (Gerring, 2008; Flyvberg, 2006). Furthermore, the case of TFU is purposively selected because a) it was part of one of the largest multinational coalitions in history and NATO’s first out-of-area and ground combat mission (NATO, 2015; Gallis & Morelli, 2008), b) it is the first NATO-mission to which the Netherlands provided a combat contribution this large since the Korean War in 1950 (Klep, 2011), and c) the analysis

3_{In July 1995, the Dutch battalion (Dutchbat) under the command of the UN peacekeeping force}

UNPROFOR failed to offer protection to the Muslim-enclave in Srebenica, Bosnia and Herzegovina (van Dam, 2011).

(22)

Page - 21 - of 38

of this case is a contribution to the academic debate as more studies on specific-outcome decisions of alliance contribution and burden-sharing, particularly with a focus on non-dominant allies, are needed (Oma, 2012). The timeframe of this single case study is set from the NATO request in 2005 until the decision of the Dutch government to deploy TFU to ISAF in 2006.

3.2 Operationalization

For each theoretical expectation, the conceptualization of the factors is presented in the theoretical framework. For the theoretical expectation of alliance value, this study examines the voluntary military support to an alliance’s operation rather than forced cooperation. Hence,

this study looks at the effort of the ally to voluntarily contribute significant resources to a mission in order to improve their relationship with dominant allies and the alliance more broadly. Indicators of these efforts are changes in the state’s foreign policy, historical record of the state’s alliance contribution, the ally’s (international) statements within and outside NATO

stages on solidarity and contribution, and/or statements and actions aiming at improving the relation with a dominant ally, such as official statements by the government.

Concerning the theoretical expectation of status enhancement, this study investigates the state’s international military status and reputation within the alliance and the contribution of troops to the alliance’s operation in comparison to past operations. An indicator for the latter are

statements of the state itself on buttressing its reputation and reliability by contributing to the alliance’s operation will be examined.

Finally, for the theoretical expectation of government institutional structure and public opinion, this study examines the executives’ strength à-vis other government entities as well as vis-à-vis society in terms of power over foreign policy. Therefore, it looks at the domestic institutional structure or system of the state (e.g. presidential or parliamentary), and legislative constraints and domestic penalties for policy failure. Respectively, indicators consist of

(23)

Page - 22 - of 38

decisions on office tenure, parliamentary motions, parliamentary voting, public opinion polls, and/or the perception of the debate by media outlet and public opinion.

3.3 Methods of data collection and analysis

This study relies mostly on primary data derived from international and domestic policies and legislation, such as NATO’s policies and official documents, state’s policies, and speeches of

executives at national and international conferences. Furthermore, secondary data includes reports of parliamentary debates of the Dutch House of Representatives as well as other relevant parliamentary and governmental documents, reports, and statements published by the Dutch government, the States-General, the Dutch Ministry of Foreign Affairs, the Dutch Ministry of Defence and the Dutch Ministry of Develop Cooperation. Also, media outlets and public opinion polls are included in this research. Finally, a selection of academic publications on the subject is analysed.

This data is studied through the qualitative approach of thematic content analysis. This method of data analysis “comprises a searching-out of underlying themes in the materials being analysed” (Bryman, 2016: 557). The content analysis will be conducted on documents relevant

to the systemic and domestic level theoretical expectations with the themes of alliance value, status enhancement, government institutional structure and public opinion. These four themes are extensively studied through examining the above presented operationalized indicators in the data. In the analysis, two levels of content are studied: “themes and main ideas of the text as primary content; context information as latent content” (Marying, 2004). This method of analysing systemic and domestic level variables may provide significant empirical insights that explain the motives behind the Dutch decision to deploy TFU in 2006.

(24)

Page - 23 - of 38

Part 4: Analysis

4.1 Alliance value: Solidarity within NATO

Mid-2005, NATO requested the Netherlands and other allies to contribute to stage 3 of the expansion in southern Afghanistan (House of Representatives, 2005). At the end of 2005, the force contribution of the allies seemed planned: the U.K. planned on contributing 3,400 troops, Canada around 1,000, the Netherlands planned approximately 1,200 to 1,400 troops, and Denmark and Estonia would provide smaller numbers of troops (Suhrke, 2008). From 2004 until 2009, NATO’s Secretary-General (SG) was the Dutch former politician of the Christian

Democratic Party (CDA) and Minister of Foreign Affairs Jaap de Hoop Scheffer. NATO’s SG has “invited the Netherlands to contribute to the deployment of ISAF in southern Afghanistan together with several other allies” (House of Representatives, 2005b: 1). Jaap de Hoop Scheffer was “an indefatigable advocate of engagement” (Suhrke, 2008: 225). NATO’s SG stated in a

speech at the 42nd Munich Conference of Security Policy that “we need more solidarity in the way we pay for our operations” (NATO, 2006: 1). In another speech, Jaap de Hoop Scheffer emphasized the value of states’ contributions in Afghanistan and their solidarity by stating:

In Afghanistan we now have NATO-led troops from 37 nations – all doing a superb job in difficult circumstances … Romania, Estonia and Denmark all have troops in the south, but over the summer months, British, Canadian and Dutch troops, bore the brunt of the fighting, and, I regret to say, the British and Canadians bore the brunt of the casualties too. But they have been successful. (NATO, 2006b: 1).

In order to assess whether the Dutch government valued the Alliance and its collaboration in Afghanistan as much as their Dutch colleague Jaap de Hoop Scheffer, a closer look at the Dutch government’s statements on NATO and Afghanistan is required.

(25)

Page - 24 - of 38

The Afghanistan war did not threaten crucial, existential interests of the Netherlands. Nonetheless, Ben Bot (Foreign Affairs), Henk Kamp (Defence), and Agnes van Ardenne (Development Cooperation) stated that “since the end of 2001, the Netherlands, as a member

of the international community, has committed itself with substantial financial and military resources to the stabilization, democratization and reconstruction of Afghanistan” (House of Representatives, 2005c: 6). Moreover, the Ministers declared that “through this great effort, the Netherlands has contributed to the promotion of the international legal order and to the fight against international terrorism” (House of Representatives, 2005c: 6). Hence, the ‘intervention of choice’ demonstrates the voluntarily character of the Dutch contribution of forces (Jockel,

2014). The Dutch government partly justified its participation by imperatives of alliance and trans-Atlantic solidarity (Weitz, 2011).

This imperative is demonstrated in the government’s answer on a question of the House of Representatives’ commissions for Foreign Affairs and Defence. The commissions asked “To

what extent has solidarity within NATO played a role in the realization of the government decision to participate in the mission?” (House of Representatives, 2006c: 18). The government’s answer acknowledges the importance of NATO’s goal and alliance collaboration as demonstrated in the following statement: “The government considers it of great importance that NATO succeed in its mission to support the democratically elected authority in Afghanistan. For that reason, the Netherlands has also agreed to the south of Afghanistan and wants to contribute to this together with other allies” (House of Representatives, 2006c: 18).

Furthermore, the government’s answer emphasizes solidarity within NATO: “Thanks to the solidarity principle of NATO, the Netherlands can, if necessary, appeal to the military deployment of other allies in Uruzgan. The Dutch government attaches great importance to this, also with a view to a proportionate alliance burden sharing” (House of Representatives, 2006c:

(26)

Page - 25 - of 38

their answer: “In the context of the preparation of the intended contribution of the Netherlands to Uruzgan, close consultations have been held with NATO and with relevant allies since June 2005. Among other things these consultations led to good agreements on mutual assistance” (House of Representatives, 2006c: 18). As these quotations of the government demonstrate, the Netherlands fulfilled an active role in contributing to NATO’s interests and promoting solidarity and alliance burden-sharing in Afghanistan.

(27)

Page - 26 - of 38

4.2 Status enhancement: Status building in Afghanistan

The Dutch security position on the national and international level can be perceived “as a medium power, pursuing good relations with the United States through, among other things, being a trustworthy member of NATO” (Grandia, 2015). Hence, the Netherlands is able to

display its dedication to the Alliance through attempting to be a trustworthy NATO member. In the last two decades, the Netherlands demonstrated its support through military contributions, including military operations in Iraq and Afghanistan (Grandia, 2015). These military contributions reflect the efforts of the Dutch Ministries of Foreign Affairs and Defence to provide the Netherlands a trustworthy military reputation in the international arena (Verbeek & van der Vleuten, 2008).

This trustworthy military reputation of the Netherlands was especially insecure after the massacre in Srebrenica in July 1995 where thousands of Muslims were killed by militias from Serbia, despite they were under the protection of the Dutch peacekeepers (Blair & Fitz-Gerald, 2009; Grandia, 2015). The fact that the Dutch forces were unable to protect the Muslims, largely shaped the Dutch perception on the use of force and to some extent initiated the desire to be perceived as a trustworthy ally (Blair & Fitz-Gerald, 2009). Therefore, the possibility to contribute substantial military forces for NATO’s expansion in a high risk area in southern Afghanistan seemed to be an opportunity for the Netherlands to demonstrate its ability as a trustworthy military ally. Hence, “an implicit desire of the Netherlands to be a key player in the international arena could be fulfilled” (Grandia, 2015: 114).

The ambition of the Netherlands to contribute to a safe area for citizens in Afghanistan can already be found prior to the deployment of TFU to ISAF in 2006. The Dutch government emphasized its active support to ISAF in Kabul, North- and West-Afghanistan by stating “Meanwhile, ISAF has achieved a lot in Kabul and North and West-Afghanistan with an active Dutch contribution” (House of Representatives, 2005c: 2). Furthermore, the Dutch government

(28)

Page - 27 - of 38

provided an example of the active Dutch contribution by declaring that “the Dutch PRT [Provincial Reconstruction Team] has not only contributed to more stability and to increasing the authority of the government, but also given an impulse to reconstruction and economic development” (House of Representatives, 2005c: 2). Moreover, the same letter of the government stated that “our country wants to contribute to the continuation of that successful model and also to offer people in South Afghanistan a safer environment and a better future” (House of Representatives, 2005c: 2). Hence, the deployment of TFU in southern Afghanistan rather provided the Netherlands with the opportunity to further enhance its status as a trustworthy military ally. However, the data is rather pointing at the intrinsic motivation of the Netherlands to contribute to the interests of the Afghan people, despite all the risks, than at the direct aim for enhancing its status within NATO. The Dutch government expressed that it “wants, despite the risks that military operations inevitably involve, to continue to do so in the future, all the more since the needs are still high and lasting support and involvement are needed for sustainable peace and development” (House of Representatives, 2005c: 6). Nonetheless, the desire to be perceived as a trustworthy military ally is reflected in the fact that the Netherlands contributed military troops to a high-risk area in order to achieve sustainable peace and development in Afghanistan.

Moreover, the Netherlands clearly aimed for influence on decision-making in the ISAF-mission. The Dutch Ministers of Foreign Affairs, Defence and Development Cooperation argued that “the Netherlands is well represented in the relevant decision-making forums and headquarters, which guarantees sufficient influence on decision-making with regard to ISAF”

(House of Representatives, 2005c: 22). Furthermore, the Netherlands acted, alongside Canada and the U.K., as one of the leading countries in southern Afghanistan (House of Representatives, 2005c).

(29)

Page - 28 - of 38

4.3 Government institutional structure and public opinion: Executives, legislators and citizens in the Uruzgan debate

The Dutch government is based on a constitutional monarchy. Therefore, the King4_{of the}

Netherlands as well as all ministers and state secretaries together form a coalition government (Rijksoverheid, 2019b). The Prime Minister, ministers and their states secretaries bear the responsibility for specific policy areas as executives of the Netherlands. This team of executives is also referred to as the Cabinet (ProDemos, 2013). In order to be effective, the coalition government needs to work on consensus building within the Cabinet as well as with the States-General. The latter institution forms the parliament of the Netherlands, consisting of the Senate and the House of Representatives. The States-General is responsible for controlling the government. Hence, the Cabinet members are not allowed to occupy a post in the Dutch parliament (ProDemos, 2013). The Dutch government, often a coalition, rules over a so-called multi-party system. Governmental decisions on behalf of the Cabinet are made by the Cabinet’s executive body, the Council of Ministers (ProDemos, 2013). Nonetheless, the ministers are bound to the ministerial responsibility and information duty as the government is obliged to inform the parliament. Moreover, in a situation of distrust in the Cabinet or individual ministers, the parliament can submit a motion of no confidence which could result in a resignation of the minister or Cabinet (Rijksoverheid, 2019c). Therefore, the Cabinet – i.e. the executives of the Netherlands – does not possesses domestic strong power over the States-General – i.e. the legislators of the Netherlands.

The government Ministers of Foreign Affairs, Defence, and Development Cooperation informed the parliament about the intention to deploy Dutch troops in southern Afghanistan

4The King of the Netherlands “is inviolate and has no actual political authority” (ProDemos, 2013:

(30)

Page - 29 - of 38

with the Article 100 Letter Uruzgan5 on which basis the parliament could debate the deployment (House of Representatives, 2005c). The Article 100 Letter send to the parliament on 22 December 2005, marked the completion of the decision process within the government (House of Representatives, 2006). In the process of informing the parliament through this

Article 100 Letter, the government clearly demonstrated their viewpoint as can be seen in the

following statement of the Ministers Ben Bot, Henk Kamp, and Agnes van Ardenne: “The government has thus concluded that it is prepared to respond to the request from NATO under the premises and conditions as formulated in the Article 100 Letter of 22 December 2005” (House of Representatives, 2006: 4). Moreover, the government argued in the same letter to the parliament that “the content, scope and structure of the Article 100 Letter make it entirely clear for the government what its willingness is: it wants to make a Dutch contribution to ISAF through the deployment of the Dutch Armed Forces” (House of Representatives, 2006: 4).

Although the government is allowed by the Constitution to deploy Dutch forces without the majority endorsement in parliament, in practice the government desires a majority of parliament members endorsing the proposal of military deployment (Grandia, 2015). The government stated that “with the imposition letter of December 22, the government has indicated that it will consider the consequences to be attached to that parliamentary opinion after the voting on the intended deployment by your Chamber” (House of Representatives, 2006: 4). Hence, the Dutch parliament has “acquired itself quite a prominent role in the endorsement of deploying military means” (Grandia, 2015: 65). On 2 February 2006, the House of Representatives voted in favour of the mission’s deployment to Uruzgan under the flag of the TFU: 126 votes in favour and 24

5_{The Article 100 Letter Uruzgan (22 December 2005) described the aim of the mission – i.e. promoting}

safety and stability, good governance and economic development, and training Afghan police and armed forces (House of Representatives, 2005c). As the mission was in accordance with the ISAF mandate, the Netherlands was thus not officially at war (de Graaf, 2010).

(31)

Page - 30 - of 38

votes against (AIV, 2006; Van der Meulen & Vos, 2011). Only one of the three coalition parties6 in the House of Representatives, Democrats 66 (D66), voted against the deployment (Parlement, 2019). Furthermore, of the opposition parties the Socialist Party (SP), GreenLeft (GL), and Godelieve van Heteren of the Labour Party (PvdA)-fraction voted against the mission (Parlement, 2019). Hence, in the end, the government of the Netherlands enjoyed legislative support by the parliament to deploy Task Force Uruzgan in Afghanistan. Nonetheless, the formal decision to deploy troops in Uruzgan took months to effectuate due to political division as the parliament was already informed about the intention of deployment in the summer of 2005. This is reflected in the emphasis on the risks of this ‘dangerous’ mission as stated in numerous political statements and news items about the expansion of ISAF to the south of Afghanistan (Hazelbag, 2009).

Whereas the Dutch government, after rather excessive political discussion, enjoyed legislative support by the majority of parliament, the support of the Dutch population remained rather moderate. The opinion polls7 for public support for the mission demonstrate a reasonable increase in public support – on 21 December 2005, 26% supported the mission, on 13 January 2006, 33%, on 30 January 2006, 45%, on 3 February 2006, 49%, on 13 June 2006, 40%, and on 29 July 2006, 48.9% (De Volkskrant, 2006; Van der Meulen & Vos, 2011). The AIV argued that “this is still not a majority of public opinion and even farther away from a large majority which the AIV considers desirable in principle for high-risk operations such as the Uruzgan mission” (AIV, 2006: 13). However, the sole focus of the parliamentary debate on 2 February 2006 was parliamentary support for the mission. Despite low public support, neither the government nor the House of Representatives referred even single time to public opinion during

6_{Cabinet-Balkenende II (2003-2006) consisted of three coalition parties: the Christian Demoratic}

Appeal (CDA), the People’s Party for Freedom (VVD) and Democrats 66 (D66) (Parlement, 2019).

7_{Data retrieved from several opinion polls by de Hond (21 December 2005), de Hond (3 February}

2006), TNS NIPO / RTL Nieuws (21 December 2005), and TNS NIPO / de Volkskrant (13 January 2006) (De Volkskrant, 2006; Van der Meulen & Vos, 2011).

(32)

Page - 31 - of 38

the debate (House of Representatives, 2006b). Hence, in the political context of this mission, the government relied only on parliamentary majority support for deploying TFU in southern Afghanistan in 2006.

(33)

Page - 32 - of 38

Part 5: Conclusion and Discussion

This study aimed at providing contextualized insight into the incentives and conditions for middle power’s burden-sharing behaviour in alliances. Hence, this thesis applied a case study

on the motivations of the Netherlands for the deployment of Task Force Uruzgan in Afghanistan, 2006. In this case study, four systemic and domestic level theoretical expectations on burden-sharing behaviour in alliances were examined. The four theoretical expectations were derived from theories on the variables alliance value, status enhancement, government institutional structure and public opinion. Hence, this study provided insights from both systemic and domestic level perspectives through a neoclassical realist approach. Moreover, rather ideational grounded systemic level expectations were chosen as the majority of scholars examined non-ideational grounded explanations for burden-sharing behaviour.

Concerning the systemic level factor alliance value, the analysis demonstrates that the Netherlands was actively engaged in contributing to NATO’s interests and promoting solidarity

and alliance burden-sharing in Afghanistan. The Dutch government greatly emphasized the need of NATO’s succession in Afghanistan and the importance of collaboration with allies.

Hence, the Dutch imperatives of alliance solidarity reflect a significant value rewarded to NATO. To that extent, the theoretical expectation on alliance value can be verified as the Netherlands valued its security alliances and contributed to the alliance’s operation.

In the analysis of the systemic level factor status enhancement it was found that, in light of the earlier failure of the Dutch troops in Srebrenica, the Dutch security policy was focused on the desire of being perceived as a trustworthy military ally in Afghanistan by contributing significant military troops to a high-risk area and being active in decision-making forums of ISAF. However, the findings do not convincingly demonstrate the search for status enhancement as the Dutch government expressed to be focused on achieving sustainable peace

(34)

Page - 33 - of 38

and development in Afghanistan instead of clearly aiming at a higher status. Hence, the theoretical expectation on status enhancement cannot be verified convincingly.

Finally, the analysis of the domestic level factor government institutional structure and public opinion showed that the Dutch institutional structure of executives has medium power. According to the theoretical expectation on government institutional structure and public opinion, the middle power is expected to enjoy legislative and public support before officially deciding on contributing military troops to an alliance’s operation. However, the findings show

that the Dutch executives enjoyed major legislative support, with a rather large delay due to earlier political division, while the public opinion was not taken into consideration at all. Hence, the theoretical expectation can only be verified on the basis of legislative support.

Hence, the analysis of the three factors revealed that not all three theoretical expectations examined in this thesis provide a convincing explanation of the Dutch incentives and conditions to deploy Task Force Uruzgan. Whereas the factor alliance value explained the deployment of Task Force Uruzgan most convincingly, the findings demonstrate that the application of the three theoretical expectations on the case of Task Force Uruzgan each exposed various dimensions of the incentives and conditions for burden-sharing in a security alliance. Therefore, this study paved another part of the way towards a better understanding of middle powers’ burden-sharing behaviour in alliances by addressing both systemic and domestic level variables through a neoclassical realist framework. Moreover, the verification of the rather ideational grounded explanation of alliance value demonstrates that drifting away from the well-established theories on burden-sharing behaviour, can provide new and additional insights on non-dominant states’ burden-sharing behaviour in alliances.

The main limitations of this study are that, due to time and resource constraints, only one contributing middle power to ISAF is researched and the study is not triangulated by interviews to confirm the validity of the process. Therefore, further research is encouraged to apply a

(35)

Page - 34 - of 38

comparative analysis on various middle powers that contributed significantly to an alliance’s operation, allowing for triangulation by interviews with relevant stakeholders. Also, based on the fact that the Dutch government collapsed four years after the decision to deploy Task Force Uruzgan due to disagreement among the ruling coalition, further research is suggested to study changes in states’ burden-sharing behaviour within an alliance’s operation.

Overall, this study provides a good starting point for research into the incentives and conditions for other middle powers that contributed to NATO or another security alliance. Particularly, as a lot of research hitherto analysed the incentives and conditions of dominant or rather influential states’ burden-sharing behaviour in alliances, without focusing on ideational motivations.

(36)

Page - 35 - of 38

References

Advisory Council on International Affairs (2017). The future on NATO and European security. Advice no. 106. Retrieved from: https://aiv-advies.nl/download/6934b2f7-1331-4af1-9be0-07c300c28bbc.pdf

Advisory Council on International Affairs (2006). Society and the armed forces. Advice no. 48. Retrieved from:

https://aiv-advies.nl/download/d40171ea-0cd8-4db0-a65c-364ba35f218c.pdf

Auerswald, D. P. (2004). Explaining wars of choice: an integrated decision model of NATO policy in Kosovo. International Studies Quarterly, 48(3), 631-662.

Bennett, A., Lepgold, J., & Unger, D. (1994). Burden-sharing in the Persian Gulf War. International Organization, 48(1), 39-75.

Blair, S., & Fitz-Gerald, A. (2009). Stabilisation and Stability Operations: A Literature Review. Centre for Security Sector Management, Cranfield University, England. Bryman, A. (2016). Social research methods. Oxford university press.

Cimbala, S. J., & Forster, P. (2005). The US, NATO and military burden-sharing. Routledge. Cooper, A. F., Higgott, R. A., & Nossal, K. R. (1991). Bound to follow? Leadership and followership in the gulf conflict. Political Science Quarterly, 106(3), 391-410.

Dafoe, A., Renshon, J., & Huth, P. (2014). Reputation and status as motives for war. Annual

Review of Political Science, 17, 371-393.

Davidson, J. (2011). America's allies and war: Kosovo, Afghanistan, and Iraq. Springer. De Volkskrant (2006). Meer Nederlanders voor missie Afghanistan. Retrieved from:

https://www.volkskrant.nl/nieuws-achtergrond/meer-nederlanders-voor-missie-afghanistan~bf11f481/

Dimitriu, G. R., & de Graaf, B. A. (2009). De Nederlandse COIN-aanpak: drie jaar Uruzgan, 2006-2009. Militaire spectator, 178, 23.

Flyvbjerg, B. (2006). Five misunderstandings about case-study research. Qualitative

inquiry, 12(2), 219-245.

Gallis, P. and Morelli, V. (2008). NATO in Afghanistan: a Test of the Transatlantic Alliance. Congressional Research Service. Washington DC.

Gerring, J. (2008). Case selection for case‐study analysis: Qualitative and quantitative techniques. The Oxford handbook of political methodology.

George, A. L., & Bennett, A. (2005). Case studies and theory development in the social

sciences. mit Press.

Graaf, B. D. (2010). The Dutch mission in Uruzgan: Political and military lessons. Atlantisch

Perspectief, 7, 18-24.

Grandia, M. (2015). Deadly embrace: the decision paths to Uruzgan and Helmand. Faculty of Social and Behavioural Sciences, Leiden University.

(37)

Page - 36 - of 38

Hazelbag, L. J. (2009). Political decision making of the mission in Uruzgan, a

reconstruction. Complex operations: studies on Lebanon (2006) and Afghanistan

(2006-present), 2009, 251-276.

Haesebrouck, T. (2017). NATO burden sharing in Libya: a fuzzy set qualitative comparative analysis. Journal of conflict resolution, 61(10), 2235-2261.

House of Representatives (2005). Bestrijding internationaal terrorisme. Kamerstuk 27925, nr 177. Retrieved from: https://zoek.officielebekendmakingen.nl/kst-27925-177.html

House of Representatives (2005b). NAVO: Brief van de minister van Defensie. Kamerstuk 28676, nr. 22. Retrieved from: https://zoek.officielebekendmakingen.nl/kst-28676-22.html

House of Representatives (2005c). Bestrijding internationaal terrorisme. Art. 100 brief.

Kamerstuk 27925, nr. 193. Retrieved from:

https://zoek.officielebekendmakingen.nl/kst-27925-193.html

House of Representatives (2006). Bestrijding internationaal terrorisme. Kamerstuk 27925, nr.

197. Retrieved from: https://zoek.officielebekendmakingen.nl/kst-27925-197

House of Representatives (2006b). Debat over de Nederlandse deelname aan de Internationale Security Assistance Force (ISAF) in Zuid-Afghanistan. Kamerstuk 27925, nr. 45. Retrieved from: https://zoek.officielebekendmakingen.nl/h-tk-20052006-3013-3035.html

House of Representatives (2006c). Bestrijding internationaal terrorisme: lijst van vragen en antwoorden. Kamerstuk 27925, nr. 201. Retrieved from:

https://zoek.officielebekendmakingen.nl/kst-27925-201.html

Hudson, V. M. (2013). Foreign policy analysis: classic and contemporary theory. Rowman & Littlefield.

Jockel, J. T. (2014). Canada, Nederland en de oorlog in Zuid-Afghanistan. Militaire

Spectator, 183(1), 4-16.

Jockel, J. T., & Massie, J. (2017). In or out? Canada, the Netherlands, and support to the invasion of Iraq. Comparative Strategy, 36(2), 166-181.

Jordaan, E. (2003). The concept of a middle power in international relations: distinguishing between emerging and traditional middle powers. Politikon, 30(1), 165-181.

Kaarbo, J. (1996). Power and influence in foreign policy decision making: The role of junior coalition partners in German and Israeli foreign policy. International Studies

Quarterly, 40(4), 501-530.

Klep. (2011). Uruzgan. Nederlandse militairen op missie, 2005-2010. Uitgeverij Boom/SUN. Kreps, S. (2010). Elite Consensus as a Determinant of Alliance Cohesion: Why Public

Opinion Hardly Matters for NATO‐led Operations in Afghanistan. Foreign policy

analysis, 6(3), 191-215.

Kupchan, C. A. (1988). NATO and the Persian Gulf: Examining intra-alliance behavior. International Organization, 42(2), 317-346.

Laatikainen, K. V. (2006). Pushing soft power: middle power diplomacy at the UN. In The

Incentives and conditions for burden-sharing in NATO. Alliance value, status enhancement, government institutional structure, and public opinion: A case study on the Dutch motivations and conditions for the deployment of Task Force Uruzgan - Afghanistan 2