SOUTHEAST ASIA’S DIVERSE INTERNAL
BALANCING BEHAVIOR
A Neoclassical Realist Approach
Tom Verbart 5755859
Masterscriptie Politicologie Juni 2014
Docent: Dr. J Bader
Faculteit der Maatschappij- en Gedragswetenschappen Universiteit van Amsterdam
1
Abstract
The validity and applicability of neorealist balance of power theory as an approach to explaining Southeast Asian state behaviour in reaction to the multidimensional rise of China is contested. Scholars assume all states have to show balancing behaviour in order for the theory to be validated. However, neorealist balance of power theory does not expect all states in the international system to balance against a rising power. Domestic-level factors are expected to be influential in the ability of individual states to pursue balancing policies, of which internal balancing is perceived to be the most likely form of balancing. Neoclassical realism, a new body of realist approaches, is proposed to incorporate domestic-level variables in neorealist theory such as the balance of power theory. Because of this, a neoclassical realist approach to Waltz’s balance of power theory makes it possible to explain the variation in the internal balancing behaviour of nine Southeast Asian states in the 2001-2012 period.
2
Table of Content
Abstract………...1
1. Introduction……… 4
2. Existing approaches to balancing behavior in Southeast Asia………. 7
2.1 Limitations to existing approaches ………. 9
3. Neorealist balance of power theory……… 12
3.1 Understanding systemic balance of power theory: the independent variable………..15
3.2 The dependent variable: balancing as state behavior………...19
3.3 Limitations to neorealist balance of power theory………...22
4. Introducing domestic-level variables in neorealist theory………. 25
4.1 Neoclassical realism: a new realist approach………. 26
4.2 Limitations to neoclassical realist approaches……… 29
5. Neoclassical realist approach to balance of power theory………. 32
5.1 A comprehensive model………...34
5.1.1 Conceptualizing the dependent variable of internal balancing………... 35
5.1.2 Domestic-level intervening variables………. 38
5.1.2.1 State strength………. 39
5.1.2.2 Regime type………... 42
5.1.3 Conceptualizing the independent variable of relative power capabilities…………...43
5.2 Case selection………...45
5.3 Alternative explanations for internal balancing………... 45
6. Operationalization and data analysis methods………... 48
6.1 Dependent variable of internal balancing……… 48
6.2 Intervening variable of state strength………...50
6.3 Intervening variable of regime type……….52
7. Analysis: explain diverse internal balancing behavior………... 53
7.1 The impact of state strength on internal balancing………..54
7.2 The impact of regime type on internal balancing………55
7.3 General conclusions and limitations of analysis………..57
3
9. Appendix... 61
4
1. Introduction
The multidimensional rise of China has ignited a scholarly debate on the theoretical explanations for the behavior of Southeast Asian states in reaction to China’s rise. Various theoretical approaches such as liberalism and constructivism have been applied to explain Southeast Asian state behavior over the past two decades (Johnston, 2012; Goh, 2007; Roy; 2005; Acharya, 2004). Southeast Asia’s response to the changing regional power relations is however particularly often analyzed from a realist perspective with neorealist balance of power theory being the most commonly applied realist theoretical approach to Southeast Asian state behavior (Johnston, 2012; Kang, 2003; Roy; 2005; Ross; 2006; Acharya; 2004; Goh; 2007).
Despite the fact that many scholars have resorted to this approach, the validity and
applicability of neorealist balance of power theory as an explanatory approach to Southeast Asian state behavior is however, severely contested (Roy; 2005; Kang, 2003; Johnston; 2012). This is primarily due to the fact that many scholars expect all Southeast Asian states to show balancing behavior in order for the theory to be validated (Kang, 2003; Roy, 2005; Goh; 2007; Johnston; 2012). However Southeast Asian states appear to respond differently to the rise of China. Scholars, however often seem to misunderstand the premises of balance of power theory. As a systemic theory, the theory expects that the anarchic nature of the
international system forces states to resort to balancing strategies in order to secure their own survival (Waltz, 1979). Not all states however are expected to be equally able to pursue those balancing policies (Waltz, 1979; Hobson, 2000). Although not extensively discussed, the internal characteristics of states have, according to the theory, an influence on the ability of states to pursue the expected balancing policies (Waltz, 1979; Taliaferro; 2006; Ruggie; 1983).
Despite being a structural theory, balance of power theory is very often perceived to be a theory of state behavior (Levy, 2003) The aim of the most prominent balance of power theorist, namely Kenneth N. Waltz, is not to provide for a theory of foreign policy (Waltz, 1979). The theorist does however provide for specific expectations concerning the behavior of states in the international system in his balance of power theory (Waltz, 1979; Levy, 2003).
5
For example, states are expected to pursue a policy of internal balancing, which is the
improvement of military abilities vis-à-vis other states, in order to protect themselves against a rising power such as China (Waltz, 1979). It is for that reason that many scholars perceive neorealist balance of power theory to be a theory of state behavior (Levy, 2003).
The impact of domestic-level factors, such as the internal characteristics of states on state behavior, have for long time been ignored by realists in neorealist theories, such as the
balance of power theory (Sterling-Folker, 1997; Rose, 1998). A relatively new body of realist approaches, namely that of neoclassical realism, made the step to introduce domestic-level variables in neorealist theory such as the balance of power theory (Rose, 1998; Lobell, Ripsman & Taliaferro; 2009). However, most neoclassical realist works have primarily focused on the theoretical possibility of introducing domestic-level variables in neorealist theory (Lobell, Ripsman & Taliaferro, 2009; Rose, 1998). Only a very limited number of neoclassical realists have tried to introduce domestic-level variables in neorealist theory and an even smaller number have empirically tested whether domestic-level variables have an influence on state behavior (Lobell, Ripsman & Taliaferro, 2009; Taliaferro; 2006).
A neoclassical realist approach to neorealist balance of power theory could possibility explain the variation in the behavior of Southeast Asian states in reaction to the rise of China. Since internal balancing is the most likely form of balancing according to the balance of power theory, I will look at internal balancing as state behavior. The main question of this thesis therefore is if a neoclassical realist approach is able to explain the variation in the internal balancing behavior of 9 Southeast Asian states in the 2001-2012 period. This thesis will contribute to the scant body of neoclassical realist approaches that make the challenging step to introduce domestic-level variables in neoclassical realist balance of power theory and to empirically test whether these domestic-level variables are able to explain state behavior.
I will first provide for an overview of the existing realist approaches to balancing behavior in Southeast Asia and their limitations. In most approaches, balancing is not well conceptualized and balancing behavior is mostly not properly empirically tested, which has led to ad hoc interpretations (Vasquez, 2003). This thesis will provide for a clear conceptualization of internal balancing and will provide a comprehensive way of empirically testing internal balancing.
6
In order to grasp a proper understanding of balance of power theory, the theory needs to be explained in more detail. Balance of power theory is open to various interpretations and theorists often disagree on what the theory holds (Jervis, 1997: 313). A clear interpretation of the prominent balance of power theory, namely that of Kenneth N. Waltz and its most
important aspects will be presented (Lobell, Ripsman & Taliaferro: 2009:10). On top of that, a number of serious limitations to the balance of power theory will be discussed. Subsequently, the most important aspects of the neoclassical realist approach introduces domestic-level variables in neorealist balance of power theory and its limitations and caveats will be discussed.
A comprehensive theoretical model that incorporates two domestic-level variables primarily based on balance of power and neoclassical realist literature in balance of power theory will be presented. Subsequently, the most important variables of the theory will be more narrowly conceptualized and the two domestic-level variables of state strength and regime type will be discussed in more detail. After making the implications of the independent of relative power capabilities clear, the case selection method will be presented. In order to test whether the two domestic-level variables are able to explain the variation in the internal balancing behavior of Southeast Asian states, the domestic-level variables and also the dependent variable of
7
2. Existing approaches to balancing behavior in Southeast Asia
The multidimensional rise of the People’s Republic, which became increasingly apparent during the early 1990s, has triggered a scholarly debate on the behaviour of primarily East and Southeast Asian states in reaction to a more powerful China over the past two decades. Not only has China experienced high economic growth rates during the past twenty years, Beijing’s military expenditure has also risen exponentially with annual average growth rates of over ten percent ‘between 1996 and 2006’ (Peou, 2014: 124). The economic and military rise of China has shifted East Asian power relations and has challenged scholars to predict and explain Southeast Asian state behaviour in reaction to this change.
Various theories and theoretical approaches, such as different branches of realism, liberalism and constructivism have been applied to predict or explain East Asian state behaviour in reaction to the economic and military ascendance of China (Busse, 1999; Peou, 2002; Roy, 2005; He & Feng, 2010). The question of how East Asian states react to China’s rise is still however perceived as debatable (Roy, 2005). Some scholars even go as far to claim that the ‘prevailing lenses in international relations theory’ are inadequate to explain Southeast Asia’s seemingly varied approach to regional security matters (Goh, 2007: 114).
Neorealist theory, also often referred to as structural realism, is among the most commonly applied theoretical approaches to analyse contemporary political developments in the Southeast Asia region. Many of the first scholarly articles published in the early and mid-1990s that dealt with consequences of the multidimensional post-Cold War rise of the People’s Republic looked at the regional political developments from a realist perspective (Friedberg, 1993; Roy, 1996, Acharya, 2004: 149). Scholars like Aaron L. Friedberg predicted that the end of the bipolar international system, together with a rise in the
importance of regional subsystems, would lead to the destabilization of the security situation in the East Asian region in the then near future (Friedberg, 1993: 5; Acharya, 2004: 149; Roy, 1996; Goh, 2005).
Friedberg’s (1993) rather bold prediction that East Asia was ‘ripe for rivalry’ and would descend into a seriously destabilized security situation has (thus far) proven to be incorrect. However, China’s rise in general and its growing military strength in particular is still
8
early 2000s on, claim that especially Southeast Asian states have applied various strategies to deal with what Evelyn Goh calls ‘the most profound strategic worry for Southeast Asian states’ (Friedberg, 1993: 5; Goh, 2007: 809). During the last decade various media have also frequently and extensively reported on Southeast Asian state behaviour in response to the rise of China (Chang, 2003). Various developments in the Southeast Asian region are not only often connected to the growing economic and military strength of China, they are also very frequently analysed from a realist perspective (Chang, 2003).
Neorealist balance of power theory, one of the oldest and most influential realist theories, has particularly been applied numerous times by scholars to explain Southeast Asian state
behaviour in reaction to an ever-stronger China (Paul 2005; 51, Goh, 2007; Goh 2008; Roy, 2005; Ross, 2006). Scholars who have resorted to neorealist balance of power theory expect states in the Southeast Asia region to pursue specific balancing strategies such as forging alliances or to increase their military expenditure as a counterbalance to the rise of China (Roy 2005; Goh, 2007; Ross 2006; Acharya; 2004). Journalists often analyse regional political development in the region from a balancing perspective, thereby implicitly or
explicitly perceiving international balancing as a law of international relations (Chang, 2013).
Despite the fact that realism in general and neorealist balance of power theory in particular are often applied by scholars as an approach to analyse state behaviour in the Southeast Asia region, the approach is at the same time still contested. (Kang, 2003; Roy, 2005; Goh, 2007; Johnston, 2012). The validity and applicability of realist approaches such as Kenneth N. Waltz’s neorealist balance of power theory, particularly in an Asian context, is contested by numerous scholars (Kang, 2003; Johnston, 2012; Roy, 2005; Goh, 2007). The validity of the balance of power theory is mostly questioned due to the fact that either no Southeast Asian states appear to balance or that only a limited amount of countries seem to show behaviour which can be qualified as balancing (Kang, 2003; Johnston, 2012; Roy, 2005; Acharya 2004).
Scholars often perceive neoclassical realist power of balance theory to be limited in scope (Kang, 2003, Goh: 2007, Roy, 2005). This is due to the fact that many scholars assume the theory to predict that all Southeast Asian states have to show signs of balancing behaviour in reaction to the rise of China in order for the theory to be validated. Southeast Asian states appear to have reacted differently to China’s rise. This has led to the application of new, often weakly conceptualized, and therefore often confusing terms such as ‘low-intensity balancing’,
9
‘hedging’, ‘engagement’, and ‘soft balancing’ to explain, (the variation), in Southeast Asian state behaviour in reaction to the rise of China (Johnston, 2012; 60; Goh, 2007: 825; Kang, 2003; Paul, 2005; Pape, 2005). These terms are based on the principle realist theory, but seem to have been applied in ex ante fashion to explain the variation in Southeast Asian state behaviour (Johnston, 2012; Kuik, 2008).
2.1 Limitations of approaches to balancing behaviour in Southeast Asia
A number of serious shortcomings are apparent in approaches to balancing behaviour in the Southeast Asian state region. Firstly, many scholars incorrectly perceive neorealist balance of power theory as having direct predictive validity which has led to incorrect conclusions regarding the validity of neorealist balancing theory. Secondly, a lot of attention is paid to the question of whether Southeast Asian states balance or bandwagon. This mostly irrelevant approach to state behaviour has led to analyses that are profoundly lacking in depth. Thirdly, scholars are often not explicit on what balancing implies, which has resulted in ad hoc interpretations and biased or ungrounded conclusions. A number of other shortcomings deserve adequate attention as well. Most balance of power approaches favour a focus on external balancing strategies while relatively little attention is paid to internal balancing strategies (Ross, 2006; Roy 2005) and the lack of clarity on the implications of balancing subsequently also has resulted in ‘a lack of sound empirical research,’ and that a significant number of analyses are not up to date (Roy 2005: 310).
Balance of power theory is discarded or criticized by a number of scholars due to the fact that not all Southeast Asian states show signs of balancing behaviour, whatever those signs may be (Kang, 2003; Johnston, 2012; Roy, 2005). Rejecting balancing theory on these grounds implies an inadequate understanding of the neorealist balance of power theory. Kenneth N. Waltz’s balance of power theory, which is still perceived to be among the most dominant balancing theories, does not assume all states to balance against rising powers such as China (Waltz, 1979; Vasquez & Elman, 2003). Various behaviours of states are perceived to be ‘logically compatible’ with neorealist balancing theory (Vasquez, 2003: 40). Not all realist balance of power theories or approaches are similar when it comes to predicting state
behaviour, but most approaches do not predict states to balance at all times (Elman, 2003: 13-16, 40; Schweller,1999). An inadequate understanding of neorealist balancing theories, such
10
as Waltz’s influential balance of power theory, has frequently resulted in incorrect conclusions concerning the validity and applicability of balancing theory. No attempt has been made to explain the variation in the balancing behaviour of Southeast Asian states in reaction to China.
The ubiquitous scholarly focus on the question of whether Southeast Asian states balance or bandwagon in reaction to the rise of China deserves adequate attention (Goh, 2008: 114-115; Roy 2005; Acharya, 2004; Kang, 2003). Kenneth N. Waltz’s influential balance of power theory does not mention bandwagoning as an alternative to balancing (Waltz, 1979; Lobell, Ripsman & Taliaferro, 2009: 10). According to the neorealist theory, either states balance against a rising power or they do not (Vasquez & Elman, 2003). The theory does not explicitly provide for an opposing strategy of strategic cooperation with a rising power (Waltz, 1979). Stephen M. Walt’s balance of threat theory, which shares many characteristics with balance of power theory, on the other hand does mention bandwagoning as an alternative to balancing (Walt, 1987). However, Walt’s theory expects bandwagoning to seldom occur (Walt, 1987; Vasquez & Elman, 2003: 13-14). Other balance of power theories such as those of Christensen and Snyder also do not expect states to bandwagon (Elman, 2003: 14).
Therefore, the question of whether states balance or bandwagon is not particularly relevant.
The most profound downside to approaches that focus on the question of whether states balance or bandwagon is the fact they frequently lead to analyses that are lacking in depth. Balancing and also bandwagoning are not well conceptualized in scholarly works, which has resulted in imprecise analysis and ad hoc interpretations. If scholars do not find proof of what they perceive as balancing behaviour by states, they often automatically assume that states bandwagon without being very explicit on what the implications of bandwagoning are (Kang, 2003; Roy, 2005; Acharya, 2004).
There are multiple interpretations of bandwagoning ranging from economic
cooperation to alignment (Ross, 2006: 363; Acharya, 2004; Schweller, 1994; Ross, 2006). David Kang (2003) perceives the growth in bilateral trade and investment as a sign of
bandwagoning with the stronger side (Acharya 2004: 151-152). However, international trade and thus economic self-interest are highly compatible with balancing strategies because according to neorealist balance of power theory, economic growth improves the relative capacities of states vis-à-vis other states in the international system (Waltz, 2003: 57). Balancing and many interpretations of bandwagoning, particularly the interpretation of
11
‘bandwagoning for profit’ do not mutually exclude one another (Schweller, 1994; Ross, 2006; Waltz, 2003: 57).
A very serious limitation to scholarly works on balancing in Southeast Asia is the fact that balancing is not properly conceptualized (Kang, 2003; Acharya, 2004; Roy, 2005; Ross, 2006). This has resulted in a lack of ‘specific indicators’ which are required to judge the balancing behaviour of states (Acharya, 2004: 141). A lack of clarity on the implications of balancing is a general critique on neorealist balance of power theory (Vasquez & Elman, 2003). However this does not imply that scholars do not have to clarify their preferred interpretation of balancing (Levy, 2003: 128-130). In balance of power theory two different ways through which states can balance are mentioned (Waltz, 1979: 118). Either through internal means, mostly understood as allocating resources to improve military strength, or through external means, which implies alliance formation with other states in order the balance against a challenger (Waltz, 1979: 118). Neither the conceptualizations of external nor those of internal balancing have received adequate attention by scholars (Goh, 2008; Acharya, 2004; Roy, 2005; Kang, 2003; Johnston, 2012). In most of the literature on balancing in the Southeast Asia region, external balancing receives significantly more attention than internal balancing (Kang, 2003; Goh, 2008; Roy, 2005; Ross, 2006; Acharya, 2004). However external balancing behaviour is more difficult to empirically measure than internal balancing (Waltz, 2000: 5-6).
The absence of a clear conceptualization of both internal balancing and external balancing in scholarly works has resulted in a lack of proper empirical analyses. No scholars made the crucial step from conceptualizing external or internal balancing strategies properly to operationalizing them. John A. Vasquez (2003: 37) criticizes neorealist balancing theory on grounds of ‘imprecise measurement, leaving open the possibility of ad hoc interpretation’. Neorealist balance of power theorist Jack Levy (2003: 129-130) also stresses the importance of a proper conceptualization of balancing. Proper conceptualizations of the theory’s most important elements are a prerequisite for an acceptable balance of power approach (Levy, 2003: 130). External and internal balancing can be conceptualized in various ways, but if scholars conceptualize balancing well and are clear on their operationalization of balancing the possibility of ad hoc interpretations diminishes significantly.
12
3. Neorealist balance of power theory
The idea of international balancing has been around for a long time (Levy, 2004: 29). The core principles of international balancing behavior, which are perceived by many as being a law in international politics, are primarily based on European historical-political experiences, and can be traced back to as early as the sixteenth century (Elman, 2003: 1; Levy, 2004: 29; Kang 2003)1. Kenneth N. Waltz’s, neorealist balance of power theory in Theory of
International Politics, is, however, the first substantial theory of balancing within the
discipline of international relations (Lobell, Ripsman & Taliaferro, 2009; Waltz, 1979). Prior to Waltz’s theory, classical realists such as Hans Morgenthau, Henry Kissinger and Arnold Wolfers already focused on aspects of international state behavior that are perceived to be an integral part of balance of power theory, such as the distribution of power among states (Waltz, 2003: 49; Levy, 2004: 29; Lobell, Ripsman & Taliaferro, 2009: 4). It was Waltz, however, who for the first time established a substantial balancing theory that dealt with the structure of the international system, among which its constraints (Lobell, Ripsman & Taliaferro, 2009; 16; Hobson, 2000).
Waltz’s balance of power theory is at this point not the only neorealist balance of power theory. Theorists such as Stephen M. Walt, Thomas J. Christensen and Jack L. Snyder have also developed balance of power theories over the following years (Walt, 1987; Christensen, 1999; Snyder, 1997; Elman, 2003; 10-15; Levy, 2003: 130-131). To this date Waltz’s balance of power theory is, however, still perceived to be the most prominent and influential
neorealist balancing theory (Lobell, Ripsman & Taliaferro, 2009: 10). Other balance of power theorists have built their theories on the Waltz’s work, primarily with the aim of improving the original theory (Lobell, Ripsman & Taliaferro, 2009: 10; Vasquez & Elman, 2003: 299; Walt, 1987; Brooks, 1997: 457-458; Glaser, 1994; Grunberg & Risse-Kappen, 1992). I will therefore analyze balance of power theory by first looking at the original Waltzian neorealist theory. However Waltz’s theory is not sacrosanct. Certain aspects of the balancing theory are severely under-theorized. Taking other balance of power theories, which do not break with Waltz theory’s most important characteristics, but often do provide for a better understanding
1The European origins of balance of power theory has led scholars like David Kang to conclude that balance of power theories are Eurocentric and of no relevance to an Asian context (Kang, 2003; 58)
13
of important aspects of the theory, is therefore a necessity. Multiple balance of power theorists have borrowed from the insights of latter balance of power theories to improve Waltz’s original balancing theory (Taliaferro, 2004; 2006; Lobell, Ripsman & Taliaferro, 2009; Brooks, 1997: 457; Walt, 1987).
However an important caveat should be mentioned. Balance of power theory is perceived to be one of the best and most prevalent theories within the discipline of international relations. However an analysis of any balance of power theory is not without its problems (Jervis, 1997: 131). Despite the prevalence and popularity of balance of power theory among scholars, theorists disagree on what the theory is exactly comprised of (Jervis, 1997: 131; Elman & Vasquez, 2003). Confusion remains over questions of what balancing is and what the theory precisely predicts (Wohlworth, 2003: 250-251; Jervis, 1997: 131). It is of grave importance to present a clear interpretation of the theory and its implications in order to avoid unnecessary confusion.
In the first place I will focus on the most prominent underlying principles of the neorealist theory that are shared by all balance of power theories. In order to grasp a proper
understanding of Waltz’s balance of power theory and avoid confusion, the theory’s most important elements and implications should be made absolutely clear. Balance of power theory is a systemic theory this implies that ‘the theory’s primary goal is to explain a ‘high degree of continuity in international politics’(Taliaferro, 2009: 206; Hobson, 2000: 19-20). Waltzian neorealist balance of power theory expects states to automatically balance against other states in the international system. However it does (paradoxically) not expect all states to do so (Vasquez, 2003: 43; Elman, 2003: 9-11; Ruggie: 266). The dependent variable of balance of power theory, namely that of relative power capabilities, is a product of the international environment. ‘The international environment in which states interact is the primary determinant of their interests and behavior’ (Taliaferro, 2009: 21; Hobson, 2000: 20-21; Sterling-Folker, 1997: 16). Balance of power theory is therefore also referred to as an ‘environmental-based theory’ (Sterling-Folker, 1997: 16). John Gerard Ruggie has
distinguished three basic features of the international political structure or environment in Waltz’s theory, which are instrumental to a better understanding of the theory (Hobson, 2000: 20, Ruggie, 1983). The first feature is the ordering principle, which Ruggie refers to as the ‘deep structure’ (Ruggie, 1983: 266). The second feature is the character of the units or states, and the third feature the scholar distinguishes is the distribution of ‘material’ capabilities,
14
which he refers to as the ‘surface structure’ (Ruggie, 1983: 135-136; Hobson, 2000: 20; Taliaferro, 2009: 208).
The second aspect of the balance of power theory that deserves adequate attention is that of the dependent variable of adaptive strategies. States resort to adaptive strategies in the international system, amongst which balancing is the most prominent (Waltz, 1979). Waltz himself is not concerned with providing a theory of foreign policy (Waltz, 1979: 121-122; Taliaferro, 2009: 206; Christensen & Snyder, 2003: 67-68). However his is neorealist theory paradoxically does provide for the expectation of individual states’ foreign policy outcomes (Elman, 2003: 8; Dueck, 2009: 140-141; Hobson, 2000: 19). Therefore ‘Most balance of power theorists predict strategies of balancing’ (Levy, 2003: 133). Either states resort to external or internal means to strengthen their position vis-à-vis other states in the international system, of which internal balancing is the most likely (Waltz, 1979: 118; Paul, 2005: 51). Another aspect that needs to be properly addressed is the issue of polarity. According to Waltz and other balance of power theorists the polarity of the international system has
consequences for the likeability of countries pursuing certain balancing policies (Waltz, 2000: 5-6; Vasquez, 2003: 36; Waltz, 1979). The issue has been raised by numerous balance of power theorists and therefore deserves proper attention.
Waltzian neorealist balance of power theory has its limitations. One of the most salient challenges to the applicability and validity of the theory is the fact that certain elements of the theory are seriously under-theorized. The independent variable of the distribution of relative power capabilities is, despite being an essential part of the theory, poorly conceptualized and therefore not entirely clear. The concept of balancing in Waltz’s theory is also seriously limited. A proper conceptualization of the independent and dependent variables is however a prerequisite for empirically testing the theory and will significantly diminish the risk of ad hoc interpretations. Other neorealist balance of power theories and approaches that share the most important characteristics of Waltz’s balance of power theory provide for a more detailed understanding of the independent variable of relative power capabilities and of the dependent variable of internal balancing. In this thesis I will therefore resort to their insights for a more specific conceptualization of the independent and dependent variables. However the most profound limitation to Waltz’s neorealist theory is the lack of a proper theory of the state (Lobell, Ripsman & Taliaferro, 2009: 13; Wohlforth, 2003: 252).
15
The theorist does mention a limited number of domestic-level factors that are thought to be of influence on the balancing behavior of states. However the theory in general leaves out domestic-level factors. Since I will be analyzing possible explanations for the variation in the internal balancing strategies of Southeast Asian states, these domestic-level variables are of great importance.
3.1 Understanding systemic balance of power theory: the independent variable
Waltz’s neorealist balance of power theory is a systemic theory of international politics (Harrison, 2002; Hobson 2000: 19-20). ‘A system is composed of a structure and of
interacting units’ (Waltz, 1979: 79). The structure refers to the international system, while the units are the states that interact within the international system. The primary aim of the theory is to explain repetitive patterns in international politics, namely the continuity of the
recurrence of the international political outcome of balancing and the absence of a single hegemonic power within the system (Lobell, Ripsman & Taliaferro; 2009: 17; Hobson 2000: 19). The fact that Waltzian balance of power theory is a systemic theory of international politics has important implications for the scope and depth of the theory. Because it is Waltz’s primary goal to explain ‘continuity’ in the international system, the complex nature of many elements of the theory are simplified (Hobson, 2000: 19-20). The independent variable of the theory is the distribution of power capabilities among states and the dependent variable is the adaptive strategies that are applied to secure state survival, among which balancing strategies receive most scholarly attention (see figure 1).
In order to come to a proper understanding of Waltz’s balancing theory it is helpful to take a closer look at the traditional realist assumptions that lie at the very core of Waltz’s neorealist balance of power theory. Lobell, Ripsman and Taliaferro (2009: 14) distinguish three primary principles of realism that underpin balancing theory. Firstly, all realist theories are ‘group-centric’ (Lobell, Ripsman and Taliaferro, 2009: 14). A basic assumption of (neo)realist theory is that states are the primary actors in the international system (Levy, 2003: 130). The primary actors in Waltzian balance of power theory are indeed states (Waltz, 1979: 118) Waltz
attributes functional similarity to states in his ‘narrow’ systemic theory, thereby implicitly discarding domestic level differences among states, which he coins as ‘units’. (Waltz, 1979; Hobson, 2000: 20-22). This is why states in neorealist theory are often referred to as billiard
16
balls or black boxes (Hobson, 2006: 520). Another basic realist assumption about states as the primary actors in the international system is that they are expected to behave in a rational manner (Levy, 2003: 130). The wish of states to survive and balancing strategies are the rational underpinnings of state behavior (Waltz, 1979: 117-118). Secondly, ‘politics is a perpetual struggle among self-interested groups under conditions of general scarcity and uncertainty’ (Lobell, Ripsman & Taliaferro: 2009: 14). At the core of self-interest lies the struggle of states to survive (Levy, 2003: 130; Waltz, 1979, 118; Hobson 2000: 25). States ‘are unitary-actors who at a minimum, seek their own preservation and, at a maximum, drive for universal domination’ (Waltz, 1979: 118). Balancing is an important strategy to secure state survival. The anarchic nature of the international system causes struggle and uncertainty and forces states to help themselves in order not to perish. Waltz states that ‘self-help is necessarily the principle of action in an anarchic order’ (Waltz 1979: 111). Thirdly,‘power is a necessary requirement for any group to secure its goals, whether those goals are universal domination or simply self-preservation’ (Lobell, Ripsman & Taliaferro; 2009: 14). The (unequal) distribution of power among states drives states to adapt strategies to secure their own state survival (Waltz, 2003: 49).
The fact that Waltz’s neorealist balance of power theory simplified the complex nature of some of its most important theoretical components does not imply that the theory is generally understood without ambiguity. (Hobson, 2000: 19-22; Jervis 1997: 131). In order to secure a proper understanding of the theory, an elementary model of Waltz’s neorealist balance of power theory based on the work of Ruggie (1983) and on previous models created by Hobson (2000: 25) and Taliaferro (2009: 208) is shown in figure 1. In order to enhance a full
understanding of the systemic theory, it is important to stress again that balance of power theory is an ‘environmental-based theory’ (Sterling-Folker, 1997: 16) ‘The international environment in which states interact is the primary determinant of their interests and behavior’ (Taliaferro, 2009: 21; Hobson, 2000: 20-21; Sterling-Folker, 1997). The
independent variable of the distribution of material capabilities is therefore perceived to be the international political structure itself (Hobson, 2000: 20-21).
17
Figure 1. Elementary model of Waltz’s balance of power theory. Based on models presented by John M. Hobson (2000: 25) and Jeffrey W. Taliaferro (2009: 208).
Ruggie, has distinguished three features or tiers of the international structure that are a great help to grasp a better understanding of Waltz’s balance of power theory (Ruggie, 1983: 264-267; Hobson 2000: 20-25; Taliaferro, 2009: 208): firstly, an ordering principle referred to as ‘deep structure,’ secondly the ‘character of states’ and lastly, the ‘distribution of material capabilities also referred as the ‘surface structure’ (Ruggie, 1983: 264-267; Hobson 2000: 20-25; Taliaferro, 2009: 208).
The first feature of the international political system is that of the ordering principle, also referred to as the ‘deep structure’ (Ruggie, 1983: 266). The term ‘deep structure’ is applied, because the ordering principle is not empirically visible, but has an influence on the other two features of the international political system (Ruggie, 1983: 266). The rational realist
assumption of state survival lies at the very core of the systemic theory. Rational states strive
Systemic definition of the international political
structure International realm of necessity: State survival ‘Deep structure’ Ordering principle (Anarchy) States have to resort to
self-help ‘Character of states’ Functional similarity ‘Surface structure’ The distribution of material capabilities (Independent variable) High state power (ambiguous assumption) Adaptive strategies to secure state survival
A) Balancing (internal or external) B) Emulation (Dependent variable) Reproduction of anarchy
18
for survival in the international system (Waltz, 1979: 118; Hobson, 2000;25) If states do not strive for their own survival, they will perish (Waltz, 1979: 118; Taliaferro, 2009: 208). This is why international realm is often characterized as a ‘realm of necessity’ (Hobson 2000: 25; Taliaferro 2009: 208). Waltz claims that ‘balance-of-power politics prevails wherever two, and only two, requirements are met: that the order be anarchic and that it be populated by units wishing to survive’ (Waltz, 1979: 121). Thus the ordering principle of the international system is that of anarchy (Waltz, 1979). The international system is anarchic due to the fact that there is no global institution that exercises authority over all states (Ruggie, 1983: 265; Glaser, 1994: 50). ‘Anarchy is seen as one end of a continuum whose other end is marked by the presence of a legitimate and competent government’ (Waltz, 1979: 114). States are ordered according to the principal of hierarchy (Waltz, 1979; Hobson, 2000: 20-21). Due to the lack of an institution of world authority, states are ‘the constitutive units of the system’ (Ruggie, 1983: 256). The struggle for state survival in an anarchic international system forces states to resort to self-help. (Waltz, 1979: 111, 118). ‘A self-help system is one in which those who do not help themselves, or who do so less effectively than others, will fail to prosper, will lay themselves open to dangers, will suffer’ (Waltz, 1979: 118). States are continuously forced to improve their relative position vis-à-vis other states within the international system (Taliaferro, 2009:). The competition of rational, selfish, survival seeking states guarantees the reproduction of an anarchic internationals system (Hobson, 2000: 22).
The second feature Ruggie distinguishes in Waltz’s balance of power theory is that of the character or differentiation of states (Ruggie, 1983: 265-266; Hobson, 2000: 22-23). Waltz perceived states to be highly similar when it comes to their functions and tasks (Waltz, 1979: 95-97). Modern states all pursue the same tasks and there is a uniform international trend in their activities (Waltz, 1979: 97). It is for this reason states can be perceived as units.
However, at the same time Waltz acknowledges that states can vary in many respects (Waltz, 1979: 96). States all perform the same tasks, but they are not alike in the ‘in their abilities to perform them’ (Waltz, 1979: 96). However the impacts of the characteristics of states are left out of the structural theory.
The third feature or tier of the international political system is that of the relative distribution of power (Ruggie, 1983: 265-266; Hobson, 2000: 20). The relative distribution of power is also the independent variable of the balance of power theory and builds on the two previously mentioned features of the international political system (Taliaferro, 2009: 208; Hobson, 2000:
19
23-25; Ruggie, 1983: 266 ). Without a proper understanding of the ‘deep structure’ and the ‘character of states,’ the independent variable of relative distribution of power cannot be properly understood (Ruggie, 1983; 266-267). Ruggie refers to the third feature as a feature that comes closest to ‘surface level phenomena’ (Ruggie, 1983: 266). The relative distribution of power is renamed after the example of Taliaferro as the distribution of material capabilities in figure 1 (Taliaferro, 2009: 208). Material capabilities can be measured empirically and the independent variable is therefore referred to as the ‘surface structure.’ Material capabilities reflect the power of states in the international system according Waltz (Waltz, 1979; 131; Wohlforth, 2003: 251-252). Selfish, survival seeking states in the anarchic international system fear that other states will use their material capabilities against them (Taliaferro, 2009: 206). States therefore continuously try to improve their relative material position vis-à-vis other states in the international system (Taliaferro, 2009: 206). It is the unequal distribution of material capabilities of states that drives states to pursue adaptive strategies such as balancing. According to Waltz, it is not the intentions of states but their material capabilities to which states react (Taliaferro, 2009: 2006). The distribution of power amongst states is a ‘system-wide concept’ because it deals with the material power relations between states and not within states. (Waltz, 1979: 97-98; Hobson, 2000: 24). Waltz does provide for some insight into the material capabilities that matter, such as resource endowment and economic capability (Waltz, 1979: 131). The independent variable distribution of power or material capabilities, however, remains under-theorized (Wohlforth, 2003: 251-252; Taliaferro, 2004, 2009).
3.2 The dependent variable: balancing as state behavior
The three previously mentioned features or tiers of the international political system all dealt with the independent variable of the relative distribution of power. The dependent variable of the theory deals with the way states react to the relative material capabilities of other states in the international system. Thus, states try to ensure their own survival at the minimum and drive for universal domination at the maximum (Waltz, 1979: 118). Firstly, balance of power theory expects states to emulate the successful practices of the most powerful states within the international system (Lobell, Ripsman & Taliaferro: 2009: 38). Secondly, the theory expects states to balance against more powerful states by internal or external means (Waltz, 1979: 118: Paul, 2005: 51). As mentioned earlier, balancing is mostly perceived as state behavior (Levy, 2003: 133; Ross, 2006: 360; Elman, 2003: 8-9). However, some theorists claim that
20
balance of power theory does not provide for expectation of individual state behavior (Ross: 2006: 360; Elman, 2003: 8-9). Therefore, this discussion deserves adequate attention. Another relevant point of discussion is the likeability of states to pursue certain balancing strategies. Since I am looking at the variation in the internal balancing behavior of Southeast Asian states in this thesis, it is important to address the likability of internal balancing as opposed to external balancing strategies. Firstly, the realist assumption of self-help in an anarchic international system influences the preferred balancing strategies of states (Waltz, 1979; Secondly, Waltz and others assume that the polarity of the international system is also of influence on the likability of certain balancing strategies.
The dependent variable of balance of power theory is that of adaptive strategies to secure the survival of states (Hobson, 2000 Taliaferro, 2009: Ruggie, 1983). Waltz expects states to emulate and to balance against other states in the international system. Emulation refers to the ambition to equal the military, governing and technological practices of the most powerful or successful states in the international system (Taliaferro, 2006: 467). In other words states adapt a strategy of innovation to keep up with the material capabilities of the international system’s more powerful players. The adaptive strategy of emulation receives relatively scant attention by balance of power theorists. Emulation is an internal strategy, but it’s mostly not perceived to be a form of balancing (Hobson, 2000: 25, Taliaferro, 2009: 208). The other adaptive strategy to secure state survival is that of balancing. Waltz’s makes a distinction between external and internal balancing strategies (Waltz, 1979: 118). External balancing strategies are perceived by Waltz as the ‘move to strengthen and enlarge one’s own alliance or to weaken and shrink an opposing one’ (Waltz, 1979: 118). Internal balancing moves to increase a state’s ‘economic capability,’ to increase the ‘military strength’ or ‘to develop clever strategies’ (Waltz, 1979: 118). Waltz does not provide for more detailed information on the balancing strategies in the neorealist balancing theory. More detailed conceptualizations of the two balancing strategies are provided by other balance of power theorists.
Waltz himself has expressed that it is not the aim of the balance of power theory to explain the individual foreign policies of states (Waltz, 1979: 117-119, Taliaferro, 2009: 210). Balance of power theory does however expect states to balance in various ways against more powerful states in the international system. Most balance of power theorists therefore
logically expect individual states to pursue strategies of balancing (Levy, 2003: 113; Ross, 2006: 360). Not all theorists however perceive balancing as a state behavior (Elman, 2003: 8).
21
A distinction has been made between those who perceive balancing as state behavior and those we perceive ‘balance’ purely ‘as an international political outcome’ (Elman, 2003: 8). Those who opt for the latter approach stick to the theory by the letter. This interpretation of the theory is not incorrect; Waltz’s theory is indeed narrow and does not seem overly
concerned with providing for a theory of foreign policy (Waltz, 1979: 121-122, Christensen & Snyder, 2003: 67-68). On the other hand, the theory does provide for certain expectations concerning the foreign policy outcomes of states (Elman, 2003: 8, Dueck, 2009: 140-141). Waltz’s balance of power works also sporadically unveil information on circumstances that are thought to be influential on the balancing behavior of states (Waltz, 1979: 1996; 2000). The self-help assumption, the polarity of the international system and also state-level factors are expected to be influential on the balancing behavior of individual states (Waltz, 1979: 96, Waltz, 2000: ). This at first glance seems like a paradox. The narrow theory itself is not concerned with domestic-level attributes of states and their influence on state behavior (Waltz, 1979). States are presumed to be similar when it comes to their functions and states are implicitly expected to be equally able to extract resources in order to pursue emulation and balancing strategies (Waltz, 1979, Hobson, 2000). Waltz’ aim is to explain repetitive patterns in international politics, namely the continuity of the international political outcome of balancing (Lobell, Ripsman & Taliaferro: 2009: 17) . Balance of power theory however does expect states to balance by various means against more powerful states in the international system. The theory even provides for insights into the circumstances and attributes of states that might be influential in the balancing behavior of states. Therefore I concur with many other balance of power theorists that balancing can be perceived as state behavior and thus as a foreign policy (Taliaferro, 2004; Levy, 2003).
Despite the fact that the descriptions of the balancing strategies in the original neorealist balancing theory are rather scant, Waltz does provide for some information on the likability of strategies of internal or external balancing. Certain types of balancing seem more preferable or logical than others. In the first place, the self-help assumption is influential. The anarchic character of the international system forces states to rely on self-help measures in order to protect themselves against stronger states (Waltz, 1979; Waltz, 2000) Thus states simply have to look out for themselves. This entails that external balancing strategies of alliance formation with other states are less attractive than internal balancing strategies (Waltz, 1979: 105-107, Glaser, 1994: 56). The anarchic nature of the international system causes fierce competition amongst states. Realists claim that international cooperation is expensive because there is a
22
high chance states will be cheated (Glaser, 1994: 56). Waltz states in Theory of International Politics that internal balancing is superior to external balancing due to the fact that is more reliable and precise (Waltz, 1979: 168). Thus the theorist implies that it is more beneficial and therefore more likely that states will pursue a strategy of internal balancing (Vasquez, 2003: 101).
The polarity of the international system is also expected to have implications for the likability of internal and external balancing strategies of states (Waltz, 2000: 5-6). The issue is raised because numerous balance of power theorists have dealt with the influence of polarity on balancing (Vasquez, 2003, Bueno de Mesquita, 2003). Polarity affects the structure of the international system and a change in polarity is therefore expected to affect the ways in which states balance (Waltz, 2000: 5). Multipolar systems are more complex and unstable according to Waltz (Waltz, 2000: 6). According to the author, this is due to the fact that it is not always clear to states in a multipolar world order which states have relatively better material
capabilities (Waltz, 2000: 6). Waltz expects internal balancing to occur more frequently in a bipolar world order (Waltz, 1979, 168, Vasquez, 2003: 36). However Waltz still presumes that under multipolarity states will still opt for internal balancing strategies (Waltz, 2000: 5-6). Because states are expected to balance against regional powers according to many scholars who defined the independent variable of relative power capabilities more narrowly, Waltz’s reasoning loses significant leverage (Ross, 2006: 361, Taliaferro, 2004: 486). This entails that multipolarity is not expected to be influential on the internal balancing behavior of states. In conclusion, internal balancing behavior primarily due to the self-help principle of the
neorealist theory is more likely to occur than external balancing behavior in a multipolar world order.
3.3. Limitations to neorealist balance of power theory
The original Waltzian neorealist balance of power theory has a number of serious limitations. Firstly, important elements of the theory are profoundly under-theorized. This has resulted in poor conceptualizations of important elements of the theory. The theory’s two most important elements, namely the independent variable of the relative distribution of power and the
dependent variable of balancing, have both been insufficiently conceptualized. Insights of other balance of power approaches with regard to the independent and dependent variables are
23
therefore crucial in order to make the leap to operationalizing them and to be able to test the balancing theory empirically. Secondly, and most importantly neorealist balance of power theory has no theory of the state (Lobell, Ripsman & Taliaferro, 2009: 13, Wohlforth, 2003: 252). Although Waltz does mention possible factors of influence on the balancing behavior of states, the attributes of states are generally left out of his theory (Waltz, 1979). Thirdly and lastly,
The independent variable of the relative distribution of power and the dependent variable of balancing are the most important elements of the balancing theory that are poorly
conceptualized. Some insights are given about the basic characteristics of the variables, but Waltz abstains from giving detailed descriptions. The downside of poor conceptualizations is that they cause serious confusion. Waltz’s concepts are often so broadly defined that multiple interpretations are possible. Consequently this leads to the problem of ad hoc interpretations which are of little value if we want to explain international phenomena such as the balancing behavior of states. As mentioned earlier, most scholarly approaches to balancing behavior in Southeast Asia have fallen into the trap of ad hoc interpretations because the dependent variable of balancing was to broadly conceptualized (Goh, 2007; Acharya; 2004). In this thesis I am looking at balancing as state behavior. It is therefore of grave importance that the independent and dependent variables are properly and thus more narrowly conceptualized. Properly conceptualized variables are a prerequisite for an empirical testing the balancing behavior of states. Because Waltz’s conceptualizations of the independent and the dependent variables are unsatisfactory, I will resort to the works of other balance of power theorists. Many theorists have significantly improved the conceptualizations of the variables. As
mentioned earlier, most balance of power theorists have established balance of power theories on the basis of the Waltz’s balance of power theory with the primary aim of improving it. Other scholars have also improved the conceptualization of the dependent and independent variables or resorted to improved conceptualizations provided by balance of power theorists. Making use of improved conceptualizations of the independent and dependent variable does not violate the fundamental implications of the theory.
The most serious limitation of the neorealist balance of power theory is that there is no theory of the state (Hobson, 2000; Ruggie, 1983). Though Waltz acknowledges that domestic-level factors are influential on state behavior (Waltz, 1979: 71, 96-98, 121-123, Sterling-Folker, 1997: 3). The theorists express more than a few assumptions concerning domestic-level
24
factors in Theory of International Politics that could be influential on the balancing behavior of states (Waltz, 1979: 96-98). States are thought to have different capabilities when it comes to performing state tasks (Waltz, 1979: 96-97). The size, wealth and power of states are expected to be influential on their ability to perform their tasks (Waltz, 1979: 96-97). In another section of his book, Waltz states that the internal structures of the states affect state policies (Levy, 2003: 132-133, Waltz, 1979: 121-123). Thus Waltz’s domestic-level assumptions are highly ambiguous. Balance of power theory is a narrow systemic theory; many complex elements are therefore simplified (Hobson, 2000: 19-20). The attributes of states are left out of the theory, because the theory focusses on a general trend of balancing in the international system over time (Glaser, 2003: 267). One of the most salient examples is the fact that Waltz assumes all states to have a similar ability to extract resources in order to pursue balancing strategies. This assumption is mostly referred to as ‘’high state power’ (see figure 1)(Hobson, 2000: Taliaferro, 2009: ) Paradoxically, Waltz then also implicitly claims that high state power is an ideal situation. The author namely points to the fact that not all states have the same ability to extract the resources that are necessary to pursue balancing policies due to their capabilities and internal structure. In short, Waltz expects states to balance against more powerful states in the international system, but not all states have the same ability to do so. The aim of this thesis is to look at those domestic-level factors that could explain the variation in the internal balancing behavior of (Southeast Asian) states. Therefore an approach that makes it possible to look at domestic-level factors of influence is needed.
25
4. Introducing domestic-level variables in neorealist theory
Waltz’s systemic neorealist balance of power theory expects states to balance against more powerful states in the international system by various means, among which the strategy of internal balancing is the most likely. The theorist however also acknowledges that not all states are equally able to pursue balancing strategies. Waltz makes clear that domestic-level factors, such as the internal structure of states and their capabilities to perform state tasks are influential to the balancing abilities of states (Waltz, 1979: 96-97). In order to explain the variation in the internal balancing behavior of Southeast Asian states an approach that makes it possible to look at the most likely domestic-level factors of influence on the balancing behavior of states is needed.
During the 1990s international relations theorists such as Jennifer Sterling-Folker (1997) acknowledged the need for approaches that incorporate domestic-level variables in systemic neorealist theory because a realist systemic theory such as Waltz’s balance of power theory is perceived to be especially accommodating of domestic level variables (Sterling-Folker, 1997). Neorealist theory is perceived to be accommodating of domestic-level variables due to the fact that domestic-level variables are seen to be consistently causal in systemic realist theory (Sterling-Folker, 1997: 3-4, Lobell, Ripsman & Taliaferro, 2009: 19, Taliaferro, 2006, Rathbun, 2008, Rose, 1998). Since domestic-level variables are consistently causal in systemic realist theory, they are even perceived to be superior to liberalism and constructivism (Sterling-Folker, 1997: 3, Rathbun, 2008: 296-297).
A relatively small group of scholars have made the jump from discussing the mere possibility of incorporating domestic-level variables into neorealist theory to approaches that look at which domestic-level variables could be influential on state behavior and how to incorporate those variables into existing realist theory. The realist approaches and theories that appeared during the 1990s in which domestic-level variables had been taken into account were for the first time coined by Gideon Rose (1998) as ‘neoclassical realist’. Neoclassical realism deviates from neorealist approaches in two profound ways: firstly, all neoclassical realist approaches deal with realism as a foreign policy (Rose, 1998). Secondly, neoclassical realist approaches focuses specifically on the influence of domestic-level factors on those foreign
26
policy outcomes. Neoclassical realist theorists however acknowledge the premises of neorealism (Rose, 1998: 146). They merely introduce intervening domestic-level variables into neorealist theory (Rathbun, 2008: 294-298). A great number of potential domestic-level factors of influence on state behavior have been mentioned by neoclassical realist scholars, despite the body of neoclassical realist approaches being rather scant. However many domestic-level variables correspond to the underlying logic of states’ abilities to extract resources in order to pursue a particular foreign policy.
Neoclassical realist approaches are not without their limitations and caveats. In the first place, it is of importance to analyse how bringing in domestic-level variables impacts the logics of neorealist theories, such as Waltz’s balance of power theory. Waltz acknowledges that domestic-level factors are of importance, but certain domestic-level variables seem to be in violation with the premises of Waltz’s balance of power theory. It is therefore of utmost importance to address this limitation. Secondly, a big critique on neoclassical realism is that domestic-level variables are brought in on an ad hoc basis. It is therefore important to be very clear on the causal expectations of domestic-level variables in order to avoid the trap of randomly bringing in variables into neorealist theory. Thirdly, hardly any neoclassical realists have empirically tested expected domestic-level influential variables on the foreign policies of states. This has led to approaches which mention domestic-level variables with an expected influence on policies of states that are empirically very difficult to test.
4.1 Neoclassical realism: a new realist approach
Realist scholars have for a long time refused to focus on the role of domestic-level factors in neorealist theory. This occurred despite the fact that most neorealist theorists, including Waltz himself, recognized the importance of domestic-level factors on state behavior (Waltz, 1979; Rathbun, 2008: 296; Sterling-Folker: 3) However, looking at the role of domestic-level factors on realist theory was perceived, and is still perceived by some, as violating the premises of neorealist theory (Sterling-Folker: 3). It is mainly because Waltz ignores the attributes of individual states in his narrow structural theory that these scholars perceive it as incorrect to take domestic-level factor of influence into account (Sterling-Folker, 1997: 2-3; Lebow, 1994: 268). However, more than a decade after Waltz’s original balance of power
27
theory was presented in Theory of International Politics scholars begin to focus on the possibility of incorporating domestic-level variables into neorealist theory.
The idea that the international domain and the domestic domain are both influential on state behavior has been around for some time, but a theoretical sound approach to combining systemic and domestic-level variables only appeared during the second half of the 1990s (Sterling-Folker, 1997: 2-3, Snyder, 1991: 319, Rise-Kappen, 1994). Jennifer Sterling-Folker was among the first scholars to seriously examine a theoretical basis for combining systemic and domestic-level variables and came to the conclusion that domestic-level variables are consistently causal in systemic neorealist theory (Sterling-Folker, 1997: 3-4; Rose, 1998). Many other realist scholars currently agree with this interpretation (Lobell, Ripsman & Taliaferro, 2009: 19, Taliaferro, 2006, Rathbun, 2008). The ‘systemic environment’ and domestic-factors both have a causal impact on the dependent variable (Sterling-Folker, 1997: 3-4). In the case of the balance of power theory this means that the independent variable of relative power capabilitiesand domestic-factors have a causal impact on the theory’s dependent variable (Taliaferro, 2006: 485-486, Rose, 1998: 154). The domestic-level variables function as a ‘transmission belt’ between the independent and dependent variables of neorealist theory (Rose, 1998: 146-147).
Figure 2. Causal logic of a neoclassical realist approach (to neorealist theory) based on a model presented by Gideon rose (1998: 154).
Realist systemic theory is perceived to be particularly accommodating of domestic-level variables (Sterling-Folker, 1997: 2-4, 16-22; Rathbun, 2008: 294, Rose, 1998: 148-150). Constructivism and especially liberalism are thought by many scholars to provide the best theoretical grounds for the inclusion of domestic variables. However, realist scholars such as Jennifer Sterling-Folker (1997: 3-5) and Brian Rathbun (2008: 296) claim that neorealist theory is far superior when it comes to the inclusion of domestic-level variables. The ‘primary determinant for actors’ interests and behavior’ in realist theory is the environment (Sterling-Folker, 1997: 3). The (international) environment forces states to pursue certain strategies
Systemic incentive (Independent variable) Internal factors (Intervening variables) Foreign policy (Dependent variable)
28
such as internal balancing in the case of the balance of power theory to secure their own survival. Domestic-level variables have an impact on how efficient states react to the imperatives of the environment (Taliaferro, 2006: 467, Sterling-Folker, 1997: 19-20). Liberalism on the other hand is a ‘process-based theory’ (Sterling-Folker, 1997: 3-4). It therefore presents domestic-processes as independent variables but also as dependent variables. This is perceived to violate the premise that states‘ interests derive from process’ (Sterling-Folker, 1997: 4).
A small number of realist scholars not only concurred with domestic-level variables being consistently causal in systemic neorealist theory, such as Waltz’s balance of power theory, they also made the leap to apply the approach to real-life cases (Lobell, Ripsman &
Taliaferro, 2009; Schweller, 1999; Taliaferro, 2006; Zakaria, 1998). Rose (1998), referred to neorealist approaches that take domestic-level variables into account as ‘neoclassical realist’. By doing so a new and distinct body of realist approaches emerged. The term ‘classical’ refers to classical realism which had a specific domestic focus (Rose, 1998; Lobell, Ripsman & Taliaferro, 2009: 13-15). Neoclassical realism builds on neorealism and is perceived by its protagonists as the ‘logical outgrowth’ of neorealism with the aim of serving it (Rathbun, 2008: 297). Neoclassical realism is distinct from neorealism in two important ways. In the first place it deals with the realism as a foreign policy. Perceiving structural neorealist theories, such as the balance of power theory, as a theory of state behavior is however not exclusive to neoclassical realists (Rathbun, 2008: 301; Levy, 2003: 133; Elman, 2003: 8-9) Many neorealists also see the balance of power theory as a theory of state behavior (Levy, 2003: 133; Elman, 2003: 8-9). The difference is that all neoclassical realist theorists deal with realism as a foreign policy, while not all neorealist theorists do so. In the second place, as mentioned earlier, neoclassical realists focus on the domestic-level factors of influence on state behavior. They perceive domestic-level variables as being consistently causal in
systemic neorealist theory. The idea is expressed that domestic-level variables can be brought in in order to make neorealist theory more determinate (Rathbun, 2008: 307; Elman, 1996: 38; Rose, 1998: 146) Waltz’s neorealist balance of power theory received a significant amount of attention by neoclassical realists. (Lobell, Ripsman & Taliaferro, 2009; Taliaferro, 2006; 2009; Rose, 1998; Schweller, 1999).
In neoclassical realist works a great number of potential domestic-level factors influencing foreign policy outcomes have been mentioned. Examples include: state structure, state
29
strength, type of government, weak-state strong-state dichotomies, interest groups, the level of elite and societal cohesion, threat perception, the extractive and mobilization capacity of politico-military institutions, nationalism and ideology (Rose, 1998: 147; Sterling-Folker, 1997: 2; Taliaferro, 2006; Schweller 2004; Zakaria, 1998; Christensen 1996). Rose (1998: 157-165) distinguished two types of domestic-level variables in a rather limited number of neoclassical realist work that occur on a more frequent basis. Firstly: variables that deal with perceptions, such as those of major decision-makers (Rose, 1998: 157-161) and secondly, variables that deal with the strength of the state apparatus and the relationship between the state and the surrounding society (Rose, 1998: 161-165).
The second type of domestic-level variables distinguished by Rose (1998) all correspond to the same underlying logic, namely that of the ability of states to extract resources in order to pursue a particular foreign policy. Also in latter neoclassical realist works, domestic-level variables are based on the idea of resource extraction (Lobell, Ripsman & Taliaferro, 2009: 38-39; Taliaferro, 2004: 479-480; 2009: 194-226; Rathbun, 2008: 302). ‘Neoclassical realism identifies states’ extractive and mobilization power capacity as a crucial intervening variable between systemic and the actual foreign and defense policies’ (Lobell, Ripsman & Taliaferro, 2009: 38). The idea of resource extraction corresponds with Waltz’s ambiguous assumption of ‘high state power’ (Hobson, 2000: 25). In his narrow systemic theory the author implicitly assumes that all states in the international system have the same extractive capacity (Hobson, 2000: 25). However at the same time Waltz acknowledges that the internal structure of states and their capabilities to perform state tasks have an impact on their ability to pursue a certain foreign policy. The extractive capabilities of states are what influence the ability of states to pursue a foreign policy, such as internal balancing.
4.2 Limitations to neoclassical realist approaches
Neoclassical realist approaches make it possible to analyze the impact of domestic-level factors on the dependent variable of balancing in the neorealist balance of power theory. However, Neoclassical realist approaches are not without their limitations and caveats. Firstly, it is of grave importance to analyze how domestic-level variables of interest intervene with the premises of neorealist theories. Rathbun (2008) is particularly optimistic about the relationship between neorealist theory and neoclassical realism, and is reluctant to take the
30
possible violation of neorealist theory into account. In the case of Waltzian balance of power theory, it would for example make less sense to look at domestic-level variables that take intention-related domestic-level factors into account. Waltz clearly expects states to balance against the power, which is expressed through material capabilities, of other more powerful states in the international system. The intentions of states are expressed to be of no
importance (Taliaferro, 2009: 206). The rational state assumption also plays an important role in Waltz’s neorealist theory and should not be so easily discarded (Bueno de Mesquita, 2003: 173-174). Analyzing resource extraction-related domestic-level variables therefore seem to fit better with the premises of the theory rather than analyzing perception related domestic-level variables. If one plans to deviate from a neoclassical realist theory by implementing domestic-level variables, it should be made very clear what the implications of such an approach are. Referring to the relationship between neorealism and neoclassical realism simply as ‘a division of labor’ (2008: 297) without mentioning potential caveats is too simplistic.
Secondly, realist and non-realist scholars have frequently mentioned the issue of ad hoc implementation of domestic-level variables into neorealist theory (Vasquez, 2003: 104-106, 288-289; Rathbun, 2008: 297; Sterling-Folker, 1997: 2). Vasquez (2003: 104) states that a theoretical improvement which is not derived from the logic of the theory or paradigm is of an ad hoc nature. This touches upon the previously discussed issue of the impact of domestic-level variables can have on the premises of neorealist theory, such as Waltz’s balance of power theory. Rathbun (2008: 297) and Sterling-Folker (1997: 2, 23) are less clear on how the ad hoc implementation of domestic-level variables can be avoided. Both authors refer solely to the theoretical possibility of implementing domestic-level variables in neorealist theory. Neoclassical realists often base their domestic-level variables on the insights of non-realist international relations theory (Rose, 1998). This is mostly seen by non-realists as problematic (Legro and Moravscik, 1999). Implementing domestic-level variables derived from non-realist international relations theory is of added value if it does not violate the premises of neorealist theory and if there is a clear expected causal relation between the intervening variable and the dependent variable.
Thirdly, although neoclassical realist approaches make it possible to look at the influence of domestic-level factors on the foreign policy outcomes of states such as internal balancing, a significant number of neoclassical realists have not tested the impact of domestic-level variables on state behavior empirically (Taliaferro, 2006: 2009). Many neoclassical realists
