What factors determine state-building success in urecognised de facto states?

(1)

A thesis submitted in partial fulfilment of the requirements for a degree of MSc in Political Science (International Relations):

What Factors Determine State-Building

Success in Unrecognised De Facto

States?

Christopher Wignall

Submitted on 22

nd

_{June 2018}

Supervisor: Dr Abbey Steele

Second Reader: Dr Imke Harbers

Word Count: 23,768

Department of Political Science

Graduate School of Social Sciences

(2)

chris.wignall@student.uva.nl

11931876

(3)

Acknowledgements

There are a number of people that I need to thank for their contribution to this work. First and foremost, I owe a debt of gratitude to Dr Abbey Steele for her expert guidance, friendly encouragement, and endless patience at every step in the writing of this dissertation. I must also offer my thanks to Dr Imke Harbers for graciously agreeing to be my second reader, to Jules Goslinga for his constructive comments on the first draft of this thesis, and to Dr Daniel Pommier Vincelli who first piqued my interest in this topic.

Secondly, I express my deep appreciation to my girlfriend Rachel, without whom this work would never have come to fruition. I shall forever be grateful for the sacrifices she has made in encouraging me to pursue this degree and I would like to thank her for her unwavering love and support throughout the course of my studies, as well as for the endless hours that she spent reviewing draft versions of this work.

Finally, I owe the greatest thanks of all to my parents, for I would not be in this position without them. I cannot begin to express my gratitude for their constant care, encouragement, and commitment to my personal development and education. I hope that this work can in some small manner begin to repay them for their ongoing support and validate the numerous sacrifices they have made for my benefit.

(4)

Abstract

De facto states exist on the margins of international society, incessantly trying to generate the international attention that might serve to complement their internal sovereignty with external recognition. Yet, other than this unifying characteristic there is not much else that binds de facto states. They exist in different worlds, with vastly different histories, paths of emergence, and internal environments. Many have been able to build effective state institutions despite their external illegitimacy, while others remain stagnant, propped up by an external patron or safeguarded only by the fact that their parent state is equally weak. Therefore, this thesis asks the question: what factors determine a de facto state’s capacity to administer their populations and mobilise their resources in pursuit of their ultimate goal of statehood? The results indicate that ethnic and religious differences, political fragmentation, external support, militarisation, and a history of armed conflict all have an influence on state-building success.

(5)

Figures and Tables

Table 1: Stated Hypotheses . . . 19

Figure 1: Hypothesised Mechanism of Militarised State-Building . . . . 28

Table 2: A History of De Facto States from 1945 to 2011 . . . 32

Table 3: Expected Direction of Influence by Variable . . . 42

Table 4: Descriptive Statistics . . . 44

Table 5: Regression Results . . . 46

(8)

1. Introduction

A cursory glance over any modern political map conveys the image of a neat and well-organised world in which each and every scrap of territory has been divided up and commonly recognised as belonging to a single sovereign power which exercises uncontested authority over that particular patch of land. As any student of political science will be aware, however, this description of the world is not quite befitting of the observed reality. All political systems are to some degree representations of a community of sub-level political orders, which all reflect unique societies existing in a state of constant change, and many of these legitimate or illegitimate sub-state actors do not wilfully sacrifice power to a state-level authority or accept that their own power flows exclusively from above. Yet, there are two very good reasons for the consistent omission of these regional orders from our collective conception of global power distribution. The first is that they are not particularly conducive to cartographic representation. If one were to take the opposing tack and sacrifice readability in the name of accuracy, the world map would be indecipherable for all the dashed lines, asterisks, and annotations that would attempt to portray some of the separated or devolved power arrangements, local power struggles, and mechanisms of political hybridisation that occur throughout the world. The second reason is distinctly more ontological than practical – political maps are not awash with notes scribbled in margins and vain attempts to represent all the different types of political configuration because only one type of political structure matters in today’s international system: states.

States constitute the basic unit of international interaction and representation; they offer individuals the structural means to trade among, travel to, and communicate with individuals in other states, as well as providing a vehicle for the expression of identity on the world stage. States form the basis for ordering, categorising, and understanding humankind to such an extent that is it now impossible to conceive of a world without them – they have become synonymous with the post-Westphalian global order. It is perhaps surprising, then, that in such a state-centric world, rival political structures still persist and, in some cases, even thrive. One such type of alternative structure is the de facto state: ‘state-like’ areas that possess many of the same ceremonial structures (such as national flags, anthems and holidays) as states, issue visas, provide passports,1_{circulate currency, conduct elections, keep public records,}

provide welfare services, sustain armies, regulate the economy, and generally enjoy popular legitimacy, but lack international recognition (Pegg, 1998; Isachenko, 2008, 2009; Byman & King, 2012; Caspersen, 2012b).

De facto states are secessionist entities that claim territories currently within the boundaries of recognised states as their own and administer them in much the same manner a sovereign state would. They develop governance structures strong enough to provide self-rule and capable enough to repel attempts made by their parent states to

1_{While some of these passports do hold limited value as means of entry to a number of foreign states (e.g. Somaliland (Pegg & Kolstø, 2015; Som}

Tribune, 2016)), most are purely ceremonial. Generally, citizens of de facto states travel using a passport issued by a parent or patron state (Popescu, 2006b; Blakkisrud & Kolstø, 2011; O'Loughlin, Kolossov, & Toal, 2011, 2014), but even these may not grant a right of entry (Kvarchelia, 2013) (see note 2 for definitions of parent and patron states).

(9)

retake control.2_{Moreover, de facto states provide public goods to their populations, promote civic order, and}

develop legal codes, and may even govern with a level of popular consent that exceeds that of some fully-fledged states (Berg, 2012). In fact, the only major characteristic of statehood they lack is recognition by other states. Despite all their protestations, appeals, and demands for their right to self-determination to be recognised, they are in the eyes of international law ‘rogue state[s]’ (Kolstø, 2006, p. 729). As a result, de facto states are forced to cope with being shunned by the rest of the world, often being compelled to survive almost completely by their own means. They exist in a state of ‘existential paradox’ (Broers, 2013, p. 59), mimicking the political structures of states in their pursuit of eventual recognition but by their very existence violating the rules of the international system through their wilful disregard for ‘de jure’ territorial borders.

Perhaps by virtue of the fact that they exist hidden from the view of most of the global elite and enjoy little representation on the highest political stage, de facto states have become magnets for misunderstanding and misrepresentation (Broers, Iskandaryan, & Minasyan, 2015).3_{They have regularly been described as ‘anomalies’}

within the international system (Harvey & Stansfield, 2011, p. 1; Caspersen, 2012b, p. 3; 2017, p. 11; Caspersen & Stansfield, 2012, p. 5; Pegg & Kolstø, 2015, p. 193) – an unwarranted term that conveys an impression of rarity and inherent eccentricity.Firstly, de facto states are present across the globe, and in large numbers. While exact figures vary, the number of post-WWII de facto states has been counted as high as 34 (Florea, 2014, 2017). Secondly, many of the odd characteristics that de facto states do exhibit are products of their external environment – namely the post-war global structure – rather than ‘individual’ peculiarities. If we want to understand de facto states it is crucial that we do not think of them as just mere accidents of the international system; they are instead produced systematically by a world that prioritises legal rights and norms above governance capacity.4_{Therefore, to use}

Iskandaryan’s (2015, p. 210) phrase, they are more ‘technical errors within the system of international law’ than ‘anomalies’.

De facto states are products of a system of conferred statehood that emerged in earnest following the First World War. This system created a standardised conception of statehood conducive to international interaction (Österud, 1997), upon which the structural determinants of state formation in the post-1945 mass decolonisation were developed. In turn it has created stability – now that the process of (formal) decolonisation is over, political borders are arguably at their most stable in the whole of recorded history. While border disputes and inter-state wars have certainly not been confined to history, the drive to seize land has been largely curtailed by the diminishing value to be gained from invasion and the growing belief that national borders should be sacred, not least because of the international backlash that would be unleashed against any state which occupies territory belonging to another.

2_{Parent states are the countries de facto states have seceded from; patron states are countries that offer de facto states financial, political, or}

military support.

3_{King (2001, p. 550) described Eurasian de facto states as ‘informational black holes’.}

4_{Indeed, Buzard, Graham, and Horne (2017) demonstrate that the dynamics of attempted secessionism make de facto statehood a natural}

(10)

However, this system also created side effects. Whereas previously states had to exhibit ‘a certain level of military,

economic, and governmental effectiveness’ to prevent opportunistic rivals from seizing parts of their territory, now

they rely on the international system as guarantors of their legitimacy (Pegg, 1998, p. 1). States confer legitimacy upon one another and collectively work to preserve existing state boundaries in the name of stability (Bahcheli, Bartmann, & Srebrnik, 2004). This is what James Mayall (1990, p. 56) has termed the ‘freez[ing] of the political map’, in which the vast majority of the earth’s territory (and so too its oceans and airspace) are ruled by the sovereign power in place following decolonisation. This system, therefore, opened a window of opportunity for aspiring states with a history of systematic subordination, and closed it for the rest.

By downplaying or wilfully ignoring political realities in favour of the helpful fiction that each square mile of territory (and every individual within it) is administered by one of the 193 sovereign states currently in existence (UN, 2018), this global system supports failed states and hurts de facto states (Jackson, 2007). It breaks away from historical Darwinian principles as the base determinants of state survival and instead places incumbent structures of power above all challengers. Thus, the potential for discord between externalised and internalised conceptions of sovereignty is created. When a state has the former but not the latter it is a quasi-state (or a failed state); when a state possesses the latter but lacks the former is can be described as ‘de facto’. As such, quasi states and de facto states are two sides of the same coin, with a shared origin story (Pegg, 1998). They are both children of the prevailing negative conception of statehood which characterises the ‘new sovereignty game’ (Jackson, 1990, p. 40). This post-war idea of statehood as a distinctly indivisible and absolute status regularly prevents failed states from being invaded by rivals that could provide better governance, and thwarts secessionist governments in their quest for recognition even when they could provide more legitimate rule over a particular territory than central governments. Undeterred by this formidable barrier that obstructs the road to recognition, de facto states do what they can to ensure their own long-term survival. Under the shadow of non-recognition, they develop state institutions, build defensive capabilities, and rally whatever support they can for their cause. Yet, some de facto states have been distinctly more successful in this regard than others. Chechnya, for example, struggled to develop independent institutions capable of extracting revenue from its population and was eventually forcefully reabsorbed by Russia after eight years of de facto statehood (Tishkov, 2004); Republika Srpska Krajina suffered a similar fate because of its chaotic organisation and inability to exercise undisputed control (Caspersen, 2012b; Kolstø & Paukovic, 2014), as did Tamil Eelam (Stokke, 2006). At the other end of the scale, Taiwan has managed to develop institutions that have since supported its rise to become a key player in the global economy, while other less noticeable de facto states, such as Transnistria, Kurdistan, and Somaliland, outperform their parent states across a range of governance indicators (King, 2001; Kolstø, 2006; Menkhaus, 2007; Jüde, 2017).

(11)

Therefore, this thesis attempts to answer the question: why are some de facto states more successful than others in developing quality state institutions? I aim to identify what conditions are most fundamental for states with no or limited external legitimacy to build a functioning institutional base – an aspect of de facto statehood that has been largely ignored up to this point. A quantitative analysis is employed, testing the influence of social, political, and economic factors both at the onset of de facto statehood and throughout the de facto state’s existence, on each’s experience of state-building. Thus, this study provides an alternative perspective on de facto states from the case study-based qualitative literature, with the intention of challenging prevailing assumptions and enhancing existing explanations.

This is a valuable endeavour for two reasons, which exist on contrasting planes of analysis. The first is that the degree of state-building has a large bearing on the social and economic opportunities afforded to residents of de facto states and thus to a large degree determines their quality of life. The second is predicated on the fact that state-building plays an intermediary role in the histories of de facto states. While the task of predicting the ultimate effects of de facto states is beyond the concern of this thesis, there is significant potential for postulating initial causal steps that cause final outcomes in de facto states through the mechanism of state-building. Currently, the predictive capacity of state-building has received far greater attention than the question of how de facto states’ institutions are affected by the social and political environments that they are born into and the motivations and intentions of de facto state leaders. By tracing the reasons for the development or underdevelopment of de facto state institutions, and filling the present gap in our understanding of de facto states, the hope is that it will become possible to explain the ultimate fate of de facto states by reference to underlying characteristics rather than the catch-all explanation that is the quality of governance and the extent of central control, thereby offering a much more tangible explanation for variation than is currently available.

I proceed as follows: I begin by reviewing the existing literature on state-building in de facto states and provide a definition for these malleable concepts; secondly, I develop the foundations of the theories that will be tested in the coming sections and state my hypotheses; thirdly, I outline the source of the data used, the coding procedures employed, and the methods of analysis that I use to test my hypotheses as part of the research design; fourthly, I present the results of this analysis, discuss their implications and theorise about why some results did not conform to my expectations; lastly, I conclude.

(12)

2. Literature Review

Despite the frequency with which they have materialised as pockets of administrative ambiguity from within the broader global patchwork of clearly-defined nation states in the decades since the end of the Second World War (Pegg, 1998; Caspersen & Stansfield, 2012; Florea, 2017), de facto states have commanded little scholarly attention until comparatively recently. In part this was because of their perceived instability. The majority of de facto states tended to last only a few years before being reabsorbed by their parent state, and consequently did little to upset the post-colonial international system. Their fleeting presence on the global stage, often anarchic methods of self-organisation, and tendency to materialise in poor African and Asian countries meant that they were not typically considered to be far enough removed from rebel groups to deserve detailed study in their own right. Even overarching differences in the nature of their struggle, such as the placement of common ethnicity above common ideology in the creation of group identity, and the concentration of conflict upon the right to change the political system in a particular area rather than the country as a whole, were not considered fundamental enough to differentiate de facto states from rebel groups, either in their strategies, capabilities, or chances of success. Only after a multitude of new de facto states emerged from the collapse of the Eastern bloc in the early 1990s did policy-makers’ interest in these ‘places that don’t exist’ (Caspersen & Stansfield, 2012, p. 2) achieve a critical mass and it left political scientists playing catch-up. Since the publication of Pegg’s (1998) seminal account of the capabilities, limitations, and future prospects of unrecognised de facto states, however, the academic attention given to these phenomena has exploded. As this burgeoning field of study developed, the area of focus quickly shifted from attempting to understand what de facto states are and how they should be categorised to theorising about why they emerged from under the stewardship of larger parent states (Bobick, 2014; Zheger, 2015), how they continue to repel forceful reintegration efforts made by more powerful ‘de jure’ authorities (Kolstø, 2006; Caspersen, 2009; Florea, 2014; Buzard, Graham, & Horne, 2017), why they enjoy persistent interactions with external actors despite their illegitimacy in this arena (Kolstø, 2006; Caspersen, 2008b, 2009; Berg & Toomla, 2009; Hoch, 2011; Frear, 2014), and how one might predict their respective chances for continued survival under the present status-quo, reabsorption into the parent state, or ‘graduation’ to recognised statehood (Coggins, 2011).

While none of these studies made describing, explaining, or predicting the process of state-building their ultimate goal, they all rely on some understanding of the power of the de facto state central government and a mechanism of assessing the de facto state’s institutionalised means of operation to say anything of any use within their area of interest. As a result, the majority of our present understanding of state-building in de facto states comes from such ‘indirect attention’ and as a result of its role as a principal vehicle of change. Consequently, the default method of analysis regarding state-building in de facto states is to compare them with internationally recognised states so as to ascertain the influence of non-recognition in general. While this is a worthwhile endeavour that has resulted in

(13)

the advancement of our understanding of how and why de facto states act in the way they do, it does invariably result in the portrayal of de facto states as identical territories wholly defined by their lack of recognition (Kolstø, 2006) – an invalid characterisation that is equivalent to describing states as identical simply because they are all internationally recognised (Caspersen, 2012b; O’Loughlin, et al., 2014; Hoch & Rudincova, 2015; Iskandaryan, 2015).5

This is now beginning to change, and scholars are starting to analyse how de facto states differ in their approach to state-building by looking at how conditions within de facto states determine individual and collective reactions to non-recognition (O’Loughlin, Toal, & Chamberlain-Creangă, 2013; Johnson & Smaker, 2014; O’Loughlin, Kolossov, & Toal, 2014). Caspersen’s (2012b) book ‘Unrecognized States: The Struggle for Sovereignty in the Modern

International System’, which provides a generalised account of how the defining characteristic of de facto states –

non-recognition – has influenced the mechanisms by which they attempt to build state structures, but also examines how de facto states’ ‘path[s] of creation’ (p. 32) affect their internal dynamics and their drive towards statehood, is perhaps the best example of this shift in analytical perspective.

Yet, before discussing the existing scholarly work on state-building in de facto states in detail, it is important to recognise that neither de facto statehood nor state-building are concrete and uncontested concepts. Therefore, it is necessary to develop an understanding of what is meant by each of them, or at least outline what each term denotes in the context of this study. The next sections summarise the debate surrounding each in turn and provide definitional frameworks that will form the basis for the forthcoming analysis.

2.1 Definitions and Classifications of De Facto Statehood

Despite the recent scholarly attention given to de facto states, no agreement has been reached on what conditions an unrecognised self-governing entity must satisfy to be considered a de facto state, or even what they should be called if they do. The first issue of categorisation and definition is decidedly more advanced than the latter, however, and different approaches generally fit into the ‘short’ or ‘long’ approach. The ‘long’ approach emerged in the name of conceptual concreteness after initial efforts at defining de facto states were deemed too ambiguous (such as that provided by Pegg (1998)). While Pegg warned against the potential to confuse power vacuums, colonies, fleeting armed rebellions, puppet states, and protest movements with de facto statehood, there was little in his ‘short’ definition designed to exclude these entities. Therefore, in this work I shall use a ‘long‘ definition provided by Florea (2014, p. 791, 2017, p. 338) which incorporates the benefits of two decades worth of debate on the subject to produce a robust characterisation of a de facto state as an entity that:

5_{This has been exacerbated by the fact that previous research has relied heavily on case studies which shared some common characteristics,} such as geographical proximity, ethnic homogeneity, extensive shadow economies, and external patrons.

(14)

‘(1) belongs to (or is administered by) a recognized country, but is not a colonial possession, (2) seeks some degree of separation from that country and has declared independence […] through a referendum or a ‘sovereignty declaration’, (3) exerts military control over a territory or portions of territory inhabited by a permanent population, (4) is not sanctioned by the parent government, (5) performs at least basic governance functions, (6) lacks international legal sovereignty, and (7) [has existed] for at least 24 months.’ 6

While this definition rightly excludes many entities that could not be considered de facto states, it is still far from perfect since many of these conditions are open to individual interpretation. Here debate tends to focus on the required consistency and unanimity of independence as a stated goal (Byman & King, 2012), the degree of acceptable outside intervention by other states, what qualifies or does not qualify as a ‘basic governance function’, the extent of control required to feasibly claim to be exerting one’s sovereign will (Caspersen, 2012b), the level of external recognition that should confer a sufficient level of de jure sovereignty such that the de facto qualification is not warranted (Li, 1979; Richards & Smith, 2015), and whether the fairly arbitrary two-year mark is justifiable (Caspersen, 2012b; Richards & Smith, 2015). Consequently, the exact dates of de facto state emergence and demise, as well as assertions over the number of de facto states that have existed throughout history, are disputed by authors employing different minimum standards for inclusion (Pegg, 2017). For example, Kolstø and Paukovic (2014) find a dozen examples of post-WWII de facto states, whereas Florea (2014, 2017) records 34 in his analysis.

Such differences in perception cloud the analysis of de facto states and ensure that researchers focus on the few examples whose credentials as de facto states are not in doubt.7_{This selectivity restricts the capacity for meaningful}

analysis by comparing a group of de facto states that are all in the same political situation. Indeed, it can limit our potential to understand how and why they may change, the process by which they ever reached their present status, or how they might develop in the future. To a large extent, then, the debate has missed the point regarding the rationale behind the categorisation of de facto states – whilst being selective in describing the present extent of de facto states has inherent value for embellishing our understanding of the global political patchwork, it is only by including less ‘de facto’ or more ‘de jure’ examples in our framework for analysis (such as that employed by Florea (2014, 2017)) that we improve our understanding of the field as a whole (Geddes, 1990).

The debate about terminology also lacks consensus. Authors have used a range of different names to describe these phenomena, including unrecognised states (Caspersen, 2011 2012b; Richards & Smith, 2015), contested states (Geldenhuys, 2009; Ker-Lindsay, 2012, 2015), quasi-states (Kolstø, 2006), pseudo-states (Kolossov & O'Loughlin,

6_{The adoption of this definition is also driven by the fact that de facto states recorded in Florea’s dataset (2014) – from which this thesis will draw}

a number of conclusions – were selected by reference to this standard.

(15)

1998), and phantom states (Byman & King, 2012), as well as amalgamations of these terms (Berg, 2007). Many of those listed have fluctuated between labels, even within the same work (King, 2001). Yet, unlike the debate over definitions, the debate over one’s choice of jargon is almost entirely lacking in terms of academic significance. Despite the fact that different terms are regularly used, there is little, if any, semantic difference between them and the devotion of more time to this issue will do little to advance the collective understanding of these entities (Pegg, 2017). Therefore, in this essay I adopt the most utilised term – de facto states – as adhered to by Pegg (1998, 2017), Lynch (2004), and Florea (2014, 2017), in the interest of general ease.

A much more valid concern about the semantic integrity of the term ‘de facto state’ concentrates on the latter aspect: whether these unrecognised political structures should be referred to as ‘states’. The fact that the word ‘state’ is always used with a qualifying adjective that indicates a degree of uncertainty or reduced status indicates that under the normal legalised procedure of acquiring statehood de facto states do not qualify as states. In the post-war international system, statehood is considered to be a distinctly legal status conferred by other sovereign nations (Crawford, 1977; Jackson, 1990), usually on the basis of a perceived moral right (Fabry, 2010; Ker-Lindsay, 2012). In other words, external judgements matter, but the quality of internal governance does not (Pegg, 1998). Therefore, in a strict legal sense, ‘failed’ or ‘quasi’ states, which lack internal legitimacy due to their inability to provide public services or justifiably rule in the eyes of their citizenry but are recognised by other states (the most frequently cited example being Somalia), hold statehood, but de facto states – for which the opposite is true – do not.

However, neither sovereignty nor statehood are fixed concepts, either in time or across perspectives (Oppenheim, 1920; James, 1984; Jackson, 1990; Krasner, 1999; Schrijver, 2000; Nagan & Hammer, 2004). Alternative definitions of sovereignty promote existing governance capacity, internal legitimacy (usually demonstrated by overwhelming support in referenda on independence and the adoption of new constitutions),8_{and historic episodes of autonomous}

rule as the factors that confer statehood (Lynch, 2002; Richards & Smith, 2015). However, the most common manner in which de facto states promote their right to statehood is through reference to the Montevideo Convention on the Rights and Duties of States (International Conference of American States, 1933, p. 25).9_{Article 1 of this document}

declares that a state should acquire legal personhood if they possess: ‘[…] (a) a permanent population, (b) a defined

territory, (c) [a] government, and (d) a capacity to enter into relations with the other States.’ While the fourth

requirement raises a point of contention regarding whether or not de facto states have the capacity to enter into relations with other states who refuse to recognise them, this definition would undoubtedly be more familiar to a

8_{Support for statehood is often overwhelming in these votes – Somaliland voted for independence with a vote percentage of 97% to 3% (African}

Elections Database, 2010), 98% of Karabakhs voted in favour of a state-forming constitution in 2006 (BBC News, 2006), and most recently 93% of Iraqi Kurds voted for independence (Financial Times, 2017) – all with some degree of external observation and endorsement.

9_{The common response to this argument is that this treaty is distinctly regional in character, being signed only by American states, and has}

(16)

political scientist than a scholar of international law, since it emphasises the empirical over the conceptual and the tangible over the titular.

If we accept this description of what makes a state a state and a government sovereign over its people, then the term ‘de facto state’ loses its oxymoronic tone. In other words, if we recognise that non-recognition undoubtedly has a deep and long-lasting impact on the de facto state but that the primary focus of this study is on the empirics of internal institution-building and not on the external mechanics of international law and the ontology of recognition, then we will encounter fewer conceptual obstacles. This is especially the case given that the subject of this thesis is state-building, a term which implies the existence of a number of different magnitudes of ‘stateness’. As such, conceptualising the state in a binary form would be a counterproductive enterprise for the measurement of state-building success. Therefore, in this study the degree of statehood is determined by internal capacity, not external recognition, which is measurable to a significantly greater extent than simply being ‘present’ or ‘not present’ (Clapham, 1998; Berg & Kuusk, 2010). This conforms with the rest of the de facto state literature written from the perspective of political science in which the term ‘state’ is used with one’s own choice of qualifier and is subject to complex comparison. Only those writing from the perspective of international law employ different terminology (Hillgruber, 1998; van Essen, 2012).

2.2 What is State-Building?

The state is an elusive concept. In many respects the state is influential over all the institutions that formulate a society, although since it does not strictly create any of these institutions itself it is very difficult to know where the state begins and ends. The state is a concept considered by many to be entirely separate from, and distinctly above, the game played by rulers, governments, and departments, and yet its very nature is entirely determined by their collective action. In a sense, it is debatable whether the state exists as an entity in itself or is simply used as a collective term for a succession of disparate governments, parties, and interest groups that compete for resources and influence under a generally consistent system of rules.10_{In this thesis I consider the state to be an entity}

developed and maintained by governments, which inherit brief jurisdiction over the instruments of public administration and have the opportunity to modify them in the manner they see fit. In most respects, therefore, it is helpful to think of a government and the state as one and the same, despite their conceptual differences (Nettl, 1968). If a government increases its powers of governance it increases the power of the state. As such, in this thesis I shall equate government with the state in the interest of clarity and readability.

Yet, understanding what the state is provides few clues as to how it should be measured. States are complex bodies that vary dramatically in the extent of their action and the types of functions that they perform for their citizens.

(17)

They range from ‘weak’ or ‘failed’ states such as Somalia and South Sudan, in which the reach of the state is minimal and authorities struggle to provide some of the most basic necessities to their people, to ‘stable’ states such as Sweden, in which virtually all aspects of life are in some way influenced by government intervention (Fund for Peace, 2017). The same is true of de facto states also – replace Sweden with Taiwan, Somalia with Chechnya, Karen State or Bougainville, and South Sudan with the pre-independence de facto version of itself, and the previous statement remains equally valid. However, while measuring states when comparing Somalia and Sweden is easy, since there is no doubt as to which is the better state-builder (Sweden outperforms Somalia in every state function), when comparing states around the same level, such as Somalia and South Sudan, the results are dependent on the aspect of state governance that one chooses to measure. Consequently, a systematic approach to conceptualising state-building is required to advance past rudimentary observations that rich, socially democratic states outperform poor, corrupt ones. To do so we need to ask the question: what is it about how states operate that makes some ‘better’ than others?

Perhaps the most intuitive answer to this question was provided by Francis Fukuyama (2004a, 2004b) in his separation of ‘state strength’ from ‘state scope’. While Fukuyama’s study was not ground-breaking, and in many ways only ‘repackaged’ existing theories of bureaucratic efficiency and public control provided by Weber (1965) and the separation of infrastructural and despotic power developed by Mann (1984), it does offer a stripped-down conception of what determines state performance. Fukuyama’s ‘state strength’ refers to what has come to be known as ‘state capacity’ – the ability of the state to execute policies and enforce laws – and ‘state scope’ refers to the range of functions a state performs. Each state exists at a unique point on the matrix created by these two measurements: the US, for example, possesses significant strength, but is culturally and constitutionally limited in its potential scope; India is the opposite – it lacks the ability to enforce laws in inaccessible rural villages but involves itself in virtually all aspects of economic and social life. Underdeveloped nations such as the majority of Sub-Saharan African states lack strength and scope, whereas rich welfare states, such as the Nordic countries, have both. Although some minimum degree of functional scope is necessary to provide enforceability, state strength without scope is preferable to scope without strength. It is better to be in the United States’ position than India’s, which is a condition ripe for inefficiency, bureaucratic confusion, and corruption. Thus, when we are considering the success of a state-building project it is generally the degree of ‘state strength’ that is of interest. Successful state-builders do not struggle to implement directives in rural areas, depend on private support to achieve their objectives, or overburden themselves by attempting to perform an excessive number of functions at the expense of overall bureaucratic quality.

Yet, the mechanism by which one determines the degree of ‘state strength’ depends on how it is conceptualised. Hillel Soifer (2008) proposes three theoretical foundations for understanding the strength of the state. The first –

(18)

the ‘national capabilities approach’ (p. 236) – considers power to be a ‘dispositional attribute’ (Morriss, 1987, p. 20), such that power is determined by the state’s potential ability to exercise mechanisms of control, rather than the normal actions of state institutions. Therefore, the resources that states could mobilise, were it necessary to ensure compliance, represent the true power of the state, not simply the observable day-to-day resources committed to maintaining state control. Despite appearances, this approach represents the most realist and tangible conception of power in its implicit assumption that an added unit of deterrence – an extra reserve police officer, soldier, or tax inspector, for example – will always lead to an added unit of state power (Snider, 1987). Despite the practical flaws of this assumption, the approach has formed a central basis for conceiving of and measuring power, not least because its simplicity lends itself to measurement. However, relying solely on this aspect of state power without considering Soifer’s two other conceptions of power will lead to a misunderstanding of the mechanisms by which the state controls its society.

Soifer’s (2008) second approach to conceptualising state power is decidedly less concrete. The ‘weight of the state’ approach (p. 239) adopts a constructivist lens in assessing the extent to which the state has influenced the behaviour of individuals within a society. By inducing individuals to adopt a culture of thought in which the legitimacy of the state and its status as the ultimate power is unquestioned, the state develops a power that requires no threat of punishment to maintain. Initially this is achieved by demonstrating the mutual gains to be made from submitting to the rule of the Leviathan (Hobbes, [1651] 1996), but it also relies heavily on the sister concept of nation-building and the social pressure to conform for the common benefit of one’s kin (Linz, 1993; Wimmer, 2013). Yet, in many respects the ability of the state to change behaviour, while important at all stages of state-building, represents the ‘end-game’ of state institutions and social capital formation. Indeed, Holsti (1996, p. 84) claimed that ‘[…] it is in the realm

of ideas and sentiment that the fate of states is primarily determined’. This is a statement supported by the

emblematic text on the importance of changing attitudes towards the state – Eugen Weber’s (1976) ‘From Peasants

into Frenchmen’ – which charts the development of the French state over the course of almost half a century – a

timespan that incidentally far exceeds the lifespans of many de facto states.

The third approach focuses on the degree of sub-national variation in the reach of the state (usually between urban and rural areas, or majority and minority populations) and assumes that those states which struggle to attain and conserve authority over the entirety of their territory are less powerful. Such an approach attempts to combine the two previous lenses of recognising state power into an overall indicator. However, it is debatable whether consistency necessarily means unconstrained centralised power, and as such this approach offers little to aid the effort to construct a means to compare state-building capacity across state borders.

These three measures are generally complimentary (Buzan, 1991; Holsti, 1996; Brautigam, Fjeldstad, & Moore, 2008; Lemay-Hébert, 2009), but not always so. Thus, authors who select only one of these approaches – usually the

(19)

capabilities approach, as measured either by reference to a state’s ‘military capacity, bureaucratic/administrative

capacity, [or] institutional coherence and quality’ (Hendrix, 2010, p. 274) – ignore a large portion of what constitutes

state-building, which can lead to systematic biases in measurement.11_{Since some states rely more heavily on}

particular mechanisms of power than others, any comparative analysis must incorporate all three of Soifer’s measurements in order to accurately reflect differences between states.

The primary determinant of ‘what counts’ – capabilities or the weight of the state – for the maintenance of centralised power is state age: new states at the beginning of the nation-building project, which have not developed ‘soft’ institutions such as state loyalty and trust in the state-building project, are largely forced to rely on ‘hard’ enforcement measures alone to dispel ideas of practical alternatives to rule by the state. As states mature, however, they start to rely more heavily on the weight of the state for implementation and enforcement.12_{By ignoring the}

weight of the state, as many authors do, the extent to which established states differ from new states is undervalued. Furthermore, state age is not the only influence on the type of state power which follows this pattern; cultures towards authority, the strength of civic identity, and natural geographical formations also affect the forms of legitimacy a state can depend upon.

Therefore, the formation of state capacity depends on the coercion of agreeable behaviour through the threat of force, the development of favourable public attitudes towards the state, and the maintenance of internal consistency. Thus, I consider state-building to be a function of all three of Soifer’s mechanisms for measuring the strength of a state: national capabilities, the weight of the state, and internal consistency. An improvement in any of these factors can be considered to be state-building.

2.3 The State in De Facto States

State-building, being a new phrase that expresses an old concept, possesses a rich history of scholarship, whereas state-building in de facto states does not. However, the literature that does exist on de facto state institution-building is in general agreement: the fundamental structures of state-institution-building in de facto states differ little to those of fully-fledged states; the major differences come from how inherent characteristics of de facto statehood influence individual and collective approaches to state-building. Foremost among these differences is the fact that de facto states possess an extra incentive to build well-functioning state institutions. What Bakke, Linke, O'Loughlin, and Toal termed de facto states’ ‘double burden’ (2018, p. 2) – a reference to the fact that de facto state governments are

11 This issue is revisited as part of the research design on pages 35-38.

12_{This could be to the benefit or detriment of state power – either allowing the state to influence society by manufacturing a generalised}

subconscious acceptance of the state’s will, or by causing the unidirectional mechanism of the state influencing society to break down as the divisions between state and non-state become increasingly imperceptible, such that state actions become as much driven by non-state actors as the other way around and the state surrenders some of its elevated authority and infrastructural power (Migdal, 2001; Soifer & vom Hau, 2008).

(20)

tasked not only with winning over their citizens by displaying their ability to offer stability and general prosperity but also convince international actors of their right to statehood – also serves as a ‘double incentive’ to develop systems of good governance (Caspersen, 2012b). Yet, the influence of non-recognition also goes much deeper than Weberian institutional development in general, and often produces specific behaviours within state-building. In this section I outline the four most prominent effects: the necessary prioritisation of military capabilities, the tendency for de facto states to engage in autonomous self-development, the demand for external support, and the influence of non-recognition on the de facto state’s choice of political system.

2.3.1 The Prioritisation of Military Capacity

One of the most important aspects of international recognition is the external security it brings to comparatively weak states. The prohibition of unauthorised foreign occupations and opportunistic land grabs protects vulnerable states from outside domination. De facto states, however, do not enjoy the same protection under international law, since they lack the legal personality of states. Consequently, they exist in an ultra-realist world in which their only defence against reabsorption by the parent state is military force (Caspersen, 2012b). If a de facto state cannot repel efforts made by the parent state to regain control of the territory on which it stands there is no institutional safety net upon which it can depend. Therefore, the ability to compete militarily is an essential requirement for de facto states and hence it is quite understandably prioritised within the general drive to state-build (Caspersen, 2012b).13

Whether this forced militarisation is beneficial for the wider state-building project is debated, however.14_Kolstø

(2006, p. 732) asserts that this prioritisation of military defence comes at the expense of ‘civilian purposes’ and

‘contributes to the weak development of welfare, educational facilities, and the building of infrastructure’, whereas

others claim that an external threat, as well as the conviction that this threat can be defended against, engenders a new ‘national’ identity based on self-defence and self-reliance (Byman & King, 2012; Pegg, 2017), and that this ‘soft’ aspect of state-building – creating or enhancing a shared history, unifying identity, and common destiny – stimulates trust in the ‘hard’ state-building process.

2.3.2 Autonomous State-Building

De facto states tend to emerge in poor areas and have very often fought protracted and costly wars in order to retain their de facto statehood. In normal circumstances (were they recognised entities) they would qualify for recovery aid and support from post-conflict peacekeepers. Given the tensions that have arisen within recent UN-led and

13 See Caspersen (2012, p. 149) for data on the size of de facto state militaries relative to population. 14 This is an issue that will be tested later on in this thesis.

(21)

backed post-conflict state-building programmes (Paris & Sisk, 2007; Berdal, 2016), however, it is not surprising that many de facto state scholars consider non-recognition as an advantage in this regard. Even when taking into account the lack of development funding given to de facto states, the fact that non-recognition prevents potential meddlers, wary that any form of sustained interaction with de facto states could constitute effective recognition, from involving themselves in the state-building process can be seen as positive. Indeed, this has been proposed by Menkhaus (2007), Bradbury (2008), Podder (2014), and Johnson and Smaker (2014) as the reason that unrecognised Somaliland and Puntland have managed to build better state institutions than recognised Somalia.15_{They claimed that the}

freedom to construct national institutions from existing clan-based hierarchies has led to the formation of a more stable, representative, and legitimate system of governance which is less likely to collapse into grievance-based conflicts than top-down systems imposed by UN missions or coalition forces. This lack of oversight also allows some de facto states to pursue methods of state-building that are ethically suspect but generally bolster the nation-building and state-nation-building processes, such as the large-scale ethnic homogenisation achieved by the previously diverse states of Abkhazia (Kolossov & O'Loughlin, 2011) and Nagorno-Karabakh (Kolstø & Blakkisrud, 2012) through mass-expulsions.

2.3.3 Seeking External Support

Despite the fact that there are benefits to be gained from autonomy in some respects, state-building cannot take place within a vacuum. De facto states need to seek out external support not only to provide them with vital resources to defend their territory and engage in state-building, but also to bypass what Pegg (1998, p. 43) termed

‘the economic cost of non-recognition’.16_{Without outside support to overcome their perceived illegitimacy and}

instability, de facto states may find themselves without the ability to export goods or attract FDI, develop a strong enough economy to finance an army capable of resisting invasion, or outline their case for recognition in international fora. Diaspora communities and patron states therefore offer de facto states short-term solutions to these issues so that they can continue to build state institutions and develop their economies while they pursue international recognition, with diasporas usually offering the use of ‘soft-skills’, such as technical assistance and local knowledge, and patron states offering the ‘harder’ elements – for instance large-scale finance, aid payments, and military support (Caspersen, 2012b). Patron states are usually neighbouring states that share historic ties or political goals; for instance, South Ossetia’s and Transnistria’s patron state is Russia, Nagorno-Karabakh depends on Armenia, and Northern Cyprus receives support from Turkey.

15 See Höhne (2011, 2013) for a contrasting perspective on Somaliland’s clan-based state formation. 16 See also Blakkisrud and Kolstø (2012)

(22)

The magnitude of the external support some de facto states have received has been substantial, to the extent that Caspersen (2009, p. 52, 2012b, p. 38) has described many as being ‘on the dole’.17_{For example, Armenia provides}

more than half of Nagorno-Karabakh’s budget and military personnel (Broers, 2005; Caspersen, 2008a; Kolstø & Blakkisrud, 2008; Ó Beacháin, Comai, Toal, & O'Loughlin, 2017), Transnistria has been allowed to run up a debt of $3.7 billion to the Russian state for the non-payment of already heavily subsidised gas supplies (Popescu, 2006b; CISR, 2010; Blakkisrud & Kolstø, 2011; Całus, 2013), South Ossetia has in the past received up to 96% of its yearly state budget from Russia (Kolossov & O'Loughlin, 2011; Ó Beacháin, et al., 2017), and Northern Cyprus has relied on Turkey for the provision of many of its army officers and much of its equipment (Solsten, 1993).

2.3.4 Choosing a Political System

As well as being an incentive to build state institutions, non-recognition also extends to the type of political system that emerges in de facto states, albeit in a much subtler manner. Here de facto states exist in a world of conflicting incentives and social pressures; on one hand they need to ensure their own survival, which generally inspires the creation of a centralised authority capable of mobilising resources quickly, but on the other they desire to present themselves as a more preferable option than rule by the parent state to key international decision-makers, which requires a pluralistic system of accountable governance and minority representation. In effect, then, they have to balance immediate pressures with long-term objectives (Kolstø, 2006; Caspersen, 2012b; Smolnik, 2012).

There is an ongoing debate over the relative strength of these counteracting pressures, as well as whether their impetus derives from civil society or is manufactured by elites. Etzioni (1993) attests that the inherent inclination of nationalist self-determination movements (of which de facto statehood is the most developed form) is an intolerance of difference that naturally lends itself to authoritarianism. Consequently, he claims that self-determination harms democracy where it exists and retards its emergence where it does not, especially within the seceding region itself. Popescu (2006a), however, writing specifically on de facto states, offers a more unique reason for the emergence of de facto state authoritarianism. He considers de facto states’ inherent post-emergence instability and the prospect of renewed warfare with the parent state as a means for incumbent powers to legitimise and retain their position by denouncing opposition voices as destabilising and potentially damaging. He claims that individuals in de facto states adopt a ‘siege mentality’ by lending support to the incumbent authority in the interest of achieving an outcome that is in everyone’s interest (in this case continued survival as a secessionist entity).18

Contemporary evidence, however, gives a mixed picture regarding regime legitimacy. In testing for the presence of siege mentalities by surveying attitudes towards the state in Abkhazia, Bakke, O'Loughlin, Toal and Ward (2014)

17 This phrase was first used by Kolstø and Blakkisrud (2008, p. 494) to describe Abkhazia. 18_{See also Coser (1956) and Caspersen (2009, 2012a, 2012b, 2015).}

(23)

found that 74% were supportive of the direction Abkhazia was moving in.19_{However, a later, more extensive, study}

by Bakke, et al. (2018) highlighted that only 28% of interviewees were supportive of the state direction in Transnistria.

The practical impact of the counteracting pressure for democratisation has also been debated. This is an argument that revolves around the perceived pervasiveness of the idea that democratisation will eventually lead to recognition within de facto state government circles and populations. When writers first started analysing the position of de facto states it was considered rather naïve to appeal to the international community on the basis of internal governance quality (Pegg, 1998), but the geo-political landscape has since undergone a seismic shift. After the UN’s adoption of the principle of ‘standards before status’ in 2003, many de facto states began to base their claims for recognition on their ‘earned’ right to sovereignty, through their governance capacity, promotion of democracy, and commitment to human rights, rather than on a ‘remedial’ right based upon persecution at the hands of their parent state and of nationalistic self-rule (Kopeček, Hoch, & Baar, 2016). They presented themselves as ‘islands of

democracy’ within authoritarian seas (Caspersen, 2012b, p. 71). However, after Kosovo’s partial recognition in 2008,

seemingly on the basis of its ‘special status’ (Ker-Lindsay, 2013, p. 847), and the subsequent lack of recognition for many de facto states that considered themselves to be at least the democratic equal to Kosovo, many lost confidence in such legitimisation strategies (Caspersen, 2008b, 2008c, 2013; Gardner, 2008). The real impact of external inducements to democratise, and whether this effect is stronger or weaker than the internal incentive to always support the incumbent, therefore remains under review.

Despite this, some writers have remarked that many de facto states do have democratic credentials (Blakkisrud & Kolstø, 2011; Berg, 2012; Ó Beacháin, 2015), and Blakkisrud and Kolstø (2012) assert that the fact that all three South Caucasian states still in existence possess good-quality democratic structures despite legacies of Soviet single party rule is testament to this ‘democratisation for recognition’ effect. Similarly, Ó Beacháin (2012), writing about Abkhazia, notes that elections are competitive affairs and that there have been multiple peaceful transfers of political power, and O’Loughlin, Kolossov, and Toal (2014) observe that a number of de facto states are politically freer than their parent states. Some scholars, however, dispute the quality of these democratic structures and liken them to ‘empty shells’, developed only for the benefit of international observers and de facto state populations that have come to accept the principle that democratisation will lead to recognition (Smolnik, 2012). Indeed, Caspersen (2011, p. 338) describes how democracy has become a ‘buzzword’ in de facto states in that everyone recognises its importance, but no one knows what it means or how to implement it. Many de facto states seemingly attempt to ‘have their cake and eat it’ by presenting a democratic and liberal image to the world but maintaining a firm grip on power behind closed doors (Popescu, 2006a; Blakkisrud & Kolstø, 2011),20_{often by using their histories and}

19 This figure was significantly higher for the ethnic Abkhaz population. 20 Blakkisrud and Kolstø (2011, p. 205) label this a ‘Potemkin democracy’.

(24)

recognition as an excuse to restrain democratisation, while still being able to claim that they possess only democratic intentions. Thus, while Tansey (2011) demonstrated that the lack of recognised sovereignty does not prohibit the emergence of democracy, it does create significant barriers (Caspersen, 2012b).

(25)

3. Theory

Below I detail the six hypotheses that I theorise have an impact on state-building success and explain my reasoning for their inclusion in the study. They are separated into hypotheses that are based on the conditions in de facto states before their emergence (hypotheses 1 and 2) and hypotheses that are based on post-emergence conditions, which then fluctuate throughout the de facto state’s existence (hypotheses 3, 4, and 5). Hypothesis 6 includes measures from each group and is therefore listed in a separate category. While these divisions are by no means concrete, given that initial conditions can be expected to affect post-emergence structures and that some initial conditions are only observable post-emergence (and may thus potentially have been affected by other factors), they form a helpful way to conceptualise my hypotheses. For the ease of the reader these six hypotheses are listed in Table 1.

3.1 Linking Theories to the Existing Literature

Much of the literature summarised in the previous chapter is markedly one-dimensional in its acceptance of non-recognition as the defining characteristic of de facto states. It often disregards the fact that de facto states are heterogeneous structures which differ substantially, not least in terms of average income, population size, and state age – characteristics which interact with non-recognition to produce a diverse range of outcomes. In an attempt to change this prevailing discourse, this chapter develops theories about how pre-emergence conditions and the fluctuating circumstances within de facto states interact with the inherent burden that is faced by all of these entities – non- or partial-recognition – to affect state-building success.

The hypotheses that I have selected serve two fairly distinct purposes, which differ depending on whether they measure the effect of initial conditions or ongoing changes. Those which measure initial conditions relate eminent theories within political science to the specific environment of de facto statehood in order to test whether these variables exhibit the same effect within the context of non-recognition (hypotheses 1, 2, and 6 in Table 1). The others, which measure how changes within and across de facto states affect state-building, draw heavily on the existing de facto state literature by selecting hypotheses that test a current debate within the field (hypotheses 3, 4, and 5 in Table 1). The intention here is to test the validity of claims made within the literature regarding the reasons why de facto states behave the way they do. Most theories of de facto state behaviour are based upon the assumption that actions regularly taken by governments improve (or are at least perceived to improve) de facto states’ chances of survival or recognition, and employ state-building success as a mechanism of this theory. In other words, by testing the link between variables that reflect how de facto state societies orientate themselves and state-building success, I indirectly test the validity of common strategies for survival or recognition, thereby analysing the initial part of the causal mechanism that is largely ignored.

(26)

(27)

3.2 Theoretical Frameworks and the Statement of Hypotheses

This section explains the theorised causal mechanisms between state-level characteristics and state-building success. In doing so it attempts to justify the inclusion of each hypothesis and lend the study conceptual weight. I develop the theories behind each in turn, beginning with those that focus on the pre-emergence conditions and then moving on to the theories that involve post-emergence environments as explanatory factors.

3.2.1 Differences of Identity

Differences of identity between parent states and secessionist minorities are a fundamental part of what stimulates the formation of de facto states and are thus present in some form across all of our cases. Equally, all de facto states claim their right to separated statehood at least partially on the basis of ideological, cultural, ethnic, and religious differences. However, both the extent and character of these differences are not equal across de facto states, with some claiming strong differences across all four of these categories, and others just a single one (Kolossov & O'Loughlin, 1998).

Previous studies have sought to test whether the type and degree of difference have important implications for the likelihood of pursuing secession and even a de facto state’s chances of acquiring external recognition. However, I hypothesise that differences also influence the state-building process. Differences between the parent and the de facto state form an important part of the discourse within de facto states and help to convince populations that if they were to be re-absorbed they would experience severe repression and marginalisation at the hands of the parent state majority – a sentiment that is aided by a general demonisation of de facto state populations within the parent state media and governance circles (Caspersen, 2012b, 2015). Differences lend credibility to the elite’s claims that the alternative to maintaining their current position is to live as second-class citizens within a reunified parent state, and that their only rational choice is to lend their full support to state-building.

Intuitively, then, the more different a de facto state population is from the population of its parent state, the more weight these claims should possess, such that individuals within de facto states very dissimilar from their parent state can be expected to develop a stronger demand for state-building, so as to protect their communal interests. Thus, in this context state-building is more likely to be driven by the people, rather than by the elites – a situation that should produce better results. De facto state populations under the impression that the survival of their identity is dependent upon the survival of their de facto state are more likely to sacrifice personal interests for the general struggle to maintain the status-quo. Public trust in the ideals and identity of the state lead to practical effects – individuals committing to the state by paying taxes, assenting to mechanisms of control, and supporting the state in

(28)

its endeavours (such as enlisting in the army) – which improve the resource base from which de facto states can develop institutional frameworks.

However, it is also not beyond reason that any relationship between differences and state-building is due to ethnic and religious differences compelling kin-states – those that share aspects of the same identity with de facto states – to involve themselves in the affairs of de facto states, thereby inducing a stronger state-building performance through the provision of goods and services in the pursuit of this goal. Yet, I do not predict this causal mechanism to be particularly strong for the simple reason that kin-state involvement usually has a strong under-current of realpolitik, wherein the self-determination struggles of the de facto states are used to the advantage of the patron state in their own nation- and state-building programmes (Caspersen, 2009). It seems likely that were they not kin-states they would still finance de facto kin-states in an attempt to weaken regional rivals and improve their relative standing, only less openly.21

This argument therefore generates the following hypothesis:

Hypothesis 1: De facto states that have populations that are substantially different from those

of their parent state are more successful state-builders than those with only marginally different populations

3.2.2 Past Experience of Self-Rule

When de facto states declare their independence from their parent state and proceed to rule over their newly-proclaimed state they are rarely completely ‘new’ formulations. Most are rooted within some idea of a shared cultural and political heritage, and some have histories as independent states or autonomous provinces. These histories usually form important parts of de facto states’ claims to legitimate statehood, but they are also important in determining the starting-point from which they begin their state-building task. Previous experience of self-rule means a period of time in which legitimate state institutions may already have been built, many of which can be carried over to de facto statehood. This is even the case when autonomy has been revoked by the parent state years before the emergence of the de facto state, albeit perhaps to a lesser degree.

Although they do inevitably decay over time, institutions can prove remarkably durable even if they are not particularly prominent within societies or have been suppressed by external powers. Indeed, this has been the experience of many de facto states that underwent a withdrawal of autonomy or were invaded by stronger powers.

21_{By contrast, diasporas, which generally have more altruistic reasons for involvement, can be seen simply as an extended version of the}

What factors determine state-building success in urecognised de facto states?