Does recognition matter? Exploring patron penetration of de-facto state structures

ABSTRACT

De facto states seek international recognition. But do patron states that extend such recognition to their de-facto state clients treat them differently than they would have done otherwise? Will patrons, having extended official recognition, be more inclined to treat their clients as ‘regular’ states? Or does recognition only make the client more beholden to the patron? To investigate this, we explore whether granting recognition affects the degree of patron-state penetration of two sectors critical to the de facto state for retaining (internal) sovereignty and agency: the executive branch, and the military.

KEYWORDS:

• De facto states
• patron states
• patron–client relations
• recognition

1. INTRODUCTION

De facto states are states that have unilaterally seceded from another state, usually as a result of a civil war which they have won (Blakkisrud, 2023). Although they are in control of all or most of the territory to which they lay claim, and also otherwise may be said to fulfil the criteria of empirical statehood, they have failed to win international recognition (Caspersen & Stansfield, 2010; Kolstø, 2006; Pegg, 1998) – at best they are recognised by one or perhaps a handful of the UN member-states. De facto states are thus denied the protection enjoyed by even the smallest and weakest internationally recognised states (Jackson & Rosberg, 1982). Consequently, their security and continued survival typically hinge on the support of an external patron (Berg & Vits, 2020; Blakkisrud & Kolstø, 2012).

In legal and practical terms, de-facto state patron–client relationships display great variation. Some patrons have extended formal diplomatic recognition to their de-facto state clients, others have not. There is no universal agreement on how to delimit the universe of de facto states. According to Pål Kolstø (2006, pp. 725–726), secessionist entities must fulfil three criteria to qualify as de facto states: 1) their authorities must control (most of) the territory to which they lay claim; 2) they must have sought but not achieved international recognition; and 3) they must have existed as non-recognised states for more than two years. Other scholars have suggested more liberal criteria and cut-off points, allowing for more entities to qualify (see, e.g., Florea, 2020; Klich, 2021; Kursani, 2021).¹ To explore the impact of patron-state recognition, we have chosen to focus on the uncontested, clear-cut cases: those entities that would qualify according to all standard definitions of a ‘de facto state’.

Applying Kolstø's conservative definition, eight cases qualified as existing de facto states at the time of data collection (end of 2020). Seven of these enjoyed the backing of a patron state and thus qualify for our study of de-facto state patron–client relations.² Three have been recognised by their patrons: the Turkish Republic of Northern Cyprus (TRNC) was recognised by Turkey in 1983, and South Ossetia and Abkhazia by Russia in 2008.³ The remaining four, despite enjoying the backing of their respective patrons, had not achieved diplomatic recognition: Nagorno-Karabakh had, notwithstanding its very close ties with Armenia, failed to win patron recognition; likewise with Transnistria, the Donetsk People's Republic (DPR) and the Luhansk People's Republic (LPR) in relation to their patron state Russia.⁴

As lack of formal recognition is part of the definition of what constitutes a de facto state and what sets it apart from other states, patron-state recognition may be assumed to represent a crucial distinction – there is presumably a major difference between recognition by no other state and recognition by one, especially if that state happens to be the patron. But how does such recognition affect the patron–client relationship? In this article, we explore this question empirically.

As discussed in other contributions to this special issue, various factors may influence the degree to which the patron intervenes in the domestic affairs of its de-facto state client. In most cases, the correlation appears fairly straightforward: if, for example, the de facto state is populated by ethnic kin, this can be expected to facilitate patron-state penetration and the circulation of elites (Caspersen, 2007). Less easy to pinpoint is the impact of formal recognition. On the one hand, it may be expected to equip the client state with more independent agency. Having extended diplomatic status to its client, the patron will in principle be obliged to treat it as a full-fledged state entity, with which it exchanges ambassadors and signs international agreements. On the other hand, patron-state recognition may also reduce the de facto state's room for manoeuvre: formalisation of the relationship may lock the client into step with its patron (Cooley & Mitchell, 2010). Accordingly, we start out with two hypotheses:

1)	Patron-state penetration is lower in client states recognised by the patron.

Or alternatively.

2)	Patron-state penetration is higher in client states recognised by the patron.

To investigate this, we examine the seven cases identified above: three based on recognition (Abkhazia, South Ossetia, the TRNC) and four on non-recognition (DPR, LPR, Nagorno-Karabakh, Transnistria). Seven cases is not a large sample, but for comparative purposes these cases can yield meaningful insights as they, according to Kolstø's definition, represent the entire universe of de-facto state patron–client relationships existing at the time of our data collection.

2. INTERNATIONAL RECOGNITION: BENEFITS – AND DRAWBACKS?

Military victory in the secessionist struggle and territorial control are no guarantee for de jure recognition. Once the active phase of fighting is over, the issue of recognition – ‘the act by which another state acknowledges that the political entity in question possesses all the attributes of statehood’ (Taras & Ganguly, 2016, p. 38) – poses the next key challenge. Recognition ‘confers an international legal personality on a state and allows it to join the international states system with full rights, privileges, and duties’ (Taras & Ganguly, 2016, p. 38). Recognition will provide a basic security guarantee through the extension of the international norm against territorial conquest (Fazal & Griffiths, 2014) and thus plays a crucial role in securing the physical survival of the new state (Ker-Lindsay, 2012, pp. 17–18). Moreover, recognition is also associated with sundry financial and material opportunities, ranging from access to international aid, to benefits from participating in the global economy (Fazal & Griffiths, 2014; Krasner, 2001), further contributing to the sustainability of the state.⁵

Despite numerous attempts to outline and analyse the conditions under which a state should be granted legal (de jure) recognition, no systematic approach to this issue has been developed in international law (Ker-Lindsay, 2012; Neuberger, 1995). Recognition of a new state has depended largely on the parent state's approval and on ideological criteria – whether a ‘would-be state’ or its government ‘is likely to be friendly towards the recognizing party’ (Taras & Ganguly, 2016, p. 38).

In this era of globalisation and cultural and economic interdependence, there has been much talk about the erosion of the state and the increasing irrelevance of territory. Many argue that no state is totally sovereign (Berg & Kuusk, 2010; Krasner, 2001), although Kal Holsti (2004, p. 136) insists that ‘Sovereignty is an institutionalized legal or juridical status, not a variable’ and that it has nothing to do with, for example, the ability to control illegal trade, or being vulnerable to financial flows. However, the fact remains: a fully recognised state is more able to act independently than an unrecognised state.

It is usually assumed that, since de facto states seek external recognition of their self-determination claim, they also aspire to ‘true independence’ and ‘internationally recognised sovereignty’ (Neuberger, 1995, pp. 298–299). In an early contribution to the literature on Eurasian de facto states, Charles King (2001) noted that leaders of these states may prefer continued unrecognised status, as life in a legal limbo creates opportunities for smuggling and other illicit economic activities. Marcin Kosienkowski (2013) concurs, adding that the ‘nominal goal’ of seeking recognition may not be the ‘real’ motive of de facto states. Even if Transnistria has twice conducted referenda on ‘independence’ (1992 and 2006), the Transnistrian authorities may indeed be more interested in being annexed by the patron state. Continued unrecognised status may also provide them with ‘the broadest scope of power’, unfettered by the need to honour international agreements and laws (Kosienkowski, 2013, pp. 57–58).

Others have argued that rather than seeking full independence, de facto states may want to emulate the position of small, dependent overseas jurisdictions that enjoy the support of a powerful and wealthy patron (Comai, 2017). There is, however, a major difference between de facto states and microstates under great-power protection: with microstates, there are usually no other states harbouring territorial claims. Hence, most authors seem to accept the aspirations of the de facto state for international recognition as a legitimate sovereign state as genuine: such recognition remains ‘the top prize’ (Caspersen, 2018).

A litmus test for assessing whether a proclamation of independence reflects ‘real’ desires in a de facto state may be whether the leaders are willing to jeopardise what they have achieved in exchange for a declaration of independence, that is, risking losing more than they gain, but still push forward (cf. the 2017 referendum in Iraqi Kurdistan). In other situations, proclamations of independence may indeed be a bargaining chip in complicated negotiation games, to be played or retracted depending upon the circumstances. Thus, for instance, the TRNC renounced this ambition in 2004 when Kofi Annan was hammering out a plan for a bicommunal federation on the island, but reactivated it when the Greek Cypriots rejected this plan.

3. SINGLE-STATE RECOGNITION: A PARTLY FULL OR AN ALMOST EMPTY GLASS?

In a discussion of contested statehood, Deon Geldenhuys (2009, p. 25) distinguishes between six levels of recognition among states without a seat in the UN General Assembly: titular, partial, paltry, patron, peer and zero. ‘Titular’ recognition refers to entities that are widely seen as having the right to international recognition but are constrained by external powers (e.g., Western Sahara). ‘Partial’ and ‘paltry’ refer to being recognised only by some UN member states, with ‘paltry’ being reserved for those recognised by ‘only a handful of existing countries’ (Geldenhuys, 2009, p. 25). ‘Patron’ is an example of single-state recognition, while ‘peer’ is limited to recognition by other equally unrecognised entities only. Geldenhuys would define the TRNC as a case of ‘patron recognition’ while South Ossetia and Abkhazia would fall into the category of ‘paltry recognition’, being recognised by Russia as well as by a sprinkling of other states. While being recognised by Nauru or Venezuela may provide opportunities for state visits to exotic places, in the world of Realpolitik their recognition does not amount to much, hence we will focus on ‘patron recognition’.

In the literature on international recognition, the issue of single-state recognition has not garnered much interest; it hardly registers on the radar of leading scholars (see, for instance, Fabry, 2010). However, with de-facto state patron–client relations, things may be different. After all, by breaking with the international consensus, a patron state that extends diplomatic recognition to a secessionist entity sends a clear signal that it intends to treat this entity differently from what other states do. It is no coincidence that Russian recognition of Abkhazia in 2008 was greeted with wild jubilation in the streets of Sukhumi; and it is also not surprising that such recognition has been fought tooth and nail by parent states (Ker-Lindsay, 2012; Pegg, 1998).

Do de facto states recognised by their patron enjoy greater sovereignty than those completely outside the international society of states? Opinion is divided. Regarding Abkhazia and South Ossetia, Alexander Cooley and Lincoln Mitchell (2010, p. 60) have argued that, after being recognised by Russia in 2008, these two entities actually experienced a loss of sovereignty: ‘perhaps ironically, the territories have gone from enjoying de facto independence as unrecognized states (…) to becoming almost de facto parts of the Russian Federation in their new status as “independent states”’. By contrast, Thomas Ambrosio and William Lange (2016, p. 674) maintain: ‘this perspective understates their separate state- and nation-building processes, their capacity for independent agency, and their ability to garner concessions from Russia during bilateral negotiations’.

Nina Caspersen’s seminal study of de facto states, Unrecognized states (2012), highlights the fact that these entities lack international recognition. She subsumes states with ‘paltry’ recognition under the catch-all umbrella of ‘unrecognized states’ – making no distinction between patronage with or without recognition. Exploring the relations between unrecognised states and their patrons is an important part of Caspersen's study, but she focuses on substantial, not formal, links between patron and client. She thus discusses Republika Srpska Krajina, Nagorno-Karabakh, Abkhazia and South Ossetia in the same breath, implicitly saying that these patron–client relationships have basically the same character, irrespective of recognition/non-recognition.

One of few researchers to address this distinction explicitly is Daria Isachenko (2012), who, in a paired study of state-building in the TRNC and Transnistria, notes a ‘major difference’ between the two cases: Turkey officially recognises the TRNC, whereas Russia officially supports the territorial integrity of Moldova. Even so,

despite the differences of context, the striking similarity in the two cases is the geopolitical element, which dominates the perception of Ankara and Moscow regarding their attachment to these informal states. (Isachenko, 2012, p. 131)

She finds that although Russia does not have an embassy in Transnistria, the country's involvement in the Transnistrian state-building project shares similarities with the Turkish case (Isachenko, 2012, p. 141). Thus, the ‘major’ formal difference does not necessarily seem to have major consequences.

To explore the possible impact of patron-state recognition, we compare patron-state penetration of the executive branch and military structures across our sample of de facto states.

4. HOW TO MEASURE PATRON-STATE PENETRATION?

Sandra Kamilova and Eiki Berg (2011) have developed an index to categorise degrees of de facto statehood, based on relations with the patron. They identify four dimensions of patron-state dependence: economic dependence, patron-state influence on state structures, loyalty to the patron state among the de-facto state population, and the patron state's interests. As we are interested in how formal recognition affects state penetration in the de-facto state client, here we focus on the patron's influence on state structures.

Kamilova and Berg operationalise such influence through four variables: ‘representatives of other origin in state institutions’, ‘influence on decision-making process’, ‘democracy level’, and ‘military presence’ (2011, pp. 158–160). We view ‘influence on decision-making process’– which they operationalise as ‘level of influence’ on a scale from one to zero – as the potential outcome of patron-state penetration, not an independent variable. Further, as demonstrated by Pål Kolstø and Helge Blakkisrud (2021), patron states are not necessarily engaged in exporting or imposing their regime type on their clients. Thus, we drop this variable and focus on patron-state penetration of state institutions and the armed forces.

In operationalising these two variables, we have slightly revised Kamilova and Berg's indicators. First, concerning patron representatives in state institutions, we trace patron-state penetration of the executive (not the legislative) branch. We consider this a more pertinent indicator, given the tendency of de facto states to adopt presidential/semi-presidential regimes (Ó Beacháin, 2017), with legislatures generally weak and marginalised in the decision-making process.⁶ Second, whereas Kamilova and Berg use ‘proportion of total defense forces’ as an indicator for ‘military presence’, we find this too crude an indicator, and, drawing on Gražvydas Jasutis (2014), measure ‘military presence’ as the presence of patron-state bases and military units on client-state territory, and the degree of integration of patron-state and client-state military structures (see Table 1).

Table 1. Patron-state penetration of state structures in de facto states.

Independent variables	Operationalisation	Indicators
Representatives of patron in state institutions	Proportion of ministers in the government	Share of ministers and deputy ministers who were born and/or spent more than half their lives in the patron state before being appointed to the de-facto state government
Military presence	Presence of patron-state bases and military units on client-state territory	Military bases [yes/no] Number of patron-state troops stationed on the territory of the client versus the size of the standing forces of the de facto state
	Degree of integration of patron-state and client-state military structures	Formal agreement on military cooperation [yes/no] Joint military training [yes/no] Joint military units [yes/no] Joint command structure and headquarters [yes/no]

Sources: Drawing on Kamilova & Berg, 2011 and Jasutis, 2014.

We start by examining patron-state penetration of the three cases in which the patron has formally recognised its de-facto state clients: Abkhazia, South Ossetia and the TRNC. The longevity of formalised state-to-state relations, 40 years in the case of the TRNC versus 15 for Abkhazia and South Ossetia, may well affect the quality and intensity of such relations. However, as we are going to compare patron-state penetration of these states with penetration of unrecognised states, we decided to use the same measuring point for all cases, comparing the situation across all seven cases as of the end of 2020.⁷ Moreover, to capture the within-case effect of patron-state recognition, we needed to establish a baseline in order to be able to compare penetration before and after recognition. Here we opted for measuring penetration four years prior to recognition as compared to the end of 2020.⁸ To contextualise the data,⁹ we add a discussion of the development trajectories of patron–client relations over time in each case.¹⁰

5. PATRON–CLIENT RELATIONS AFTER RECOGNITION

5.1. Abkhazia

Abkhazia's war of secession (1992–1993) resulted in a major uprooting of the population, with ca. 250,000 (mainly ethnic Georgians) of a pre-war population of 525,000 being internally displaced. In 1993, Russia assisted in brokering a ceasefire; in the following summer, Moscow provided soldiers to what was officially a Commonwealth of Independent States (CIS) peacekeeping mission along the de facto border between Abkhazian-controlled territories and the rest of Georgia.

In the first post-war years, at least officially Moscow leaned more towards Georgia than Abkhazia, imposing a blockade on the secessionists (Blakkisrud et al., 2021, p. 352). Only from the turn of the millennium did Russia gradually assume a more formal role as Abkhazia's patron, a process that culminated in the 2008 Russo-Georgian war, when Russia recognised Abkhazia as an independent state. Since then, Abkhazia has gradually – albeit often reluctantly – been drawn closer into Moscow's embrace, signing a series of agreements on closer cooperation and homogenisation of legislation and administrative procedure (Ambrosio & Lange, 2016).

5.1.1. Representatives of patron in state institutions

Although it took some time before Russia sided unequivocally with Abkhazia, Russian citizens have played an active role in Abkhazian politics ever since the onset of the conflict. A prominent example is Sultan Sosnaliev, who, as head of the military department of the ‘Confederation of Mountain Peoples of the Caucasus’, played a central role in the Abkhazian fight for independence. He went on to serve as Abkhazian Minister of Defence 1993–1996 and then again 2005–2007, during his second term doubling as Deputy Prime Minister (Regnum, 2007).

With Russia gradually becoming more involved, dynamics shifted from private initiatives to more state-backed supply of cadres – especially to the local security apparatus, including law enforcement, intelligence and military institutions (see, e.g., Komakhia, 2017; Popescu, 2006). One example is Lieutenant General Anatolii Khrulev, who served as commander of the Russian 58th Army in the August 2008 Russo-Georgian war. Upon retiring from the Russian armed forces, he was in 2015 appointed First Deputy Minister of Defence and Chief of the General Staff of Abkhazia.¹¹ Three years later, he was succeeded as Chief of the General Staff by Vasilii Lunev, a Russian Major General still in service, who had already served as South Ossetian Minister of Defence – an interesting example of cadre rotation between the patron and these two clients (Komakhia, 2017).

Table 2. Patron-state penetration of Abkhazia.

		Four years prior to recognition	As of the end of 2020
Representatives of patron in state institutions	Share of ministers	2/14	6/14
	Share of deputy ministers	Incomplete data	4/25
Military presence	Bases	Yes*	Yes
	Patron versus de-facto state troops	1600*/2200	4000/2200
	Degree of integration	N/A**	Formal agreement on military cooperation: yes Joint military training: yes Joint military units: yes Joint command structure and headquarters: no

Notes: *Officially: CIS peacekeepers. **Due to non-recognition, formal integration not applicable.

Sources: In compiling Tables 2–7, we have relied on a range of different sources – official websites, local newspapers, analytical reports, databases, etc. – in order to obtain and crosscheck data. We draw on sources published in English, Georgian, Russian and Turkish. The main sources for each table are listed in the online supplemental data.

Russia has also been active in staffing the Abkhazian Ministry of Internal Affairs. In 2017, after several attacks on Russian tourists in Abkhazia, Moscow exerted significant political pressure on Sukhumi, and Garri Arshba was appointed as Minister of Internal Affairs (Komakhia, 2017). Arshba is an example of a local who made a career in Russia before returning to Abkhazian politics: at the time of his appointment, he was working for the local branch of the Russian Ministry of Internal Affairs in the North Caucasus Federal District (Kavkazskii uzel, 2017). This shows that patron influence may operate in various forms – not only as parachuting in what Christofer Berglund and Ketevan Bolkvadze (2024) refer to as ‘servants of the empire’, but also through cadre rotation between client and patron. In our sample we find an increase of such ‘servants of empire’ over time (see Table 2).

Table 3. Patron-state penetration of South Ossetia.

		Four years prior to recognition	As of the end of 2020
Representatives of patron in state institutions	Share of ministers	0/19	2/12
	Share of deputy ministers	1/19	2/18
Military presence	Bases	Yes*	Yes
	Patron versus de-facto state troops	450*/1250	4000/1250
	Degree of integration	N/A**	Formal agreement on military cooperation: yes Joint military training: yes Joint military units: yes Joint command structure and headquarters: yes

Notes: *Officially: the Russian battalion of the trilateral Joint Peacekeeping Force. **Due to non-recognition, formal integration not applicable.

Sources: see online supplemental data.

5.1.2. Military presence

With the 2008 Russo-Georgian war, the CIS/Russian peacekeeping mission was abrogated. Two years later, Moscow agreed with the Abkhazian authorities to formalise the Russian military presence by taking over the military base in Gudauta. Some 4000 Russian troops are stationed at this base (Menabde, 2019).¹²

A further step towards integration of the armed forces came in 2014 with the signing of a bilateral Alliance and Strategic Partnership agreement that envisioned a ‘common space of defense and security’ (Ambrosio & Lange, 2016, p. 684). The treaty obliged Russia to engage in the modernisation of Abkhazian armed forces, to equip Abkhazia with modern weapons and the ‘phased unification of the standards of military command, logistics, cash allowance and social guarantees of military personnel’ (Menabde, 2019). However, the initial idea of creating ‘a common defense infrastructure’ under Russian command was dropped, on Sukhumi's insistence (Ambrosio & Lange, 2016, p. 684).

Russia has also engaged in securing and protecting the borders of the breakaway region. In April 2009, the two parties signed an agreement on joint efforts to protect the Abkhazian state border. This paved the way for introducing the Border Police Department of the Russian FSB in Abkhazia.

5.2. South Ossetia

Simmering tensions among ethnic Georgians and Ossetians in South Ossetia led to armed conflict in early 1991. The war lasted until June 1992, when a Russia-brokered ceasefire resulted in the introduction of a trilateral peacekeeping operation consisting of Russian, Georgian and South Ossetian troops.

As in the case of Abkhazia, Russia did not immediately assume patronage. However, unlike the Abkhazians, the Ossetians were clearly oriented towards integration with Russia: their goal was to unite with their ethnic kin in Russian North Caucasus – and that would require becoming part of the Russian Federation (Blakkisrud & Kolstø, 2012).

The South Ossetian secessionists were also on relatively good terms with the central authorities in Tbilisi. Hence, instead of economic blockade, the main crossing between South Ossetian-controlled territories and the rest of Georgia soon became a giant market for smuggled goods. This all changed with Georgia's 2003 Rose Revolution and President Mikheil Saakashvili's ambition to reintegrate the breakaway territories, which, in the case of South Ossetia, led to the August 2008 Russo-Georgian war.

After the war, Russia shed all pretences of being a neutral negotiator and recognised its client as an independent state. Since then, South Ossetia has become increasingly integrated with its patron across all sectors of society – but Russia has not yet accommodated repeated calls from South Ossetian politicians for formal unification (Kolstø & Blakkisrud, 2021).

5.2.1. Representatives of patron in state institutions

South Ossetia's state institutions are considered even more susceptible to patron-state penetration than those of Abkhazia (Berglund & Bolkvadze, 2024; Popescu, 2006). All presidents have been locals, but at the time of Russian recognition, three prime ministers in a row hailed from Russia – Yurii Morozov (2005–2008), Aslanbek Bulatsev (2008–2009), and Vadim Brovtsev (2009–2011) – none of them with any previous connection to South Ossetia (Gerrits & Bader, 2016). They were succeeded by Rostislav Khugaev (2012–2014), who was born in South Ossetia but had spent some 40 years in Russia before taking up the premiership. Since then, however, all subsequent prime ministers have been local.

The case of Brovtsev demonstrates that parachuting in patron representatives is not always easy, even in the most pliant of de facto states: Brovtsev was the protégé of Russian Minister for Regional Development Viktor Basargin, who was responsible for financing post-conflict reconstruction in South Ossetia. However, soon after his arrival, the local media launched a smear campaign, accusing Brovtsev and his staff of inefficiency and embezzlement of budget funds. In the end, President Putin had to intervene, ordering South Ossetian president Eduard Kokoity to halt the harassment (RFE/RL, 2010).

As in Abkhazia, there has been considerable patron penetration of the security structures. For example, during Kokoity's presidency (2001–2011), South Ossetia had no less than eight defence ministers, six of whom were Russian career officers (Kommersant, 2010). Berglund and Bolkvadze (2024) found a rapid increase of patron penetration from 2003 and until official recognition in 2008, whereafter it returned to a low but stable level. We note something similar: Four years prior to Russia's recognition, there was only a single patron-state representative in the South Ossetian government (a deputy minister). By 2020, there were four, including Viktor Fedorov, first deputy minister of defence and chief of the General Staff (see Table 3). Beyond the security sector, locals still seemed to dominate the top positions.

5.2.2. Military presence

The 2008 August war spelled an end to the trilateral peacekeeping mission. In April 2010, Russia signed an agreement on establishing a military base in South Ossetia on similar terms as in Abkhazia (49 years, with the possibility of automatic extension every 15 years). The 4th Guards Military Base has detachments in Tskhinvali and Java, with some 4000 troops.

In March 2015, Russia and South Ossetia signed an Agreement on Alliance and Integration. The slight difference in wording as compared to the Abkhazian equivalent is not accidental (in the Abkhazian case, ‘integration’ was replaced by ‘strategic partnership’), but reflects the balance of power between the pro-independence and pro-Russian forces in the two societies (Ambrosio & Lange, 2016). According to the agreement, the South Ossetian armed forces and security services were to be integrated into the relevant structures of the Russian military and security forces. In March 2017, this process commenced with the incorporation of several South Ossetian military units into the Russian armed forces (Tass, 2018).

Moreover, since 2009, Moscow has also assumed responsibility for South Ossetia's de facto border, with Russian guards patrolling and securing the separation line with Georgia proper. In that capacity, Russian troops have engaged in a contested ‘borderization’ process, pushing the de facto border further into Georgian territory (Toal & Merabishvili, 2019).

5.3. Turkish Republic of Northern Cyprus

In 1974, after a Greek Cypriot coup d’état aimed at uniting Cyprus with Greece, Turkey intervened on behalf of the Turkish Cypriot population, invading present-day Northern Cyprus. The conflict led to a wave of displacement – over 150,000 Greek Cypriots were evicted to the southwest, while some 50,000 Turkish Cypriots were displaced to the Turkish-controlled northeast – resulting in the partition of the island (Gürel et al., 2012).

In 1975, the Turkish part declared the establishment of the Turkish Federated State of Cyprus (TFSC), attempting to create a federation with the Greek Cypriots. In October 1983, however, the local authorities went for full independence as the Turkish Republic of Northern Cyprus (TRNC).

Turkey immediately recognised the TRNC as an independent state. No other states have followed suit, making the TRNC heavily dependent on Turkey in terms of security and economic assistance. A sense of growing subordination and inequality has, however, weakened popular support of patron influence in this de facto state, especially regarding the privatisation of TRNC economic entities and the growing presence of Turkish companies (Dayioğlu et al., 2021, pp. 451–452).

5.3.1. Representatives of patron in state institutions

With the patron representing ethnic kin, a high degree of elite circulation might have been expected between Turkey and the TRNC. However, the questions of whether the TRNC should be considered as a junior partner or equal in relations to Turkey, and whether the locals are ‘Turks’ or ‘Turkish Cypriots’, have remained contested dividing lines in TRNC politics. The TRNC's founding father, long-term president Rauf Denktaş (1983–2005), dismissed the idea of a Cypriot identity, claiming that the local population was thoroughly Turkish (Dayioğlu et al., 2021, p. 454). However, his successor as president, Mehmet Ali Talat (2005–2010), held that ‘Turkey may be some Cypriots’ ancestors’ land but it cannot be their motherland’ (Dayioğlu et al., 2021, p. 456).

Since 1974, there has been a massive influx of immigrants from Turkey. Today, Turkish immigrants and their descendants are thought to comprise almost half the population of the TRNC (Hatay, 2017), but politics remain dominated by ‘sons of the soil’: All five presidents of the TRNC since 1983 have been locals, although most of them have received higher education in Turkey – and in some instances, also worked there for some time. Current president Ersin Tatar (2020–), for example, after having graduated, spent some 10 years working in the Turkish business sector before returning to the TRNC. The same is true for the long string of prime ministers who have served since the declaration of independence.

As of late 2020, despite the heavy presence of Turkish immigrants in the TRNC, the then newly-appointed government of Ersan Saner was composed of homegrown politicians: only two out of 23 ministers and undersecretaries were born in Turkey (see Table 4). Moreover, these two, while coded here as ‘patron-state representatives’ due to place of birth, had arrived in the TRNC with their parents during the first wave of Turkish migration after the division of the island, and thus had strong local credentials.

5.3.2. Military presence

Ever since the 1974 invasion, Turkey has maintained a military garrison in the TRNC, the Cyprus Turkish Peace Force Command in Northern Cyprus. According to Ankara, there are currently about 40,000 Turkish troops stationed there (Ertan, 2022). In addition, Turkey holds annual military exercises on and off the island. Ankara views this military presence on Cyprus as strategically important to Turkey. As noted by Bülent Ecevit, Turkish prime minister at the time of the 1974 Turkish invasion, ‘Anyone who takes a look at the map (…) will see how crucial Cyprus is [for Turkey]. The security of our southern shores, ports, and future pipelines depends on our military presence in the “TRNC”’ (cited in Dayioğlu et al., 2021, p. 457).

The TRNC has, in close cooperation with Turkey, set up its own armed forces, the Security Forces Command, with an estimated 8000 men. However, not only is this dwarfed by the patron-state presence, the TRNC forces are also under the command of a general from the Turkish Army, and the military presence of the patron state is perceived as the ultimate guarantor of the security of the Turkish Cypriots (Ministry of Foreign Affairs, 2011).

Table 4. Patron-state penetration of the TRNC.

		Four years prior to recognition	As of the end of 2020
Representatives of patron in state institutions	Share of ministers	N/A*	2/11
	Share of deputy ministers	N/A*	0/12
Military presence	Bases	Yes	Yes
	Patron versus de-facto state troops	40,000/0	30-40,000/8000
	Degree of integration	N/A**	Formal agreement on military cooperation: yes Joint military training: yes Joint military units: no Joint command structure and headquarters: yes

Notes: *As Turkey recognised the TRNC on the same day as the local authorities in the TFSC declared independence, there is no equivalent to the de-facto state government prior to recognition. **TFSC had not declared independence, and military integration was thus not on the table.

Sources: see online supplemental data.

6. COMPARING PATRON PENETRATION IN UNRECOGNISED DE FACTO STATES

Thus far, we have examined the impact of patron-state recognition on de-facto state patron–client relations by comparing the situation before and after recognition. How does post-recognition patron-state penetration compare to the situation in de facto states where the patron has not extended formal recognition? To explore this, we compare our three cases of patron-state recognition with four cases of de facto states that as of late 2020 enjoyed strong patron-state support, but de jure remained fully unrecognised: the Donetsk and Luhansk People's Republics (DPR and LPR), Transnistria and Nagorno-Karabakh.

6.1. Donetsk and Luhansk People's Republics

In the wake of Euromaidan and the February 2014 regime change in Kyiv, pro-Russian forces mobilised in the easternmost parts of Ukraine. In May they organised referenda on ‘the declaration of state independence’, followed by the promulgation of the Donetsk People's Republic (DPR) and the Luhansk People's Republic (LPR) as independent states. Although the Russian president officially requested the separatists to postpone the vote – and thus the declaration of independence – Russia, as the future patron, was involved in the conflict from the very start.

With the declaration of independence, the new de facto authorities started setting up state structures (Kasianenko, 2019). However, the two entities remained heavily dependent on Russian support, and were gradually drawn into Moscow's embrace. In 2017, for example, the Kremlin decided to recognise various types of documents issued by the DPR and LPR authorities, including ID cards, diplomas and vehicle registration, thus granting residents of the DPR and LPR the right to travel, study and work in Russia. Two years later, Moscow enabled bearers of locally issued passports to obtain Russian citizenship through a fast-track procedure.

As of the end of 2020, when we conducted our cross-case data collection, the DPR and LPR seemed to be aiming less for independence and more for integration with Russia. In 2017, the head of the DPR Aleksandr Zakharchenko (2014–2018) expressed hopes that the two de facto entities would soon ‘reintegrate’ with Russia; in 2018, the head of LPR, Leonid Pasechnik (2017–2022) stated that the goal was to achieve a ‘strong and independent Republic (…) in a single political, cultural, economic and civilizational space with the Russian Federation’ (cited in Kasianenko, 2019, p. 122). However, Russia as patron did – as in the case of South Ossetia – push for integration, but flatly rejected all calls for formal annexation. That changed in 2022: first, on the eve of its full-scale invasion of Ukraine, Russia decided to recognise DPR and LPR as independent states; then, in September, it welcomed the two as new federal subjects.

6.1.1. Representatives of patron in state institutions

Patron-state representatives were plentiful in the first governments set up by the secessionists. For example, the first head of the new government of DPR was Aleksander Borodai, a Russian citizen who, inter alia, had participated in the secessionist war in Transnistria in 1992. He was joined by fellow veteran from the Transnistrian war – and Russian citizen – Igor Girkin (also known as Igor Strelkov), a former FSB officer, who became Minister of Defence. Several prominent field commanders and warlords, including Igor Bezler and Arsen Pavlov (better known by his nom de guerre Motorola), were also Russian citizens (Wilson, 2016, p. 648).

Also in the LPR, the first head, Valerii Bolotov, was Russian-born. He had, however, grown up in Luhansk and thus held strong ‘local’ credentials. Whereas there were examples of parachuting in patron-state representatives also in LPR, Russia as a patron was keen to present the secessionist conflict in Donbas as home-grown. When several members of the first cohort of government officials were killed in a series of assassinations that shook the DPR and LPR, Moscow supported an ‘indigenization’ of the leadership with several infamous Russian leaders now being replaced by locals with less radical reputations. With unruly figures replaced by more pliant ones, the patron could more effectively control decision-making processes in the two clients (Fischer, 2019).

As of the end of 2020, despite the ‘indigenization’, there was still a visible patron-state presence in the DPR government, including Deputy Prime Minister Vladimir Pashkov, a former deputy governor of the Irkutsk Oblast in Russia. Out of 20 ministers, four were Russian citizens, and two others were born in Russia (see Table 5). In the LPR, six out of 19 ministers were Russian citizens, and a seventh member was born in Russia (see Table 5).

Table 5. Patron-state penetration of DPR and LPR.

		As of the end of 2020
		DPR	LPR
Representatives of patron in state institutions	Share of ministers	4/20	7/19
	Share of deputy ministers	Incomplete data	Incomplete data
Military presence	Bases	No	No
	Patron versus de-facto state troops	Incomplete data*	Incomplete data*
	Degree of integration	Formal agreement on military cooperation: no Joint military training: no Joint military units: no Joint command structure and headquarters: no	Formal agreement on military cooperation: no Joint military training: no Joint military units: no Joint command structure and headquarters: no

Note: *Estimates vary greatly, no official information available.

Sources: see online supplemental data.

Still, the patron seemed to prefer to channel its influence on the decision-making processes in the de facto states through a system of ‘curators’ (kuratory), patron-state officials responsible for coordinating and overseeing patron-state interests (Isachenko, 2019).¹³ In the DPR and LPR, Vladislav Surkov – presidential assistant and former deputy head of the Presidential Administration – held this position 2014–2020, before being replaced by another long-time Putin confidant, Dmitrii Kozak.

6.1.2. Military presence

In addition to political support, DPR and LPR also depended on Russia in the military sphere. In 2014, both de facto states created ‘people's armies’ – military units set up to control the territory within the de facto borders of DPR and LPR. These armies consisted of volunteer paramilitary forces and cooperated with international volunteers (Kasianenko, 2019, p. 123). As of our measuring point (end of 2020), the exact numbers were not publicly available. Local leaders openly admitted that the international volunteers included thousands of Russian citizens, hereunder professional soldiers, who chose to ‘spend their holidays not on the beach but shoulder-to-shoulder with their brothers, fighting for the freedom of Donbass’ (BBC, 2014).

The emerging military forces of the two separatist entities were totally dependent on Russian weapons, equipment and military training. Concerning Russian participation, although Putin had admitted in 2015 that there were Russian military personnel in Donbas (ICG, 2016), the lack of official recognition of the client states and the ongoing fighting along the front made Moscow not very forthcoming with information on its military presence. Estimates range from a few thousand to upwards of 40,000 Russian troops (see, e.g., Gramer, 2017).

6.2. Nagorno-Karabakh

The secessionist conflict in Nagorno-Karabakh predates the break-up of the Soviet Union. In 1988, local authorities in the Nagorno-Karabakh Autonomous Oblast appealed to Moscow to have the territory transferred to the Armenian SSR. The request resulted in communal clashes and ethnic violence in Nagorno-Karabakh and beyond. With the dissolution of the Soviet superstructure and Nagorno-Karabakh's subsequent declaration of independence in 1992, the conflict evolved into a war of secession. To defend ethnic kin, Armenia sent troops into the region. The result was one of the bloodiest conflicts in connection with the break-up of the Soviet Union, with an estimated 25,000–30,000 casualties (de Waal, 2013, p. 326) and about 1 million IDPs. The war spilled across the borders of Nagorno-Karabakh proper; by the time of the Russia-brokered 1994 ceasefire, Karabakhi and Armenian forces had partly or fully occupied seven adjacent previously Azerbaijani-populated regions.

The close relations between Nagorno-Karabakh and its patron Armenia were enshrined in the Karabakhi constitution: ‘The Republic of Artsakh [Nagorno-Karabakh] shall implement a policy aimed at political, economic and military cooperation and ensuring comprehensive ties and security with the Republic of Armenia’ (Constitution of the Republic of Artsakh, n.d.). However, Armenia refrained from recognising its client. In connection with the – from a Karabakhi perspective – disastrous 2020 Second Karabakh War, Prime Minister Nikol Pashinyan broached recognition as a potential countermeasure to shore up the rump de facto state (Prime Minister, 2020), but after the 2023 Azerbaijani Karabakh offensive, this was no longer an option: Suffering military defeat, the Karabakhi authorities were forced to dissolve the de facto state.

6.2.1. Representatives of patron in state institutions

The close ties between Nagorno-Karabakh and its patron were reflected in the extensive circulation of elites between Stepanakert and Yerevan. Many ministerial posts and other high-profile positions have been held by former Armenian officials (Kopeček, 2017). However, a unique feature of Armenian-Karabakhi patron–client relations is the reverse trend that can be observed here: a client penetration of patron-state politics. For two decades, the Armenian presidency was held by former Karabakhi officials: Robert Kocharyan, who had started his political career as Prime Minister (1992–1994) and then President (1994–1997) of Nagorno-Karabakh, went on to serve two consecutive terms as President of Armenia (1998–2008). Similarly, his successor as President, Serzh Sargsyan (2008–2018), had played a central role in the secessionist war, serving as Minister of Defence, before moving on to ministerial positions in Yerevan that included two brief stints as Prime Minister.

It was the 2018 Velvet Revolution that broke the client state's hold on patron politics: Sargsyan fell victim to popular unrest and was replaced by a ‘local’ Armenian, Pashinyan. The 2020 Second Karabakh War further reduced the power and influence of the Karabakh clan. Despite the upheaval caused by this war, the political elite in Stepanakert still included a nominal share of persons who either hailed from or had spent most of their careers in the patron state before taking up ministerial positions with the client. As of the end of 2020, two of the ten ministers were identified as patron-state representatives (see Table 6).

Table 6. Patron-state penetration of Nagorno-Karabakh.

		As of the end of 2020
Representatives of patron in state institutions	Share of ministers	2/10
	Share of deputy ministers	0/3
Military presence	Bases	No
	Patron versus de-facto state troops	8500*/5500
	Degree of integration	Formal agreement on military cooperation: yes Joint military training: yes Joint military units: yes Joint command structure and headquarters: no

Note: * As part of the 2020 ceasefire agreement, Armenian armed forces were officially to withdraw. Exact numbers are disputed.

Sources: see online supplemental data.

6.2.2. Military presence

As noted above, the Karabakhi Constitution stresses the military and security cooperation with the patron. From the very onset of the conflict, this cooperation was extremely close. During the war of secession, regular Armenian forces fought side-by-side with the separatists. After the 1994 ceasefire, Nagorno-Karabakh commanded a standing army of almost 20,000 men, made up of Karabakhi and patron-state recruits (ca. 8500 and 10,000 men, respectively) (ICG, 2005). Hence, during the interwar period (1994–2020), although there were no patron-state bases as such, there was a high level of integration, making it difficult to distinguish between Karabakhi military structures and patron-state presence.

In its national security doctrines, Armenia has provided explicit security guarantees to its client. According to the version adopted only a few weeks prior to the Second Karabakh War, ‘The Republic of Armenia is the guarantor of the security and liberty of the Armenians of Artsakh’ (Government, 2020). Hence, the Nagorno-Karabakh Defence Army – backed by Armenian guarantees and the natural defences provided by the de facto state's mountainous terrain – was seen to hold a strong position vis-à-vis parent state forces.

All this changed in a matter of weeks during the 2020 Second Nagorno-Karabakh War. In the face of a massive Azerbaijani attack, Karabakhi defences crumbled. With Azerbaijani forces entering the outskirts of Stepanakert, Armenia had to sign a humiliating ceasefire agreement on behalf of the Karabakhis, according to which all Armenian forces were to withdraw from Nagorno-Karabakh, and Armenia was to halt its supply of armaments to its client (ICG, 2021). Instead, Russia deployed 1960 peacekeepers under the ceasefire agreement. The Armenian full withdrawal was officially completed in September 2022 (Mejlumyan, 2022), leaving the decimated client army – now consisting of an estimated 5500 troops – to fend for itself. In the 2023 Azerbaijani offensive, the latter proved no match for the Azerbaijanis.

6.3. Transnistria

As the Soviet Union started to implode, local leaders in Moldova's heavily Russified eastern parts began protesting what was perceived as an attempt by Chisinau to ‘Romanianize’ the republic. In September 1990 these leaders proclaimed the Transnistrian Moldavian Soviet Socialist Republic and pledged allegiance to the Soviet Union. Once the Soviet superstructure was gone, Moldovan authorities sought to re-establish control over the territories across the Dniester River with military means. A brief war ended in a Russia-brokered ceasefire in July 1992.

As Transnistria's patron, Russia has been involved in the secessionist conflict from the very beginning. The Russian 14th Army, stationed in Tiraspol, provided crucial support to the secessionists. Ever since the 1992 ceasefire, Russia has been part of the trilateral Russian-Moldovan-Transnistrian peacekeeping mission under the Joint Control Commission.

At times Transnistria and its patron appear to have had differing visions on the future of the de facto state. The Transnistrian authorities have been clear on their pro-Moscow orientation. Once the Soviet Union was no longer an option, they switched their allegiance to the Russian Federation. In a 2006 referendum, 98% of the votes cast supported Transnistrian independence and ‘possible future integration in the Russian Federation’ (Kolstø & Blakkisrud, 2017, p. 520). A 2010 survey revealed that nearly half of the population preferred integration with the patron, whereas only a third opted for full independence (O’Loughlin et al., 2013, p. 253).

Russia's 2014 annexation of Crimea raised new hopes in Tiraspol that Russia might also annex Transnistria (Ministerstvo inostrannykh del, 2015). However, Russia has vacillated. In the early 2000s, it even seemed like Moscow was more interested in establishing friendly relations with Moldova than in propping up the de facto state (Popescu, 2006). After Chisinau in 2003 turned down the Russian Kozak Memorandum on the basic principles of a reunited Moldova, Moscow has sided more unequivocally with its client – although officially still supporting its reintegration with the parent state (Kosienkowski, 2019).

6.3.1. Representatives of patron in state institutions

Nicu Popescu (2006) has discussed how Eurasian secessionist entities have ‘outsourced’ state- and institution-building to their Russian patron, arguing that this tendency has been most prevalent in South Ossetia, somewhat less in Abkhazia, and relatively limited in Transnistria (beyond the security services). This trend is also reflected in the parachuting in of patron-state representatives. Transnistria's first leader, President Igor Smirnov (1991–2011) was not a local: born in the Russian Far East, he spent most of his life in Ukraine before arriving in Tiraspol in 1987 to take up a position as a factory director. Although a newcomer at the time of the secessionist conflict, Smirnov can hardly be seen as a patron-state representative – he was more a product of Soviet labour-market policy and migration patterns.¹⁴ He was succeeded by a local, Yevgenii Shevchuk (2011–2016); and the current president, Vadim Krasnoselskii (2016–), although born in Russia, moved to Bendery (the second largest city of Transnistria) at the age of eight.

This is not to say that there have not been cases where patron-state representatives took up office in Transnistria. The security sector has had several patron representatives. For almost two decades (1992–2011), the odious Vladimir Antyufeev (also known under the aliases ‘Vadim Shevtsov’ and ‘Vladimir Aleksandrov’) served as Minister of Security. He had previously been involved in supressing national mobilisation in the Latvian SSR – and in 2014, he resurfaced as First Deputy Prime Minister of the DPR in charge of security. Also some of his successors have had close ties with Russian security structures. Tellingly, the only patron-state representative in the government in 2020 is the Minister of State Security, Valerii Gebos, who had spent more than 20 years in Soviet and Russian security structures before arriving in Tiraspol to take up the position as Deputy Minister of State Security (2007–2011) and then as Minister from 2017 onwards (see Table 7).

6.3.2. Military presence

As noted, during the war of secession, the Transnistrian paramilitary forces benefited greatly from the supply of arms and manpower from the Russian 14th Army. After the war, personnel from the 14th Army were instrumental in building up, training and equipping the Transnistrian armed forces (Kosienkowski, 2019). In 1995, the 14th Army was scaled down and transformed into the Operational Group of Russian Forces (OGRF). Since then, the OGRF has provided Russian forces to the peacekeeping mission as well as guarded the old Soviet ammunition depot in Cobasna (said to be one of the biggest in Eastern Europe). While Moldovan authorities have repeatedly protested against the presence of a Russian military base on what is de jure Moldovan territory, and Moscow in 1999 agreed to pull out all troops except for the peacekeepers, an estimated 1500 Russian troops are still deployed in Transnistria (some 400 as peacekeepers) (Ria Novosti, 2020). As for military cooperation, Russia's failure to recognise Transnistria and formal insistence on Moldova's territorial integrity prevents this from being formalised. However, many soldiers serving in the OGRF are recruited locally among Transnistrians with Russian citizenship, thus contributing to ‘informal’ integration.

Table 7. Patron-state penetration of Transnistria.

		As of the end of 2020
Representatives of patron in state institutions	Share of ministers	1/12
	Share of deputy ministers	0/28
Military presence	Bases	Yes
	Patron versus de-facto state troops	1500/7500
	Degree of integration	Formal agreement on military cooperation: no Joint military training: no Joint military units: no Joint command structure and headquarters: no

Sources: see online supplemental data.

7. CONCLUDING DISCUSSION

Despite the changing nature and understanding of ‘sovereignty’, it is still widely assumed that a recognised state can act more independently and enjoy a greater degree of agency (Krasner, 2001). Above we have explored the potential impact of patron-state recognition on de-facto state sovereignty, asking whether the act of recognition has resulted in greater independent agency – or, to the contrary, locked the client to its patron. From the seven cases we have analysed it is difficult to give a clear answer.

First, patron-state recognition may – or may not – be followed by an increase in penetration of the executive (although we find no examples of a decrease). Even if Abkhazia fiercely defends its independence, we note a greater degree of patron-state penetration of the executive in the aftermath of the Russian recognition. In the TRNC, by contrast, patron-state recognition is hardly evident in this respect: in 2020, the two members of the government coded as ‘patron-state representatives’ were actually ‘homegrown’. These findings seem to align with those of Eiki Berg and Raul Toomla's ‘normalization index’ (Berg & Toomla, 2009). Berg and Toomla's sample partly overlaps with ours;¹⁵ and they find that the TRNC, which we argue display very low formal political penetration, scores the highest on their ‘recognition scale’, while Nagorno-Karabakh, where we describe a reverse penetration, scores the lowest (Berg & Toomla, 2009, p. 33). Patron-recognised Abkhazia here performs only marginally better than non-recognised Transnistria. Given the trend indicated in our data, were Berg and Toomla to repeat the ranking exercise, the ranking of the latter two would probably be reversed.

Second, whereas recognition entails the advantage of the patron being able to formalise military cooperation and security guarantees, it may also, as in the case of South Ossetia, be the first step towards a full merger of the armed forces and the complete outsourcing of security.

Finally, patron-state recognition appears to be no guarantee for client survival. Although the two newest de facto states in our sample, DPR and LPR, were fully recognised by Russia in 2022, their ‘independent statehood’ lasted for only half a year. When the security situation deteriorated in the face of Ukrainian advances, Moscow responded by absorbing these ‘states’ into the Russian Federation.

Subsequently, there seems to be no clear pattern to indicate that patrons that extend formal recognition to their de-facto state clients actually treat these entities much differently from those that remain unrecognised. To test the robustness of our findings, one way forward would be to include historical cases that fit our conservative definition – or, alternatively, to broaden the case selection to include less clear-cut cases of de facto states, cases with more variable degrees of recognition, less territorial control or no explicit declarations of independence. Another option for further exploring the impact of patron-state recognition would be to compare how the patrons examined above treat client states that are fully internationally recognised (see Kolstø & Blakkisrud, 2021).

Based on the conservative definition applied here, however, we may conclude that formal recognition does not seem to matter much when it comes to de-facto state patron–client relations. Although the de facto states have been seen to approach international recognition as a panacea that would resolve all problems (Caspersen, 2018; Pegg, 2019), single-state recognition by the patron may in fact prove a double-edged sword.

ACKNOWLEDGEMENTS

We would like to thank the editor and three anonymous reviewers for their careful reading and thoughtful suggestions. We are also grateful to Darmen Koktov, İzzet Yalın Yüksel and Ismoil Sadullozoda for their assistance in data gathering.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research has been carried out with support from the Research Council of Norway, grant number 301277.

Notes

1 While for example Sebastian Klich (2021) would define Kosovo and Taiwan as de facto states on the basis of the fact that these entities have not secured sufficient recognition to obtain full membership in the UN, Nina Caspersen (2012) refers to these as ‘borderline cases’, and Kolstø dismisses them, arguing that they are recognised by too many states to qualify.

2 The only exception in our sample is Somaliland, which has survived without having a clearly defined patron. However, the state from which it seceded, Somalia, is itself widely considered a failed state (Pegg & Kolstø, 2015).

3 South Ossetia and Abkhazia have also been recognised by a handful of states with close ties with Russia: Nicaragua (2008), Venezuela (2009), Nauru (2009) and Syria (2018). No other state has yet recognised the TRNC.

4 On the eve of the Russian military attack on Ukraine in February 2022, the Kremlin extended recognition to the DPR and LPR – and then absorbed them into the Russian Federation in September that year.

5 However, international recognition is not the panacea which it is sometimes perceived as in de facto states; existing economic and social problems will not disappear overnight (Pegg, 2019).

6 Despite the criteria we introduce in Table 1 (being born and/or having spent more than half their lives in the patron state), it is not always straightforward to determine whether an individual should be considered a representative of the patron state. Such cases were discussed by the authors, and the final decisions to include them made on a case-by-case basis.

7 An alternative would have been to compare patron-state penetration after recognition in the patron-recognised entities and after declaration of independence in the unrecognised ones. This would, however, entail other challenges: If, for example, we opt for five years after recognition/declaration of independence, this would mean that we would compare Russian penetration of Abkhazia in 2013 to that of Transnistria in 1996 – but the patron Russia was a very different entity in 1996 than in 2013.

8 We opted for four years to allow sufficient time to pass between our measuring point and the act of official recognition.

9 In de-facto state conflicts, reliable data are often hard to obtain. To identify and cross-check data, we draw on a wide range of local, patron- and parent-state online resources, reports of international and nongovernmental organisations and media sources. The main sources for each case are presented in the online supplemental data.

10 As the DPR and LPR, the two newest de facto states in our sample, emerged from the same process and display closely overlapping development trajectories with their patron state, we have merged them into a single case.

11 With the exception of Aleksandr Ankvab (Chief of Staff 2010–2011), this position has, since 2003, been held by what we define as patron-state representatives.

12 The agreement is valid for 49 years, with a possibility of automatic extension every 15 years. Some of the current facilities had previously been used by CIS/Russian peacekeepers.

13 The system of curators applies to all of Russia's de facto clients. In the case of the TRNC, the Turkish ambassador initially played a similar role.

14 Indeed, Russian authorities played a greater role in Smirnov's final ouster in 2011 than in his initial instalment (Kosienkowski, 2019).

15 They do not include South Ossetia and, for obvious reasons, DPR and LPR.