Lost and Found: The WikiLeaks Of De Facto State-Great Power Relations

Article Contents

Abstract

There is a growing recognition that sovereign states increasingly share the political stage with a wide variety of entities including autonomous regions, de facto states, dependent, internationalized or leased territories, and governments-in-exile (McConnell 2009; Berg and Kuusk 2010). Yet, McConnell’s (2009:344–345) twin observations that these geopolitical anomalies “are generally overlooked … with their existence under-theorized” and that framing them “inconsistently negative terms (as illegal, pathological, and clandestine) and with regards to what they fail to achieve (sovereign territorial statehood) ultimately restricts analysis of these polities and denigrates their achievements” remains valid today.

However, a series of events taking place from 2008 to 2012 has brought sovereign anomalies to new prominence. Kosovo declared its independence in February 2008 and is today recognized by 106 members of the United Nations, albeit not by permanent Security Council members China and Russia or by other important countries including Brazil, India, and Spain. Subsequently, Russia recognized the independence of Abkhazia and South Ossetia in August 2008 and “announced that after lying dormant for several years, the game of recognition had returned to global politics” (Caspersen 2012:2). In May 2009, the Sri Lankan government completed its forcible eradication of Tamil Eelam. Most recently, in November 2012, the UN General Assembly upgraded Palestine to “non-member observer state” status in the United Nations.

This study takes advantage of the public dissemination of recent US diplomatic cables via WikiLeaks to investigate diplomatic relations between a great power (the United States) and a subset of sovereign anomalies, which is termed henceforth as de facto states (see Pegg 1998; Chapter 2 for a detailed definition). We prefer this terminology both because it is “the most appropriate and most neutral” term (O’Loughlin, Kolossov, and Toal 2011:2) to describe secessionist entities with contested sovereignties that control territory and provide governance over an extended period of time, and because it is the most commonly used (see, for example, Pegg 1998; Bartmann 2004; Lynch 2004; Popescu 2006; Berg and Kuusk 2010; Kolstø and Blakkisrud 2012).

De facto states are typically seen as marginal actors in the international system, which are either ignored or viewed with hostility by the vast majority of sovereign states. However, we argue that hostility/support and isolation/engagement often go hand in hand, yet vary significantly depending on the US interests at stake. In this article, we show that de facto states are frequently engaged by sovereign states and that at least some of them are not typically engaged in a hostile manner. Apparently, de facto states are not treated as homogenous entities, and US foreign policy is quite capable of discriminating between them. The central research question we address in this paper is: what determines the degree of US support or hostility toward specific de facto states.

Although de facto states are not “informational black holes,” it still remains difficult to find good empirical data on these entities for a variety of reasons including remote locations, secretive regimes, and a lack of coverage by most international and non-governmental organizations. While using WikiLeaks cables as an information source may seem questionable since Chelsea (Bradley) Manning, a US Army soldier who leaked these restricted documents to the public was convicted in July 2013 of violations of the Espionage Act and other offenses and sentenced to 35 years in prison, this documentation is now in the public domain and easily accessible to all interested parties.

Moreover, a January 2014 search of the Social Science Citation Index for WikiLeaks produced 91 results. Increasingly, the WikiLeaks cables are being assessed in terms of specific diplomatic or foreign policy questions (see, for example, Heisbourg 2011; Page and Spence 2011; Pieterse 2012). We see this study as building on a much smaller subset of the WikiLeaks literature that uses these cables as a valuable source of data from which to test specific hypotheses or address scholarly questions of interest, similar to what Linke, Witmer, and O’Loughlin (2012), O’Loughlin et al. (2010) and Zammit-Mangion et al. (2012) did while utilizing the WikiLeaks-released Afghanistan and Iraq war logs to engage in spatial geographic analyses of those conflicts or what Lefebvre (2012) has done using the WikiLeaks cables to examine three specific US policy decisions in the Horn of Africa.

The article starts with theoretical and methodological outlines. First, we review discussions over engagement perspectives in the de facto state literature. Second, we put forward four specific hypotheses that we test against our data set to ascertain what determines the level of US support or hostility to de facto states. In this section, we also describe our four cases, the case selection criteria, data set formation, and coded variables. The next section presents our findings that we use to test the four specific hypotheses and provides illustrative examples from specific WikiLeaks cables to offer texture to the quantitative analysis. The conclusion summarizes our findings and assesses how they support, challenge, or enrich various theoretical arguments put forward in the de facto state literature. Collectively, the portrait painted by our analysis of the WikiLeaks cables reveals a previously hidden tapestry of great power–de facto state relations that is far richer and more variegated than typically assumed.

Theorizing De Facto State Engagement

This study builds upon recent attempts to bring more empirical rigor to the study of de facto states. Examples here include such things as detailed fieldwork (Kolstø and Blakkisrud 2008, 2012; Caspersen 2012), the generation of relevant new empirical measures or scales (Berg and Toomla 2009; Berg and Kuusk 2010), and the use of survey data both within de facto states (O’Loughlin et al. 2011) and between de facto states and the parent states they are trying to secede from (Berg 2012). In our case, the use of WikiLeaks US diplomatic cables provides us with a new and empirically rich window on the foreign relations of de facto states and an ability to test specific theoretical hypotheses that might explain variations in how sovereign states engage de facto states.

Since at least 1945, the international community has provided strong diplomatic, military and financial support to existing sovereign states. The principle of self-determination has been interpreted quite narrowly and emphasis has instead been placed on maintaining respect for the sovereignty and territorial integrity of widely recognized states even if, as with Somalia, such states have largely ceased to exist in empirical terms. In contrast, de facto states are usually labeled as “pariahs, excluded from the mainstream channels of international diplomacy, existing in conditions beyond the pale of normal international intercourse” (Bartmann 2004:12). Some of them manage, without belonging to the most privileged club of confirmed states, to enjoy either “a degree of toleration” or “limited accommodation” from the members of the international community (Pegg 1998:98, 198). All this speaks to the generally unfavorable legal conditions and widespread political disengagement that de facto states face in the postcolonial international system in particular.

Given that parent states’ argumentation has largely been accepted—“the entities are illegal, they represent a de facto occupation, they are based on ethnic cleansing and their leaderships lack any popular legitimacy” (Caspersen 2012:41), then “engagement remains in most cases tentative and restricted, informed by a variety of considerations” (Geldenhuys 2009:47). More often than not, this pariah status brings with it political isolation: the absence of official bilateral representation and exclusion from inter-governmental organizations; economic isolation: sanctions aimed at restricting the flow of goods and services to and from the de facto state; and finally sociocultural isolation which excludes contestants from sporting events or restricts its nationals’ ability to travel abroad (Geldenhuys 2009:47).

Although, Caspersen may note that legally speaking any international engagement with unrecognized states has been deemed unlawful (2012:31) and that therefore “the default position has been non-engagement” (2012:40), in reality, “the de facto states have not been either consistently ignored or comprehensively embargoed by the international community” (Lynch 2004:112). Even within the context of the generally strong support for the status quo in international relations, Pegg (1998:177) argues that de facto states have been treated “in three main ways: actively opposing them through the use of embargoes and sanctions; generally ignoring them, and coming to some sort of limited acceptance of their presence.” He also notes that the fact that Northern Cyprus can be cited in all three of these “categories shows that they are not necessarily mutually exclusive” (Pegg 1998:181). To this list of three main approaches, Lynch (2004:104) adds a fourth option, usually adopted only by parent states: “active opposition and the attempt to eliminate the de facto state by force.” Arguably, neither legalizing all de facto states nor approaching Lynch’s fourth option of forceful reintegration as being “statistically the most likely outcome for unrecognized states” (Caspersen 2012:126) are realistic policy options.

The idea that de facto states can receive varying levels of acceptance from international society is not new. Berg and Toomla (2009), for example, focus on multiple variables derived from political, economic, and public spheres to reveal both the international community’s willingness to integrate de facto states and, correspondingly, the willingness of de facto states to open up to the outside world. They argue that their composite “normalization” measure can tell us more about the relative differences in treatment than can measures of either internal or external sovereignty. Thus, even though Taiwan is gradually losing international recognition while Kosovo is gradually gaining it, the approach from the rest of the world to both of them is quasi-recognition. A bit surprisingly, Northern Cyprus is tolerated by the international community more than “partly recognized” Abkhazia. The rest of the de facto states in this study (Abkhazia, Transnistria, and Somaliland) remain in the boycott zone, with Nagorno-Karabakh receiving the lowest score. Not a single de facto state under scrutiny was found to be completely negated by international society because of its illegality (Berg and Toomla 2009:31–33).

If it wanted to do so, international society could accommodate the existence of de facto states in a variety of different ways, some of which are demonstrated by the many creative ruses by which post-1971 Taiwan has been accommodated to such an extent that it has become the world’s 18th largest trading nation. De facto states could be accorded certain rights and responsibilities and denied others because their sovereignty is not widely recognized. One could argue that the caveat to all such speculation, however, is that these theoretical possibilities do not equate with contemporary realities (Pegg 1998:191–192). We argue that it is an empirical question to sort out the conditions necessary for hostile or friendly de facto state engagement. In this paper, we test four theoretically derived hypotheses to examine what determines US support or hostility for specific de facto states.

Methodology

We use a subset of 448 WikiLeaks cables to generate an original data set to investigate how US foreign policy deals with four de facto states (Abkhazia, Nagorno-Karabakh, Somaliland, and Northern Cyprus). Our four case studies are all widely accepted in the academic literature as de facto states and they have been selected because they exhibit significant variation in terms of such things as their geopolitical affiliations, domestic political systems, and their ultimate end goals. On the one hand, we believe that this variation represents all possible versions of de facto states existing in contemporary world politics. On the other hand, we assume that these variations also affect the way the US interacts with them which enables us to test four competing theoretical hypotheses. We code each cable across multiple dimensions including whether or not there was direct engagement, the level of US/de facto state participation, the types of issues covered and what, if any, US actions were taken or recommended.

Hypotheses Tested

In order to address our central research question of what determines the degree of US support or hostility toward de facto states, we advance four specific hypotheses drawn from the IR literature. All four of these hypotheses were initially developed to explain sovereign state behavior or relations between sovereign states but we believe that all four can be extended or extrapolated to explain sovereign state—de facto state relations.

Our first hypothesis derives from realist IR theory and, more specifically from what Holsti (1996:207) has criticized as “the theoretical paraphernalia of realist and neo-realist studies, in which virtually everything is explained by great-power behavior or systemic characteristics such as global balances of power … .” In Waltz’s (1979:72–73) famous phrasing, “Theories that apply to self-help systems are written in terms of the system’s principal parts … . The fates of all the states … in a system are affected much more by the acts and interactions of the major ones than of the minor ones … . A general theory of international politics is necessarily based on the great powers.” Applied to the world of de facto states, this hypothesis directs our attention not toward de facto states themselves as they are all relatively minor powers but rather toward their external patron state backers who are great powers or at least regional great powers. Thus, H₁ is that: US support or hostility is explained not by any characteristics internal to de facto states but rather by US relations with their patron state great-power supporters.

Our second hypothesis derives from liberal IR theory more generally and from the democratic peace thesis more specifically. In contrast to realists like Waltz (1979:96) who believe that states are “like units” in the sense that they “are alike in the tasks that they face, though not in their abilities to perform them,” liberals believe that the internal characteristics of states matter. According to Moravcsik (1997:513, 520), “For liberals, the configuration of state preferences matters most in world politics—not, as realists argue, the configuration of capabilities …” with the result being that “state behavior reflects varying patterns of state preferences.” Doyle (1986:1151–1152) argues in the context of democratic peace that “liberalism does leave a coherent legacy on foreign affairs” because “liberal states are different.” Specifically, the normative logic underlying democratic peace suggests that “democracies both trust and respect one another when a conflict of interest arises between them” whereas “non-democracies are neither trusted nor respected” (Doyle 1986:1161; Rosato 2003:586). Applied to de facto states, this perspective suggests that the level of democracy found within de facto states should affect US foreign policy toward them. We assume here that “international democratization” continues “to lie at the heart of American grand strategy” (Rosato 2003:600) and thus H₂ is that: US support or hostility toward individual de facto states is explained by their level of democratization.

Our third hypothesis derives from constructivist IR theory more generally and from securitization theory more specifically. For Buzan, Wæver, and de Wilde (1998:26), when it comes to security, “it is the utterance itself that is the act. By saying the words, something is done ….” As explained by Elbe (2006:124–125), security is not viewed “as something that exists independently of its discursive articulation, but rather as a particular form of performative speech act ….” Extrapolating from security to foreign policy more generally, we assume that US foreign policy is also not “something that exists independently of its discursive articulation.” Words, language, and presentation matter. Thus, referring to de facto state leaders as presidents or prime ministers is a fundamentally different thing than referring to them as “presidents” or “prime ministers” just as labeling their countries as Republic of … is different than labeling them as the “so-called” or “self-proclaimed” Republic of …. Thus, H₃ is that: US support or hostility toward individual de facto states is explained by how these entities are referred to by US diplomats.

Our final hypothesis derives from work on the domestic determinants of US foreign policy more generally and from work on ethnic interest groups more specifically. Putnam (1988:432) has argued that “A more adequate account of the domestic determinants of foreign policy and international relations must stress politics: parties, social classes, interest groups (both economic and non-economic), legislators, and even public opinion and elections, not simply executive officials and institutional arrangements.” He famously suggested conceiving diplomacy as a “two-level game” where “domestic groups pursue their interests by pressuring the government to adopt favorable policies” at the national level while “national governments seek to maximize their own ability to satisfy domestic pressures while minimizing the adverse consequences of foreign developments” at the international level (Putnam 1988:434). In this article, we specifically focus on one subset of domestically based foreign policy actors which are ethnic interest groups. The existing literature is divided both on whether or not ethnic interest groups, in general, are successful at shifting US foreign policy in directions they favor (Rubenzer 2008) and whether or not they are successful in specific high profile cases (Mearsheimer et al. 2006). There are enough suggestions that ethnic interest groups can influence US foreign policy, though; thus H₄ is that: US support or hostility toward individual de facto states can be explained by the strength (or weakness) of the ethnic interest groups intervening on their behalf in the US domestic political process.

Case Selection

The Republic of Abkhazia (ABH) declared its independence from Georgia in July 1992 and has been de facto independent since 1993. Abkhazia benefits from significant Russian patronage and support (Popescu 2006; Kolstø and Blakkisrud 2008) and Russia and five other states formally recognized its sovereignty after the 2008 Russo-Georgian war. Abkhazia has made some progress toward democratization (Kolstø and Blakkisrud 2008; O’Beachain 2012) and its non-Georgian population is divided on whether its end goal should be independence or some form of association with Russia (O’Loughlin et al. 2011).

The Nagorno-Karabakh Republic (NKR) declared its independence from Azerbaijan in September 1991 and has been de facto independent since then, even controlling more territory than it actually claims for itself. Nagorno-Karabakh benefits from significant Armenian patronage with the end result that “when it comes to economy, culture, and defense, Nagorno-Karabakh and Armenia can be seen as a single space” (Caspersen 2012:56). NKR performs better and scores higher in Freedom House rankings than its parent state Azerbaijan (Kolstø and Blakkisrud 2012). Yet, it remains insecure and entirely unrecognized with many of its residents favoring unification with Armenia over independence as their preferred end goal.

The Republic of Somaliland (SL) declared its independence from Somalia in May 1991 and has been de facto independent since then, although its control over the eastern part of its claimed territory is tenuous and contested (Hoehne 2009:271). A former British colony, the combination of Somaliland’s separate colonial existence, its five days of sovereign independence, and its respect for former colonial borders gives it a unique degree of legitimacy in terms of the contemporary interpretation of self-determination that no other secessionist entity approaches. Yet, in spite of this and its generally impressive accomplishments in terms of democratization and maintaining peace, Somaliland remains entirely unrecognized. It is also unique among our cases in lacking the support of an external patron state. Somaliland remains committed to the goal of sovereign independence.

The Turkish Republic of Northern Cyprus (TRNC) proclaimed its independence from the Republic of Cyprus in November 1983. Tens of thousands of Turkish troops and a UN buffer zone have secured the territory claimed by the Turkish Cypriot de facto state since 1974. Northern Cyprus has high levels of internal democratic legitimacy which have been found to be comparable to those enjoyed by citizens in the internationally recognized Republic of Cyprus (Berg 2012). The Turkish Cypriots carefully formulated their declaration of independence in such a way that it would not preclude a settlement to the Cyprus problem (Pegg 1998:105) and their leadership remains officially committed to a bi-communal, bi-zonal Cyprus settlement.

Our four case studies were selected specifically to address variations that would enable us to test our four hypotheses (see Table 1). In terms of H₁, we find significant variation across our case studies in terms of patron states and the geopolitical environment. If H₁ is correct, we would expect to see the most US support for the TRNC (Turkey is a US ally) and the least US support for Abkhazia (Russia is increasingly seen as a US rival). NKR (with Armenia as its patron state) would fall closer to Abkhazia as Turkey supports Azerbaijan and Russia supports Armenia. Somaliland could potentially go either way. As it has no patron-state backer, this could leave it the most vulnerable and exposed of all of our cases (Kolstø 2006) and hence generate significant US hostility. As Ethiopia (a US ally) has friendly relations with Somaliland, though, this could push it closer to the expected TRNC end of this continuum.

In terms of H₃, the TRNC is the only entity consistently referred to in quotes throughout every US diplomatic cable covered in this study. The TRNC, its institutions (“Central Bank,” “Ministry of the Environment”) and its officials (“president,” “foreign minister”) are always referred to in quotes. The only exceptions here are that Turkish Cypriots as an ethnic group are not placed in quotes and occasionally North Cyprus is also used without quotes. Nagorno-Karabakh is treated somewhat similarly with its name often put in quotes and sometimes proceeded by the modifiers so-called (2006-11-24, Baku) or self-proclaimed (2009-04-23, Yerevan). Yet, one finds frequent references to Nagorno-Karabakh or the Nagorno-Karabakh conflict without quotes being used. Abkhazia receives somewhat more nuanced treatment. Its officials often have their titles put in quotes (2006-10-10, Moscow; 2007-07-06, Tbilisi as examples) but it also frequently receives quote-free treatment on more neutral terms such as de facto government (2007-03-12, Tbilisi), de facto officials (2007-04-13, Tbilisi), or separatist authorities (2008-06-10, Tbilisi). Linguistically, Somaliland receives by far the best treatment. There is only one cable (2005-09-02, Nairobi) where Somaliland is treated like the TRNC with everything in quotes. There are a few cables where Somaliland is qualified by the modifiers self-proclaimed (2004-01-15, Djibouti) or self-declared (2006-06-16, Nairobi) but neither these nor its name are placed in quotes. In the overwhelming majority of the cables, it is just referred to as Somaliland, and its officials are listed as Somaliland Foreign Minister or Somaliland President without any quotes or modifiers. If H₃ is correct, we would expect to see the United States be most supportive of Somaliland and least supportive of the TRNC with Nagorno-Karabakh closer to the TRNC and Abkhazia closer to Somaliland in their treatment.

In terms of H₄, Greeks and Armenians are regularly cited as two of the most successful ethnic interest groups trying to influence US foreign policy (Rubenzer 2008:178). The high levels of material support flowing from the Armenian diaspora in the United States to Nagorno-Karabakh have also been noted (Kolstø and Blakkisrud 2008:495). Somaliland has also received significant support from its diaspora but this is concentrated more heavily in the United Kingdom and in the Gulf States than it is in the mass protests in cities around the U.S. against an executive order that would block millions of people from entering the United States. Abkhazia and Somaliland should fare much worse than NKR while the TRNC (with active Greek-American opposition) should fare worst of all if H₄ is correct and ethnically based domestic interest groups exercise a significant influence on US foreign policy.

Data Set

In total, our data set comprises 448 leaked US diplomatic cables with a date range from April 24, 2003, to February 24, 2010. We generated our data set using the cables.mrkva.eu browser. Our data set comprises the population of cables that have one of our case studies’ names in the title of the cable (in the case of the TRNC we additionally searched “Northern Cyprus”; “North Cyprus”; “Turkish Cypriot”; and “Turkish Cypriots” as US diplomats frequently use these terms as well). In total, we ended up with 174 cables for Abkhazia, 53 cables for NKR, 93 cables for Somaliland, and 128 cables for the TRNC.

Searching for these terms on the “official” WikiLeaks Web site (http://www.cablegatesearch.net) where the browser searches the entire contents of the cables and not just the titles will produce dramatically higher figures. For example, our 174 Abkhazia cables compare to 1,878 cables and our 93 Somaliland cables compare to 449 cables that are generated via the cablegatesearch.net browser. Yet, we believe that our data set is justified because the vast majority of “missed” cables are of only peripheral or tangential relevance. Cables such as a “missed” Somaliland cable noting that the US embassy has secured an agreement to distribute HI! Magazine as free in-flight reading material on Djibouti Airlines, some of whose passengers come from Somaliland (2004-08-26, Djibouti), and another “missed” Abkhazia cable which shows up only because one of the candidates for a position at the OSCE was a “former UN Special Representative to Abkhazia” (2003-08-22, The Hague), are not really those we wanted to target. In short, we do not claim that our data set comprises the entire universe of cables that mention one of these entities in passing. By limiting our search to cables that include one of these entities in their title, though, we believe we have captured the overwhelming majority of relevant cables.

We take these cables as a legitimate source of invaluable data on great power– de facto state relations. Some of the blunt, entertaining, and offensive language used throughout these cables demonstrates that they were obviously not intended for public disclosure. US Ambassador to Cyprus Ronald Schlicher was particularly colorful in his language, lambasting one Turkish Cypriot Mufti as “a mid-level ‘TRNC’ bureaucrat who couldn’t find Mecca with a map” (2007-08-17, Nicosia) and describing TRNC Deputy Prime Minister and Foreign Minister Serdar Denktash as “a nationalist wing-nut” (2006-03-24, Nicosia) who only gets “symbolic policy scraps like OIC meetings and relations with Azerbaijan or the Gambia” (2006-06-20, Nicosia).

Beyond language, it is clear that the information contained within these cables was clearly not intended for public release. One Somaliland cable (2005-09-02, Nairobi), for example, lists the personal cell phone numbers for Somaliland’s Minister of the Interior. There are also at least five separate TRNC cables (deliberately not referenced here) which identify specific named individuals with the designation “(please protect).”

Variables Coded

We coded four different variables across our data set of WikiLeaks cables, most directly relevant to answering our research question. First, the type or level of engagement refers to direct engagement (direct contacts between US officials and de facto state government officials, politicians, citizens, or diaspora members) and indirect engagement (meetings or contacts between US officials and other actors about the de facto state). Examples here include parent state (the state the de facto state is trying to secede from) or patron state (the de facto state’s principal backer) officials, international organization officials, and other governments the death of four U.S. Special Forces soldiers in Niger on Oct. 4 has lead some in the United States is coordinating policy with. None signifies a lack of direct or indirect engagement, most often seen in diplomats summarizing developments. Relatively higher levels of direct and indirect engagement would presumably indicate that the [24] Peter J. Schraeder, “Why the United States was paying more attention to a specific de facto state, although they would not in and of themselves indicate whether that attention was supportive or hostile.

Second, the de facto state level of participation (categories yes/no) reveals US diplomatic contacts with the president, prime minister, foreign minister, opposition party leader, other cabinet officials, or private actors like businessmen or non-governmental organization employees. Relatively high and frequent contacts with leading de facto state officials would again signal US interest in the de facto state and they would presumably also indicate higher levels of US support for that de facto state. Following the same logic, the participation of other involved parties, such as parent state actors, patron state actors, and third party actors was also coded.

Third, we coded whether the US action could be characterized as supportive, hostile, or neutral toward the de facto state. This is the most direct evidence addressing our central research question. Neutral, which was defined broadly here to include restatements of existing US policy toward the de facto state, accounted for the vast majority of cables in all of our cases, with wide variation in the ratios of supportive to hostile cables across individual cases.

Finally, we coded the primary issue dealt with in the cables to see if the dominance of any particular issues lent support to one or more of our hypotheses. Examples of the primary issue dealt with in the cables include domestic politics about internal politics within the de facto state; status about the de facto state’s quest for sovereignty; conflict management about resolving the conflict; and economics about investment, trade, or business conditions in the de facto state. Through the use of an innovative data set generated via the WikiLeaks diplomatic cables, we probe in this article just how far the United States is or is not willing to engage de facto states and how supportive or hostile it is toward them.

Uncovering US Diplomatic Activities

In broad terms, US foreign policy has consistently been not to recognize any de facto state and to promote conflict settlements in all of them. US diplomatic activity is by no means uniform or standard across de facto states. Instead, we find certain types of engagement and issues prevailing over others. Specific events or crises like the Russo-Georgian war in the case of Abkhazia or repeated postponements of a presidential election in Somaliland also substantively determine the US diplomatic agenda with these entities.

Engagement: How?

The US engagement with de facto states as reflected in these cables can be divided into three categories: direct (the US representatives are meeting with or talking to de facto state government officials, political parties, citizens, or diaspora groups), indirect (the US representatives are meeting with other countries or third parties about the de facto state), and none (there is no contact with anyone). Without question, the biggest single divide in terms of engagement is that the United States frequently engages directly with two de facto states (Somaliland and the TRNC) and almost never engages directly with two others (Abkhazia and NKR; see Figure 1).

Fig. 1.

Although US diplomatic practice regularly puts all mentions of the TRNC, its institutions and its officials in quotes (“President” or “Central Bank”), US diplomats met directly with Northern Cypriot officials or residents in more than half (52.3%) of the WikiLeaks cables, including, for example, the US ambassador to Cyprus meeting with the TRNC Minister of the Environment (2006-11-21, Nicosia), with the governor and deputy governor of the TRNC’s Central Bank (2007-09-13, Nicosia), and with TRNC President Mehmet Ali Talat at his home (2006-03-28, Nicosia).

In the case of Somaliland, direct contact represented nearly two-thirds (64.5%) of the WikiLeaks cables: examples include US diplomats meeting with five separate Somaliland ministers in Djibouti (2003-10-05, Djibouti; 2004-12-06, Djibouti), with both opposition party presidential candidates (2009-08-07, Nairobi), and with the Speaker of Parliament (2009-09-01, Nairobi). At the same time, US officials never met directly with anyone in NKR and only rarely (1.7% of cables) met directly with anyone from Abkhazia.

In contrast, Abkhazia (94.8%) and NKR (60.4%) had by far the highest percentage of cables that featured indirect meetings with other parties. Abkhazia’s figures here are particularly inflated given significant US efforts in 2008–2009 to both ascertain other countries’ positions on recognizing Abkhazia and South Ossetia and to discourage them actively from doing so. NKR’s relatively high figure here reflects various discussions with the parent state, patron state, and other regional powers about the NKR conflict. The TRNC (29.7%) and Somaliland (19.4%) both have significant amounts of indirect contact as well, although these figures are comparatively lower given the United States’ greater willingness to engage them directly.

As might be expected, many of our cables (coded none) reflect diplomats’ traditional function of monitoring developments and gathering information, often by summarizing local news reports or reporting on major developments. In three of our four cases, substantial minorities of cables (39.6% for NKR; 18.0% for TRNC; and 16.1% for Somaliland) merely report on local events or developments without reflecting any form of direct or indirect engagement. Abkhazia (3.5%) has by far the lowest number of such cables, a clear reflection of the indirect US diplomatic campaign against its recognition.

De Facto State Involvement

In terms of the level of de facto state involvement, our findings here mirror the stark dichotomy noted above that the United States frequently engages directly with Somaliland and the TRNC and almost never engages directly with Abkhazia or NKR. Apparently, US diplomats traveled to Abkhazia on only three occasions during the period covered by the WikiLeaks cables. In 2005, they met with high-ranking Abkhaz officials including the president, prime minister, and foreign minister (2005-12-12, Tbilisi). On subsequent trips, those high-ranking officials were sidelined in favor of meetings with the head of the Tourism Commission, the President of the Abkhazian Chamber of Commerce and the mayor of the city of Gagra (2007-10-17, Tbilisi), and “mid-level de facto officials” and civil society representatives (2009-11-09, Tbilisi). Nagorno-Karabakh had only one direct contact with US embassy officials when its foreign minister complained in his letter to US embassy staff about the presentation of information in the State Department’s 2008 Human Rights country reports (2009-04-23, Yerevan).

In contrast, Somaliland and the TRNC are characterized not only by multiple meetings with top de facto state officials but also by the sheer breadth of people the United States engages with. In terms of high-level meetings, the US ambassador to Cyprus met with TRNC President Talat at least five times in 13 months (2008-09-17; 2009-05-07; 2009-07-17; 2009-07-20; 2009-10-29, all Nicosia) and met with TRNC Prime Ministers at least four times during a similar 13-month period (2008-11-25; 2009-06-17; 2009-07-20; 2009-12-04, all Nicosia). US direct contacts with Somaliland officials are even more noteworthy. The United States had direct contact with Somaliland’s President Rayale at least 11 times (see, for example, 2007-01-09, Addis Ababa; 2008-11-05, Nairobi; 2009-11-02, Nairobi). It had direct contact with two different Somaliland Foreign Ministers at least 23 times (see, for example, 2004-09-01, Djibouti; 2007-08-22, Nairobi; 2010-02-04, Addis Ababa) and it had direct contact with Somaliland opposition party leaders at least 17 times (see, for example, 2009-09-23, Nairobi; 2009-11-06, London) during the period of time covered in the WikiLeaks cables. In addition to these high-level official contacts both in the TRNC and Somaliland, the United States also frequently engages with private citizens (2004-06-17, Djibouti; 2006-10-13, Nicosia) and businessmen (2009-06-02, London; 2006-06-27, Nicosia) in both locations and with diaspora members from Somaliland (2008-11-10, London).

In partial contrast to the dichotomy noted above where the United States directly engages some de facto states (Somaliland and the TRNC) and not others (Abkhazia and NKR), and in direct contrast to the typical portrayal of de facto states as peripheral actors in the international system, all four of our cases received direct attention from high-ranking US officials, including the Secretary of State. Brigadier General Robeson from the Combined Joint Task Force-Horn of Africa met Somaliland’s foreign minister in Djibouti (2005-01-15, Djibouti) and US Secretary of State Hillary Clinton intervened to ascertain Somaliland’s willingness to receive two Guantanamo Bay detainees (2009-11-13, Washington, DC). US Congressman Adam Schiff (D-California) visited Azerbaijan to discuss NKR with its foreign minister (2008-06-26, Baku) while Vice President Dick Cheney visited Baku in the wake of Russia’s invasion of Georgia (2008-09-05, Baku) to signal support for Azerbaijan. Abkhazia received direct attention from the US Secretary of State six times, more than our other three cases put together, either urging the host governments to oppose the participation of representatives from Abkhazia and South Ossetia in the UN Security Council meetings on Georgia (2008-08-27, Washington DC) or rejecting and condemning Russia’s August 26 decision to recognize Abkhazia and South Ossetia as independent states (2009-03-13, Washington, DC).

Other Involved Parties

Given that its juridical parent is routinely cast as the paradigmatic example of a failed state, it is perhaps not surprising that the United States only engaged Somalia once in the period covered by these cables (2007-07-13, Addis Ababa). Perhaps more surprising are the small number of TRNC cables (7.8%) that indicate any contact with Greek Cypriots. Typically, the United States seems to engage Greek Cypriots in areas where it is trying to promote functional cooperation between the two sides, touching upon property claims issues (2008-04-29, Nicosia), intellectual property rights protection (2009-04-30, Nicosia), or anti-money laundering initiatives (2007-07-28, Nicosia). The “Cyprus problem” is also frequently discussed separately with both Greek Cypriot and Turkish Cypriot officials with our sample of cables reflecting only the latter set of discussions.

In contrast, both Abkhazia and NKR feature comparatively more frequent contacts with their respective parent states (see Figure 2). With Abkhazia, 15.5% of the WikiLeaks cables included some form of engagement with Georgia, for example, expressing Georgia’s strong opposition to a meeting with the Abkhaz in New York (2007-05-23, Tbilisi) or Georgian Foreign Minister Grigol Vashadze asking for urgent assistance to pressure Burundi, the Central African Republic, Guinea-Bissau, and Mali not to recognize Abkhazia and South Ossetia (2009-09-01, Tbilisi). Nagorno-Karabakh featured by far the highest percentage of parent state engagement at 35.9% of its cables. More often than not, Azeri officials discussed Nagorno-Karabakh as one issue among many covered including Azerbaijan’s growing role as an energy supplier, its regional relationships, and its accession to membership in the World Trade Organization (2006-10-31, Baku).

Fig. 2.

Nagorno-Karabakh Republic which had the highest percentage of cables reflecting parent state involvement has the second-lowest percentage of cables (5.7%) reflecting patron state involvement. There are only three NKR cables which involve Armenia and two of these (2006-07-25, Yerevan; 2007-08-30, Yerevan) feature no direct contact with the patron state but merely report on developments within it. There is only one NKR cable which reflects direct Armenian involvement (2009-12-10, Yerevan).

In contrast, Abkhazia (19% of cables) and the TRNC (21.9% of cables) both reflect much higher levels of patron state involvement. One example is Russian Deputy Foreign Minister Grigoriy Karasin invoking the US ambassador to request “a more constructive and flexible” US approach to a Russian draft UN Security Council Resolution on renewing the mandate for the UN Observer Mission in Georgia (2006-10-10, Moscow). Examples of TRNC-related topics discussed with Turkey included the significance of the Turkey–TRNC customs union agreement (2003-08-15, Ankara), and details about Turkey’s estimated US$ 600 million annual assistance to the TRNC with Turkish officials complaining about the TRNC’s “prolific spending” (2009-12-15, Ankara).

In addition to meetings or contacts with de facto, parent, and patron state officials, the United States also frequently engages third parties on issues related to all four of our case studies. Here, NKR (13.2%) and the TRNC (22.7%) cluster at the lower end of a continuum on third-party engagement. Third-party engagements brought together State Department officials and Russia’s Deputy Foreign Minister Grigoriy Karasin on the prospects for Nagorno-Karabakh settlement negotiations (2007-10-25, Moscow) as well as led to discussions between US Deputy Assistant Secretary of State Laura Kennedy and officials from the Dutch Presidency and the EU Commission and Council on ending the economic isolation of the TRNC (2004-07-30, Brussels).

Somaliland (34.4%) has the second-highest level of third-party engagement of our four cases. We notice the US ambassador to Thousands upon thousands of cassette tapes and master reels were quickly removed from the soon-to-be targeted buildings. They were dispersed to neighboring countries like Djibouti and Ethiopia meeting privately with the British Ambassador and four British MPs (2007-09-19, Addis Ababa), as well as meeting with the President of Puntland about Somaliland–Puntland relations (2007-10-16, Addis Ababa). Moreover, during Somaliland’s electoral crisis, the US regularly coordinated positions with a variety of European actors including the European Union, Norway, Sweden, and the UK through the Nairobi-based Democratic Steering Committee (2009-06-11; 2009-07-30; 2009-09-28; 2009-11-30, all Nairobi).

Without question, though, Abkhazia (66.7%) features the highest percentage and the highest absolute number of cables reflecting third-party engagement. Before the Russo-Georgian war of 2008, this activity largely reflected attempts to manage the tensions associated with Abkhazia’s de facto statehood and included six different meetings at the UN or in the UN Security Council about the Georgian-Abkhaz situation (2006-07-17; 2006-11-23; 2007-01-23; 2007-03-17; 2008-01-30; 2008-04-17, all New York). After the war, the overwhelming majority of US third party engagement relates to ascertaining other countries’ positions on recognizing Abkhazia and South Ossetia and/or actively discouraging them from recognizing these entities.

US Positioning

The volume and variety of cables discussed so far demonstrates that the commonly held belief that international society ignores de facto states is incorrect. Coding the WikiLeaks cables in terms of US support, hostility, or neutrality further demonstrates that the commonly held belief that when the international community engages de facto states it is typically in a hostile manner is also at least partially incorrect. Reflecting the generally informative nature of most diplomatic cables, the overwhelming majority of the WikiLeaks cables for all four of our de facto state case studies can best be described as neutral, with neutral defined as merely presenting information or restating long-held US positions (see Figure 3). The percentage of WikiLeaks cables where the United States can be characterized as neutral range from a low of 71% in Somaliland to a high of 86.8% in NKR with Abkhazia (73.6%) and the TRNC (80.5%) more toward the center of this continuum.

Fig. 3.

The number of cables where the United States can be characterized as actively supporting or being hostile toward a de facto state accounts for nearly one-fifth of our entire data set (87 cables or 19.4%). The two cases that most closely correspond to the view that de facto states are treated with hostility are Abkhazia and NKR. In the case of Abkhazia, there are 27 hostile cables versus just two supportive cables. Examples of hostile US cables toward Abkhazia include opposition to a UN Security Council invitation for Abkhaz or South Ossetian leaders to participate in UN discussions (2008-08-27, Washington, DC) and the Secretary of State underscoring the United States’ “continuing rejection of Russia’s recognition of Abkhazia and South Ossetia” and urging other governments “to avoid actions that could be viewed as conferring recognition on the separatist governments” (2009-03-19, Washington, DC). One supportive US cable argued that “1 year after the US ceased nearly all aid to the breakaway regions in the aftermath of the Russia–Georgia war, the time is right to re-engage with Abkhazia. The Georgian government and the Abkhaz de facto authorities both support the idea, and US long-term goals are better served with an active presence in Abkhazia” (2009-09-08, Tbilisi).

Examples of hostile cables toward NKR include the US ambassador to Azerbaijan urging the State Department to issue a “strongly worded statement rejecting the December 10 referendum in the so-called ‘Nagorno Karabakh Republic’” (2006-11-24, Baku) and urging the State Department “to consider a strongly worded statement from the podium, affirming the US Government’s support for Azerbaijan’s territorial integrity, including Nagorno-Karabakh and the seven occupied territories” (2007-04-23, Baku). The one supportive cable related to NKR recommended “that the international community be encouraged to involve Karabakhis in conferences, trainings, virtual communities and exchanges that put them in contact with the rest of the world …. Contrary to the Azeri strategy, the current isolation engenders not greater sympathy for a return to Azerbaijani rule, but instead ungrounded expectations about NK’s viability as a potential independent state” (2009-11-02, Yerevan).

In contrast to Abkhazia and NKR, Somaliland has an almost evenly balanced number of supportive (13) versus hostile (14) cables. One recurring area where US support is expressed is in terms of security and counter-terrorism cooperation. We learn that an attack on a Chinese oil facility in lamented Beijing’s economic engagement model, saying it undermined democracy and mired African countries in debt. When he landed in Ethiopia near Somaliland underscores “the value of enhancing counter-terrorism cooperation with Somaliland Authorities” (2007-05-08, Addis Ababa) while “reports of terrorist threats coming from groups crossing Somaliland into Ethiopia underscore the need for developing an information-sharing relationship with Somaliland” (2007-06-05, Addis Ababa). Another consistent area of US support is for Somaliland’s continued democratization. At one point, the US ambassador to Kenya states that “Somaliland’s continued democratic development should be a top U.S. policy priority for the Horn of Africa” (2009-06-11, Nairobi).

US hostility toward Somaliland shows up in two main and related areas. First, specific concerns are expressed about the arrest of both opposition politicians (2007-08-03, Washington, DC) and journalists (2009-07-17, Nairobi). Second, the bulk of US hostility concerns repeated attempts by the Rayale administration to postpone or manipulate Somaliland’s presidential election originally scheduled for 2008. Ultimately, the United States actively contemplated targeted sanctions against key members of President Riyale’s inner circle “including travel bans and asset freezes, if they continue to undermine the political process” (2009-09-23, Nairobi).

Perhaps surprisingly, since it is under the most extensive embargo and sanctions campaign of any de facto state and since US diplomatic practice always puts its officials and institutions in quotations (“Prime Minister” or “TRNC Property Commission”), the TRNC is our only case where supportive cables (22) clearly outnumber hostile cables (3). In one cable, the US ambassador to Cyprus notes that “it is important for us to maintain close, if informal, cooperation with those in the banking system who are interested in pursuing these matters” and urges “a positive Washington response to the ‘Central Bank’s’ request for assistance in investigating the suspect transactions in question” (2007-09-13, Nicosia). In another cable, he observes that cooperation with the Turkish Cypriot “Financial Intelligence Unit” or MEBEB “has been excellent although they complain about their inability to interact directly with international law enforcement because of the non-recognition of the ‘TRNC’ and its agencies” (2009-04-24, Nicosia).

In the aftermath of the Turkish Cypriot yes vote on the Annan Plan, the United States expresses support repeatedly for “ending” (2004-06-08, Istanbul; 2004-08-27, Ankara; 2005-10-20, Ankara; 2008-08-13, Istanbul) or “easing” (2005-12-28, Nicosia) “the isolation of Turkish Cypriots.” In this context, the United States also clearly expresses its support for EU Commission efforts to increase aid to Northern Cyprus (2004-07-30, Brussels), with the US ambassador to Cyprus going so far as to recommend that the State Department “ask the Embassies in the seven abstaining countries [on a prior EU vote] to lobby for approval of the EU assistance at the next PHARE Committee meeting” (2006-07-14, Nicosia).

The US ambassador to Cyprus also recommended allowing TRNC President Talat to have a photo opportunity with Secretary of State Hillary Clinton during his visit to Washington just ahead of TRNC elections where the pro-settlement Talat appeared to be losing to a more anti-settlement candidate. The ambassador observed that “the Embassy fully understands the sensitivities involved, especially amongst the Greek- and Greek Cypriot-American community. That said, allowing a quick photoshoot would be consistent with past State practice and go a long way toward helping Talat electorally” (2009-04-14, Nicosia; see Figure 4).

Fig. 4.

Issues Addressed

The United States has engaged these four de facto states on a wide variety of different issues (see Figure 5). To take the TRNC as an example, some of the diverse issues addressed in the WikiLeaks cables include the TRNC’s deteriorating water supplies (2008-01-07, Nicosia); the prognosis for its university sector (2006-10-13, Nicosia); the prospects for TRNC exports through Greek Cypriot ports (2006-06-27, Nicosia; 2008-02-28, Nicosia); and problems with the TRNC census (2006-05-02, Nicosia). The diversity of issues covered with Somaliland includes refugee repatriation (2004-03-04, Djibouti); the status of Somaliland’s financial services and banking legislation (2009-06-02, London); requests for security clearances for US officials to travel to Hargeisa (2008-09-23, Nairobi); and discussions of “interim status” or “semi-recognition” (2004-12-06, Djibouti; 2009-02-02, Addis Ababa).

Fig. 5.

That said, one can clearly see the importance of dominant issues in US foreign policy with three of our case studies. In the case of NKR, the dominant issue accounting for 58.5% of all of its cables is managing or settling the conflict. Perhaps the most interesting cable here is one where Azeri President Ilham Aliyev tells the US ambassador to Azerbaijan that the Minsk Group format is “useless,” and asks “that the United States now impose a settlement on Armenia and Azerbaijan as he believes, it has on the Balkans” (2007-06-15, Baku). Before the 2008 war, conflict management was also the dominant issue in Abkhazia and it accounts for nearly one quarter (23.6%) of all the Abkhazian cables. The Russian recognition of Abkhazia and South Ossetia, however, quickly made status the dominant theme in US foreign policy toward Abkhazia. Status or, more accurately, non-recognition accounts for 53.2% of all the Abkhazian cables. In Somaliland, the dominant issue of concern was the 2008–2010 presidential election crisis when President Riyale correctly feared losing a free and fair election and postponed that election repeatedly (he ultimately conceded defeat and peacefully transferred power after an election in June 2010). The United States and other Western donors frequently engaged opposition party leaders and pressured the Somaliland administration to hold internationally acceptable elections. At the peak of the electoral crisis, US diplomats issued 11 cables a month on Somaliland’s domestic politics in both August and September 2009. Cables relating to Somaliland’s domestic politics ultimately accounted for 45.2% of all its WikiLeaks cables.

The only exception to this pattern of a single dominant US foreign policy focus was the TRNC. In this case, the single largest issue, the TRNC’s economy and ways to ease the economic isolation of the Turkish Cypriots after they voted yes on the Annan Plan accounted for more than one-third (38.3%) of all the TRNC WikiLeaks cables. Conflict management (30.5% of all cables) and the domestic politics of the TRNC (21.1%) were also important topics for US foreign policy.

Conclusions

Our findings based on the WikiLeaks cables indicate that at least some of the traditional narrative on how sovereign states deal with de facto states needs to be revised. We do find some evidence of the traditional hostility toward de facto states, particularly in the active and global US diplomatic campaign to discourage recognition of Abkhazia after the 2008 Russo-Georgian war. Yet, instead of active opposition through the use of embargoes and sanctions, we repeatedly find the United States offering various forms of diplomatic support to ease the economic isolation of Turkish Cypriots. Our findings also make it hard to sustain the argument that de facto states are ignored. Perhaps the closest any of our case studies come to this is Nagorno-Karabakh where NKR leaders are seldom granted contact of any form with US diplomats. Albeit within the general context of non-recognition, our cables reveal broad, sustained, and often friendly US interactions with Somaliland and the TRNC.

Our findings also call into question the repeated argument in the de facto state literature that the international community lacks clear direction or purpose in its engagement with de facto states. Kolstø (2006:734), for example, opines that the problem with the involvement of the international community in de facto state conflicts “is indecision and inconsistency” which, in his view, “clearly reflects the low priority these conflicts have in Western capitals.” Similarly, Lynch (2004:10) argues that “current international approaches, lacking coordination and strategy, work against one another and thus sustain the status quo.” Our cables reveal the consistent US pushes for conflict settlements in Abkhazia and Northern Cyprus and a very active global push against recognizing Abkhazia. They also reveal the United States regularly coordinating with its European allies on Northern Cyprus and on Somaliland.

Berg and Kuusk (2010:46) argue that “on the basis of external sovereignty, de facto states form a more homogenous unit than for internal sovereignty.” This is probably still true but the WikiLeaks cables reveal significantly different levels of acceptance from or engagement with external actors. In particular, our findings reveal a sharp dichotomy between Abkhazia and NKR on the one hand and Somaliland and the TRNC on the other with the latter pair consistently being engaged more directly and more favorably than the first pair.

In terms of our hypotheses, the evidence gleaned from the WikiLeaks cables is broadly consistent with H₁‘s realist emphasis on the importance of great-power relations. As expected, the United States is most supportive of the TRNC (Turkey) and least supportive of ABH (Russia) and NKR (Armenia). Somaliland, without a patron state, presumably benefits from its friendship with elected with 55 percent of the vote. At the time of our meeting, he’d already met leaders in neighboring Djibouti and Ethiopia, a strong US ally in the Horn of Africa. The dominant focus on conflict management (ABH and NKR) and status or non-recognition (ABH) with our two least supported cases also lends support to H₁.

Our findings are also at least somewhat consistent with H₂‘s liberal emphasis on the importance of democracy. The most democratic de facto state, the TRNC, is also the most supported de facto state. The dominance of domestic politics in the Somaliland cables and its relative importance in the TRNC cables also lend some support to H₂. H₂ is, however, somewhat problematic in that there is not a lot of democratic distance between ABH, NKR, and Somaliland and yet the latter case receives far more favorable treatment from the United States than either of the two former cases. This arguably lends additional support to H₁.

The WikiLeaks cables provide little evidence to support H₃‘s constructivist emphasis on the importance of linguistic presentation. There is some support for H₃ in that Somaliland receives the most favorable linguistic presentation and some of the most favorable diplomatic treatment. The TRNC’s linguistic disparaging, however, seemingly has no relation to its diplomatic treatment by the United States. In contrast to H₃, our findings suggest that language means nothing and quotation marks (for example, “President” Talat) do not indicate the depth or attitudinal aspects of engagement.

Finally, our findings suggest rejecting H₄‘s emphasis on the importance of domestically based ethnic interest groups. NKR is treated comparatively poorly despite strong support from Armenian-Americans while the TRNC, language aside, is treated comparatively well despite strong opposition from Greek- and Greek Cypriot-Americans. Somaliland is also treated much better than Abkhazia despite the fact that neither has significant diaspora support in the United States.

In short, our findings do not necessarily fundamentally overturn insights generated by the existing de facto state literature. They do, however, call some conventional wisdom into question and qualify, limit, or modify other theoretical arguments advanced. While uncovering the WikiLeaks cables, great-power–de facto state relations become far richer, more variegated, and, in places, far more passionate and emotional than is typically assumed.