CC BY
73
No. 5(47), 2007 CENTRAL ASIA AND THE CAUCASUS
REGIONAL SECURITY I 4
TERRITORY,
POPULATION, ETHNOSES,
AND SECURITIZATION:
ON THE ENDOGENOUS FACTORS OF SECURITY IN THE REGIONAL SYSTEMS OF THE CAUCASUS AND CENTRAL ASIA
Jannatkhan EYVAZOV
Deputy Director,
Institute of Strategic Studies of the Caucasus, Executive Secretary of the Central Asia and the Caucasus journal of social and political studies (Baku, Azerbaijan)
I n t r o d u c t i o n
This is an attempt to assess the endogenous security factors of the regional political systems of the Caucasus and Central Asia in order to find out, in particular, how the region’s territorial-demographic and ethnic factors affect the basic perceptions of security, as well as the securitization1 processes occurring in this context. I
1 The securitization phenomenon originally interpret-
ed as a process of comprehension by society/the state of cer-
do not claim to provide exhaustive answers here to all the questions relating to the problems I in-
tain phenomena as an existential security threat was theoretically substantiated by the Copenhagen School. For more detail, see: B. Buzan,O. Weaver, J. De Wilde, Security. A New Framework for Analysis, Rienner Publishers Boulder, London, 1998; O. Weaver, B. Buzan, M. Kelstrup, P. Lemaitre, Identity, Migration and the New Security Agenda in Europe, Pinter, London, 1993, and also B. Buzan, O. Weaver, Regions and Powers. The Structure of International Security, Cambridge University Press, Cambridge, 2003, etc.
tend to investigate. At the same time, the specifics of the territorial, demographic, and ethnic structures, especially in the regions in which “modern states”2 predominate (Central Asia and the Caucasus belong precisely to this category), directly affect the perceptions of the threats and vulnerabilities created by the foreign political environment and domestic sub-national groups. This approach might clarify the reasons for the post-Soviet conflict dynamics in the Caucasus and Central Asia.
This assessment is based on certain specific conceptual-categorial provisions calling for preliminary explanation. First, this article looks at the regional political systems of the Caucasus and Central Asia as Regional Security Complex-
2 By way of assessment of the states’ sociopolitical development, B. Buzan and O. Weaver have identified three types (levels): pre-modern states (with a very low level of internal sociopolitical cohesion and state organization; weak government control over the territory and population); modern states (strong government control over society, limited openness, sanctity of sovereignty and independence and related attributes, including territory and borders, and stakes placed on self-sufficiency, self-reliance, and national identity), and post-modern states (relatively moderate treatment of the sovereignty, independence, and national identity issues, openness in economic, political, and cultural contacts with the outside world) (for more detail, see: B. Bu-zan, O. Weaver, op. cit., pp. 22-24).
es (RSC).3 Second, I do not intend to operate using the fairly limited traditional spatial political division of the Caucasus into two segments— the Northern and the Southern Caucasus. I am proceeding from a relatively recent, yet much more adequate structure (the Northern, Central, and Southern Caucasus) better suited to objectively reflect the region’s most important geopolitical and ethnocultural specifics.4
3 B. Buzan’s model of the regional security complex is based on the interdependence of the national security interests of a geographically close group of states. The original interpretation speaks of RSC as “a group of states whose primary security concerns link together sufficiently closely that their national securities cannot realistically be considered apart from one another” (B. Buzan, People, States and Fear. An Agenda for International Security Studies in the Post-Cold War Era, Second edition, Lynne Rienner Publishers Boulder, Colorado, 1991, p. 190).
4 This conception divided the Caucasus into three parts: the Northern (the autonomous entities of the Southern Federal District of the Russian Federation); the Central (the independent republics of Azerbaijan, Georgia, and Armenia) and the Southern (the Northeastern ils of Turkey, or the Southwestern Caucasus, and Northwestern ostans of Iran, or Southeastern Caucasus) (see: E. Ismailov, Z. Ken-gerli, “The Caucasus in the Globalizing World: A New Integration Model,” Central Asia and the Caucasus, No. 2 (20), 2003; E. Ismailov, V. Papava, The Central Caucasus: Essays on Geopolitical Economy, CA&CC Press® AB,
Stockholm, 2006.
The States, Borders, and Formation of the Basic Security Perceptions within the Caucasian and Central Asian RSC
The processes that led to the security perceptions at the national level in the RSC were very different, mainly because they required different periods of time to come to fruition, which could not but affect the stability of the perceptions themselves. The autonomous regional political systems took shape during different historical periods divided by a fairly wide time gap.
The RSC in the Caucasus is not a post-Soviet reality to be described as an aftermath of the Soviet Union’s disintegration: it was an earlier structure, the dynamics of the relations of which were revived in the post-Soviet period.5 The basic interdependence vectors in the security sphere and the
5 See also: B. Coppieters, “Conclusions: The Caucasus as a Security Complex,” in: Contested Borders in the Caucasus, Ed. by B. Coppieters, Brussels, Vubpress, 1996, pp. 194-195; S.E. Cornell, Small Nations and Great Powers. A Study of Ethnopolitical Conflict in the Caucasus, Curzon Press, 2001, p. 24.
corresponding dynamics of relations appeared in the pre-Soviet period; the total Soviet control that took shape in the 1920s merely “conserved” them; B. Buzan describes this as “overlay.”6
The security relations system in Central Asia is post-Soviet realia; the difference between the two regions is easily explained by the different periods of statehood formation in the Caucasus and Central Asia. In the former, the national states came into being in the first quarter of the 20th century, after the Bolshevik revolution and disintegration of the Russian Empire (the Azerbaijan Democratic Republic of 1918-1920; the Georgian Democratic Republic of 1918-1921, and the Armenian (Ararat) Republic of 1918-1920).7 In Central Asia, on the other hand, the national republics proper are the product of Soviet power; they became sovereign states in the early 1990s, during the post-Soviet period. This means that in the Caucasian RSC, the endogenous security factors were first activated in the first third of the 20th century, which helped to create stable perceptual security constructions that were revived as soon as outward “overlay” disappeared and the Central Caucasian republics restored their independence in the 1990s.
The basic constructions of political relations of Central Caucasian states of the late 20th century (both among themselves and with the external actors) almost completely coincided with the constructions of the first independence period.
In the early 20th century, territorial issues were very much prominent in the three independent republics; Azerbaijan and Georgia managed to resolve their territorial disputes more or less promptly and painlessly, while their territorial disagreements with Armenia developed into an armed conflict. In fact, moderation and the desire to cooperate that Azerbaijan and Georgia demonstrated in their mutual relations during the first and the second independence periods offered a striking contrast to the fairly complicated relations with Armenia and its relative alienation.
The same applies to the nature of the perceptual constructions in the three republics as friendly/ hostile toward other states. During both periods, Armenia demonstrated an obvious pro-Russian bias irrespective of who ruled Russia: be it Denikin’ Volunteer Army, or the Bolsheviks. Today this is the Russian Federation. In every case, Armenia tended to neglect the harm this bias might do to its independence. Azerbaijan and Georgia, on the other hand, placed their stakes on cooperation with Turkey and the West.
The five Central Asian republics acquired their official territories and institutions within the Soviet Union, which explains why the autonomous conceptualization of the security sphere at the national level could have occurred only in the post-Soviet period. This could not help to create stable perceptions of the threats and vulnerabilities conditioned by the regional geopolitical environment. The territory and the borders as the main objects of securitization of modern states could not develop in Central Asia into an active and determining security relations function. As distinct from the Caucasus, where the autonomous national-state formations have been settling the national territorial issue since the first third of the 20th century, with all the ensuing conflict potential, in Central Asia this function belonged to the Soviet Center.8
A mere glance at the political map reveals the striking differences of delimitation in the two regions. In the Caucasus, the land border meanders to form enclaves and exclaves, while the Central Asian borders are mainly lineal and very simple. This fully applies to the 2,203-km-long9 bor-
6 See: B. Buzan, op. cit., pp. 219-221.
7 It should be said that in the wake of the Russian Empire’s disintegration, several more republics appeared in the Caucasus, including the Gorskaia (Mountain) Republic in the north (1917-1919); the Southwestern Caucasian (Kars) Democratic Republic (1918-1919), and the Araz-Turkic Republic (1918-1919) in the southwest; the Republic of Azadistan (1920) and the Gilian Republic (1920-1921) in the southeast.
8 For more details about the Central Asian borders, see: S. Golunov, “The Post-Soviet Borders of Central Asia: Security and Cooperation,” Central Asia and the Caucasus, No. 5 (11), 2001, pp. 141-152.
9 See: The CIA WorldFactbook 2006—Kazakhstan, available at [https://www.cia.gov/cia/publications/factbook/geos/ kz.html], 3 December, 2006.
der between Kazakhstan and Uzbekistan, the 379-km-long10 border between Kazakhstan and Turkmenistan, and the 1,621-km-long11 border between Uzbekistan and Turkmenistan. This could have been explained by the terrain; however, it reveals the importance of the territories with different terrains and climates (or, rather, the possession of such territories) for securitization in the states concerned.
Securitization of the “Territorial Deficit” and the Conflict Level
The specific features of the Caucasian borders speak of the very high importance of the territories enclosed by them, and of the never-ending struggle for possession of the territories, as well as of the much higher securitization level of the geographic conditions and factors (geo-securitization) in the Caucasus compared with Central Asia. Modern states assess the territories from the point of view of their natural conditions conducive to material and economic development. This much is obvious and should not surprise anyone. In the Caucasus, where the nearly entire territory was highly favorable, attempts to enlarge the “Lebensraum” were frequent. In the periods when the region was controlled by outside forces, territorial rivalry went on at the social level, which logically and inevitably created a “historical insults” complex in the minds of the regional ethnoses, as well as the “lost lands” perception together with the threats and vulnerabilities related to potential occupation of their “historical lands.”
In Central Asia the situation was different: deserts, which occupy a large part of the region, could not be perceived as territories suitable for the local ethnoses’ material and economic development. The Ferghana Valley was the only area that could be regarded as such. Today it is divided among three Central Asian states—Kyrgyzstan, Tajikistan, and Uzbekistan. In the late 1980s, it became the most dangerous seat of regional tension.
We can argue that the conflict level should be higher in regions with an obvious deficit of desirable territories. Anyone wishing to compare the dynamics of security relations in the Caucasian and Central Asian RSC from the viewpoint of this stimulating element will be forced to conclude that, under identical conditions, the negative dynamics in the Central Asian RSC generated by the endogenous territorial contradictions would have been more pronounced than in the Caucasus. Such an analysis, however, requires that other, and highly important, factors should also be taken into account. In the final analysis, they change the anticipated situation by 180 degrees. It should be said here that the territory/population size correlation is different in the two regions (see Table 1).
Table 1 shows that if we take into account the autonomous central part of the Caucasus (Azerbaijan, Georgia, and Armenia) and exclude the corresponding parameters for Afghanistan from our investigation of Central Asia (for the same reason we exclude the Northern, Southeastern, and Southwestern Caucasus, due to the absence of de facto fairly developed autonomous political behavior, including independent security policy), we shall discover that the average population density of the Caucasian RSC is more than five times larger than the same parameter for Central Asia. Even if we
10 Ibidem.
11 See: The CIA World Factbook 2006 — Uzbekistan, available at [https://www.cia.gov/cia/publications/factbook/geos/ uz.html], 3 December, 2006.
Table 1
General Correlation between Population and Territory (Population Density) in the Caucasian and Central Asian RSC12
| No. Regional Security Complex Territory (thou. sq km) Population Size (million) Density A
1. Central Asian (including Afghanistan) 4,564.8 91.2 19.8
2. Central Asian (without Afghanistan) 3,917.3 60.1 15.4
3. Caucasian (entire region) 905.1 50.5 56.1
4. Caucasian (limited to the independent republics—Azerbaijan, Georgia, Armenia) 184.2 15.6 84.6
discuss the indices for the entire region (the Central, Northern, Southeastern, and Southwestern parts of the Caucasus, as well as the five Central Asian republics plus Afghanistan), the difference remains large (nearly three-fold). All other conditions being equal, this means that the risk of social conflicts in Central Asia produced by “territorial deficit” securitization is at least four times lower. The case of the Ferghana Valley confirms the point: the region, which since the late 1980s has been the seat of potential conflicts, is marked by the highest population density (up to 500 per 1 sq km).13 This is the only area in the region where territorial rivalry is high. Central Asia differs in this respect from the Caucasus, where there are much more similarly densely populated areas.
Ethnic Homo/Heterogeneity
Ethnic and religious specifics of the competing entities are additional, and very important, factors that could affect any of the territorial conflicts in the regions discussed. The natural and logical conclusion is that the likelihood of rivalry over favorable living conditions developing into a conflict is higher in ethnically and confessionally heterogeneous populations. Regional realities demonstrate considerable differences, which can be used to explain the general stimuli for regional security relations and the current differences in their relative negative/positive nature within the Caucasian and Central Asian RSC.
12 The figures for Central Asia and the Caucasus are based on information taken from The CIA World Factbook 2006. The figures (2002) are based on calculations for the sub-regional divisions of the Caucasus (Northern, Southwestern and Southeastern) found in E. Ismailov, V. Papava, op. cit., pp. 67, 73.
13 See: N. Ziadullaev, “Central Asia in a Globalizing World: Current Trends and Prospects,” Central Asia and the Caucasus, No. 6 (42), 2006, pp. 125-126.
Let’s have a look at the ratio between the conventionally titular and non-titular ethnoses, as well as the average population density of the two regions’ political expanse (see Table 2). This ratio may affect the domestic bias toward conflicts in the countries of the Caucasian and Central Asian RSC and their relations. We all know that the conflictive nature of the region came to the fore on the eve of the Soviet Union’s disintegration and the weakened “overlay” regime. The table takes this factor, which is dated to 1989 (based on the last Soviet population census), and the current data into account.
Table 2
Correlation between the Titular and Non-Titular Ethnoses and Population Density in the Countries of the Caucasian and Central-Asian RSC (1989 and 2006)14
H State Total Population Titular Ethnos (percent) Non-Titular Ethnoses (percent) Coefficient of Population Density
1989 2006 1989 2006 1989 2006 1989 2006
1. Azerbaijan 7,021,178 7,961,619 82.7 90.6 17.3 9.4 81.5 92.4
2. Armenia 3,304,776 2,976,372 93.3 97.9 6.7 2.1 116.3 104.8
3. Afghanistan — 31,056,997 — 42 — 58 — 47.9
4. Georgia 5,400,841 4,661,473 70.1 83.8 29.9 16.2 77.4 66.8
5. Kazakhstan 16,464,464 15,233,244 39.7 53.4 60.3 46.6 6.2 5.7
6. Kyrgyzstan 4,257,755 5,213,898 52.3 64.9 47.7 35.1 22.2 27.2
7. Tajikistan 5,092,603 7,320,815 62.3 79.9 37.7 20.1 35.7 51.3
8. Turkmenistan 3,522,717 5,042,920 72.0 85 28 15 7.2 10.3
9. Uzbekistan 19,810,077 27,307,134 71.3 80 28.7 20 46.6 64.1
The above suggests that although the three states of the Caucasian RSC are ethnically less heterogeneous, this relative homogeneity is accompanied by their much higher population density, which, theoretically, should have exacerbated securitization of “territorial deficit” in this region compared with the Central Asian states. The table offers several very interesting figures: Azerbaijan and Armenia are the two most densely populated countries. We all know that the Armenian-Azeri conflict remains the most acute and bloodiest across the post-Soviet expanse. There is an obvious correlation
14 The table does not contain data relating to the ethnic situation in the Northern, Southeastern and Southwestern Caucasus, even though the heterogeneous nature of the Caucasus is associated more with these three sub-regions (particularly with the Northern Caucasus) than with the three Central Caucasian states. I explain this by the fact that it is very difficult to identify the titular or the non-titular status of the ethnoses in the three parts of the Caucasus. The sub-regions are not politically independent parts of the Russian Federation, Iran or Turkey. Under these conditions, the titular/non-titular label is meaningless. The figures are based on data taken from The CIA World Factbook 2006, as well as on the data of the U.S.S.R. population census of 1989 (see: “Rasselenie narodov SSSR po soiuznym respublikam po perepisi 1989 g.,” Soyuz, No. 32, August 1990).
between the acutest conflict dynamics between Armenia and Azerbaijan and the “territorial deficit” as perceived by the public of the two countries, even though there are other factors that triggered the conflict and are keeping it going. The late 1980s was a very important period and turning point in the development of the post-Soviet bias toward conflict in the relations between the two nations. It was during this period that the two republics reached the highest level of ethnic heterogeneity: in Azerbaijan, it was 82.7/17.3 and in Armenia 93.3/6.7 percent.
The “territorial deficit” as perceived by the public is becoming an independent securitization object and, as such, affects the states’ behavior in the security sphere. This is best illustrated by Armenia and its behavior in the Nagorno-Karabakh conflict and in its geopolitical context as a whole. The conflict was initiated by Armenia’s territorial claims (its population density coefficient in 1989 was 116.3, the highest in the Central Eurasian region) to the neighboring republic. By 1994, by the time the cease-fire agreement was signed, Armenia had occupied not only Nagorno-Karabakh with its predominantly Armenian population (the war’s original cause), but also seven districts of Azerbaijan with a total area of about 9,988 sq km15 and a predominantly Azeri population that was driven away from the captured areas. The facts of illegal settlement of Armenians on the Armenia-occupied territories of Azerbaijan exacerbated the conflict in late 2004 and early 2005 to the extent that a Special OCSE Mission had to come to the conflict area. It registered that at least 15-16,000 Armenians had settled illegally in the occupied Azeri territories.16
This example is illustrative in another respect: “territorial deficit” securitization in Armenia turned out to be powerful enough to affect the country’s relations with nearly all of its close neighbors. Indeed, its relations with three (Azerbaijan, Turkey, and Georgia) out of four of its neighbors (Azerbaijan, Turkey, Georgia, and Iran) are strained to different degrees because of territorial issues; there are other factors behind this as well. The hostility perceptions in Armenian’s relations with Azerbaijan and Turkey are especially strong: it is at war with the former and has no diplomatic relations with the latter; in the case of Turkey, the conflictive level is high enough to develop into an armed confrontation.
To assess the domestic ethnopolitical dynamics, let’s have a look at the ethno-territorial descriptions of the two republics on the eve of the Soviet Union’s demise. These countries, now part of the Caucasian RSC, are today, as they were in 1989, the most densely populated. By the end of the 1980s, that is, on the eve of the heightened ethnopolitical conflict, the titular/non-titular population ratio was different from what we have today. It would be logical to expect that the presence of ethnic minorities in the densely populated states and other social problems (economic decline, the ideological vacuum, and the still unfinished quest for national identity) may create a conflictive domestic ethnopolitical situation. Whether or not ethnic minorities have been living in compact groups on definite territories for a long time may prove to be very important: population mobility within a state or its permanent settlement in definite places may affect the intensity of ethnopoliti-cal conflicts to different extents.
On the one hand, the prospects for the ethnopolitical situation in Azerbaijan, seen in the context of the correlation of domestic demographic and territorial factors, are much more complicated than in Armenia, which is essentially a mono-ethnic state. In Azerbaijan, on the other hand, there are compact ethnic communities. This is one side of the coin. “Territorial deficit” securitization in Armenia is much higher than elsewhere in the Caucasian RSC, which will affect its security policy
15 They are the Lachin (1,835 sq km), Kelbajar (3,054 sq km), Agdam (1,154 sq km), Fizuli (1,386 sq km), Jebrail (1,050 sq km), Gubadla (802 sq km), and Zangilan (707 sq km) districts.
16 According to the calculations made in Azerbaijan, there are 23,000 Armenians illegally residing in the occupied districts (see: Republic of Azerbaijan, Ministry of Foreign Affairs [http://www.mfa.gov.az/img/map_eng.gif], 2 August, 2007).
and maintain tension in its relations with its neighbors. This may trigger negative dynamics within the region.
The case of Georgia offers no less pertinent generalizations. The population density coefficient on the eve of the end of the Soviet era and in 2006 was much lower than in Azerbaijan and Armenia (77.4 in 1989 and 66.8 in 2006), but higher than in any of the post-Soviet Central Asian states. At the same time, today, as in the late 1980s, Georgia remains ethnically the most heterogeneous of the states in the Caucasian RSC.
The ethnic minorities of Georgia comprise a considerable part of its population, even after the post-Soviet migration wave. They mainly live in compact groups along Georgia’s borders. Two such areas, Abkhazia and South Ossetia, were highly autonomous parts of the Georgian S.S.R. both in theory and in practice. This, as well as the weakened central power structures and the acute political instability of the late 1980s channeled “territorial deficit” securitization within rather than beyond the country (the cases of Azerbaijan and Armenia). In other words, the two autonomous republics vs. central power in Tbilisi became securitizing actors; the role of referent object applied to Georgia’s territory as a whole. The ethnic groups of the two autonomous formations, as well as those in other parts of the country (the Georgians of Abkhazia and South Ossetia, the Abkhazians and Ossets living elsewhere in Georgia), were regarded as sources of danger. Here is an interesting fact: with a total population of 525,061 in the late 1980s17 and a total territory of 8,600 sq km (population coefficient of 61.0), the numerical strength of the Abkhazian population proper was merely 93,267, or 17.8 percent.18 The number of local ethnic Georgians was 239,872, or 45.7 percent,19 of the republic’s total population. This ratio might have strengthened securitization of the “territorial deficit” in the autonomy, which had been striving for independence from Georgia for some time, and encouraged the Abkhazian authorities to evict the Georgians from the republic. By the late 1980s, South Ossetia reached the same threshold: of the total population of 98,500 living on a territory of 3,900 sq km, 29 percent were Georgians and 66.2 percent Ossets.20
The Central Asian territorial-demographic picture offers no less interesting conclusions. Tajikistan is a pertinent example: its ethnic heterogeneity/population density coefficient-conflict potential correlation is the most indicative. The newly independent republic had to cope with the bloody civil war of 1992-1996, the first manifestation of trouble in the more or less stable post-Soviet situation in the region as a whole. It remains specific to the same extent to which the ethno-territorial situation in the republic is specific.
It is the smallest in the Central Asian RSC with the third largest population (see Table 2). The ethno-linguistic situation and the degree of the population’s heterogeneity are also highly specific. First, Table 2 shows that today, as it was in 1989, the population density in Tajikistan is one the highest among the former Soviet Central Asian republics (35.7 and 51.3, respectively): the country comes second after Uzbekistan. Second, the difference between the shares of the titular and non-titular population in the total population is relatively insignificant: the republic comes third in this respect after Kazakhstan and Kyrgyzstan. Third, the republic’s titular population belongs to the Iranian Indo-European ethno-linguistic group, which means that Tajikistan is the only non-Turkic republic in an otherwise Turkic region.
This led to a relatively high level of “territorial deficit” securitization yet, like in Georgia of the late 1980s, its vector was directed within the country rather than beyond it. Just like in Georgia, eth-
17 See: G. Hewitt, “Abkhazia, Georgia and the Circassians (NW Caucasus),” Central Asian Survey, No. 18 (4), 1999,
p. 463.
18 Ibidem.
19 Ibidem.
20 See: Georgia: Society, Language and Culture, Population, available at [http://webzone.imer.mah.se/projects/geor-gianV04/DEMO/GeoLINK/Intro2.html], 2 August, 2007.
no-political conflict trends accelerated in Tajikistan because central power proved to be weak; local ethnic and clan groups acted as securitizing actors. The civil war ended in the mid-1990s, probably not so much under the impact of endogenous factors: the Russian armed forces deployed in Tajikistan obviously did not limit themselves to peace keeping and border guarding, which means that the domestic conflict dynamics were suppressed or overlaid by Russia’s military presence.
The external vectors of Tajikistan’s “territorial” securitization are closely connected with Uzbekistan, the most populous post-Soviet Central Asian state with the third largest territory in the region, which makes the republic the regional leader in terms of population density coefficient: 46.6 in the late 1980s and 64.1 today. In this respect and in the titular/non-titular population ratio, Uzbekistan comes close to the states of the Caucasian RSC. Table 2 shows that in both periods, the difference between the shares of the titular and non-titular populations in the total population of Uzbekistan was one of the region’s highest.21 Logic suggests that judging by the first parameter, Uzbekistan should have a high level of “territorial” securitization and, correspondingly, a high conflict level inside the republic and in its relations with its neighbors. For the same reason, the second index should have quenched the domestic conflict potential. On the other hand, however, whereas the first parameter is fairly prominent, the second parameter cannot always reduce the conflict level of the country’s relations with its closest neighbors. Destabilization of such relations is highly possible. In this case, the possible scenario depends on the other factors involved in “territorial” securitization of the countries discussed. The ethnic minorities in one state, which are in the majority (titular ethnoses) in a neighboring state, the extent to which they tend to live in compact groups, the distance between such places and their historical homeland, the level of political autonomy, etc. are all ethno-territorial factors.
Uzbekistan’s strained relations with its neighbors—Tajikistan, Kyrgyzstan, and to a considerable extent, Kazakhstan—are caused, among other things, by the specifics of population distribution within its borders. The largest part of its 25 million-strong population is crowded in the east, while the west and the center with their harsh climates are sparsely populated. The western parts of Kyrgyzstan and Tajikistan, which border on the densely populated parts of Uzbekistan, are also overpopulated. Here I particularly have in mind the Ferghana Valley, an overpopulated and ethnically variegated area, which in the late 1980s was the scene of one of the most tragic events in the region’s history: an ethnic conflict that drove away over 90,000 Meskhetian Turks.22 This population distribution on both sides of the border (in Uzbekistan, on the one hand, and in Tajikistan, Kyrgyzstan, and Kazakhstan, on the other) presupposes a high level of “territorial” securitization in all of these countries. The local nations regard one another and the compact ethnic groups on their territories ethnically related to the ethnic majorities across the border as a threat.
On the whole, according to Table 2 the population density coefficient in Turkmenistan, Kyrgyzstan, and especially in Kazakhstan is much lower than in the other two Central Asian states, which somewhat defuses the impact of the “territorial deficit” on the general securitization process. This means that these states are less inclined to military-political expansion. The above, however, does not exclude the possibility of them taking up arms against their neighbors to defend themselves. Internal ethnopolitical dynamics might cause an international conflict: because they have the highest level of ethnic heterogeneity in the region, Kyrgyzstan and Kazakhstan are potentially more likely to be involved in such conflicts.
Turkmenistan is the region’s most stable state in terms of the population density coefficient/ethnic structure ratio. Indeed, its population density coefficient23 is one of the lowest in Central Asia (even
21 In this respect, Uzbekistan comes second after Turkmenistan (see Table 2).
22 See: A. Yunusov, Meskhetinskie Turki: Dvazhdy deportirovanniy narod, Zaman Publishers, Baku, 2000, p. 95.
23 In this respect, Turkmenistan comes second after Kazakhstan.
if we take into account the fact that a large part of its territory is covered by desert), which means that the republic’s vast territory, second to Kazakhstan’s, is one of the least densely populated of the Central Asian countries (see Table 2). This creates a low level of “territorial deficit” securitization and limits contacts with neighbors on territorial issues. The domestic ethnopolitical conflict potential is likewise very low mainly due to the country’s relatively simple ethnic structure. Today, as in the Soviet past, Turkmenistan is the least ethnically heterogeneous republic among its Central Asian neighbors. Table 2 shows that the titular/non-titular correlation in these periods was 72/28 and 85/15, respectively. The share of the non-titular population will continue decreasing, which will soon make Turkmenistan the only mono-ethnic Central Asian republic.
I n L i e u o f a C o n c l u s i o n
The territorial-demographic correlation and ethnic heterogeneity obviously cannot serve as the only, or a sufficient basis for a comprehensive assessment of the securitization processes now underway in the states of the Caucasian and Central Asian RSC. At the same time, these factors are obviously important for forming basic security perceptions within the regional political systems. The degree of development of such systems and the level of its “maturity” affect the degree of securitization of all the factors mentioned above.
The development level of the Caucasian and Central Asian RSC does not allow us to talk about a considerable degree of de-securitization of the geography- and ethnicity-connected factors, which can be observed in Western Europe. It means that they should be taken into account as the key endogenous factors of the security relations within these two regional complexes of Central Eurasia.
The differences in the territorial-demographic and ethnic structures can explain why the security relations in post-Soviet Central Asia are not developing into open armed conflicts, as distinct from what is going on in the Caucasus. As on the eve of the post-Soviet conflict situation, the level of “territorial deficit” securitization is higher today in the Caucasian RSC than it is in the Central Asian RSC. The conflict potential is somewhat alleviated by the populations’ fairly active mobility and labor migration as its part.
The situation might change because of different demographic dynamics and population mobility24 or, to be more precise, due to a stronger “territorial deficit” factor in the securitization process in the Central Asian states and, correspondingly, to its weakening within the Caucasian RSC, with all the repercussions for regional security. No matter what, this is possible, all other things being equal, which, as we know, very rarely remain equal.
24 For example, the demographic differences in the two regions suggest very interesting ideas. According to UNDP Human Development Report 2006, an annual population increase in the Central Caucasian republics in 2004-2015 will be: in Azerbaijan (+0.8), in Armenia—(-0.2), in Georgia—(-0.7); in the republics of Central Asia: in Kazakhstan—(0.0), in Kyrgyzstan—(+1.1), in Tajikistan—(+1.5), in Turkmenistan—(+1.3), and in Uzbekistan—(+1.4) (see: UNDP Human Development Report 2006, Demographic Trend, Annual Population Growth Rate (%), available at [http://hdr.undp.org/hdr2006/ statistics/indicators/39.html], 11 June, 2007).
