A.  Introduction

1  Secessionist tendencies are a phenomenon that most State[s], if not all, are faced with at some point. And most States, though not all, try to prevent such attempts from succeeding. The objection by the State turns separation, if it occurs, into secession. The unilateral withdrawal from a State of a constituent part, with its territory and its population, constitutes secession stricto sensu. As a consequence of secession, the existing State splits in two: the State continues to exist, but a new State comes into existence concurrently (New States and International Law). In other words, what was formerly a constituent part of a State becomes independent—at least from a legal, though not necessarily a factual perspective. Rather than create a new State, the separating part of a State may choose to join an existing State. Such a case also amounts to secession.

2  Secessions in this sense of the term are a rare occurrence. James Crawford has pointed out that since 1945 there has not been a single separation of a constituent part from a State to which the latter has not, sooner or later, given its consent: ‘In fact no new State formed since 1945 outside the colonial context has been admitted to the United Nations over the opposition of the predecessor State’ (Crawford 415) (see also Decolonization; History of International Law, since World War II). Crawford’s statement shows that the role of the existing State’s consent in the context of secession is difficult to capture. One could also understand the term secession as encompassing a situation in which a constituent part separates from a State and the existing State gives its consent at a later stage. According to this understanding, the elements of secession are normally fulfilled at an earlier point in time. The notion then becomes broader, but also hazier, due to the time criterion—unless one couples the term secession, somewhat artificially, with a specific point in time, such as the occasion on which independence is unilaterally declared. However, the stricter notion of secession—the one based on the absence of consent of the pre-existing State—faces similar uncertainty, for when could one tenably claim that a separation is definitely unilateral and complete, that consent is definitely withheld? The stricter notion of secession could only deal with this uncertainty in a credible way if it either: (i) admitted that secession was progressive rather than a one-time, clear-cut event, which would mean that a secession turns into either a separation or a dismemberment (Dismemberment of States) once it has been approved by the pre-existing State; or (ii) accepted that, as a notion, it was hollow, in that it can never be said to have occurred with absolute certainty; in other words, that it is a negative term the elements of which are impossible to fulfil, and the spectre of which only serves to stigmatize attacks on the integrity of States.

3  Take Kosovo as an example: does the separation of Kosovo from Serbia, following the declaration of independence proclaimed by Kosovo on 17 February 2008, now constitute secession because Serbia does not recognize Kosovo’s independence? Or is it too early for this to be the case? Or is it secession until Serbia agrees—if it ever does—and is it then to be regarded as a withdrawal covered by consent, or a separation? (See also Yugoslavia, Dissolution of).

4  In light of these uncertainties, it might be possible to widen the term secession even further. One could also regard secession as every action that leads to a part of a State being separated off, regardless of whether or not this happens with the consent of the existing State. Based on such a definition, secession is then a frequent occurrence. However, such a wide understanding of the notion of secession conflicts with the general understanding of the term and with one’s intuition. Secession seems to be a unilateral process, one that often has negative connotations and that needs to be distinguished from an agreed withdrawal or a dismemberment of a State.

5  Such are the difficulties with the notion of secession. They are important to note, because there is so far no authoritative definition of the notion of secession. They also serve to show that different authors, when writing about secessions, may understand the term in different ways. Notwithstanding this, it is at least possible to situate the phenomenon of secession in general: it is of importance in various areas of international law, such as State sovereignty, territorial integrity (Territorial Integrity and Political Independence), self-determination, minority rights (Minorities, International Protection), and State succession (State Succession in Treaties; State Succession in Other Matters than Treaties).

6  Two thoughts underlie the present review. First, secession, more than other areas of international law, seems to be an amalgamation of legitimacy and legality. To avoid confusion, a special effort is made to keep the two concepts apart. Second, as the role and the limits of international law in situations involving secessions seem to be unclear, the review, in its structure and content, makes an attempt at clarification in this regard. Apart from these two efforts, much of the main body of the text is devoted to a review of international practice, by which secession is largely governed.

B.  The Legitimacy of Secession

7  In any case where secession is becoming a threat, there is a strong inclination to blend normative arguments with legal reasoning. Because legitimacy and legality should generally be distinguished, the following look at the most common reasons put forward to justify the legitimacy of a secession may prove helpful, while bearing in mind, however, that secessionist claims are always embedded in their specific context and that the reasons why such claims are made can only be generalized to a limited extent.

8  An overall distinction can be made between reasons that seemingly justify secession and others that militate against it. Among the strongest arguments in favour of secession is oppression. Systematic oppression of one part of the society of a State can give rise to a secession campaign that is widely felt to be legitimate. The oppression may range from the denial of participatory rights, to serious and systematic discrimination and other violations of human rights of the members of one part of the population. The regime that commits these violations and which oppresses a part of the population it governs thus appears to be illegitimate. Pro-secession legitimacy can be accentuated further by considerations that are unrelated to oppression, such as the geographic isolation of a territory, the ethnic differentness or the size of the population of a territory (Ethnicity), and, most basically, the will of this population as proclaimed in a plebiscite or a referendum. The latter arguments were all discussed in the Åland Islands dispute (The Aaland Islands Question [Report presented to the Council of the League of Nations by the Commission of Rapporteurs 16 April 1921]; see paras 17, 25 below). Of course, they may also serve, mutatis mutandis, in other cases to contest the legitimacy of a secessionist movement, when there is no oppression, no ethnic difference, etc.

9  The main reason why secessions are regarded as illegitimate is stability. Stability is considered to be one of the foundations of international co-operation and law:

10  Of course, a secession, taken in isolation, primarily threatens the stability of one State—that from which a part secedes. This is the intra-State aspect of the stability argument. Here, the argument is fed by the characteristics of the specific State, such as the size of the State and its weight in the international arena, access to the State’s natural resources, various ties—historical, ethnic, etc—between the different parts of the population, and also by considerations of power. But there is also an inter-State aspect to the stability argument. It is based on the reasoning that a runaway reaction might be the result, ie that once a single secession has been allowed to happen, it could well start a rampant backlash in other, third States, which in principle have nothing to do with the first incident. One secession would then lead to a surge of secessionist claims in these third States that could not easily be brought under control. This, the argument goes, would gravely inhibit international co-operation and even threaten the international order as a whole.

11  There is, of course, some truth in this argument. However, the argument could also be seen the other way around. If secession is illegitimate because it causes instability, the opposite could probably be said too: secession causes instability, because it is considered to be illegitimate. The stigma attributed to secessions would, in this perspective, be a cause, and not a consequence of the instabilities following secession. If the stigma were removed, in other words, if secessions were not considered as illegitimate per se, they would possibly not cause instability. Consequently, one could also imagine taking a less conservative approach. Secessions would, in a more dynamic understanding, be handled more freely and, for instance, accepted, if the desire to secede was expressed by the population of a territory in a free and clear way. This would not be totally unprecedented, as can be seen in the Supreme Court of Canada’s ruling that the federal State of Canada and its constituent entities, based on the constitution, would be obliged to negotiate secession in good faith (bona fide), if the population of Quebec, having been asked the unequivocal question, expressed its wish to secede by a clear majority (Reference re Secession of Quebec Supreme Court of Canada [20 August 1998] (1998); see the citation of the paragraphs and the discussion below para. 34). However, this is not an approach that finds much support in academia. On the contrary, it is widely rejected, typically with reasoning that is based on the stability argument. Tomuschat, for instance, expects that ‘such merry jurisprudence would unavoidably pave the way to chaos and anarchy’ (Tomuschat [2006] 24 [note that Tomuschat does not explicitly refer to Reference re Secession of Quebec when making this statement]). Nevertheless, one can probably accept that the concepts of (il)legitimacy and (in)stability are more intimately connected than it might seem at first glance.

12  There is, however, an additional, more compelling argument against the legitimacy of secessions. It is based on a functional perspective of conflict resolution. The conflicts that underlie secessionist tendencies are often strongly influenced by ethnic or national components, especially so in the Balkans. According to Koskenniemi, what then occurs is the ‘“onion problem” of nationalism’ (Koskenniemi 260). The problem stems from the fact that there is always some degree of intertwinement—both territorial and personal—between ethnic groups and nations, and that there is no clear separation line between them. Even though secession results in the creation of a new ethnic or national entity, this new entity, much like the former State, is not ethnically or nationally uniform. It seems that no matter how many smaller entities are created, the ethnic conflict will continue to exist in the ever-smaller entities. Secession does not solve the underlying conflict, be it ethnic or national. Rather, it pushes the conflict one level lower, thus drawing it out and sometimes, in reversing the minority–majority balance, even aggravating the resentments that fuel the conflict. As a means of conflict prevention and resolution, secession is therefore inappropriate and rather clumsy. Hence, the weight of the ethnic or national reason for secession—and with it, it seems, the pro-secession argument as a whole—is very limited. In no case should it be taken as more than just one element to be considered when weighing and balancing the options in a conflict.

C.  The Legality of Secession

13  In an ideal legal order, legitimacy and legality are always the same. With secession, however, such a general congruency is unlikely from the outset, because views differ on what is legitimate and what is not. Furthermore, it can be suspected that international law adheres to the States’ perspective of legitimacy, for it is mainly States and not peoples, which make international law. Unsurprisingly, international law, as will be seen below, does indeed deal with secession in a way that emphasizes the stability of existing States. Nevertheless, the oppression argument has found its way into positive international law.

14  In brief, the international legal situation seems to be that secession in the strict sense of the term is not explicitly forbidden. It is not illegal. But it runs counter to the principle of territorial integrity and the latter ultimately prevails. International law provides no unambiguous basis for a right to secession.

15  In detail, the principle of self-determination enshrined in Art. 1 (2) United Nations Charter as one of the purposes of the UN—‘[t]o develop friendly relations among nations based on respect for the principle of equal rights and self-determination of peoples, and to take other appropriate measures to strengthen universal peace’—could be considered as a legal basis for a right to secession. The right to self-determination, looked at in isolation, certainly means that a people has a right to its own State, for the people that truly determines itself may also choose a State as the appropriate vessel for its fate. However, such an unrestricted reading of the principle of self-determination does not find much support. Rather, the principle is to be balanced with the territorial integrity and sovereignty of existing States. Hence, the Friendly Relations Declaration (1970), after having elaborated in detail on the principle of self-determination, limits it:

16  The clause was probably intended as a monument to the territorial integrity of States. But one cannot help but notice its fallacy: it essentially holds that the principle of self-determination does not enable any assault on the sovereignty of a State (‘nothing in the foregoing paragraph’), provided that the principle of self-determination is complied with (‘conducting themselves in compliance’). In other words, when the principle of self-determination is not complied with, it may allow the dismemberment of a State, which may well be the opposite of what the drafters of the clause originally intended. In spite of this equivocality, the most common reading of self-determination, typically advocated by proponents of States, upholds the territorial integrity of States and thus restricts the principle of self-determination to an internal dimension. Construed in this way, the principle of self-determination perhaps entitles a people to minority rights and structures enabling autonomy or similar arrangements, such as those in federal States, but does not give them a right to secession.

17  However, a more progressive interpretation of the clause in the Friendly Relations Declaration is possible, one which puts more emphasis on the second part of the paragraph. According to this reading, external self-determination—ie the right to secession—is usually dormant, but may be activated in exceptional circumstances. Such an exception would notably apply when internal self-determination is violated. In this understanding, the right to secession is a conditional right, with the violation of the principle of (internal) self-determination being the condition. As a consequence, the right is endowed with a punitive character in the sense of ‘if you misuse your power, you lose it’. The idea of forfeiture is obviously prominent in this approach. Such reasoning is, however, on the fringes of legal analysis and is strongly inspired by legitimacy considerations. Indeed, there is little support for it in positive law. The formula in the Friendly Relations Declaration—‘possessed of a government representing the whole people belonging to the territory without distinction as to race, creed or colour’—is in fact the only positive legal basis. Note that the formula was repeated on the occasion of the Vienna World Conference on Human Rights (1993) in the Vienna Declaration and Programme of Action ([25 June 1993] UN Doc A/CONF.157/23, Sec. 1, para. 2) and again in the Declaration on the Occasion of the Fiftieth Anniversary of the United Nations ([24 October 1995] UNGA Res 50/6 GAOR 50th Session Supp 49, 13, para. 1). It is based on the idea of representation and, in an e contrario argument, can be understood as asserting that if a government does not represent the whole population without discrimination, the part of the population that is not represented may be entitled to a right to secession. A lack of appropriate representation as a violation of internal self-determination is thus understood to be a catalyst for the right to secession. Apart from this positive—though indirect—manifestation, other reasons for the activation of the right to secession can be found, if free associative reasoning is applied. As early as 1921, the Commission of Rapporteurs in the Åland islands dispute found that it is possible to reach a different conclusion, ie to recommend the separation of the Åland islands from Finland to the benefit of the Åland islanders, ‘when a State lacks either the will or the power to enact and apply just and effective guarantees’ (The Aaland Islands Question 28). The Supreme Court of Canada in Reference re Secession of Quebec identified three potential circumstances in which external self-determination may be considered: decolonization, the case of alien ‘subjugation, domination or exploitation’ (para. 133, based on the passage of the Friendly Relations Declaration, ie ‘and bearing in mind that subjection of peoples to alien subjugation, domination and exploitation constitutes a violation of the principle [of self-determination], as well as a denial of fundamental human rights, and is contrary to the Charter’ Principle of equal rights and self-determination of peoples), and ‘when a people is blocked from the meaningful exercise of its right to self-determination internally’ but then only ‘as a last resort’ (para. 134). The last circumstance was not given clear consideration by the Court, however. Moreover, when human rights are seriously and persistently violated—when the oppression of a people is extreme, for example—this is widely perceived in academia as activating the right to secession, although only as an ultima ratio (for instance Tomuschat [2006] 35 [with further references]).

18  As to such a right to secession, three comments are in order. First, it is important to stress that the impact of such a right to secession would be limited ab initio. Situations in which secession is at stake are highly politicized, and a whole plethora of political arguments are used to support or undermine secessionist claims. In such a setting, a right to secession, while certainly bolstering the position of that part of the State that wishes to secede, would be only one argument among many and its concrete application would be subject to fierce debate. Therefore, a higher, impartial implementing authority would be at least as important as the right itself.

19  Second, apart from the tenuousness of the legal basis, the right to secession in the sense of a ‘remedial secession’ (Buchheit 222) raises other serious concerns. A remedial right to secession based on external self-determination would have to tackle all the difficulties of the right to self-determination, such as the questions of what is ‘a people’, who belongs to ‘a people’, and how the will of ‘a people’ is determined (see Self-Determination). Then there are the concerns that are very familiar from the debate on humanitarian intervention: serious issues relating to threshold levels would have to be addressed if one needed to fix the level of oppression—the seriousness of human rights violations, the degree of the lack of participation, etc—that would entitle a people to secession. The ultima ratio qualification of the right to secession—as a last resort—would not clarify things very much, for the ultimacy of a measure could always be disputed. It probably only indicates that, in general, the threshold would have to be high. The duration of the oppression needed in order to activate the right would pose additional challenges. In particular, it would be necessary to answer the intricate question as to whether the right to secede would continue to exist, once the oppression had ended.

20  Third, there is another fundamental argument which militates strongly against a right to secession. It is clear that secessionist claims are born out of difficult, multi-faceted circumstances. In such circumstances, it seems that various options would be available to address the underlying issues and, in most cases, secession would only be one of these options. Here, a right to secession seems to be a crude device—a black and white instrument applied to an area dominated by shades of grey. The right would confirm a bias in favour of one solution, namely secession, in a situation where many options should be available on an equal footing. This argument is clearly not against the legitimacy of action in face of oppression, but the crux is that there must be different means of addressing situations in which secessionist claims arise. To limit the options a priori to secession, even if only as an ultima ratio, seems to be an unbalanced approach that does not take the complexity of such situations into account. It sends out the wrong signals to the parties involved. And it includes a threat: the loss of a part of sovereignty in the event that basic international obligations, such as human rights etc, are not complied with. One would expect that there are more subtle ways of ensuring compliance than via a threat to break down territorial integrity—note the similar argument discussed in Reference re Secession of Quebec (para. 91).

21  The Supreme Court of Canada showed a way to address these concerns in the first, lesser-known part of its opinion in Reference re Secession of Quebec. Here, it held that Quebec’s—and probably any other Canadian province’s—clear will to secede from Canada would entail an obligation to negotiate the separation bona fide, based on Canadian constitutional law. The Constitution would not prescribe the outcome of these negotiations (see para. 17 above). Clearly, this approach is tailored to the case of peaceful relations between the State and the part wanting to secede (Peaceful Change). Yet, it is equally clear that, even in case of oppression, it is very difficult to reach a solution without negotiations. One can conclude from this that it might be an option to shape international law following Canada’s example. For those who are unable to adhere to such a solution because the possible sacrifice of State unity is too costly, it might still be acceptable to condition the remedial right to secession in the way proposed in UN Special Envoy Martti Ahtisaari’s plan for Kosovo, (UN Special Envoy for the Future Status Process for Kosovo, ‘Comprehensive Proposal for the Kosovo Status Settlement’; see paras 37–38 below). However, it seems that such an approach fails to properly justify the direction it gives: why is the right to secession conditional while the preservation of territorial integrity is not?

D.  Secession in Practice

22  The tendency to secession is extraordinarily frequent, if declarations of independence, plebiscites, and other, less outspoken efforts can be taken as yardsticks. An account of only the most prominent cases associated with secession is given below (for a more extended review of cases involving secession see Crawford 374–448; Crawford discusses other cases which are relevant for the concept of secession and which are not mentioned here: Senegal, Singapore, Czech Republic, and Slovakia [Czechoslovakia, Dissolution of] Tibet, Kashmir, East Punjab, the Karen and Shan States, North Cyprus, Tamil Elam, South Sudan, Somaliland, Bougainville, Kurdistan, Republika Srpska, Chechnya, Anjouan, Gaugauzia, Nagorny-Karabach, and Yemen). Only the aspects of each case which are most relevant to the legal approach to secession are stressed here.

23  In 1970, UN Secretary-General U Thant famously stated:

24  The dictum is a symbol, not only of the attitude of the United Nations, but also of the attitudes of its Member States towards secession. The formula, which still holds true today, inspires the following grouping of cases relating to secession into pre-decolonization, decolonization, post-decolonization, and non-decolonization cases.

25  The case from the pre-decolonization period that is still most relevant today—the dispute of the Åland islands, sometimes also called Aaland islands—took place during the time of the League of Nations. The Åland islanders, mostly Swedish speakers that inhabited a group of islands in the Baltic Sea, expressed their desire in a referendum to separate from the newly (re-)constituted State of Finland and to join Sweden. The International Commission of Jurists (ICJ) designated by the Council of the League of Nations first found in 1920 that the case was within the purview of the League and not a purely domestic affair of Finland (International Commission of Jurists, ‘Report of the International Commission of Jurists Entrusted by the Council of the League of Nations with the Task of Giving an Advisory Opinion upon the Legal Aspects of the Aaland Islands Question’ (1920)). The Commission of Rapporteurs, subsequently entrusted by the Council to consider the substance of the case, denied the Åland islanders the option to join Sweden and referred them to a state of autonomy within Finland, for which it gave some guidelines. The Rapporteurs’ finding was based mainly on the sovereignty of Finland and disregarded ethnic, geographical, and plebiscitary arguments. The ruling of the Rapporteurs was also founded on ideas of justice and recompense for the victors of World War I, among them Finland, but not neutral Sweden: ‘It would be an extraordinary form of gratitude…to wish to despoil her [ie Finland] of territory to which she attaches the greatest value’ (The Aaland Islands Question 30). Although the dispute was mainly about the extent of Finnish sovereignty, which was uncertain at the time, the Commission of Rapporteurs famously opened the possibility of secession, ‘when a State lacks either the will or the power to enact and apply just and effective guarantees’ (at 28). This part of the ruling continues to fuel today’s discussions about remedial secessions (see also Remedies).

26  The majority of States existing today were created during decolonization under the auspices of the United Nations and in application of the principle of (external) self-determination. However, the process of separation of colonies or other non-self-governing territories from the metropolitan State is not to be considered as a series of secessions. The reason for this is that, at the time of decolonization, these territories were no longer considered to be integral parts of the metropolitan States. This is expressed by the Friendly Relations Declaration when it proclaims that such a territory ‘has, under the Charter, a status separate and distinct from the territory of the State administering it’ (Principle V). The Declaration continues:

27  The emergence of new States as a result of decolonization must therefore be understood as a process through which the ties between the administering powers and administered territories were severed, rather than as the breaking up of uniform States through secessions.

28  Yet this way of looking at decolonization does not mean that decolonization as a whole is irrelevant to the concept of secession. On the contrary, decolonization gave rise to many secessionist movements. However, these movements did not primarily target the relationship between the administered territory and the administering State, but rather the relationship between the formerly administered territories themselves. They were formed within the newly created States in the wake of decolonization and mainly concerned the relations within or among the latter. These can be described as post-decolonization cases.

29  Each of these cases of post-decolonization is marked by a wide variety of aspects, but the borders of the newly established States played a significant role in most of them. These borders had been pre-drawn during colonization as external or administrative borders, without much regard to the relations, the ethnicity, or the sense of belonging of the inhabitants. During decolonization, these borders were preserved for practical reasons. Legally, this perpetuation of the boundaries was justified by the uti possidetis doctrine. However, when the administering power was beginning to leave the stage, the arbitrary nature of the borders was an important factor and one that fuelled secessionist claims.

30  Most prominent among the post-decolonization cases are Katanga and Biafra (Biafra Conflict). Attempts at secession were made in both cases. In the Katanga case, the regime of the province Katanga declared its independence from the newly established Republic of Congo in 1960 (Congo, Democratic Republic of the). In the case of Biafra, the Ibo population endeavoured to secede from Nigeria and to establish its own State by declaring independence on 30 May 1967. Both attempts failed, mainly because the secessionist movements were not, or not sufficiently, endorsed by the international community. Only five States recognized the State of Biafra—Gabon, Haiti (see also Haiti, Conflict), Ivory Coast, Tanzania, and Zambia—and none did so with Katanga. It is also noteworthy that in 1964—one year after the secessionist attempt in Katanga had failed—the African heads of State and government, in the assembly of the Organization of African Unity in Cairo, reiterated their commitment to uphold the existing borders as established by the colonial powers: ‘The assembly of Heads of States and Government…solemnly declares that all Member States pledge themselves to respect the borders existing on their achievement of national independence’ (Organization of African Unity Assembly of Heads of State and Government, ‘Border Disputes Among African States’ [17–21 July 1964] AHG/Res 16(I), Art. 2; see also Boundary Disputes in Africa). And it was a question about the Biafra incident that led UN Secretary-General U Thant to clarify the position of the United Nations in his famous dictum in the press conference of 1970.

31  The lasting Western Sahara conflict, another case of post-decolonization, is only marginally relevant to the concept of secession, although prima facie the contrary seems to be the case, considering that the population of Western Sahara is still striving for its own State. The International Court of Justice (ICJ) held, in the Western Sahara (Advisory Opinion), that neither Morocco nor the ‘Mauritanian entity’ (see the term used in the question of the UN General Assembly repeated in para. 1 of the opinion), had any sovereign rights over the territory of Western Sahara at the time of colonization, although some legal ties did exist ([1975] ICJ Rep 12, para. 162; Western Sahara (Advisory Opinion)). The continuing dispute about the referendum to be held in Western Sahara therefore appears as a dispute about the right to self-determination in the setting of decolonization, and not about the secession of Western Sahara, be it from Spain, Morocco, or Mauritania.

32  Eritrea was admitted as a member of the United Nations by consensus in 1993 (UNGA Res 47/230 [28 May 1993] GAOR 47th Session Supp 49 vol 2, 6). It is difficult to explain the development of the State of Eritrea in terms of secession. Italian colonial domination over Ethiopia and Eritrea came to an end in World War II. The two territories were then administered by Great Britain and finally coupled into a federation under United Nations patronage in 1952, with substantially more powers being accorded to Ethiopia than to Eritrea (UNGA Res 390 (V) [2 December 1950] GAOR 5th Session Supp 20, 20: laying out the details of the federation). In 1962, Ethiopia dissolved the federation by annexing Eritrea (Annexation). The Eritrean liberation movement fought a long civil war against Ethiopian forces and ultimately helped overthrow the military government of Ethiopia in 1991 (National Liberation Movements; Armed Conflict, Non-International). Following acknowledgement of Eritrea’s right to self-determination by the new government of Ethiopia, and after Eritrea’s wish to become independent had been expressed in a UN-sponsored plebiscite by an overwhelming majority—99.8%—in April 1993, Eritrea celebrated its independence on 24 May 1993. Several facts seem to suggest that the outcome of the whole affair constitutes the final, co-ordinated settlement of a decolonization process rather than a secession: the fact that Eritrea’s distinctness was already recognized in the first, unsuccessful formula in the process of decolonization; the federation with Ethiopia; the role of Eritrean forces in the regime change in Ethiopia; and the ensuing consent by the Ethiopian government to Eritrean independence. Nevertheless, although not a secession stricto sensu—Ethiopia agreed to Eritrea becoming a State—elements of secession appear, from a formal perspective, in that a new State emerged from the clutches of an existing sovereign State.

33  The case of Bangladesh shows some similarities with that of Eritrea, and it probably comes closest to secession in the strict sense of the term. Bangladesh initially constituted the eastern part of Pakistan, which, since decolonization, was geographically separated from the western part of Pakistan by the territory of India. When a mainly eastern Pakistani party, the Awami League, won the majority of seats in the election for the parliamentary assembly of Pakistan in 1970, the government of Pakistan refused to accept the result and launched a military operation in the eastern part of Pakistan. In the ensuing civil war, India intervened in support of the Awami League. The Indian–Pakistani war that followed was ended by the surrender of Pakistan on 17 December 1971. Even before the defeat, India had recognized eastern Pakistan as the sovereign State of Bangladesh, which had formally been proclaimed in April 1971. Recognition by several States followed. After Pakistan had finally given its consent by recognition on 22 February 1974, Bangladesh was admitted to the United Nations (UNGA Res 3203 [17 September 1974] GAOR 29th Session Supp 31 vol 1, 2). While the case, which was also set in a post-decolonization context, is an incidence of secession, it also demonstrates the significance, from a political point of view, of foreign sponsorship and military support for the chances of success of secessionist movements.

34  Outside the context of decolonization, one of the most prominent cases concerning secession is that of Quebec. Secession of Quebec from Canada has been looming for a long time. The famous opinion of the Supreme Court of Canada (Reference re Secession of Quebec 1998) on whether Quebec had a right to secede from Canada was only one episode in a mainly political incident, but from a legal point of view, it was the most interesting one. The Supreme Court considered the issue of Quebec’s right to secede from Canada from two angles. It opined that, under Canadian constitutional law, if the population of Quebec expressed its will to secede by a clear majority, this would entail an obligation to negotiate in good faith the change of the Canadian Constitution necessary for a withdrawal of Quebec, without any outcome of these negotiations being pre-determined by constitutional law (paras 32–108, in particular para. 102). It held that international law, on the other hand, does not grant Quebec the right to secede unilaterally from Canada (paras 109–46, in particular para. 138). However, the Court acknowledged that in principle such a right could arise under specific circumstances—decolonization, alien ‘subjugation, domination or exploitation’ (para. 133), and ‘when a people is blocked from the meaningful exercise of its right to self-determination internally’ (para. 134). Since the Supreme Court gave its opinion in 1998, the population of Quebec has not expressed the clear wish to secede, which, according to the opinion, would be required under Canadian constitutional law.

35  The demise of the Soviet Union and the Socialist Federal Republic of Yugoslavia are other prominent cases in which the idea of secession played a certain role. In both cases, a State from which power had radiated out to the neighbouring States and beyond, came to an end. As a result, a number of States came—or came back—into existence, and each would merit a separate, extensive analysis under the subject of secession. To put it briefly here, the dismemberment of the Soviet Union enabled the satellite States of Eastern Europe, which had remained formally sovereign during Soviet rule, to shake off Russian influence and orient themselves towards the European Union. The Baltic States, which had been annexed by the Soviet Union in 1940, re-asserted their sovereignty. Belarus, Ukraine, and the Russian Federation—the latter being the successor State of the Soviet Union—founded the Commonwealth of Independent States (CIS) under the Minsk agreement of 8 December 1991, thus ending the Soviet Union (Agreement establishing the Commonwealth of Independent Nations [done and entered into force 8 December 1991] (1992) 31 ILM 143). The other former Soviet Republics (Armenia, Azerbaijan, Kazakhstan, Kyrgyzstan, Moldova, Turkmenistan, Tajikistan, and Uzbekistan) gained recognition by the Russian Federation and joined the Commonwealth of Independent States by virtue of the Alma Ata Declaration ([done and entered into force 21 December 1991] (1992) 31 ILM 148). However, Georgia, which joined in 1993, notified its withdrawal on 18 August 2008. This break up of the Soviet Union as a whole is not to be seen as a series of secessions. It is clear, however, that the States which emerged from, or survived, the process have been consolidated to such an extent that, at least from their perspective, future structural changes would have to be assessed in terms of secession, as in the case of Chechnya, South Ossetia, and Abkhazia.

36  That the Socialist Federal Republic of Yugoslavia was also disintegrating was confirmed by the Badinter Commission (for the Former Yugoslavia) in its Opinion No 1 (Conference on Yugoslavia Arbitration Commission, Opinion No 1). When asked whether the claims to sovereignty and independence of the formerly constituent entities of the Socialist Federal Republic of Yugoslavia amounted to secessions, the Badinter Commission opined that the Federation ‘is in the process of dissolution’ (para. 3). It thus applied a factual approach that confirmed that Slovenia, Croatia, the Former Yugoslav Republic of Macedonia, and Bosnia-Herzegovina did not secede from the Socialist Federal Republic of Yugoslavia. It is noteworthy that part one of the preamble of the Constitution of the Socialist Federal Republic of Yugoslavia of 21 February 1974 contained a right of the constituent republics to secede, but such a right was not given to provinces such as Kosovo (Constitution of the Socialist Federal Republic of Yugoslavia [adopted 21 February 1974], in WB Simons, The Constitutions of the Communist World [Sijthof and Noordhoff Alphen aan den Rijn 1980] 444–577). Later on, Montenegro dissolved the newly constituted union with Serbia with the consent of the latter (based on Art. 60 Constitutional Charter of the State Union of Serbia and Montenegro [adopted and entered into force 4 February 2003]), after the population of Montenegro had voted for independence in the referendum of 21 May 2006. While the dissolution of the Socialist Federal Republic of Yugoslavia, like that of the Soviet Union, was a factual rearrangement rather than a series of secessions—or at least, in the case of Montenegro, an agreed solution—the case of Kosovo is different.

37  Kosovo was the first case, at least in the immediate past, in which the separation of a part of a State was promoted by a large number of States, even though the existing State, of which the separating entity was a part, expressly opposed the separation. How did the declaration of independence of Kosovo of 17 February 2008 and the ensuing recognition by 60 States as of June 2009 come about? The most recent events may be summed up as follows. In the Socialist Federal Republic of Yugoslavia, Kosovo only enjoyed the status of an autonomous province, not that of a republic. As such, it had no constitutional right to secede from the federation. But even this limited status was rescinded by the Milosevic regime in 1989. During the early 1990s, the Kosovars resisted pressure from the Serbian government by peaceful means, eventually building up a shadow administration in Kosovo. The situation was militarized, however, when Serbia resorted to ethnic cleansing in Kosovo. After diplomatic efforts to settle the situation had proved to be in vain, the North Atlantic Treaty Organization (NATO) and its Member States intervened in Kosovo, despite the fact that the UN Security Council had not authorized the action. After the surrender of Serbia, UN Security Council Resolution 1244 of 10 June 1999 established an international civil and military presence in Kosovo—UNMIK and KFOR—while ignoring both the legality of the prior use of force, as well as the final status of Kosovo (UNSC Res 1244 [1999] [10 June 1999] SCOR 54th Year 32). There followed almost ten years of international administration and countless attempts at finding an agreeable solution (International Administration of Territories). The latest effort, led by UN Special Envoy Martti Ahtisaari, made it plain that Serbian and Kosovar positions could not be reconciled: Serbia insisted on Kosovo’s remaining a part of the Serbian State—albeit with substantial autonomy—while Kosovo accepted nothing short of independence (UN Special Envoy for the Future Status Process for Kosovo, ‘Comprehensive Proposal for the Kosovo Status Settlement’). Due to Serbian opposition, represented by Russia, the UN Security Council failed to adopt Ahtisaari’s proposal. Kosovo proclaimed its independence soon afterwards. However, the status of Kosovo remains controversial, as some States have recognized it as a State and others have refused to do so.

38  In many senses, the case of Kosovo is illustrative of the role of international law in the domain of secession. In the face of an attempt at secession, States try to avoid creating a precedent. As evidenced by the analysis of State practice, virtually no case is considered to amount to secession. Each and every case is qualified as special and unique—even Kosovo, which, although from an impartial point of view plainly constituting secession, perhaps even an example of remedial secession, is still seen as singular, due to the long-standing presence of an international administration. This may well be the case, because international law is perceived to be an impediment in this case. Practice as a creative element of international law, namely of customary international law, is therefore construed in a way that inhibits its consolidation into, and the creation of, a legal standard. In other words, a non-principled, pragmatic approach is preferred in the domain of secession. This case-by-case approach usually comes down to upholding the territorial integrity of existing States—though not always, for when a State no longer exists, as with the Soviet Union or the Socialist Federal Republic of Yugoslavia, nothing can be upheld—and, in extremis, it may mean that independence becomes an option if all other alternatives fail. The approach also means that it is not non-compliance with legal standards that is decisive for the success or failure of an attempt at secession, but rather non-legal factors, such as diplomatic abilities, or the ability to raise international support, or influence the media. Quite plainly, military power is often crucial and third States, in the aftermath of a secession crisis, tend to base their attitude on a ‘winners and losers’ rationale and on practical considerations. All these factors can undoubtedly be seen at work in the most recent secession episode in South Ossetia and Abkhazia in Georgia, although the latest facts of this situation are not well enough established, so that any verdict would be premature.

E.  After Secession

39  While the birth of a new State through secession is governed by facts rather than legal rules, the latter play a more important role after secession has taken place. It is important to note, however, that even after secession the pre-existing State—the one from which the part seceded—does not become irrelevant. On the contrary, it continues to play an important role, notably in two regards. First, it is involved in the process of succession, since the rules on State succession apply to the new State emerging from secession (Zimmermann 213). But the two existing conventions in the area of State succession concern only two, albeit important domains (Vienna Convention on Succession of States in Respect of Treaties [23 August 1978], in particular Arts 34–38; Vienna Convention on Succession of States in Respect of State Property, Archives and Debts [8 April 1983], in particular Arts 17, 30, and 40; note also UN ILC, ‘Draft Articles on Nationality of Natural Persons in relation to the Succession of States’ [1999], in particular Arts 24–26), and the settlement of matters depends to a large extent on negotiation (see Conference on Yugoslavia Arbitration Commission, Opinion No 9 [4 July 1992] para. 4). Here, the co-operation of the first State seems to be necessary if an agreeable solution is to be reached. The number of issues to be settled through these negotiations is vast and diverse issues, such as the partitioning of the foreign debt of the pre-existing State (Debts) and the citizenship of inhabitants, are involved. An illustration of the scope of the issues involved can be seen in the UN Special Envoy Martti Ahtisaahri’s plan for the final status of Kosovo. In Ahtisaahri’s proposal, particular importance is accorded to minority protection and the plan puts forward the establishment of an elaborate regime of guaranteed representation, home areas, and cultural rights. It may be presumed that similar issues relating to minority protection would have to be addressed in other cases involving secession.

40  Second, the pre-existing State’s role is crucial when it comes to recognition of the newly emerged State. Recognition of a new State is, of course, at the discretion of each State. Its nature is declaratory, in that the legal existence of a State does not depend on recognition by other States. However, the factual existence of a State is hampered when fewer States recognize it, and to this extent the declaratory nature of recognition is relative. Rather, the reality of a State is to be tested against the four criteria of the Montevideo Convention ‘a) a permanent population; b) a defined territory; c) government; and d) capacity to enter into relations with the other states’ (Art. 1 Convention on Rights and Duties of States [signed 26 December 1933, entered into force 26 December 1934] 165 LNTS 19); the last two elements can be considered to embody the principle of effectiveness. But practice shows that States regard the recognition by the pre-existing State as paramount. They are hardly ever prepared to recognize a new State without the consent of the pre-existing State. The cases of Bangladesh and Kosovo are certainly exceptional in this regard. This reluctance to grant recognition seems to be due to two factors: on the one hand, it seems that the follow-up issues in a State succession can hardly be tackled without the co-operation of the pre-existing State (see above para. 39 (i)). On the other hand, States probably prefer to recognize a new State at a late stage, because a premature recognition, much like early support for a secessionist movement, could be considered as illegal interference in the domestic affairs of an existing State.

41  As to recognition in general, and in the context of secession, it is noteworthy that recognition no longer seems to depend solely on the question of whether a new State has come into existence—in other words whether the four elements of the Montevideo Convention have been fulfilled. Other circumstances seem to have become relevant, too. To be sure, recognition as a discretional act of each State has always depended on factual circumstances, such as the existence of a diaspora of the population claiming statehood. But apart from these factual considerations, new conditions, which are applied to statehood and thus to recognition, seem to be evolving. These conditions involve elements of democracy, rule of law, human rights, minority protection, and possibly the principle of self-determination. Evidence of this can be seen in the guidelines on the recognition of new States adopted by the Council of Ministers of the European Community on 16 December 1991, or in the fact that the acts of recognition of Kosovo were usually made conditional on an effective protection of minorities in Kosovo (EC, ‘Declaration on the “Guidelines on the Recognition of New States in Eastern Europe and in the Soviet Union”’ [1992]; the text of all recognitions of Kosovo is available). It will be interesting to see the effectiveness of these conditional recognitions of Kosovo. In substance, this will have to be examined in light of the Ahtisaari plan—notwithstanding Serbia’s disagreement with it.

F.  Outlook

42  Secession affects the fundament of the international order: the State and the guarantee of its integrity. Whenever a secession is about to occur, the international order trembles, and that is why there is a stigma attached to it. Kosovo, South Ossetia, and Abkhazia are only the most recent and most important episodes in the history of secession. International law, however, does not play a very prominent role in it. As is most often the case in the origins of States, other, mainly political, considerations prevail. International law remains neutral vis-à-vis secession and neither prohibits nor permits it. International law becomes important, however, when the facts have been consolidated and the new State has come into existence, in particular when it comes to matters of State recognition and succession. The limited role of international law must be borne in mind in the argument relating to the legality of remedial secessions, as much of the argument is normatively tinged. In fact, there is little safe legal ground on which to base it. Nonetheless, the recent secession of Kosovo from Serbia, while not in itself giving rise to a customary right to secession, appears to motivate other, seemingly oppressed, populations to voice their grievances and their claims to statehood, regardless of the apparently unique nature of the Kosovo example. It remains to be seen whether such claims will be more successful than in the past, and what the repercussions on the international legal order will be. For now it can only be said that the possible success of these attempts, from a neutral standpoint, may not be such a bleak prospect after all, and the repercussions on the international legal order may well prove to be just another element of the general change in the structure of the international order. The significance of States would thus be put into a new perspective—a perspective in which States will have lost some of their air of untouchability, since they are being influenced more than ever from the top—supranational organizations, international courts—as well as from the bottom—civil society, minorities, and transnational corporations.