Secessionism Compared — A Global Comparative Study

Methodology

This is a structured, focused comparison (George & Bennett) using a most-similar/most-different design. It extends the original three-case template (Catalonia, Scotland, Hong Kong) to 39 cases. Each case is coded on six explanatory variables (0 = absent/low, 3 = high) and one outcome. The codings are analytic judgements drawn from the cited sources. They are not survey statistics, so read them as ordinal and open to argument.

G: Grievance & identity: Distinct national identity, history of independent statehood, cultural or religious discrimination.
E: Economic differential: The region is richer than the state (a fiscal-transfer grievance) or holds key resources, or is extracted from.
C: Constitutional pathway: Is there a legal route to a binding referendum? 0 = forbidden, 3 = agreed or guaranteed.
R: State repression: Intensity of coercion by the parent state against the movement.
I: International support: Recognition, a patron state, a UN decolonisation framework, or third-party mediation.
V: Violence: Level of armed conflict linked to the claim.
Outcome: Independent · De facto · Autonomy · Integrated/Suppressed · Defeated · Unresolved.

Map

Comparative matrix

Region Search

Click a column header to sort. Scores run 0–3.
Case	Parent state	Region	G	E	C	R	I	V	Outcome	Type

Typology

1. Civic-democratic

Catalonia, Scotland, Quebec, Flanders. Wealthy, non-violent and electoral. The deciding factor is whether the constitution allows a referendum (C). Where it does, voters have so far chosen No. Where it doesn't, the result is a legal standoff.

2. Decolonisation

East Timor, New Caledonia, Western Sahara, West Papua, Puerto Rico. UN self-determination norms apply. Outcomes depend on whether the administering state accepts a free ballot.

3. Negotiated after war

South Sudan, Eritrea, Bougainville, Aceh, Bangsamoro. A war ends in a settlement that sequences a referendum or autonomy.

4. Patron-state / frozen

Transnistria, Abkhazia, South Ossetia, Crimea/Donbas. Survival depends on Russian military backing, and Western recognition is nil.

5. Remedial / great-power backed

Kosovo. Mass atrocity, then intervention, then partial recognition.

6. Repressed ethno-national

Tibet, Xinjiang, Hong Kong, Balochistan, Biafra, Ambazonia, Cabinda, Kashmir, Tamil Eelam. A strong state, no legal pathway and little external support. Outcomes are suppression or military defeat.

7. De facto states

Somaliland, Taiwan. Sovereignty works in practice but has little recognition. Geopolitics, not internal capacity, sets the ceiling.

8. Autonomy-absorbed

Iraqi Kurdistan, Basque Country, Corsica, Santa Cruz, Syrian Kurdistan. The claim is converted into, or forced into, self-rule within the state.

Cross-case findings

International support (I) is the strongest single correlate of independence. Every fully independent case in the set (South Sudan, Eritrea, East Timor, Montenegro, and Kosovo partially) had I ≥ 2. A movement can have high grievance and high economic differential and still fail without it.
Consent from the parent state is close to necessary. Successful separations came through agreed referendums (Montenegro, South Sudan), state collapse or defeat (Eritrea), or an externally imposed exit (East Timor, Kosovo). No unilateral declaration made against a functioning democracy (Catalonia 2017, Kurdistan 2017) has produced a state.
Legal referendums tend to defuse secession. Where C is high in wealthy democracies (Scotland, Quebec, New Caledonia), voters have repeatedly chosen No. Closing off the pathway (Catalonia) tends to radicalise rather than end the claim.
Repression decides outcomes in authoritarian states but does not settle the claim. In Hong Kong, Tibet and Xinjiang, open mobilisation has been suppressed while grievance scores stay at the maximum.
Economic differential cuts both ways. Rich regions (Catalonia, Flanders) and resource enclaves (Cabinda, Balochistan, Aceh) both mobilise. But richer regions mostly end up with fiscal autonomy, not exit.
Violence rarely wins on its own. Armed movements without a patron were defeated (Tamil Eelam, Biafra) or ended in negotiated autonomy (Aceh, Bangsamoro, Basque).
Geopolitics sets the recognition ceiling. The great powers' rival precedents (Kosovo vs Crimea/Abkhazia) and Israel's 2025 recognition of Somaliland show recognition following strategic interest more than legal merit.
2025–26 trend: de-escalation through absorption. PKK dissolution (2025), SDF dissolution (Aug 2026), New Caledonia's stalled compromise and Bougainville's pending ratification all show states channelling claims into autonomy formulas.

Outcome distribution

Case profiles

Sources

Each case card links its primary references. General frameworks used: Buchanan, Secession (1991); Coggins, Power Politics and State Formation (2014); Griffiths, Age of Secession (2016); Sorens, Secessionism (2012); ICJ Kosovo Advisory Opinion (2010); Reference re Secession of Quebec (1998).