¹ For reasons to be stated shortly, movements for separate territorial identity will be referred to as “secessionist”, whether the demand is for independence or only for autonomy from a larger territorial unit. I shall also use secession and separation, as well as secessionist, interchangeably throughout. For an equally inclusive conception of separatism, see Nagel, Joane, “The Conditions of Ethnic Separatism: The Kurds in Turkey, Iran, and Iraq,” Ethnicity, Vol. 7, No. 3 (09 1980), pp. 279–97Google Scholar. It will also be apparent that the materials for this article are drawn from Asia and Africa, but occasionally other cases are referred to in order to supplement these materials. This is particularly true of the Basque case, which exemplifies a pattern not generally found in Asia and Africa.

² Although it makes no difference to the analysis that follows, I assume for the sake of argument that there has been an upsurge of separatism in the last two decades. That assumption squares with my own observation and with that of most close students of the subject.

³ This is not on any logical grounds. In principle, there is no reason why conditions responsible for an upsurge in separatist activity could not also differentiate between positive and negative cases. That is, a variable accounting for an increase in separatism might tend to be present in the positive cases, absent in the negative ones. In practice, however, variables adduced to explain aggregate trends tend to be so widely diffused that they are present in both kinds of cases, as shown in the text.

⁴ The relationship of the idea of self-determination to secession is admirably traced by Connor, Walker, “Self-Determination: The New Phase,” World Politics, Vol. 20, No. 1 (10 1967), 30–53CrossRefGoogle Scholar. See also Shibutani, Tamotsu and Kwan, Kian M., Ethnic Stratification (New York: Macmillan, 1965), pp. 444–45Google Scholar. For some of the philosophical roots of national selfdetermination, see Cohler, Anne M., Rousseau and Nationalism (New York: Basic Books, 1970).Google Scholar

⁵ Connor, , “Self-Determination: The New Phase,” p. 30Google Scholar. Isaacs, Harold, Idols of the Tribe: Group Identity and Political Change (New York: Harper & Row, 1975), p. 13Google Scholar, puts the point graphically: “The holy grail of self-determination in anti-colonial politics became the poison potion of group conflict, secession, rebellion, and repression in the postcolonial era.”

⁶ Cf. Connor, , “Self-Determination: The New Phase,” p. 46Google Scholar; Connor, Walker, “Ethnology and the Peace of South Asia,” World Politics, Vol. 22, No. 1 (10 1969), 70CrossRefGoogle Scholar; Issacs, , Idols of the Tribe, pp. 175–78.Google Scholar

⁷ Birch, Anthony H., “Minority Nationalist Movements and Theories of Political Integration,” World Politics, Vol. 30, No. 3 (04 1978), 325–44.CrossRefGoogle Scholar

⁸ Ibid., p. 335. Birch suggests that the existence of the American nuclear umbrella in Europe makes it less necessary than previously to secure protection against external attack by remaining part of a large, strong state. The Netherlands is protected just as France and Germany are. The Common Market, he argues, likewise makes it possible for groups like the Corsicans or the Bretons to salvage their prosperity even if they opt for independence. An independent Corsica, for example, could become a voting member of the European Economic Community and receive all of its benefits. Both of these conditions constitute an alteration of the earlier balance of advantages.

⁹ The particular changes identified by Birch, of course, apply only to Europe. In Asia and Africa, presumably, it remains advantageous in external security terms for groups to belong to large states. Likewise, though the economic advantages of large states are subject to debate, no recent change in the balance of those advantages can be detected for Asia and Africa. Even within Europe, however, there are wide variations in the enthusiasm of potential separatist groups for the prospect of separatism.

¹⁰ Or, in a large federation like India, by forces beyond the affected state or province.

¹¹ “No one in Baluchistan wants to break away [from Pakistan]. All the Baluchis want is not to lose their identity,” commented a Baluch spokesman. “Who in his right mind would want to join Afghanistan? We'd be worse off there than we are in Pakistan.” Washington Post, February 8, 1976, p. A17Google Scholar. This is also the theme of Sayeed, Khalid B., “Pathan Regionalism,” South Atlantic Quarterly, Vol. 63, No. 4 (Autumn 1964), 478–506.Google Scholar

¹² Far Eastern Economic Review (Hong Kong),08 17, 1979, p. 28.Google Scholar

¹³ Ibid., September 14, 1979, p.30.

¹⁴ See Kyle, Keith, “The Southern Problem in the Sudan,” The World Today, Vol. 22, No. 12 (12 1966), pp. 512–20Google Scholar. An article in a journal published by the Southern Sudan Association in Britain during the Sudanese civil war illustrates the point: “What, then, are we fighting for? We are fighting for freedom; freedom to unite with the North; freedom to federate with the North; freedom to reject the North; freedom for the people of the South Sudan to determine their own future without interference from the Arabs or any other people.” Akol, Jacob J., “What We Are, and Are Not, Fighting For,” The Grass Curtain (London), Vol. 2, No. 2 (10 1971), 26.Google Scholar

¹⁵ Africa: The Politics of Independence (New York: Vintage Books, 1961), p. 88.Google ScholarPubMed

¹⁶ I am well aware that this terminology is less than optimal, but alternatives (e.g., traditional-modern, disadvantaged-advantaged) seem even less apt. Of course, the term backward is not used in a pejorative way, and its frequently pejorative connotations among observers are offset for our purposes by the fact that the groups concerned very often use the term in an objective sense, referring to themselves as backward and to their antagonists as advanced.

¹⁷ There are some variations from country to country in the pattern of group stereotypes, though not in pervasiveness or in the general tendency to dichotomize. For example, most advanced groups are regarded as progressive and backward groups as traditional, but in the Philippines the “traditional” label is applied to the Ilocano, who score high on virtually all objective and subjective indices of group advancement. Bulatao, Rodolfo A., Ethnic Attitudes in Five Philippine Cities (Quezon City: University of the Philippines Social Research Laboratory, 1973), p. 58Google Scholar. It would take us far afield here to examine the origin, content, and correlates of such stereotypes. For a recent study, see Brewer, Marilynn B. and Campbell, Donald T., Ethnocentrism and Intergroup Attitudes: East African Evidence (New York: Halsted Press, 1976).Google Scholar

¹⁸ One complexity of which I shall not take adequate account concerns differences of opinion within given ethnic groups on the advisability of secession. Sometimes secessionist sentiment is virtually unanimous, but very often there are debates on whether to secede. See, e.g., Dudley, B.J., “Western Nigeria and the Nigerian Criss,” in Panter-Brick, S.K., ed., Nigerian Politics and Military Rule: Prelude to the Civil War (London: Athlone Press, 1970), pp. 106–08Google Scholar, identifying at least five Yoruba opinion strains ca. 1966–67. More often than not, I shall ignore such differences, dealing instead with central tendencies or merely with the outcomes of such debates.

¹⁹ Williamson, Jeffrey G., “Regional Inequality and the Process of National Development,” Economic Development and Cultural Change, Vol. 13, No. 4, Pt. 2 (07 1965), 14, 17.CrossRefGoogle Scholar

²⁰ See Anderson, Charles, Mehden, Fred R. von der, and Young, Crawford, Issues of Political Development (2d ed.; Englewood Cliffs: Prentice-Hall, 1974), p. 75Google Scholar. When the Uganda government refused to accede to demands for a separate district, the Rwenzururu movement in Western Uganda became secessionist very quickly, for reasons that are probably related to span of control. “The Ruwenzori mountain areas are extremely inaccessible; effective administration had never become established in the higher altitudes and, in a sense, anybody could set up an independent government there without facing the consequences for some time.” Doornbos, Martin R., “Protest Movements in Western Uganda: Some Parallels and Contrasts,” unpublished paper, 1966, p. 12Google Scholar. But the emergence of the movement in the first place had little to do with these geographic conditions.

Distance, of course, is a condition that can cut both ways. While great distance may make secession easier - or at least make its suppression more difficult - distance may also reduce the intrusiveness of central government penetration of peripheral areas.

²¹ See, e.g., Maw, Ba, Breakthrough in Burma: Memoirs of a Revolution, 1939–1946 (New Haven: Yale University Press, 1968), p. 187Google Scholar; Mehden, Fred R. von der, Religion and Nationalism in Southeast Asia (Madison: University of Wisconsin Press, 1963), p. 193Google Scholar; Bulatao, , Ethnic Attitudes in Five Philippine Cities, pp. 57–62Google Scholar; Gray, Hugh, “The Demand for a Separate Telengana State in India,” Asian Survey, Vol. 11, No. 5 (05 1971), 463–74, esp. 464.CrossRefGoogle Scholar

²² In wealthier states, however, even this subsidy may not be enough to bring per capita spending in poor regions up to levels proportionate to their share of the state's population. In such a case, a demand for per capita proportionate spending is likely to be received most unsympathetically by the center. See, e.g., Benjamin, Charles M., “The Kurdish Non-State Nation,” unpublished paper presented at the annual meeting of the International Studies Association, February 19–22, 1975, p. 6.Google Scholar

²³ See, e.g., Mehden, von der, Religion and Nationalism in Southeast Asia, p. 171Google Scholar; Khan, Rasheeduddin, “Political Participation and Political Change in Andhra Pradesh (India)” (Hyderabad: Osmania University Department of Political Science, mimeographed paper, 06 1969), p. 33.Google Scholar

In some cases, even elites that were ahead of ethnically-differentiated competitors saw separatism as a way of reducing the competition. The agitation for a separate Pakistan in the 1930s and '40s was disproportionately led by Muslims in what was then called the United Provinces. As a whole, Indian Muslims were backward, and they feared domination by educationally more advanced Hindus. But, in the United Provinces, Muslims were ahead of Hindus in government employment, the professions, and the modern private sector. Still, U. P. Muslim elites feared their minority position in an undivided India, and they demanded a separate state to protect their position. Brass, Paul R., “Muslim Separatism in United Provinces: Social Context and Political Strategy Before Partition,” Economic and Political Weekly (Bombay), Vol. 5, Nos. 3, 4, and 5 (01 1970), 167–86.Google Scholar

²⁴ Kimball, Lorenzo Kent, The Changing Pattern of Political Power in Iraq, 1958 to 1971 (New York: Robert Speller & Sons, 1972), pp. 141–42Google Scholar; Raoof, Abdul H., “Kurdish Ethnic Nationalism and Political Development in Republican Iraq,” unpublished paper presented at the annual meeting of the Middle East Studies Association, Denver, November 11–13, 1971, p. 10Google Scholar.So conscious are the Kurds of their backwardness that these demands sometimes make provision for exceptions when no qualified Kurds can be found for particular positions.

²⁵ Beshir, Mohamed Omer, The Southern Sudan: Background to Conflict (London: C. Hurst & Co., 1968), p. 72Google Scholar; Gray, Richard, “The Southern Sudan,” Journal of Contemporary History, Vol. 6, No. 1 (1971), p. 117.CrossRefGoogle Scholar

²⁶ Oduho, Joseph and Deng, William, The Problem of the Southern Sudan (London: Oxford University Press, 1963), p. 14Google Scholar. For reports of similar sentiments, see Kyle, , “The Southern Problem in the Sudan,” p. 513Google Scholar; Zartman, I. William, Government and Politics in Northern Africa (New York: Praeger, 1963), p. 140Google Scholar; Collins, Robert O. and Tignor, Robert L., Egypt and the Sudan (Englewood Cliffs: Prentice-Hall, 1967), pp. 147–64.Google Scholar

²⁷ Pledge, Robert, “France at War in Africa,” Africa Report, 06 1970, pp. 16–19Google Scholar; Ballard, John A., “Four Equatorial States,” in Carter, Gwendolen M., ed., National Unity and Regionalism in Eight African States (Ithaca: Cornell University Press, 1966), pp. 272–74Google Scholar; Lewis, William H., “Francophone Africa,” Current History, Vol. 60, No. 335 (03 1971), 143Google Scholar; Lemarchand, Rene, “Sisyphus in Chad: The MRA as a Development Partnership,” unpublished paper, n.d., ca. 1974, p. 4Google Scholar; Africa Report, November 1969, pp. 10–12.Google Scholar

²⁸ Maxwell, Neville, India and the Nagas (London: Minority Rights Group Report No. 17, n.d.), p. 9.Google Scholar

²⁹ Gray, Hugh, “The Demand for a Separate Telengana State,” Asian Survey, Vol. 11, No. 5 (05 1971), 463–74.CrossRefGoogle Scholar

³⁰ See Williamson, , “Regional Inequality and the Process of National Development.”Google Scholar

³¹ Cook, C.P., “India: The Crisis in Assam,” The World Today, Vol. 24, No. 10 (10 1968), p. 446Google Scholar; Collins, and Tignor, , Egypt and the Sudan, p. 159Google Scholar; Smith, Donald Eugene, Religion and Politics in Burma(Princeton: Princeton University Press, 1966), p. 230Google Scholar; Zartman, , Government and Politics in Northern Africa, p. 140Google Scholar; Kirk, George E., Contemporary Arab Politics (New York: Praeger, 1961), p. 145.Google Scholar

³² Quoted in Schwarz, Walter, Nigeria (New York: Praeger, 1968), p. 249Google Scholar. For an explanation, cast in terms of welfare economics, of “why even individuals who will probably lose in terms of tangible rewards through increased political autonomy may nevertheless be willing to invest in its attainment,” see Hartle, Douglas G. and Bird, Richard M., “The Demand for Local Political Autonomy: An Individualistic Theory,” Journal of Conflict Resolution, Vol. 15, No. 4 (12 1971), 443–56.CrossRefGoogle Scholar

³³ Lambert, Richard D., “Factors in Bengali Regionalism in Pakistan, ” Far Eastern Survey, Vol. 28, No. 4 (04 1959), p. 54CrossRefGoogle Scholar. Tinker, Hugh, India and Pakistan: A Political Analysis (rev. ed.; New York: Praeger, 1968), p. 167Google Scholar, reports that none of Pakistan's share of the ICS was Bengali.

³⁴ Young, Crawford, The Politics of Cultural Pluralism (Madison: University of Wisconsin Press, 1976), p. 482.Google Scholar

³⁵ The Biafrans suggested that influential foreigners (presumably the British) discouraged the idea of a Northern secession. Nigerian Crisis 1966 (Enugu: Eastern Regional Government, n.d.), pp. 41–42, 49Google Scholar. First, Ruth, Power in Africa (Harmondsworth: Penguin Books, 1970), p. 320Google Scholar, attributes the suppression of secessionist sentiment to a group of Northern civil servants, British and American diplomats, and Middle Belters in the army “who saw in Northern secession the danger that they would be a perpetual and unbearable minority in the North.”

³⁶ This is not invariably so. Before independence, the Ceylon Tamils, for example, sought additional parliamentary representation to compensate for their numerical weakness. See Wriggins, W.Howard, Ceylon: Dilemmas of a New Nation (Princeton: Princeton University Press, 1960), p. 91.Google Scholar

³⁷ Caplin, Gerald L., The Elites of Barotseland, 1878–1969 (London: C. Hurst & Co., 1970), Chaps. 6–8Google Scholar; Caplin, , “Barotseland: The Secessionist Challenge to Zambia, ” Journal of Modern African Studies, Vol. 6, No. 3 (10 1968), 343–60CrossRefGoogle Scholar; Bates, Margaret Rouse, UNIP in Post Independence Zambia (Ph.D. diss., Harvard University, 1971).Google Scholar

³⁸ This fits economic expectations quite well. See Williamson, , “Regional Inequality and the Process of National Development,” pp. 5–6Google Scholar

³⁹ Caplin, , The Elites of Barotseland, pp. 175–76.Google Scholar

⁴⁰ Wriggins, , Ceylon: Dilemmas of a New Nation, p. 234.Google Scholar

⁴¹ Coleman, James S., Nigeria: Background to Nationalism (Berkeley & Los Angeles: University of California Press, 1958), pp. 332–34.Google Scholar

⁴² Ibid., pp. 338–39.

⁴³ Caplin, , The Elites of Barotseland, p. 194.Google Scholar

⁴⁴ Wriggins, , Ceylon: Dilemmas of a New Nation, p. 146.Google Scholar

⁴⁵ Crozier, Brian, The Rebels: A Study of Post-War Insurrections (Boston: Beacon Press, 1960), p. 89.Google Scholar

⁴⁶ This is not an inflexible rule, or course, but it does hold for large parts of the developing world, especially less industrialized countries. Often the migration such groups undertake is temporary or seasonal, as in the case of agricultural labor. That tendency is altered, however, as industrialization proceeds, creating a need for large, unskilled and semi-skilled labor forces. In Spain, for instance, the poor Southern region of Andalusia exports much unskilled labor to Northern industry. The distribution of the population of backward groups may be a major difference - with implications for secession of backward regions - between developing and developed countries.

⁴⁷ Turner, Thomas, “Congo-Kinshasa,” in Olorunsola, Victor A., ed., The Politics of Cultural Sub-Nationalism in Africa (Garden City: Anchor Books, 1972), pp. 217–24.Google Scholar

⁴⁸ Sklar, Richard L., Nigerian Political Parties: Power in an Emergent African Nation (Princeton: Princeton University Press, 1963), pp. 327–28.Google Scholar

⁴⁹ Whiteman, K., “Enugu: The Psychology of Secession, 20 July 1966 to 30 May 1967,” in Panter-Brick, , ed., Nigerian Politics and Military Rule: Prelude to the Civil War, p. 117Google Scholar; Young, , The Politics of Cultural Pluralism, pp. 467–68Google Scholar; Olorunsola, Victor A., “Nigeria,” in Olorunsola, , ed., The Politics of Cultural Sub-Nationalism in Africa, pp. 35–36.Google Scholar

⁵⁰ Kearney, Robert N., “Language and the Rise of Tamil Separatism in Sri Lanka,” Asian Survey, Vol. 18, No. 5 (05 1978), 521–34CrossRefGoogle Scholar. See also Phadnis, Urmila, “Keeping the Tamils Internal,” Far Eastern Economic Review, 25 03 1972, 21–22Google Scholar; Warnapala, W.A.Wiswa,“Sri Lanka in 1972: Tension and Change,” Asian Survey Vol. 13, No. 2 (02 1973), 217–30.CrossRefGoogle Scholar

⁵¹ Kearney, , “Language and the Rise of Tamil Separatism in Sri Lanka,” p. 531Google Scholar; Schwarz, Walter, The Tamils of Sri Lanka (London: Minority Rights Group Report No. 25, 1975), pp. 12–13.Google Scholar

⁵² Report on the Survey of Consumer Finances, Pt. 1 (Colombo: Central Bank of Ceylon, 1974), p. 32.Google Scholar

⁵³ See Horowitz, Donald L., Coup Theories and Officers' Motives: Sri Lanka in Comparative Perspective (Princeton: Princeton University Press, 1980), Tables 2–1, 2–2.Google Scholar

⁵⁴ Salisbury, William T., “Some Aspects of the Regional Issue in Contemporary Spanish Affairs,” unpublished paper presented at the annual meeting of the International Studies Association, Toronto, February 26, 1976, p.6, drawing on data developed by Juan Linz.Google Scholar

⁵⁵ For these resentments, see Medhurst, Kenneth, The Basques (London: Minority Rights Group Report No. 9, 1972), p. 5Google Scholar; Douglas, William A. and Silva, Milton da, “Basque Nationalism,” in Pi-Sunyer, Oriol, ed., The Limits of Integration (Amherst: University of Massachusetts Department of Anthropology, Research Report no. 9, 1971), p. 149.Google Scholar

⁵⁶ Douglas, and da Silva, , “Basque Nationalism,” pp. 149–50.Google Scholar

⁵⁷ For a discussion of some of these reasons, see Dudley, , “Western Nigeria and the Nigerian Crisis,” p. 109.Google Scholar

⁵⁸ Kasfir, Nelson, “Cultural Sub-Nationalism in Uganda,” in Olorunsola, , The Politics of Cultural Sub-Nationalism in Africa, pp. 123–28.Google Scholar

⁵⁹ On the Baganda, see Young, , The Politics of Cultural Pluralism, pp. 149–56.Google Scholar

⁶⁰ See Blasco, Pedro González, “Modern Nationalism in Old Nations as a Consequence of Earlier State-Building: The Case of Basque-Spain,” in Bell, Wendell and Freeman, Walter, eds., Ethnicity and Nation-Building (Beverly Hills: Sage Publications, 1974), p. 347.Google Scholar

⁶¹ So far as tariff protection is concerned, much is likely to depend on whether the prosperity of the region is based on production of finished goods for the domestic market or on production of primary products or extraction of minerals for export. For elaboration, see p. 191, below.

⁶² Blasco, González, “Modem Nationalism in Old Nations as a Consquence of Earlier State-Building: The Case of Basque-Spain,” in Bell, and Freeman, , Ethnicity and Nation-Building, p.366.Google Scholar

⁶³ “Catalan and Basque Nationalism,” Journal of Contemporary History, Vol. 6, No. 1 (1971), 15–51, esp. 50CrossRefGoogle Scholar. The reference to Spanish industry appears on 38.

⁶⁴ See Horowitz, Donald L., “Cultural Movements and Ethnic Change,” The Annals of the American Academy of Political and Social Science, Vol. 433 (09 1977), 6–18.CrossRefGoogle Scholar

⁶⁵ Campbell, John Franklin, “Background to the Eritrean Conflict,” Africa Report, May 1971, pp. 19–20Google Scholar. Campbell's estimate is three to four times the Ethiopian average. Asmara, the Eritrean capital, had a literacy rate of 50 percent in the early 1970s, higher than any other area of Ethiopia, including Addis Ababa, which had a 4 3 percent rate. Provisional Military Government of Ethiopia, Central Statistical Office, Population and Housing Characteristics of Asmara (Addis Ababa: Central Statistical Office, Statistical Bulletin no. 12, 12 1974), p.5.Google Scholar

⁶⁶ Ethiopiawi, (pseud.), “The Eritrean-Ethiopian Conflict,” in Suhrke, Astri and Noble, Lela Garner, eds., Ethnic Conflict in International Relations (New York: Praeger, 1977), p. 131.Google Scholar

⁶⁷ There are many indirect indications of this. Asmara, by far the largest city of the region, is 85 percent Christian. Literacy in those districts of Asmara with the heaviest Muslim concentration (Akria and Geza Berhano) was below the average for the city, and housing in those districts was also of less than average quality. Provisional Military Government, Central Statistical Office, Population and Housing Characteristics of Asmara, pp. 68–69, 76–78.Google Scholar

⁶⁸ Clapham, Christopher, Haile Selassie's Government (New York: Praeger, 1969), pp. 75–76, 83.Google Scholar

⁶⁹ Turner, , “Congo-Kinshasa,” in Olorunsola, , p. 224.Google Scholar

⁷⁰ Ibid., p. 226; Gérard-Libois, Jules, Katanga Secession, Young, Rebecca, trans. (Madison: University of Wisconsin Press, 1966), pp. 12–13, 27–28Google Scholar; Lemarchand, René, Political Awakening in the Belgian Congo (Berkeley & Los Angeles: University of California Press, 1964), pp. 235–36, 241Google Scholar; Young, Crawford, “The Politics of Separatism: Katanga, 1960–63,” in Carter, Gwendolen M., ed., Politics in Africa: 7 Cases (New York: Harcourt, Brace & World, 1966), pp. 172–74Google Scholar; Lemarchand, René, “Congo (Leopoldville),” in Coleman, James S. and Rosberg, Carl G. Jr., eds., Political Parties and National Integration in Tropical Africa (Berkeley & Los Angeles: University of California Press, 1964), p. 581.Google Scholar

⁷¹ In the 1960 Katanga provincial elections, Conakat did well. It also did better than Balubakat, the party of the Katanga Baluba in the North of the province, in the national elections, but the Balubakat leader, Jason Sendwe, was nonetheless named by the central government to be High Commissioner for Katanga. This appointment triggered the Katanga secession. Turner, , “Congo-Kinshasa,” in Olorunsola, , The Politics of Cultural Sub-Nationalism in Africa, p. 227.Google Scholar

I leave aside the role of Katanga's European settlers in supporting the secession. For the settlers' role, see Gérard-Libois, , Katanga Secession. Lemarchand's judgment, which seems well-supported by the evidence, is that the settlers' secessionist “dispositions could not have led to the secession of the province unless they were shared and abetted by a substantial segment of the African population.” Political Awakening in the Belgian Congo, p. 233.Google Scholar

⁷² In Katanga, this meant, concretely, the exclusion of Kasai Baluba from political and administrative positions and the expulsion of many from the province. Lemarchand, , Political Awakening in the Belgian Congo, p. 239Google Scholar; Gérard-Libois, , Katanga Secession, p. 28.Google Scholar

⁷³ Gérard-Libois, , Katanga Secession, pp. 3–5.Google Scholar

⁷⁴ Ibid., p. 51.

⁷⁵ Ibid., p. 41.

⁷⁶ See n.4 6, above. If anything, a backward group in an advanced region may have the opposite problem: a population so compact it does not cover the whole region. If so, the secession will be of a limited area, as the Katanga secession was and as the Rwenzururu movement of Western Uganda was confined to the areas inhabited by Baamba and Bakonjo. Doornbos, “Protest Movements in Western Uganda: Some Parallels and Contrasts;” Kasfir, “Cultural Sub-Nationalism in Uganda,” 99. In both cases, these happened to be areas within which major mineral resources were also located. Eritrean Muslims are also confined largely to lowland areas.

⁷⁷ It would be a mistake, however, to minimize the policy dilemmas created by the coexistence in a single state of various kinds of regional groups. Occasionally, the claims are in direct opposition to each other, so that whan can salve the apprehensions of one group will simultaneously precipitate secessionist action by another.

⁷⁸ See Foster, Philip J., “Ethnicity and the Schools in Ghana,” Comparative Education Review, Vol. 6, No. 2 (10 1962), pp. 127–35.CrossRefGoogle Scholar

⁷⁹ Weinstein, Brian, Gabon: Nation-Building on the Ogooué (Cambridge: M.I.T. Press, 1966). pp. 220–25Google Scholar. The Bateke case, however, may fall within a caveat stated earlier regarding PakistaniBaluch and Pathan reluctance to do anything that might link them with Afghanistan. There is a similar reluctance among Bateke to do anything that might result in their annexation by neighboring Congo (Brazzaville).

⁸⁰ All else being equal, there will also be more subgroup amalgamation among advanced groups than among backward groups. See Horowitz, Donald L., “Ethnic Identity,” in Glazer, Nathan and Moynihan, Daniel P., editors, Ethnicity: Theory and Experience (Cambridge: Harvard University Press, 1975), pp. 111–40Google Scholar. On these grounds, too, late seceders will be more cohesive, their fighting forces less likely to fight with each other than with the forces of the rump state.