^{page 47 note 1} Good starting points are Resnick, Idrian N. (ed.), Tanzania: revolution by education (Arusha, 1968),Google Scholar and Cameron, J. and Dodd, W. A., Society, Schools and Progress in Tanzania (Oxford, 1970).Google Scholar

^{page 47 note 2} See Nyerere, Julius K., ‘Education for Self-Reliance’, in his Freedom and Socialism (Dar es Salaam, 1968),Google Scholar and John S. Saul. ‘High Level Manpower for Socialism’, in Resnick (ed.), op. cit. pp. 93–105.

^{page 47 note 3} I use class differentiation here, rather than class formation, being for the moment less concerned with the emergence of objective conflicts of interest between classes than with the mechanisms by which new sets of leaders are recruited from particular classes. G. Carchedi has argued that ‘the reproduction of social classes depends on the reproduction of both positions and agents’; ‘Reproduction of Social Classes at the Level of Production Relations’, in Economy and Society (London), IV, 4, 11 1975, pp. 361–417.Google Scholar My use of class differentiation is concerned with the latter, the reproduction of agents. See also Joel Samoff, ‘Class, Class Conflict, and the State: notes on the political economy of Africa’, African Studies Association, Houston, 1977.

^{page 48 note 1} Since it is often the case that authors — Tanzanian or expatriate — who attempt a critical analysis of Tanzania's development strategy are perceived to be aligning themselves with either the anti-socialist or the doctrinaire Marxist critics of Tanzania, it needs to be reiterated here that Tanzania's efforts to overcome its underdevelopment are among the most impressive, and perhaps among the most successful, in the Third World.

My analysis is developed in the spirit of ‘Tell No Lies, Claim No Easy Victories’, in Cabral, Amilcar, Revolution in Guinea (London, 1969), pp. 70–2,Google Scholar ‘Leaders Must Not be Masters’, in Nyerere, op. cit. pp. 136–42, and ‘To Plan is to Choose’, in Nyerere, Julius, Freedom and Development (Dar es Salaam, 1973), pp. 80–107.Google Scholar

^{page 48 note 2} Cf. Dundas, Charles, Kilimanjaro and its People (London, 1924; reprinted 1968);Google ScholarStahl, Kathleen M., History of the Chagga People of Kilimanjaro (The Hague, 1964);Google Scholar and also Samoff, Joel, Tanzania: Local Politics and the Structure of Power (Madison, 1974),Google Scholar ch. I.

^{page 48 note 3} As used here, Kilimanjaro corresponds roughly to the arena of political action that immediately surrounds the mountain. Formerly known as Kilimanjaro District, this is now administratively organised into Moshi, Rombo, and Hai Districts, all in Kilimanjaro Region.

^{page 48 note 4} United Nations Visiting Mission to Trust Territories in East Africa, 1954, Report on Tanganyika (New York, 1955), p. 4,Google Scholar and United Nations Visiting Mission to Trust Territories in East Africa, 1957, Report on Tanganyika (New York, 1958), p. 64.Google Scholar

^{page 49 note 1} Kilimanjaro refers to Kilimanjaro District as defined in 1969, and the statistics for other years have been adjusted to correspond. Sources: the 1948, 1957, and 1969 Censuses in Tanzania, , Statistical Abstract, 1970 (Dar es Salaam, 1971), pp. 42, 44, 176, and 204.Google Scholar Also data taken from the records of the Kilimanjaro District Education Office, and the Tanzania National Archives, Dar es Salaam, 5/47/14, as well as Cameron and Dodd, op. cit. pp. 102 and 104.

^{page 49 note 2} Cf. Samoff, op. cit. pp. 40–58, and Joel, and Samoff, Rachel, ‘The Local Politics of Underdevelopment’, in The African Review (Dar es Salaam), VI, 1, 1976, pp. 69–97.Google Scholar

^{page 49 note 3} Cf. Samoff, op. cit. pp. 42–3.

^{page 50 note 1} Refers to assisted schools only, all denominations. Sources: Kilimanjaro District and Regional Education Offices. The separate categories of schools with Standards I–IV, V–VII, and I–VII, have been combined. If unassisted and unrecognised ‘bush’ schools were added, the church rôle would be substantially greater.

^{page 51 note 1} Government of Tanzania, Department of Planning, Prime Minister's Office, Hali ya Uchumi na Maendeleo ya Mikoa, Juni 1973 (Dar es Salaam, 1973).Google Scholar

^{page 51 note 2} Tanzania, , The Economic Survey, 1971–72 (Dar es Salaam, 1972), p. 70.Google Scholar

^{page 51 note 3} Church official, Kilimanjaro, 1974. Unless otherwise specified, information and quotations are taken from research in Kilimanjaro during 1968–9 and 1974, with the anonymity of the sources preserved.

^{page 52 note 1} Sources: Kilimanjaro District Book and data from the Regional Education Office. Kilimanjaro refers to the District as defined in 1969, with the figures for other years adjusted to correspond—those for 1954 include Teacher-Training Colleges.

^{page 52 note 2} Comparing Kilimanjaro and the rest of Tanzania in this regard is a difficult task, since secondary-school students are often placed outside their region of origin. Thus, much of the enrolment in government institutions in Kilimanjaro is non-local, while many Kilimanjaro students are in secondary schools elsewhere. Yet since as noted below the bulk of the intake to the private schools is local, the trend is clear, and it is reasonable to expect that once established they will increase their enrolment, just as the government schools have done.

^{page 52 note 3} Five per cent of the students in Kilimanjaro primary schools are to be offered places in government secondary schools, compared to a national average of over 10 per cent.

^{page 52 note 4} Tanzania, , Annual Manpower Report to the President, 1973 (Dar es Salaam, 1974), p. 22.Google Scholar

^{page 53 note 1} Touring Kilimanjaro in 1974, the Minister of National Education praised the private schools for their self-reliance and their contribution to national education; Tanzania Daily News (Dar es Salaam), 29 04 1974, p. 1.Google Scholar

^{page 53 note 2} The nature and behaviour of this group are discussed in more detail below. Cf. Samoff, op. cit. pp. 158–9; also Samoffs, loc. cit. pp. 90–92.

^{page 54 note 1} Cf. Samoff, op. cit. pp. 42–4, and Dey, Ajoy K. and Mogil, Norman, ‘Literacy and Education’, in Egero, Bertil and Henin, Roushdi A. (eds.), The Population of Tanzania (Dar es Salaam, 1973), pp. 119–30.Google Scholar

^{page 54 note 2} Samoffs, loc. cit. pp. 82–3.

^{page 55 note 1} Cf. Foster-Carter, Aidan, ‘Neo-Marxist Approaches to Development and Underdevelopment’, in Kadt, Emanuel de and Williams, Gavin (eds.), Sociology and Development (London, 1974),Google Scholar which includes a useful bibliography.

^{page 55 note 2} See especially Poulantzas, Nicos, Political Power and Social Classes (London, 1973 trans.),Google Scholar ‘The Problem of the Capitalist State’, in Blackburn, Robin (ed.), Ideology in Social Science (New York, 1973), pp. 238–53,Google Scholar reprinted from New Left Review (London), 58, 11–12 1969,Google Scholar and ‘The Capitalist State: a reply to Miliband and Laclau’, in ibid. 95, Jaunary–February 1976, pp. 63–83. Also Miliband, Ralph, The State in Capitalist Society (London, 1969),Google Scholar and ‘The Problem of the Capitalist State: reply to Nicos Poulantzas’, in Blackburn (ed.), op. cit. pp. 253–62, reprinted from New Left Review, 59, 01–02 1970;Google ScholarLaclau, Ernesto, ‘The Specificity of the Political: around the Poulantzas-Miliband debate’, in Economy and Society, IV, 1, 02 1975, pp. 87–110;Google ScholarWolfe, Alan, ‘New Direction in the Marxist Theory of Politics’, in Politics and Society (Los Altos, Calif), IV, 2,Google Scholar Winter 1974, pp. 131–59; Amy Beth Bridges, ‘Nicos Poulantzas and the Marxist Theory of the State’, in ibid. pp. 161–90; and Jean-Claude Girardin, ‘On the Marxist Theory of the State’, in ibid. pp. 193–223. edited and abridged from an article in Les Temps modernes (Paris), 09–10 1972.Google Scholar

^{page 56 note 1} Lenin, 1916, quoted in Shivji, Issa G., Class Struggles in Tanzania (New York, 1976), p. 112.Google Scholar

^{page 56 note 2} That much of the current social science writing reflects all of these equations is obvious. Unfortunately, even many of the radical critics, perhaps attempting to impose a positivist methodology on a dialectical theory, use similar constructs. Shivji, ibid. is one example; see also Anderson, Charles H., The Political Economy of Social Class (Englewood Cliffs, 1974), especially pp. 125–3.Google Scholar

^{page 57 note 1} Samoffs, loc. cit. especially p. 90–92. That terminology is unfortunately awkward, since as the thrust of my argument thus far suggests, we are only just beginning to understand the nature of this class. As used here, the petty or petite bourgeoisie refers to small-scale capitalists, and includes the educated élite, some of whom hold political or administrative positions with little direct involvement in capitalist enterprise. Although the origins of the class are now clear, the dynamic of the larger context permits some autonomy of action – either in the consolidation of its capitalist rôle or towards a progressive, even socialist rôle. M. von Freyhold prefers the term ‘Nizers’, from those whose position stems from Africanization; quoted by Saul, John S., ‘The State in Post-Colonial Societies: Tanzania’, in Miliband, Ralph and Saville, John (eds.), Socialist Register, 1974 (London, 1974), pp. 349–72.Google Scholar

^{page 57 note 2} Cf. Feldman, Rayah, ‘Rural Social Differentiation and Political Goals in Tanzania’, in Oxaal, Ivar et al. (eds.), Beyond the Sociology of Development: economy and society in Latin America and Africa (London, 1975), pp. 154–82.Google Scholar

^{page 58 note 1} Not all estates were nationalised. Those affected were in the core of the coffee-growing area (thus excluding estates in West Kilimanjaro, and those on which coffee was not the major crop) and larger than 50 acres, thereby including only the largest African holdings.

^{page 58 note 2} Cf. Rogers, Susan Geiger, ‘The Search for Political Focus on Kilimanjaro: a history of Chagga politics, 1916–1952’, Ph.D. dissertation, University of Dar es Salaam, 1972;Google ScholarMcCarthy, Dennis M. P., ‘The Politics of Economic Change in Tanganyika, 1919–1939’, Ph.D. dissertation, Yale University, New Haven, 1972,Google Scholar ch. VII; and Samoff, op. cit. pp. 19–31.

^{page 59 note 1} Samoffs, loc. cit.

^{page 59 note 2} Ibid. pp. 94–5.

^{page 60 note 1} Bienen, Henry, Tanzania: party transformation and economic development (Princeton, 1970 edn.), pp. 43–8.Google Scholar

^{page 60 note 2} What is involved is not mass emigration but rather a movement of individuals towards economic opportunity outside the region. Although this movement is described by local leaders as large in scale, an analysis of the 1967 census indicated a new extra-regional migration of only 1·7 per Cent of the total population of Kilimanjaro Region. See Claes-Fredrik Claeson and Bertil Egero, ‘Migration’, in Egero and Henin (eds.), op. cit. p. 66.

^{page 60 note 3} See Saul, loc. cit. pp. 357–9.

^{page 61 note 1} See Saul, loc. cit. pp. 354–9; Cabral, ‘Brief Analysis of the Social Structure in Guinea’ and ‘The Weapon of Theory’, op. cit. pp. 46–61 and 73–90; and Rodney, Walter, ‘Some Implications of the Question of Disengagement from Imperialsim’, in Maji Maji (Dar es Salaam), I, 1, 01 1971,Google Scholar reprinted in Cliffe, Lionel and Saul, John S. (eds.), Socialism in Tanzania, Vol. II, Policies (Nairobi, 1973), pp. 350–3.Google Scholar

^{page 61 note 2} Girardin, loc. cit.

^{page 61 note 3} Miliband, op. cit. pp. 161–95.

^{page 61 note 4} See Ake, Claude, ‘Explanatory Notes on the Political Economy of Africa’, in The Journal of Modern African Studies (Cambridge), XIV, 1, 03 1976, p. 3Google Scholar and Harris, Richard, ‘The Political Economy of Africa – Underdevelopment or Revolution’, in Harris, (ed.), The Political Economy of Africa (New York, 1975), pp. 30–3.Google Scholar

^{page 61 note 5} Coulson, A. C., ‘A Simplified Political Economy of Tanzania’, Universities of East Africa Social Science Conference, Dar es Salaam, 1973.Google Scholar

^{page 61 note 6} It is important here to maintain a distinction between class power and state power – see Laclau, loc. cit. pp. 93 and 101. Due to the dependent structure of African economies, control over state power by the ruling class is the primary condition for its own reproduction.

^{page 62 note 1} See Shivji, op. cit. pt. III, and Harris, op. cit. pp. 24–5.

^{page 62 note 2} Cf. Cliffe and Saul (eds.). op. cit. pp. 331–58.

^{page 62 note 3} Cf. Samoff, op. cit. pp. 19–31 and pt. IV.

^{page 62 note 4} Feldman, loc. cit. p. 155, and Raikes, P. L., ‘Ujamaa and Rural Socialism’, inReview of African Political Economy (London), 3, 05–10 1975, p. 45.Google Scholar

^{page 63 note 1} Cf. McCarthy, op. cit. pp. 247–53, and ‘A Warning from the Mountain: unrest in Northern Tanganyika in the 1930s’, 1975, manuscript.

^{page 63 note 2} The structural interpenetration of government and party in Kilimanjaro makes to reasonable to regard both as state institutions. See Samoff, op. cit. pp. 161–4 and 195–200.

^{page 63 note 3} When constructs developed for national analysis are applied at the local level there may be some difficulties of terminology. The African petty bourgeoisie during the period of European rule (often called the ‘middle class’ and/or the ‘educated élite’) constituted the leadership of the nationalist anti-colonial movement. Thereafter that class became the national segment of the bourgeoisie — the other, dominant, segment remaining in the metropole. Although individuals from Kilimanjaro have become part of that national segment Tanzanian bourgeoisie, I am concerned here with the local ruling class which has remained a distinctly petty bourgeoisie. Hence, that branch whose power stems from political and administrative positions can appropriately be described as the ‘local petty bureaucratic bourgeoisie’.

^{page 63 note 4} Samoff, op. cit. ch. 3.

^{page 63 note 5} Samoffs, loc. cit.

^{page 63 note 6} Cf. Feldman, loc. cit. p. 161.

^{page 63 note 7} Cf. Dundas, op. cit. p. 301.

^{page 64 note 1} Annual Reports of District and Provincial Commissioners rarely failed to mention it.

^{page 64 note 2} The projections of political turmoil due to land scarcity that fill the historical record have led several researchers to attempt to locate and describe a landless group of residents. Those efforts have thus far been unsuccessful. In fact, many local leaders suggest that the problems of land scarcity have been greatly overstated, that technological advances permit an adequate income from smaller coffee plots, and that there is unclaimed land available in the lower areas (unsuitable for coffee but suitable for annual food crops). In a survey of Kilimanjaro farmers, 62 per cent reported there was no land shortage at all. See Mbilinyi, Simon M., ‘The Economics of Peasant Coffee Production: the case of Tanzania’, Dar es Salaam, 1973,Google Scholar and also D. F. Conyers et al., ‘Agro-Economic Zones of North-Eastern Tanzania’, University of Dar es Salaam, Bureau of Resource Assessment and Land Use Planning, Research Report No. 13, 1970.

^{page 64 note 3} See Samoff, op. cit. p. 167.

^{page 64 note 4} Mbilinyi reported that 55 per cent of the families in the coffee-growing areas hired labour; op. cit. pp. 159–60.

^{page 65 note 1} Source: Kilimanjaro Native Co-operative Union, December 1973.

^{page 66 note 1} Source: Chr. Smith, ‘Diversification of the Rural Economy on Mt. Kilimanjaro’, University of Dar es Salaam, Bureau of Resource Assessment and Land Use Planning, Research Report No. 18, 1970. 368 farms were surveyed.

^{page 66 note 2} The transport needs of 1 ton of maize and 50 kilograms of coffee, with approximately equivalent market values in 1969, are very different.

^{page 66 note 3} To encourage the production of more maize the price paid to the farmers was raised by 6 per cent in 1973–4 and by 43 per cent in 1974–5; see Hali ya Uchumi wa Taifa katika Mwaka 1973–74 (Dar es Salaam, 1974), p. 56.Google Scholar

^{page 67 note 1} Source, : Tanzania. 1967 Population Census, Vol. IV (Dar es Salaam, 1971), Table 321, p. 413.Google Scholar

^{page 67 note 2} Source: Smith, op. cit. p. 16—a sample of 368 farmers.

^{page 67 note 3} Source: Hanson, C.J. and Sinker, N. D., ‘An Economic Survey of Small Holder Coffee Farming on the Slopes of Mt. Kilimanjaro’, Moshi, Coffee Research Station, Lyamungu, 1970, p. 3.Google Scholar Note that the sum of the percentages exceeds 100 per cent, since farmers were asked to describe all their non-coffee land and livestock.

^{page 68 note 1} See Ake, loc. cit. pp. 4–9, and Molteno, Robert, ‘Cleavage and Conflict in Zambian Politics: a study in sectionalism’, in Tordoff, William (ed.), Politics in Zambia (Manchester, 1974), pp. 62–106.Google Scholar

^{page 68 note 2} Wayne, Jack, ‘Some Notes on the Sociology of Dependency: the underdevelopment of Kigoma Region, Tanzania’, Universities of East Africa Social Science Conference, Dar es Salaam, 1973, pp. 7–10.Google Scholar

^{page 68 note 3} Feldman, loc. cit. p. 167. See also Mbilinyi, Marjorie J., ‘The Problem of Unequal Access to Primary Education in Tanzania’, in Rural Africana (East Lansing), 25, Fall 1974, pp. 5–28,Google Scholar also presented to the Universities of East Africa Social Science Conference, Dar es Salaam, 1973.

^{page 68 note 4} See Rodney, Walter, How Europe Underdeveloped Africa (Dar es Salaam, 1972),Google Scholar and Harris, loc. cit.

^{page 69 note 1} Cf. Arrighi, Giovanni and Saul, John S., ‘Socialism and Economic Development in Tropical Africa’, in The Journal of Modern African Studies, VI, 2, 08 1968, pp. 141–69,CrossRefGoogle Scholar reprinted in Arrighi, and Saul, , Essays on the Political Economy of Africa (New York, 1973);Google Scholar and John S. Saul, ‘African Socialism in One Country: Tanzania’, Symposium on Strategies for Development — Africa Compared with Latin America, Dakar, 1972, reprinted in Arrighi and Saul, op. cit. pp. 237–335.

^{page 69 note 2} Cf. Ake, loc. cit. p. 6; Harris, loc. cit. pp. 45–7; and Saul, loc. cit. 1974.