^{page 99 note 1} Bienen, Henry, ‘The Background to Contemporary Studies of Militaries and Modernization’, in Bienen, (ed.), The Military and Modernization (Chicago, 1971), p. 12Google Scholar. The locus classicus of this approach is Levy, Marion J. Jr, Modernization and the Structure of Societies (Princeton, 1966), pt. III, ch. 4, pp. 571–605Google Scholar. See also Pye, Lucien W., ‘Armies in the Process of Political Modernization’, in Johnson, John J. (ed.), The Role of the Military in Underdeveloped Countries (Princeton, 1962), pp. 69–89Google Scholar; Gutteridge, W. F., The Military in African Politics (London, 1969)Google Scholar; and Lefever, Ernest W., Spear and Scepter (Washington, 1970)Google Scholar.

^{page 99 note 2} Johnson, John J., The Military and Society in Latin America (Standford, 1964)Google Scholar; Pauker, Guy J., ‘Southeast Asia as a Problem Area in the Next Decade’, in World Politics (Princeton), XI, 3, 04 1959Google Scholar; Manfred Halpern, ‘Middle Eastern Armies and the New Middle Class’, in Johnson (ed), op. cit. pp. 277–315; Halpern, Manfred, The Politics of Social Change in the Middle East and North Africa (Princeton, 1962)Google Scholar; and Daalder, Hans, The Role of the Military in the Emerging Countries (The Hague, 1962)Google Scholar. See also Edward Shils, ‘The Military in the Political Development of New States’, in Johnson (ed.), op. cit. pp. 7–67.

^{page 99 note+ 3} Finer, S. E., The Man on Horseback (London, 1962), ch. 3Google Scholar; Janowitz, Morris, The Military in the Political Development of New Nations (Chicago, 1964)Google Scholar; Luckham, Robin, The Nigerian Military (Cambridge, 1971)Google Scholar; Price, Robert M., ‘A Theoretical Approach to Military Rule in New States: reference group theory and the Ghanaian case’, in World Politics, XXIII, 3, 04 1971, pp. 399–430CrossRefGoogle Scholar; Huntington, Samuel P., Political Order in Changing Societies (New Haven, 1968), pp. 219–37Google Scholar; and Lee, J. M., African Armies and Civil Order (London, 1969)Google Scholar.

^{page 100 note 1} Nun, Jose, ‘The Middle-Class Military Coup’, in Veliz, Claudio (ed.), The Politics of Conformity (Oxford, 1967)Google Scholar; McAlister, Lyle N., ‘The Military’, in Johnson, John J. (ed.), Continuity and Change in Latin America (Standford, 1964), pp. 145–60Google Scholar; Needler, Martin C., ‘Political Development and Military Intervention in Latin America’, in American Political Science Review (Menasha), LX, 3, 09 1966, pp. 616–26Google Scholar; Lieuwin, Edwin, Arms and Politics in Latin America (New York, 1960)Google Scholar; Price, Robert M., ‘Military Officers and Political Leadership: the Ghanaian case’, in Comparative Politics (New York), III, 3, 04 1971, pp. 361–80Google Scholar; First, Ruth, The Barrel of a Gun: political power in Africa and the coup d'état (Harmondsworth, 1972)Google Scholar; and Decalo, Samuel, Coups and Military Rule in Africa (New Haven, 1976)Google Scholar.

^{page 101 note 1} Nordlinger, Eric A., ‘Soldiers in Mufti: the political impact of military rule upon economic and social change in non-western states’, in The American Political Science Review, LXIV, 4, 12 1970, p. 1144Google Scholar.

^{page 101 note 2} Tufte, Edward R., ‘Improving Data Analysis in Political Science’, in World Politics, XXI, 4, 07 1969, pp. 641–54CrossRefGoogle Scholar; and Jackman, Robert W., ‘Politicians in Uniform: military governments and social change in the Third World’, in The American Political Science Review, LXX, 4, 12 1976, pp. 1089–90Google Scholar.

^{page 102 note 1} Jackman, loc. cit. esp. pp. 1095–6.

^{page 103 note 1} Ibid. p. 1092.

^{page 103 note 2} McKinlay, R. D. and Cohan, A. S., ‘A Comparative Analysis of the Political and Economic Performance of Military and Civilian Regimes’, in Comparative Politics, VIII, 1, 10 1975, pp. 1–30CrossRefGoogle Scholar.

^{page 104 note 1} Ibid. pp. 21 and 26, footnotes 35 and 36.

^{page 105 note 1} In 1975, African states spent an average of 3·02 per cent of their G.N.P. on the military compared to a mean for all L.D.C.s of 6·10 per cent; military expenditure per capita of population was $11.85 compared to $24.65; and the armed forces per 1,000 population ratio in Africa was 2·79 in contrast to 5·12 for all L.D.C.s. U.S. Arms Control and Disarmament Agency, World Military Expenditure and Arms Trade, 1966–75 (Washington, D.C., 1977), pp. 15–17Google Scholar.

^{page 106 note 1} O'Donnell, Guillermo, Modernization and Bureaucratic Authoritarianism (Berkeley, 1973)Google Scholar.

^{page 106 note 2} The single exception is Equatorial Guinea which gained its independence on 12 October 1968 but has not regularly reported statistical data to the principal multilateral agencies. Four other military régimes were in office for an insufficient length of time during the period under consideration for meaningful data to be collected, namely: two in Sierra Leone, 1967–8; one in Benin, as a result of the coup of 26 October 1972; and the Acheampong régime in Ghana which assumed office on 13 January 1972. The design employed in this study resembles that used by McKinlay and Cohan, except for the inclusion of the ‘mixed’ régime category, and the choice of population and performance variables.

^{page 107 note 1} For details of this procedure, see Jae-on-Kim, and Kohout, Frank J., ‘Special Topics in General Linear Models’, in Nie, Norman H. et al. , Statistical Package for the Social Sciences (New York, 1975 edn.), pp. 368–71Google Scholar. Where countries gained independence after 1960, those dates were used as the base for calculating rates of growth achieved by the first post-independence régimes. Where changes occurred from civilian to military régimes or vice versa, the data for the last complete year in office of the previous régime were used as the base against which the performance of its successor was compared.

^{page 107 note 2} Data for G.N.P., G.D.I., and exports were taken from I.B.R.D. World Tables, 1976 (Baltimore, 1976)Google Scholar, while international reserves were derived from U.N. Statistical Yearbook (New York, 1961–1976)Google Scholar. No data on international reserves were available for Burundi, Guinea, Liberia, Botswana, Lesotho, and Swaziland. The last four countries used the national currencies of other states as their domestic legal tender during the period under consideration; see d'A Collins, Francis et al. , ‘The Rand and the Monetary Systems of Botswana, Lesotho, and Swaziland’, in The Journal of Modern African Studies (Cambridge), XVI, 1, 03 1978, pp. 97–121CrossRefGoogle Scholar.

^{page 108 note 1} Data for these indicators were obtained from U.N. Economic Commission for Africa, African Statistical Yearbook (Addis Ababa), 1972–1976Google Scholar; U.N.E.S.C.O., Statistical Yearbook (Paris), 1961–1976Google Scholar; and U.N. Statistical Yearbook (New York), 1961–1976Google Scholar. No data for F-P.I. were available for Benin, Burundi, Guinea, and Rwanda.

^{page 108 note 2} Clark, Edmund, ‘Socialist Development in an Underdeveloped Country: the case of Tanzania’, in World Development (Oxford), III, 4, 04 1975, pp. 223–8Google Scholar.

^{page 109 note 1} For an interesting attempt to combine objective indicators of régime performance with a survey of domestic opinion, see Olorunsola, Victor A., Soldiers and Power: the development performance of the Nigerian military regime (Standford, 1977)Google Scholar.

^{page 110 note 1} All figures for military expenditure are in constant dollars. Source: for military expenditure, and military expenditure as a percentage of G.N.P., 1963–73, U.S. Arms Control and Disarmament Agency, World Military Expenditures and Arms Trade, 1963–73 (Washington, 1975)Google Scholar; for the four variables for the period 1971–3, the source was ibid.1965–75 (Washington, 1977).

^{page 112 note 1} This was confirmed by re-analysing the data using the multiple regression design employed by Jackman. Similarly, an analysis of the data excluding the ‘mixed’ régime group did not produce any significant results.

^{page 113 note 1} T = – 1·98, p = 0·052.

^{page 113 note 2} T = – 2·79, p = 0·039.

^{page 113 note 3} Kaldor, Mary, ‘The Military in Development’, in World Development, IV, 6, 06 1976, pp. 459–82CrossRefGoogle Scholar.

^{page 114 note 1} Mean figures for the four régimes were: G.N.P., 7·08; G.D.I., 14·31; EX., 10·63; P-S.E., 3·28; F-P.I., 4·27. No data were available for I.R. since the three Southern African countries were members of the Rand monetary union.

^{page 114 note 2} Mean figures for the 15 remaining C.R. were: G.N.P., 4·53; G.D.I., 5·07; EX., 6·96; and P-S.E., 6·72.

^{page 115 note 1} Decalo, op. cit. pp. 32–3.

^{page 115 note 2} Confirmed by re-analysing the data using a multiple regression design with a dummy variable. These findings confirm those of McKinlay and Cohan, loc. cit. pp. 13–14

^{page 118 note 1} A suggestion made by Decalo, op. cit. p. 238.

^{page 115 note 2} McKinlay and Cohan, loc. cit. p. 2.

^{page 115 note 3} This apparently lends support to the analysis presented by Laitin, David D., ‘The Political Economy of Military Rule in Somalia’, in The Journal of Modern African Studies, XIV, 3, 09 1976, pp. 449–68CrossRefGoogle Scholar.

^{page 120 note 1} For an earlier analysis of the performance of the Ghanaian military régime, see Price, ‘A Theoretical Approach to Military Rule in New States’, loc. cit.

^{page 120 note 2} This tendency is evident in some of the statements made by Decalo, op. cit. esp. pp. 25–6.

^{page 121 note 1} Régimes in the ‘negative’ column recorded growth rates significantly lower than the mean on the variable indicated (but not necessarily negative rates of growth), except for F-P.I. where the ‘negative’ column represents those régimes which recorded significantly higher rates of growth. The converse applies to the ‘positive’ column.

^{page 122 note 1} To guard against distortions in the data being produced by virtue of certain régimes performing well over a limited period of time, a second analysis was made using only the scores for C.R. and for ‘mixed’ régimes, which allowed comparisons between the performance of all countries for the complete period 1960–73. Again neither the Ivory Coast nor Kenya registered performances significantly different from the group mean. A composite picture of overall performance can be obtained by ranking the performance of each régime on all economic/socioeconomic variables (in reverse order for F-P.I.) and then aggregating the scores for each régime. This has been omitted owing to space limitations.

^{page 122 note 2} Kalleberg, Arthur L. emphasises that the solution of the classification problem is a prerequisite for comparative analysis; ‘The Logic of Comparison: a methodological note on the comparative study of political systems’, in World Politics, XIX, 1, 10 1966, pp. 69–82CrossRefGoogle Scholar.

^{page 122 note 3} Bienen, Henry, ‘Military Rule and Political Process: Nigerian examples’, in Comparative Politics, 01 1978Google Scholar, reprinted in Bienen, , Armies and Parties in Africa (New York, 1978), p. 195Google Scholar.

^{page 122 note 4} The principal problem is that Bienen fails to provide an adequate definition of ‘political process’. He acknowledges, however, that ‘politics’ does not stop with the coming to power of the military; ibid. p. 209.

^{page 123 note 1} Cf. Bienen, Henry, Tanzania: party transformation and economic development (Princeton, 1970)Google Scholar, and also Zolberg, Aristide R., Creating Political Order: the party states of West Africa (Chicago, 1966)Google Scholar.

^{page 123 note 2} Bienen, ‘Military Rule and Political Process’, loc. cit. p. 196.

^{page 123 note 3} For an interesting comparison of the attitudes of the two Ghanaian military régimes towards political process, see Rothchild, Donald, ‘Ethnicity and Purposive Depoliticization: the public policies of two Ghanaian military regimes’, 21st Annual Meeting of the African Studies Association, Baltimore, Maryland, 1–4 11 1978Google Scholar.

^{page 124 note 1} This scepticism would accord with the sentiments of an early critic of ideal-type conceptions of military organisation, namely that ‘One may well doubt whether soldiers’ political activities can be distinguished analytically from those of any other political group; whether from the viewpoint of systems analysis or of functional analysis, or alternatively from a problem-solving perspective, anything unique to the military is likely to emerge.’ Dowse, Robert E., ‘The Military and Political Development’, in Leys, Colin (ed), Politics and Change in Developing Countries (Cambridge, 1969), p. 213Google Scholar.

^{page 124 note 2} Bienen, ‘The Background to Contemporary Studies of Militaries and Modernization’, loc. cit.

^{page 124 note 3} Gutteridge, William F., Military Regimes in Africa (London, 1975), p. 183Google Scholar.

^{page 124 note 4} Zartman, I. William, ‘Europe and Africa: decolonization or dependency?’, in Foreign Affairs (New York), 54, 01 1976, pp. 325–43Google Scholar.

^{page 125 note 1} Weaver, Jerry L., ‘Assessing the Impact of Military Rule: alternative approaches’, in Schmitter, Philippe C. (ed.), Military Rule in Latin America (Beverly Hills, 1973), p. 95Google Scholar.