^{page 265 note 1} While the word ‘communal’ is perhaps more frequently used in discussions of this topic, I use the term ‘collective’ in order to distinguish between purely ‘communal’ or ‘communistic’ production, in which the product is distributed according to the needs of the community, and ‘collective’ or ‘socialistic’ production in which the product is distributed according to individual labour contributions.

The distinction follows Marx, Karl, ‘The Critique of the Gotha Program’, in Tucker, Robert C. (ed.), The Marx-Engels Reader (New York, 1972), pp. 387–8.Google Scholar Since the product of the ujamaa plot was in principle - see Nyerere, Julius K., ‘Socialism and Rural Development’, in Freedom and Socialism/Uhuru na Ujamaa (Dar es Salam, 1968), p. 359–Google Scholar and in the most widespread practice, to be distributed according to the work contributions of the wajamaa (co-operators), the term ‘collective’ seems more appropriate. That actual distribution followed an intermediate pattern, however, will be pointed put below.

^{page 264 note 1} See, for example, Myrdal, Gunnar, The Political Element in the Development of Economic Theory, translated by Streeten, Paul (London, 1953).Google Scholar

^{page 264 note 2} For example, Lange, Oskar, ‘On the Economic Theory of Socialism’, in Lippincott, B. E. (ed.), On the Economic Theory of Socialism (New York, 1938).Google Scholar

^{page 264 note 3} This argument is similar to the classical argument concerning share-cropping, which has become the object of a theoretical counter-attack in recent studies. As will be seen below, it is strictly relevant only to co-operatives having egalitarian distribution systems, or those in which the differentiation of rewards according to work done is made exceedingly imperfect by poor monitoring and recording practices. In the terms that are introduced in footnote 1, p. 263, the argument is thus relevant to ‘communal’ enterprise only. However, the implicit assumption of ‘communal’ or egalitarian distribution in producers’ co-operatives was frequently made by earlier writers, and continues to be made by current non-specialists.

^{page 265 note 1} See Ward, Benjamin, ‘The Firm in Illyria: market syndicalism’, in American Economic Review (Nashville), 48, 1958, pp. 566–89.Google Scholar

^{page 265 note 2} See, for example, Vanek, Jaroslav, The General Theory of Labor-Managed Market Economies (Ithaca, 1970).Google Scholar

^{page 266 note 1} For example, see Vanek, Jaroslav, ‘The Yugoslav Economy Viewed Through the Theory of Labor Management’, in World Development (Oxford), 1, 1973, pp. 39–56.Google Scholar

^{page 266 note 2} See Domar, Evsey D., ‘The Soviet Collective Farm as a Producers' Cooperative’, in American Economic Review, 56, 1966, pp. 734–57Google Scholar; Sen, Amartya K., ‘Labour Allocation in a Cooperative Enterprise’, in Review of Economic Studies (Edinburgh), 33, 1966, pp. 361–71;Google ScholarBonin, John P., ‘Work Incentives and Uncertainty on a Collective Farm’, in Journal of Comparative Economics (New York), 1, 1977, pp. 77–97;Google Scholar and Chinn, Dennis L., ‘Team Cohesion and Collective-Labor Supply in Chinese Agriculture’, in Journal of Comparative Economics 3, 1979, pp. 375–94.Google Scholar

^{page 266 note 3} The original version of Nyerere, 1968 (cited on p. 263). The Swahili title means ‘rural socialism’ and not, as sometimes stated, ‘ujamaa villages’.

^{page 266 note 4} This proposition is the subject of much recent research and debate. For a discussion with references to the literature, see Putterman, Louis, ‘A Modified Collective Agriculture in Rural Growth-with-Equity: reconsidering the private, unimodal solution’, Working Paper No. 81–25, Department of Economics, Brown University, Providence, 1981.Google Scholar

^{page 267 note 1} For example, households may earn 20 per cent or more of their incomes from their private plots in Chinese communes, and some evidence shows net income per hectare of private plot land to be much higher than that of communally farmed land in China. (These estimates were computed using cash values for income in kind.) See Chinn, Dennis L., ‘Income Distribution in a Chinese Commune’, in Journal of Comparative Economics, 2, 1978, pp. 246–65.CrossRefGoogle Scholar

^{page 267 note 2} Nyerere, op. cit. p. 357, my emphasis.

^{page 268 note 1} See von Freyhold, Michaela, Ujamaa Villages in Tanzania: analysis of a social experiment (New York and London, 1979), pp. 72–7.Google Scholar

^{page 268 note 2} See Mkuu, Ofisi ya Waziri, Sheria ya Kuandikisha Vijiji uya Ujamaa (Dodoma, 1975)Google Scholar, and Nyerere, Julius K., The Arusha Declaration Ten Years After (Dar es Salam, 1977).Google Scholar

^{page 268 note 3} The article on the use of village land in the directive of the Prime Minister under the Village Registration Act makes it quite clear that land is to be used by individual families only at the discretion of the Village Council, and that no household may consider land as property that can be sold or transferred to others. See ‘Maagizo’, in Ofisi ya Waziri Mkuu, op. cit. pp. 44–8.

^{page 269 note 1} The use of the term ‘profit’ with reference to the earnings of collective members may be objectionable from the standpoint of classical or Marxian economics, in which the basic income categories (wages, profits, and rents) have specific social references. Here ‘profit’ means nothing other than revenue net of specific costs. While the revenue shares going to labourers or to factor-owners dealing with the enterprise can always be considered as costs regardless of the terms of contract, in one sense, they are profit shares from the present viewpoint whenever their level (per unit of service or input) depends on total enterprise revenue, rather than being contractually specified.

^{page 270 note 1} This analysis is still incomplete, however. If the members have no other means of livelihood and the ‘communal’ rule of distribution is assumed to be institutionally irrevocable, then they may realise a common interest in production, despite the individual incentives to shirk. They may therefore attempt to ‘collude’ to work hard. See below, pp. 272 4, and Putterman, Louis, ‘Incentives and the Kibbutz: toward an economics of communial work motivation’, Working Paper No. 8 81–24, Department of Economics, Brown University, Providence, 1981.Google Scholar

Although some individuals may break away from this social solution (i.e. they may shirk) for individual benefit, there will be a group interest in ‘policing’ it by means of social pressure, and, in the long run, an attempt may be made to socialise members for communal co-operation, so that internal rather than external motivation prevails. These factors do not guarantee that communes will be highly productive organisations, but they are likely to be of some effect, and they help to explain such successful cases as the Hutterite commune and the kibbutz. See Bullock, Kari and Baden, John, ‘Communes and the Logic of the Commons’, in Hardin, Garrett and Baden, John, (eds.), Managing the Commons (San Francisco, 1977), pp. 182–99;Google ScholarMelman, Seymour, ‘Industrial Efficiency under Managerial versus Cooperative Decision-Making’, in Horvat, Branko, Markovic, Mihailo, and Supek, Rudi (eds.), Self-Governing Socialism: a Reader, Vol. II, Sociology and Politics, Economics (New York, 1975), pp. 203–20.Google Scholar

^{page 270 note 2} See Putterman, Louis, ‘Voluntary Collectivization: a model of producers' institutional choice’, in Journal of Comparative Economics, 4, 1980, pp. 125–57.CrossRefGoogle Scholar

^{page 272 note 1} Von Freyhold, op.cit. p. 85, concludes that ‘in the longer run it was the consensus of the members which determined the size of the communal enterprise’. She thus enunciates a central thesis of the present analysis.

^{page 273 note 1} Unless we can assume that all hours of labour are homogeneous, which is not generally the case either across labourers, or when comparing different hours worked by a single individual (especially, under alternative organisational règimes), it is necessary to think in terms of hypothetical labour units of precisely equal effectiveness. In the present dicussion, it is implicity assumed that individuals have the same capacity to contribute such ‘efficiency labour units’, and that variation therefore reflects choice. This abstracts from other issues, while allowing concentration on the incentive problem as such.

^{page 275 note 1} I.e. ‘communal’ distribution, or equal shares to all members.

^{page 275 note 2} On the agricultural history of the People's Republic of China, see Chao, Kang, Agricultural Production in Communist China, 1949–65, (Madison, 1970), and other source noted in Putterman, ‘A Modified Collective Agriculture in Rural Growth-with-Equity’.Google Scholar

^{page 275 note 3} See Mushi, Samuel S., ‘Modernization by Traditionalization: ujamaa principles revisited’, in Taamuli (Dar es Salaam), 1, 2, 1971, pp. 13–29.Google Scholar

^{page 277 note 1} Sen, loc.cit.

^{page 278 note 1} Of course, the villager may also be simply incapable of thinking about these matters, and may follow whatever he is told to do because he has no other basis for action. This may well characterise some individuals, but the evidence collected by both orthodox economists and radical social scientists invariably supports the view that peasants in Tanzania and elsewhere are, in fact, quite astute and capable of reacting in a self-interested manner to changes in their opportunities and information.

^{page 278 note 2} This is, in fact, a controversial question in Tanzania's case. I find von Freyhold's discussion of this matter – op.cit. pp. 22–31 – to be closely parallel to my own views.

^{page 279 note 1} On public goods and the ‘free rider’ problem, a classic and relatively non-technical exposition is Olson, Mancur, The Logic of Collective Action: public goods and the theory of groups (Cambridge, Mass., 1965).Google Scholar

^{page 279 note 2} See Putterman, Louis, ‘On Optimality in Collective Institutional Choice’, in Journal of Comparative EconomicsGoogle Scholar, forthcoming, and ‘The Producer's Organizational Choice: theory and the case of the Tanzanian villages’, Ph.D. dissertation, Yale University, New Haven, 1980.Google Scholar

^{page 279 note 3} That is, if the median number of earned work-points is equal to the average number of work-points, so that the numbers of above- and below-average workers are identical.

^{page 279 note 4} See Maeda, Justin, ‘Popular Participation, Control and Development’, Ph.D. dissertation, Yale University, New Haven, 1976, p. 214; also Putterman, ‘The Producers’ Organizational Choice’.Google Scholar

^{page 280 note 1} Concrete hypotheses suggested by the more formal theoretical analysis to which the present exposition refers, were statistically and qualitatively tested on the basis of the author's research conducted in 11 villages of Arusha and Morogoro regions in 1979. The results of quantitative examination of interview data, and the informal findings from participant observation, seemed to generally confirm these hypotheses and the overall value of the approach. Interested readers are urged to see Putterman, ‘The Producers’ Organizational Choice'.

^{page 281 note 1} See Barker, Jonathan, ‘The Debate on Rural Socialism in Tanzania’, in Mwansasu, Bismarck and Pratt, Cranford (eds.), Towards Socialism in Tanzania (Toronto, 1979), pp. 95–124.Google Scholar

^{page 281 note 2} See Hyden, Goran, Beyond Ujamaa in Tanzania: underdevelopment and an uncaptured peasantry (London, Berkeley, and Los Angeles, 1980).Google Scholar

^{page 281 note 3} The argument of peasant conservatism based on the security of private property cannot be neglected, but it can be treated in a framework of long-term pursuit of objectives, including security, by a risk-averse population. Although change by its nature involves risks, collective development may offer a combination of security and growth which is attractive from the viewpoint of the peasants. Nyerere's concept of ujamaa is, in fact, aimed precisely at their desire for security, since the problem whith capitalistic development is that many peasants are likely to be left behind in the process, which includes loss of security due to the loosing of clan ties. Hyden's argument that neither ‘capitalism’ nor ‘socialism’ may lure the peasants from their own alternative economic mode should be assessed in terms of the prospects offered by each system, and not merely the assumed character of the peasantry.

^{page 283 note 1} Renè Dumont, University of Dar es Salaam, 1 August 1979.

^{page 283 note 2} The assumption that most villagers would leave the villages if free to do so is not necessarily justified, of course. There is reason to believe that many peasants have come to see real advantages in ‘living together’, as the Government puts it, so that the pattern of rural settlement in Tanzania may be seen to have been permanently affected by villagisation, whatever turns future policy takes.

^{page 284 note 1} Coulson, Andrew, ‘Agricultural Policies in Mainland Tanzania’, in Review of African Political Economy (London), 10, 1977, p. 95, cites David Leonard's prediction to this effect.Google Scholar

^{page 284 note 2} The legislation with regard to the villages already contains provisions which could facilitate such a collectivisation by giving the President and other officials power to issue direct orders to Village Councils. See Ofisi ya Waziri Mkuu, op. cit. pp. 31–4.

^{page 284 note 3} Power over village lands is already officially entrusted to the Village Councils. This power, in fact, extends to ‘all large productive property’ (‘zana zote kubwa za kuzalisha mali’). See ibid. pp. 44–5.

^{page 285 note 1} Freire, Paulo, Pedagogy of the Oppressed (Harmondsworth, 1972), p. 20.Google Scholar