^{page 595 note 1} Lobamba National Archives, RCS 798/36, 7–13 November 1942.

^{page 595 note 2} Interviews were conducted by Nikiwe Mbatha, Robert Dlamini, Vincent Sithole, and Makhosazana Ntshingila of the Rural Development Research Project, University of Swaziland, 1983.

^{page 597 note 1} Cf. Fion de Vletter, ‘Migrant Labour in Swaziland: characteristics, attitudes and policy implications’, World Employment Programme Working Paper, 2–26/WP22, Geneva, 1978; ‘A Socio-Economic Profile of Swazi Rural Homesteads’, in Vletter, de et al. , The Swazi Rural Homestead(Social Science Research Unit, University of Swaziland, Kwaluseni, 1983);Google Scholar and ‘The Swazi Rural Homestead: a case study of subsistence, wage dependency and misguided rural development’, Carnegie Conference Paper No. 285, University of Cape Town, 1984.

^{page 597 note 2} According to Sylvie K. Kamalkhani, ‘A Household Subsistence Level and Budget Survey in Manzini, July 1983’, University of Swaziland, 1984, p. 38, ‘68% of all respondents belonged to an umuti in the rural areas’; two-thirds of these visited their rural homesteads at least once and up to five times a month. In a random sample of 377 urban households in Mbabane in 1983, Jan Testerink and Margo Russell found that 46 per cent of the owners and 66 per cent of the tenants were also members of rural homesteads, and that three-quarters of them returned there at least once a month; ‘Nkwalini-Mahwalala, Mbabane: socio-economic characteristics and housing preferences’, Social Science Research Unit Paper No.9, University of Swaziland, Kwaluseni, 1983, p. 43.

^{page 597 note 3} Fion de Vletter, ‘Subsistence Farmer, Cash Cropper or Consumer? A Socio-Economic Profile of a Sample of Swazi Rural Homesteads. Programmes for Better Family Living’, Department of Research and Planning, Ministry of Agriculture, Mbabane, 1979. See also R.J. Szal and R. van der Hoeven, ‘Inequality and Basic Needs in Swaziland’, Income Distribution and Employment Programme, International Labour Organisation, Geneva, 1976.

^{page 597 note 4} de Vletter, ‘Migrant Labour in Swaziland’, pp. 25–46.

^{page 597 note 5} My own early pilot survey of Swazi workers attempted unsuccessfully to tap remittance behaviour with two questions: one about contribution to annual agricultural production, and a second conventional question on monthly support for those at ‘home’. I now recognise the results as incomplete; Margo Russell, ‘Boundaries and Structures in the Swazi Rural Homestead’, Social Science Research Unit, Research Paper No. 6, University of Swaziland, Kwaluseni, 1983.

^{page 598 note 1} Cf. M. V. Gandar and N. Bromberger, ‘Economic and Demographic Functioning of Rural Households: Malhabatini District, KwaZulu’, University of Cape Town, 1984, Carnegie Conference Paper, No. 56, p. 12: ‘in an answer to a question on the reliability of remittances many respondents said they could not rely on regular payments, but the question seems to have been misunderstood’ (my emphasis).

^{page 598 note 2} de Vletter, ‘A Socio-Economic Profile of Swazi Rural Homesteads’, p. 46.

^{page 598 note 3} de Vletter, ‘Migrant Labour in Swaziland’, p. 43.

^{page 599 note 1} de Vletter, ‘Subsistence Farmer, Cash Cropper or Consumer?’, p. 51. A ‘guesstimate’ ofE100 worth of goods received per annum was included by de Vletter, in loc. cit. 1983, p. 50, in his estimate of average homestead incomes.

^{page 599 note 2} de Vletter, ‘A Socio-Economic Profile of Swazi Rural Homesteads’, pp. 18, 21, and 49.

^{page 599 note 3} This was not a percentage calculated by de Vletter, albeit derived from his data (ibid. Tables 2 and 14, pp. 18 and 49) as follows: (1) For those homesteads receiving remittances in 1978, the average was E40.2 from two workers, namely E20.1 from each. (2) On the assumption that the wages (unrecorded) of those absent were the same as for those who were not, and since the mean number of the latter was 1.4 while their mean monthly contribution was E117, then the average monthly wage in 1978 was E83.86.

^{page 599 note 4} Ibid. p. 46.

^{page 599 note 5} Ibid. p. 57.

^{page 599 note 6} Gandar and Bromberger, op. cit. p. 24.

^{page 599 note 7} The similarities should not be pressed too far. There are almost no local opportunities of earning wages in the KwaZulu District. Remittances are the major source of income. Household incomes seem to be about half those of Swaziland. In 1981, the monthly income per capita in KwaZulu was R14.8 (ibid. p. 8), compared to R26.6 in Swaziland (de Vletter, ‘A Socio-Economic Profile of Swazi Rural Homesteads’, p. 46), adjusted for inflation since 1978 at 10 per cent per annum.

^{page 600 note 1} They use ‘household’ for what in Swaziland is termed ‘homestead’.

^{page 600 note 2} Gandar and Bromberger, op. cit. p. 26, note 9.

^{page 600 note 3} Medical expenses were traditionally very heavy. Paid in livestock, they were borne collectively by the homestead herd, implicit acknowledgement of shared responsibility for, and vulnerability to a social ill. In 1976, it was suggested that average medical expenses in Swaziland per household be estimated at E7.25 per person for visiting a traditional doctor, and 25 cents for visiting a modern medical facility; Szal and van der Hoeven, op. cit. p. 39. Modern medical facilities are, of course, very expensive, but subsidised by the state and/or Christian missions.

^{page 600 note 4} Gandar and Bromberger, op. cit. p. 22.

^{page 600 note 5} Ibid. p. 12.

^{page 601 note 1} Ibid. p. 26, note 9.

^{page 601 note 2} If the essence of proletarianisation is total dependence on wages, usually through landlessness, then there are degrees which correspond to the strength of alternative economic resources available to the worker–amongst which in Africa land would be conspicuous. I suggest that the worker with a productive economic rural base, who is consequently able to accumulate wages as savings for investment, is less intensively proletarianised than one whose wages are consumed in subsistence. By these standards, KwaZulu is probably more proletarianised than Swaziland.

^{page 601 note 3} Ruth First, ‘The Mozambican Miner: a study in the export of labour’, Centro de Estudos Africanos, Eduardo Mondlane University, Maputo, 1977, mimeographed, p. 128, citing evidence from the Homoine District.

^{page 601 note 4} Ibid. p. 130.

^{page 601 note 5} Beth Rosen-Prinz and F. Prinz, ‘Migrant Labour and Rural Homesteads’, Working Paper WEP2/26 WP31, World Employment Programme, International Labour Organisation, Geneva, 1978.

^{page 604 note 1} Swaziland Government, Wages and Employment, 1982 (Mbabane, 1983).Google Scholar

^{page 604 note 2} de Vletter, ‘A Socio-Economic Profile of Swazi Rural Homesteads’, p. 46; Rosen-Prinz and Prinz, op. cit. p. 38; Margo Russell and Makhosazana Ntshingila, ‘Sources of Swazi Rural Homestead Income’, Social Science Research Unit Paper No. 13, University of Swaziland, Kwaluseni, 1984; and Margo Russell, ‘The Production and Marketing of Swazi Rural Women's Handicrafts: a report commissioned by the Rural Employment Policies Branch of the International Labour Organisation’, presented to the I.L.O. African and Asian Interregional Workshop on Strategies for Improving Employment Conditions of Rural Women, held in Arusha, August 1984.

^{page 605 note 1} Margo Russell, Nikiwe Mbatha, and Vincent Sithole, ‘A Sample Survey of Maize-Growing in Swaziland’, Social Science Research Unit Paper No. 1, University of Swaziland, Kwaluseni, October 1982, and de Vletter, ‘A Socio-Economic Profile of Swazi Rural Homesteads’.

^{page 606 note 1} Marlene Capetta, ‘Population, Food and Nutrition, Swaziland, 1940–1982’, in de Vletter et al. op. cit. Xolile P. Guma, ‘Cash Income and Expenditure’, in ibid. p. 141, Table 16, lists the mean monthly food expenditures for a sample of 87 rural Swazi homesteads during June to August 1978; 80 per cent or more bought sweets, cold drinks, jam, sugar, biscuits, bread, and flour each week. See also de Vletter, ‘Subsistence Farmer, Cash Cropper or Consumer?’, p. 60.

^{page 606 note 2} Allen, Carol J., ‘Dimensions of Swazi Households in Rural and Urban Areas’, Ph.D. dissertation, University of Massachusetts, Cambridge, 1974, pp. 343–7. She also found that 12 per cent of the children in her sample of 100 households had been tsenga'd by their fathers.Google Scholar

^{page 607 note 1} Nikiwe Mbatha, ‘Ambiguities of Swazi Marriage’, Social Science Research Unit Paper No. 5, University of Swaziland, Kwaluseni, 1983.

^{page 607 note 2} Kuper, Hilda, An African Aristocracy: rank among the Swazi (London, 1947).Google Scholar

^{page 608 note 1} de Vietter, ‘Migrant Labour in Swaziland’, found 41.6 per cent of homesteads sold crops, but in such small amounts that they contributed only 8.7 per cent of annual cash income. Five years later, Roland Freund and Basil Maphalala, ‘Economic Circumstances ofSwazi Nation Land Homesteads, 1982/1983’, Swaziland Cropping Systems Research and Extension Training Projects, Malkerns, 1984, found a very similar incidence of cash crops in four development areas. See also Guma, loc. Cit. pp. 130–1.

^{page 609 note 1} According to Marwick, Brian A., The Swazi (London, 1966), p. 49:Google Scholar ‘If there is no heir in a house upon the death of a man, his brothers will take over (ngeng) the wife of that house in order to produce an heir…Even when a man dies unmarried the brothers will conceive it to be a duty to lobola one of his lovers and raise a family to the deceased.’

^{page 609 note 2} Kasanene, Peter, ‘Religion and the Aged’, Proceedings of a Conference on Ageing,University of Swaziland,Kwaluseni,1983.Google Scholar

^{page 610 note 1} Philip Myer, ‘Legal Aspects of the Aged’, in ibid.

^{page 611 note 1} A fuller analysis of women in work is being made by Makhosazana Ntshingila, University of Swaziland, whose fieldwork transcriptions I am grateful to have shared. See her ‘Rural Women in Wage Employment: a report on a small sample in Central Swaziland’, Kwaluseni, 1984.Google Scholar

^{page 612 note 1} Gandar and Bromberger, op. cit. p. 23.

^{page 612 note 2} Understanding and treatment are separate actions. The pressing problem is to convert understanding into effective procedures for data collection. Western models are not particularly helpful.

^{page 612 note 3} Allen found, op. cit. pp. 261–3 and 400, that the male head was the sole source of monetary support in only a quarter of her sample of rural homesteads, and that he was one support amongst others – wives, children, brothers, sisters, daughter's husbands, and ‘wives’ relatives and distant relatives’ – in half of these homesteads. In 4 cases, money came from non-residents who were not members of the immediate family or domestic group. Similarly, Kamalkhani records, op. cit. pp. 35 and 36, that in one month 12 of her sample of 100 households received an average of E20 as ‘money from relative other than rural umuti [homestead] or absentees’, 14 received cash or goods from rural umuti, and 10 received cash from absent members.

^{page 612 note 4} According to Allen, researchers realise that homesteads are extended in unusual ways but know who to count as dependants.

^{page 613 note 1} For a discussion of some of the problems of deciding what those assumptions should be for calculation of income per capita of homesteads, see Social Science Research Unit, Research Paper No. 12, University of Swaziland, Kwaluseni, 1984, pp. 21–7.

^{page 613 note 2} Kamalkhani, op. cit. p. 5.

^{page 614 note 1} For a debate on the appropriate unit for analysis of Swazi data, see A. Black-Michaud, ‘Homesteads versus Households’, Malkerns Research Station, Ministry of Agriculture, 1981; Russell, ‘Boundaries and Structures in the Swazi Rural Homestead’; and Funekile Simelane, ‘Problems of Data Collection in Rural Swaziland’, Workshop on Sociological Contributions to Farming Systems Research, Lusaka, November 1984.

^{page 614 note 2} This ideology is expressed by a man who said, ‘In Swazi custom if a woman…brings anything of hers into the home of the husband, everything automatically becomes the property of her husband. We say so. But he cannot sell them out or give it to someone else without the consultation of the wife. It is like that also between the father and his sons because he is the boss of the homestead.’

^{page 615 note 1} See, for example, the Proceedings of the Second Southern African Universities Social Science Conference,Dar es Salaam,1979, Vol. III.Google Scholar

^{page 615 note 2} de Vietter, ‘Subsistence Farmer, Cash Cropper or Consumer?’.

^{page 615 note 3} The elective affinity of the intelligentsia for ‘the bad news’, a phenomenon of our time, might be understood as an attempt to assuage feelings of guilt about their personal privileges by public statements of their awareness that disparities exist, and by open denunciation of the system which patronises them.