Development Economics on Trial: The Anthropological Case for a Prosecution

Hill (1970: 6) lamented that

[t]he Beatrice Webb tradition of qualitative field observation and experiment has been out of fashion for some half a century, so it is no wonder that field enquiries were neglected in the British colonies. But even had this spirit of scientific enquiry not fallen into desuetude, it is likely that conventional British assumptions about West African economic behaviour would have inhibited field investigation.

She ridiculed at length three of these conventional British assumptions: that West Africans had learned the elementary facts of economic life from expatriate traders, that indigenous economies are too simple to be of interest to Western economists, and, notwithstanding the belief about simplicity, that indigenous economies are too complex to be of interest to economists: until modernization eliminates such complexities as local systems of land tenure, kinship, inheritance, and communal work, ‘indigenous economies operate on too small a scale, or on too local a basis, to be of interest to economists—and are anyway incomprehensible’ (ibid.: 7).

These beliefs were transmitted to African economists: ‘Convinced as they have been of the simplicity of indigenous economic organization, few of them [visiting economic advisers], or, indeed, of the growing band of prominent West African economists (most of them Nigerians), have used their influence to urge the need for more “grass roots research’” (ibid.: 9)—a remark made in a chapter (‘A Plea for Indigenous Economics’) first presented to a conference of the Nigerian Institute of Social and Economic Research in 1962. She proceeded, with equal tact, to rebuke (future Nobel Laureate) W. Arthur Lewis’s ‘much-read’ 1953 report on industrialization in the Gold Coast for ‘an over-simplified image of agricultural organization based on the presumption that a simple technology necessarily implies simple socio-economic organization, “inefficiency” (in some absolute, though undefined, sense), and small-scale production—though none of these notions is necessarily true or meaningful’ (Hill 1970: 9).

Hill (ibid.: 12-13) also criticized the ‘mystique about official statistics which withstands much battering’, particularly ‘occupational statistics, which appear so accurate because they are so detailed’, such as the 1960 Ghana population census, which failed to distinguish cocoa farmers from their labourers and which reported nearly all of the wholesalers of Ghanaian foodstuffs in southern Ghana to be men when most were women (but she praised others of the published reports on the 1960 census as ‘real landmarks of organization and scholarship’ (ibid.: 13)). Perhaps, she suggested, the census enumerators had counted as wholesalers ‘only those who handled imported goods or who had office premises’ (ibid.: fn. 2). Chapter 3 of Hill (1986a), the second of three chapters on ‘The Vain Search for Universal Generalizations’, was subtitled ‘The Poor Quality of Official Statistics’, and had an appendix on a 1981 World Bank report on accelerating development in sub-Saharan Africa, ‘a good example of a recent tendency to base weighty discussions on official statistics which are simultaneously admitted to be of doubtful quality’ (ibid.: 49). This critique of the quality of African official statistics has been pursued at greater length by Morten Jerven, who cites Hill (1986a) on the United Nations Food and Agriculture Organization’s use of inappropriate statistical categories and methods in developing countries (Jerven 2013: 77-78) but not her earlier work, which he would find congenial.

Hill (1986a: 21-24), arguing that ‘the grand development theories are far too generalized to be testable in the field’ (ibid.: 21), was sharply critical of a striking piece of long-term fieldwork in an Indian village by leading economic theorists Christopher Bliss and Nicholas Stern, who began believing that ‘theories of development economics should have the power to explain what one finds in a village such as Palanpur’ (Bliss and Stern 1982: 4) but concluded that ‘We are unable to confirm that neo-classical economics is alive and well and residing in Palanpur, at least in its simplest form’ (ibid.: 291; see also Lanjouw and Stern 1998). Hill (1986a: 22, fn. 16) remarked that ‘So far as I can tell, there are no publications by anthropologists in this book’s bibliography’. Such neglect was in keeping with such precedents as Frank Knight (1941) and Richard Posner (1980), who viewed any interaction between economists and anthropologists as almost exclusively a matter of economists lecturing anthropologists on basic microeconomic theory (cf. Herskovits 1941, a reply to Knight 1941). However, the Palanpur project and reactions to it (including Hill 1986a) ultimately stimulated more balanced and fruitful conversations between anthropologists and development economists (see Bardhan 1989; Bardhan and Ray 2008).

Hill (1970: 15, cf. 1986a) objected that (non-indigenous) economists ‘are apt to be so much less sophisticated than their subjects, the farmers’, insisting that ‘the Marxist textbooks are no more relevant than the capitalist variety’. As befitted a student of Ghana’s migrant cocoa farmers, Hill rejected the myth that shifting cultivation was ‘a ramshackle, wasteful, obsolete state of affairs’ (Hill 1970: 15), insisting that on the contrary, ‘systems of recurrent cultivation, if understood, could often be represented as respectable, even scientific—they usually, of course, involve crop rotation’ (ibid.). Nor did she accept that it was inefficient for West African farmers to grow food for their own consumption: ‘[C]ocoyam and plantain are necessary cover crops for young cocoa seedlings, so that cocoa-farmers with new farms cannot avoid producing these foodstuffs’ (ibid.: 16) and, more generally, ‘labour, rather than land, may be the scarce factor, so that to get the maximum output farmers must distribute the available manpower between crops with different growing cycles’ (ibid.) (see also Hill 1992a). Celebrating ‘those enterprising and wily migrant cocoa farmers of southern Ghana’ (Hill 1986a: 12; see also Hill 1957-1960, 1963, 1966, 1970), and later the Ewe Seine fishermen and shallot farmers (Hill 1986b), Hill always stressed the importance of local knowledge and upheld the rational agency of people in developing countries against the presumptuous ignorance of outside ‘experts’. As Hart (2005) wrote in his obituary of Hill, ‘It would be hard to exaggerate the contrast between her Ghanaian findings and the conventional thinking of development economists and administrators at the time and since. Her work has barely been absorbed into mainstream anthropology because it contradicts deep-seated convictions about western economic leadership and African backwardness’.

Her critical attitude towards the prescriptions of mainstream neoclassical economics was matched by an equally critical distrust of socialist panaceas. Hill (1986a: 172, fn. 3) emphasized that

contrary to common prevailing belief, there is no evidence that communal arable farming (as distinct from communal grazing) has ever found favour in Africa or south India, and that ‘socialist communal farming’ has always failed wherever it has been imposed on African people—even, on [long-time Tanzanian President Julius] Nyerere’s own admission, in Tanzania.