¹ (New York: Alfred A. Knopf, 1962.)

² Economics for Development (Englewood Cliffs, N. J.: Prentice-Hall, 1963), p. 125Google Scholar.

³ Preface to Yudelman, Montague, Africans on the Land (Cambridge: Harvard University Press, 1964)CrossRefGoogle Scholar. See also Ngcobo's, S. review of Yudelman, Journal of Modern African Studies, Vol. II, No. 4 (1964).Google Scholar

⁴ Lamartine-Yates, P., Forty Years of Foreign Trade (London: Allen and Unwin, 1968), p. 238Google Scholar. Cf. also Harwitz, M., “Subsaharan Africa as a Growing Economic System,” in Herskovits, and Harwitz, , eds., Economic Transition in Africa (Evanston, Ill.: Northwestern University Press, 1964)Google Scholar.

⁵ Cf. Harrop, Sylvia, “The Economy of the West African Coast in the Sixteenth Century,” Economic Bulletin of Ghana, No. 3 (1964)Google Scholar; Talbot, P. A., The Peoples of Southern Nigeria (Oxford, Engl.: The University Press, 1962), Vol. I, passimGoogle Scholar; Jones, W. O., “Manioc: An Example of Innovation in African Economies,” Economic Development and Cultural Change, V (Jan. 1957), 99Google Scholar. The growth of production for African urban markets has also been substantial and, at least since 1945, very rapid; but less evidence for analysis is available. However, for data on large-scale domestic food farms in Ghana, see Bray, Frank F., Yam Farming in North Mampong-Ashanti (Legon: Department of Agriculture, University College of Ghana, 1958)Google Scholar.

⁶ See Furnivall, J. S., Colonial Policy and Practice (Cambridge, Engl.: The University Press, 1948), pp. 323–34Google Scholar, for a discussion of parallel policies. Malaya aimed at the transference of “peasant” production from the more profitable rubber to the less profitable rice. Furnivall's material also shows the significantly greater and more effective attention paid to European plantation agriculture as opposed to local farmer production, as does also Bauer's, P. T.The Rubber Industry: A Study in Competition and Monopoly (Cambridge: Harvard University Press, 1948).Google Scholar A detailed study of Ugandan farmers by D. G. Belshaw and T. Othieno illustrates that the Agricultural Department's proposals were often rejected because either the technical conditions for applying them or the working capital for meeting the extra cash production costs they involved were not present. “Technical Innovation in Two Systems of Peasant Agriculture in Bukedi District” (Kampala: East African Institute of Social Research, Conference Papers, 1965).Google Scholar Study of response to agricultural proposals in a Tanzanian district shows a clear pattern of acceptance for those proposals (and only those proposals) which were seen to offer substantial short- or medium-run benefits. It also shows a very different set of reactions to attempts to explain proposals as against attempts to enforce change without demonstrating its value. J. Kesby, “Warangi Reaction to Agricultural Change,” in East African Institute Conference Papers, cited above.

⁷ For example, Ghana cocoa acreage estimates—a by-product of swollen-shoot disease control surveys—give an average yield per mature acre of 250 pounds. However, sample data and agricultural service estimates range from 400 to 800 pounds per acre. In short, either almost all the micro cost-return studies or the acreage estimate (or both) must be highly unreliable. Nonetheless, the current Ghana Agricultural Census will not cover cocoa, because the official view is that existing data on acreage, yields per acre (average and range), and age distribution are adequate.

⁸ Colonial Report 1903 (Accra: Gold Coast Government). The same report estimates acreage at 44,000, which is too low by a multiple of between five and ten. In 1911 exports were 39,700 tons, implying at least 200,000 acres in full bearing (8 years old and over), assuming 400 lbs. per acre, and nearly 400,000 acres 3 official estimates of 200 lbs. per bearing acre were correct. This is a dramatic illustration of how little was known.

⁹ (11th ed.; London, 1914), p. 204.

¹⁰ A. W. Hill, letter to Under-Secretary of State, Sept. 10, 1915, cited in Gold Coast Sessional Paper, II, 1916–1917 (Accra, 1916), p. 62. This letter was placed in evidence by the Colonial Agriculture Department of the Gold Coast, indicating its support of this point of view.

¹¹ Hill, Polly, Migrant Cocoa Farmers of Southern Ghana (Cambridge, Engl.: The University Press, 1964), especially pp. 161–78Google Scholar, from which many of the data in the following paragraphs are drawn. Her account, based on extensive investigation of available sources, is an excellent reconstruction of the process of growth of this industry.

¹² The Basel Mission had begun efforts to introduce cash crops for its Christian villages by 1843 with cocoa included at least as early as 1857, but in the next two decades its efforts appear to have had slight success.

¹³ Cf. West African Lands Committee, Draft Report (London: His Majesty's Stationery Office, Africa [West], No. 1,046, 1917), p. 96Google Scholar. See Hill, p. 172, for a summary and evaluation of the evidence on Tetteh Quarshie.

¹⁴ Cf. ibid, for fuller details on the establishment of palm-oil farms and the beginning of the land purchase and migrant company systems later used by cocoa migrants. Cf. the Report on the Population Census, 1891 (Accra, 1891), p. 24, for a discussion of the outflow of skilled craftsmen.

¹⁵ Cocoa exports reached 5,000 tons in 1904, indicating that at least 25,000 and perhaps 50,000 acres had been planted by 1896.

¹⁶ Oral reports collected by Ivor Wilks of the Institute of African Studies, University of Ghana, cited by Polly Hill, p. 167.

¹⁷ The proportion of Christians among early farmers appears to have been substantially above that for the Akwapim as a whole. The direction of causation is not clear, however, in that Christianity was related to desire for access to education, health, and other European amenities—so that the Christian community was probably both more eager to advance economically and more willing to seek out new activities than the majority of the population.

¹⁸ Various dispatches of Governor Bradford Griffith (1888–1892), Report of the Commission on Economic Agriculture in the Gold Coast (1889), cited by Hill pp. 173–76.

¹⁹ Dispatch dated Nov. 9, 1891, cited ibid., p. 176.

²⁰ Estimated from Gold Coast Department of Agriculture Reports and Gold Coast Colonial Reports; cf. also Beckett, W. H., Koransang: A Gold Coast Cocoa Farm (Accra: Government Printer, 1945)Google Scholar, introduction.

In a letter, Beckett gives the 1900–1920 distribution of seedlings as “hundreds of thousands.” From 750 to 1,000 seedlings are needed to plant an acre; this would suggest that by 1920 not more than 10,000 to 15,000 acres of at least 1,000,000 could have been planted from government stock. The recurrent citation of seedling and seed totals, in themselves impressive, gives no indication of their relatively minor share in total planting.

²¹ For a more detailed description of this factor, see Hancock, W. K., “West Africa: The Traders' Frontier,” in Survey of British Commonwealth Affairs, Vol. II, Part II (Oxford, Engl.: The University Press, 1940)Google Scholar.

²² The same phenomenon has been studied in greater detail in relation to a similar switch of tea from a limited-market luxury good to a mass-consumption item two decades earlier; cf. S. Rajaratnam, “The Ceylon Tea Industry, 1886–1931,” Ceylon Journal of Historical and Social Studies (July-Dec. 1961).

²³ For example, the twelfth edition of the Encyclopaedia Britannica in revision of its previous description stated: “The cocoa industry has throughout been entirely a native enterprise, Europeans acting only as carriers, purchasers and shippers” (London, 1919), p. 296. The article was written by Sir Hugh Clifford, a former Gold Coast Governor who had opposed the Department of Agriculture's proposals for limiting industry expansion and enforcing changes in planting methods which are described later.

²⁴ (Accra: Gold Coast Government, 1936.)

²⁵ Cf., for example, Shephard, Peasant Agriculture, and Tudhope, W. S. D., “Enquiry into the Gold Coast Cocoa Industry,” Interim and Final Report, Gold Coast Sessional Papers II and IV of 1918–19 (Accra: Gold Coast Government)Google Scholar; Cocoa (Paris: OEEC, 1956), p. 20; Cacao: A Review of Current Trends in Production, Prices, and Consumption (Rome: FAO, 1955), pp. 16–21Google Scholar.

²⁶ Akokoaso: A Survey of a Gold Coast Village (London School of Economics, Monographs on Social Anthropology, No. 10; London: L.S.E. Press, 1944)Google Scholar. In fact, Beckett's second major study, Koransang: A Gold Coast Cocoa Farm (Accra: Government Printer, 1945)Google Scholar, does analyze a large migrant farm. However, the image of Akokoaso, not Koransang, has predominated.

²⁷ Personal communication from and discussion with Polly Hill; also Bray, F. R., Cocoa Development in Ahafo (West Ashanti: Faculty of Agriculture, University College of Ghana, 1959)Google Scholar. Expansion in the western and central regions has also involved substantial migration by eastern farmers—again a continuation of 1920's patterns made possible in this case by capsid control. Further confirmation of the effort and initiative required in cocoa production is found in an unpublished survey by J. H. Mensah and in a comprehensive study for western Nigeria: Dina, O., Galletti, R., and Baldwin, K., Nigerian Cocoa Farmers (Oxford, Engl.: The University Press, 1956)Google Scholar.

²⁸ In 1954, their errors were “authoritatively endorsed” by the Food and Agricultural Organization which, basing its study on colonial reports, produced a summary of the West African industry rehearsing the old theme of the lazy, noneconomically motivated, small-scale peasant cultivator (Cocoa: A Review of Current Trends). FAO's manner of providing “expert advice” on cocoa deserves mention. The organization has never conducted field research on the production side of the industry. Rather it has sent queries to agriculture departments. These have usually been filled in at the central office on the basis of prevailing guesses or impressions; the results are then printed as an “authoritative” study which is cited by the same agriculture departments in support of their opinions. FAO has recently become aware of the weakness of this process, perhaps because the early postwar reports led to prediction of a massive cocoa shortage by 1960–1965, by which point, on the contrary, a phenomenal increase in output and cocoa glut has resulted from massive 1948–1958 plantings in response to high prices and from farmers' purchases of sprays and sprayers to control diseases.

²⁹ Polly Hill, p. 174, citing Despatch 28 (Aug. 1888), ADM 1/489.

³⁰ “Letter to the Editor,” The Times (London), Feb. 25, 1929, quoting diary of Oct. 1, 1890.

³¹ “Report of the Committee on Agricultural Policy and Organization,” Sessional Paper XVII of 1927–1928, Part II-D. That the chairman of the Committee was the Deputy Director of Agriculture, G. G. Auchinleck, suggests that its very sharp criticism of the 1890–1927 record was shared by at least some senior Agriculture Department staff.

³² The Department staff in 1905 consisted of three curators, three African overseers, and two African garden assistants; even in 1915 it numbered only one administrator, seventeen garden staff (seven of these European), one inspector, thirteen African field instructors, and two technically trained agricultural research and technical officers. By 1922 the technical officers had risen to six (ibid, and Annual Report of 1905, 1915, 1920, 1921, and 1922).

Examination of the Gold Coast Civil Service Lists for the 1920's and 1930's further reveals that a distinct minority of the officers had any West Indian, or indeed any tropical, experience before being posted to West Africa. In 1922, for example, one had West Indian experience and one had served briefly in Ceylon. The majority came direct from British agricultural or technical colleges.

³³ Ibid.

³⁴ Agriculture Department, Annual Report, 1919 (Accra: Gold Coast Government, 1921).

³⁵ Ibid., 1920, p. 8. The question of allocation is important because the persistent concentration on non-cocoa (and in most cases, noneconomic) crops renders rather unconvincing the defense frequently advanced that the Department sought additional funds for cocoa research. Quite clearly its pre-1937 cocoa research program could have been increased severalfold had not sisal, palm oil, coconut, cotton, and mechanized rice cultivation been viewed as having higher priority than cocoa work.

³⁶ Sessional Taper XVII of 1927–28.

³⁷ Sessional Paper II of 1916–17; and II and IV of 1918–19. The position of the Department was dominated by the views of Director W. S. D. Tudhope, who prepared the Sessional. He wrote at one point, “I live in constant dread of disaster overtaking the industry”(ibid., II of 1916–17).

³⁸ Cf. annual reports of the Department of Agriculture.

³⁹ These were not the only issues raised: the Department was concerned with the problem of rural indebtedness. This problem will not be discussed here in detail, but the overall conclusions would not be affected if it were, for here as else-where the Department's evaluation was in large part erroneous. Unlike Asia, which probably served the Department as a model, the African cocoa farmers were not ground down by usurers. As subsequent investigations have shown, there is a high correlation between total output and debt, indicating that it is the richer farmers who are the most heavily indebted and that the overall ratio of debt to assets is not high. There are some cases where the value of debt is equal to the value of the farm, but there is a good deal of evidence to suggest that the loans in this case are often a device for disguised sales and that in most cases the debtor has other farms for income. Cf. Polly Hill, pp. 186–87; and idem, The Gold Coast Cocoa Farmer (Oxford, Engl.: The University Press, 1955)Google Scholar. Report of the Committee on Agricultural Indebtedness (Accra: Department of Agriculture, 1958)Google Scholar includes an historical summary of past reports and proposals.

⁴⁰ Sessional Paper XI of 1916–17.

⁴¹ Cf. Cardinall, A. W., The Cold Coast (Accra: Government Printer, 1931)Google Scholar; Hill, Migrant Cocoa Farmers and The Gold Coast Cocoa Farmer; Dina, Galletti, and Baldwin.

⁴² When an expert evaluation of productivity was made in 1936–37, the author, C. Y. Shephard, found: “The brief description of methods of cocoa cultivation in the Gold Coast shows that the farmer has adopted few of those expensive operations which planters in Trinidad and Grenada consider necessary for the maintenance of yields and profits … [but] it is a fairly safe assumption that the yield in the Gold Coast is at least twice that obtained in Trinidad and equivalent to that obtained by intensive methods in Grenada” (Peasant Agriculture, p. 16).

⁴³ Ibid., pp. 19–20.

⁴⁴ Cf. Dina, Galletti, Baldwin; Urquhart, D. W., Cocoa (London: Longmans Green, 1961)Google Scholar.

⁴⁵ Sessional Paper II of 1916–17.

⁴⁶ Cf. Urquhart; also Entwhistel, P. F., “Minor Insect Pests,” in Wills, J. D. (ed.), Agriculture and Land Use in Ghana (Oxford, Engl.: The University Press, 1962)Google Scholar.

⁴⁷ W. H. Beckett, in his comments, cites sankonuabe (Twi for “go back to oil palms”) as having been controlled by Department action. Department reports suggest that the attacks were first overestimated and then receded. In fact these were largely early capsid outbreaks. The level of loss remained high, although never adequately estimated, until control was developed and instituted—in the 1950's and 1960's, not in the 1920's or 1930's.

⁴⁸ Hammond, “Cocoa Agronomy,” in Wills (ed.) estimates loss of farms (seedlings killed) from capsid as perhaps 50 per cent prior to control in the late 1950's—an excellent reason for scattering new farms. Before 1937, little research was done on this pest, and control measures date to 1954. Its importance appears to have been gravely underestimated. Beckett, for example, relates the low growth and indeed the decline of cocoa in the Western Region to lack of transport. It now appears from interviews with farmers by Polly Hill that the basic reason was heavy capsid infestation killing new and sometimes even established cocoa. Certainly, since capsid control was achieved the planting and output of the Western Region have risen rapidly.

⁴⁹ There exists an important misunderstanding of the role of plantations in West Africa. They were forbidden in Nigeria; in the Gold Coast they were encouraged, but failed.

⁵⁰ Annual reports of 1922 through 1934, passim; also see Report by the Hon. W. G. A. Ormsby-Gore, M. P.… on His Visit to West Africa During the Year 1926 (London: 1926, Cmd. 2744)Google Scholar; and Hancock, William Keith, Survey of British Commonwealth Affairs, Vol. II, Part 2 (London: Oxford University Press, 1940), pp. 188, 192.Google Scholar

Unsatisfactory plantation experience has been the rule, not the exception, in West African cocoa and in European Ivory Coast and Cameroon plantations. Despite provision of conscripted labor, almost all failed in the 1930's; but large as well as small African farms survived.

⁵¹ Discussion with WACRI officers from both Ibadan and New Tafo stations.

⁵² Cf. Economic Survey of Ghana (Accra: Government Printer, annual) for 1955 and successive years.

⁵³ Sampson, H. C. and Crowther, E. M., “Report on Crop Production and Soil Fertility Problems,” in the West African Commission 1938–39 Technical Reports (London: Leverhulm Trust, 1943), pp. 39–40Google Scholar.

⁵⁴ Hill, Migrant Cocoa Farmers, p. 179. The theme is reiterated and amply documented in her many writings on the subject. From her analysis, it emerges that “agricultural capitalist” rather than “peasant” provides the more appropriate description of the Gold Coast farmer.

⁵⁵ See G. Dudgeon, C., The Agricultural and Forest Products of British West Africa (Empirical Institute Handbook; London: John Murray, 1911), p. 50Google Scholar and following.

⁵⁶ Botanical and Agricultural Department, Annual Report, 1907. No effort was made to estimate the costs of the special preparation. The Department merely noted that they received a London price of £67 7s. per ton for their few tons of highquality cocoa as opposed to the “prevailing” price of £65 per ton for Gold Coast cocoa. As they sold in five lots at prices ranging from £65 to £70 per ton in a rising market, the existence of any differential seems hypothetical.

⁵⁷ Agricultural Department, Annual Report, 1908.

⁵⁸ One can see the persistence of the fallacy as late as 1930–31 in “Memorandum on the Creation of a Fund for Improving the Quality and Marketing of Cocoa,” Sessional Paper XVIII of 1930–31Google Scholar, paragraph 19.

⁵⁹ See Urquhart, ch. x; Hammond, “Cocoa: Agronomy,” in Wills (ed.).

⁶⁰ Sessional Paper II of 1916–17.

⁶¹ Ibid, and Gold Coast Government, Legislative Council Debates, 1927–28, p. 51.

⁶² Colonial Office, Report of the Commission on the Marketing of West African Cocoa (London, 1938 Cmd. 5845).Google Scholar

See also Appeadu, K. K., Notes on the History of the Gold Coast Cooperative Movement (Accra: Department of Agriculture, 1956)Google Scholar, who shows that before the Second World War cooperatives failed to achieve a volume of 5 per cent of the crop, whereas after the war they rapidly rose to one third when the emphasis was changed to making profits on buying commissions instead of losses on quality. See also Shephard.

⁶³ Report … on the Marketing of West African Cocoa.

⁶⁴ Similar pleas, reminiscent of the lost battle of England to save the Corn Laws, were endemic in a number of colonies. Ceylon and Malaya in particular sought to reduce rice imports (from another British colony, Burma), sometimes even at the expense of higher income-yielding smallholder export crops.

⁶⁵ Ironically, the Annual Plan, 1965, p. 15, cites the falling off in cocoa planting (and the consequent reduction in interplanted-food output) as a source of pressure on the food supply.

⁶⁶ Computed from Survey of Population and Budgets of Cocoa Producing Families in the Oda-Swedru-Asamankese Area, 1955–56, and from Survey of Cocoa Producing Families in Ashanti, 1956–57 (Statistical and Economic Papers nos. 6 and 7, respectively; Accra: Government Printer, 1960).

⁶⁷ A point emphasized by Polly Hill in discussion with the authors.

⁶⁸ At a forest-zone agricultural station visited by one of the authors in 1961, the chief crop was indeed a root crop—Irish potatoes. For two decades these had been grown with poor size, low yield, and constant need for new planting materials, and at a probable cost of £ 1 to £ 2 per pound, with the only apparent gain being to the local senior officers' diet.

⁶⁹ Gold Coast, Legislative Council Debates, 1927–28, p. 276.

⁷⁰ Sessional Paper IV of 1918–19, p. 20.

⁷¹ Cf. Bauer (cited in n. 6); Silcock, T., The Economy of Malaya (Singapore: D. Moore, 1956)Google Scholar.

⁷² Legislative Council Debates, 1919–20, p. 7.

⁷³ This is not to the discredit of the Gold Coast colonial staff as such. Imperial economic organization did not envisage such changes in West Africa. Under Governor Guggisberg, the Gold Coast did have an economic plan of some sophistication. However, its basic dynamics were increases in trade leading to increases in government revenue expended on public works to promote further export expansion and on health and education to raise the level of African job potential and welfare. Cf. Governor Guggisberg's annual addresses to the Legislative Council, in Legislative Council Debates, 1919–20 to 1927–28. The Volta River dam was considered but dropped, largely because no use for the volume of power necessary to make it economic could be foreseen.

⁷⁴ See Lamartine-Yates, passim, on price levels and trends for various crops. Even with the post-1960 price collapse, Ghana's cocoa and cocoa-product export earnings have risen steadily from £67.8 million in 1960 to £76–78 million in 1965, while total export earnings have stagnated. The 1960–1965 cocoa series is 67.8, 70.7, 70.8, 71.7, 72.8, 76–78; Economic Surveys, 1961, 1964.

A far more sensible solution to the problem of primary-product stabilization would have been cooperation on a sterling-area basis in some type of insurance scheme. This would have required coordination from London, and though it would seem that the empire would be an ideal basis for planning an international division of labor, it was never seriously considered. Instead, the colonial governments encouraged coffee and rubber in the Gold Coast, cocoa and oil palms in Malaya and Ceylon, and so on for the other colonies.

⁷⁵ In addition to the annual reports, see, for example, “Report on Communal Coconut Plantations,” Sessional Paper X of 1921–22Google Scholar; “Correspondence Relating to the Development of the Oil Palm Industry,” Sessional Paper IV of 1923–25Google Scholar; Despatch No. 665, “Relating to the Oil Palm Industry,” 1929.

⁷⁶ Relevant here are all papers cited in footnote 75. Also see the annual reports of 1921 and following years, and legislative council debates of 1924 and following years—especially 1927–28, pp. 46–60. The 1932–33 Annual Report summed up the plantation record fairly when it commented: “There have been but a few plantation ventures in the Gold Coast and they have in general not been sufficiently successful to encourage further development. The fall in the price of raw material products has now rendered such propositions unattractive.”

⁷⁷ “Agricultural Commodity Projections, Real Growth and Gains from Trade,” in Stewart, I. G. and Ord, H. W., eds., International Seminar on African Primary Products and International Trade (Edinburgh: The University Press, 1965), p. 111.Google Scholar

⁷⁸ The authors are personally familiar with parallel cases in Nigeria, Uganda, the Cameroon Republic, Ceylon, Burma, and Malaysia. In general, independent states have been more sympathetic to farmers but not necessarily better informed on rural institutions or on techniques. They have frequently embarked on large-scale institutional and technical changes borrowed from temperate countries with neither substantial local (tropical) testing nor serious cost studies.

⁷⁹ The result of such a policy is likely to be the creation of skepticism and caution on the part of fanners. If the past record of agricultural “expert” proposals is poor, and especially if the farmer in question has suffered by introducing one, the maxim “once burnt, twice shy” will tend to govern. This, however, is a criticism of attempted generalization of unproven innovations, not of farmer attitudes.