Hostname: page-component-586b7cd67f-tf8b9 Total loading time: 0 Render date: 2024-11-27T20:39:24.292Z Has data issue: false hasContentIssue false
  • English
  • Français

The Relevance of Spatial Analysis for African Economic History: the Sierra Leone–Guinea System

Published online by Cambridge University Press:  22 January 2009

Allen M. Howard
Get access Rights & Permissions [Opens in a new window]

Extract

This article suggests ways in which a spatial perspective is relevant to questions of African economic history prior to formal colonization. Aspects of spatial analysis can help to identify and compare economically dynamic centres and subregions and to understand the qualities of regional and inter-regional systems. Spatial analysis can also illuminate various economic relationships between Africa and Europe; thus it is relevant to such issues as African adaptation to ‘legitimate’ trade and economic dependency. Among the main concepts discussed are central place theory, ‘growth centres’, port gateways, and dendritic marketing systems. The article focuses upon the Sierra Leone–Guinea plain and the larger Sierra Leone–Guinea inter-regional commercial system during the second half of the nineteenth century. Brief comparisons are made with other areas.

Type
Research Article
Information
The Journal of African History , Volume 17 , Issue 3 , July 1976 , pp. 365 - 388
Copyright
Copyright © Cambridge University Press 1976

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Research for the substantive sections of this article was made possible by a generous grant from the Comparative Tropical History Program, University of Wisconsin-Madison. Writing and additional research were carried out while I held a fellowship from the Program in African Economic History, University of Wisconsin-Madison. Earlier versions were read at the African Economic History Workshop, July 1974, sponsored by the latter program, and at the 89th Annual Meeting of the American Historical Association, Chicago, Dec. 1974. Philip D. Curtin, John Friedmann, and Eric E. Lampard greatly stimulated my interest in spatial analysis. I have benefited much from my association with them and with members of the Department of Geography, Fourah Bay College, University of Sierra Leone; Social Science Research Council Committee on Comparative Urbanization; and Department of Urban Planning and Policy Development, Rutgers University. Members of the African Economic History Workshop commented on earlier versions of this article, and detailed criticism was given by Sara Berry, Donald Krueckeberg, and Salah El-Shakhs. They, of course, bear no responsibility for my attempts to apply what I have learned from them.

2 J. A. Jakle has surveyed recent studies which apply geography to United States history, ‘Time, Space, and the Geographic Past’, Amer. Hist. Rev., lxxvi (1971), 1084–103. For an insightful comparison of United States and Latin American experiences, see Morse, R. M., ‘The Development of Urban Systems in the Americas in the Nineteenth Century’, J. of Inter-American Studies and World Affairs, xvii (1975), 4–26CrossRefGoogle Scholar. For widesweeping discussions of the spatial approach to history, see the works of E. E. Lampard, e.g., ‘Historical Aspects of Urbanization’, in Hauser, P. M. and Schnore, L. F., eds, The Study of Urbanization (New York, 1965), 519–54Google Scholar. For an early and influential study of regional development carried out by co-operating scholars in various disciplines, see Perloff, H. S. et al. , Regions, Resources, and Economic Growth (Baltimore, 1960).Google Scholar

3 Mabogunje, A. L., Urbanization in Nigeria (London, 1968)Google Scholar; Dickson, K. B., An Historical Geography of Ghana (Cambridge, 1969)Google Scholar. Only a few quantitative studies of spatial change in the colonial and national periods have seriously looked at pre-1900 patterns. See Soja, E., The Geography of Modernization in Kenya (Syracuse, N.Y., 1968).Google Scholar C. Berberich has recently completed a doctoral dissertation at Northwestern University which examines spatial developments in the Asante hinterland, prior to 1900. I have not yet seen this important work.

4 An excellent overview of the spatial approach is found in Kolars, J. F. and Nystuen, J. D., Geography: The Study of Location, Culture, and Environment (New York, 1974)Google Scholar. For more technical treatments of theory and methodology, see Berry, B. J. L. and Horton, F. E., Geographic Perspectives on Urban Systems (Englewood Cliffs, N.J., 1970)Google Scholar; and Hurst, M. E., A Geography of Economic Behaviour (North Scituate, Mass., 1972).Google Scholar

5 An Economic History of West Africa (New York, 1973), 124–66.

6 Rodney, W., How Europe Underdeveloped Africa (London & Dar es Salaam, 1972), 103–61 ffGoogle Scholar. Also, see the works of S. Amin: for example, L'Afrique de I'Ouest bloquée: l'économie politique de la colonisation, 1880–1970 (Paris, 1971). Many geographers and planners have been bound too tightly to the political datelines, beginning their studies at the onset of formal colonialism. Because of clear historical continuities, however, research on marketplaces often has cut across political time periods. See Hodder, B. W. and Ukwu, U. I., Markets in West Africa (Ibadan, 1969)Google Scholar; and various articles in Bohannan, P. and Dalton, G., eds., Markets in Africa (Evanston, Ill., 1962).Google Scholar

7 Howard, A. M., ‘Big Men, Traders, and Chiefs: Power, Commerce, and Spatial Change in the Sierra Leone-Guinea Plain, 1865–1895’ (Ph.D. dissertation, University of Wisconsin-Madison, 1972).Google Scholar

8 Clarke, J. I., ed., Sierra Leone in Maps (London, 1966), 12–27Google Scholar. The coastal littoral was, in a sense, another region, but can be treated as a subregion of the plain. Within the plain, there was significant variation in quality of soil, nature of vegetation, access to rivers, and other features which provided the basis for subregional specialization of production and also meant that potentialities for commercial production varied considerably. Other regions also had subregional variation.

9 Historiographically, spatial analysis is valuable because it raises the issue of the most appropriate level(s) for study—given a particular question. Ideally, it forces an investigator to examine several levels and the connexions between them. An excellent example of a functional economic region would be a currency zone which bridged various states or ecological belts. See Lovejoy, P. E., ‘Interregional Monetary Flows in the Precolonial Trade of Nigeria’, J. Afr. Hist., xv (1974), 563–85.CrossRefGoogle Scholar

10 Howard, , ‘Big Men, Traders, and Chiefs’, chap. iGoogle Scholar. For commerce in the area beyond the Kaloum peninsula (site of Conakry), see Mouser, B. L., ‘Trade and Politics in the Nunez and Pongo Rivers, 1790–1865’ (Ph.D. dissertation, Indiana University, 1971)Google Scholar. For commerce in Kpa Mendi and Sherbro, see Davidson, J., ‘Trade and Politics in the Sherbro Hinterland, 1849–1890’ (Ph.D. dissertation, University of Wisconsin-Madison, 1969)Google Scholar. A. Abraham has undertaken research on nineteenth-century trade in eastern Mendi areas.

11 Howard, , ‘Big Men, Traders, and Chiefs’, chaps. 1 and 7Google Scholar, contains statistics on over seas exports and a more detailed description of inter-regional commerce. On the question of slaves being sold from the hilly intermediate zone and other parts of what is now Koinadugu District, see Lipschutz, M. R., ‘Northeast Sierra Leone after 1884: Responses to the Samorian Invasions and British Colonialism’ (Ph.D. dissertation, University of California, Los Angeles, 1973), 48–52.Google Scholar

12 Howard, , ‘Big Men, Traders, and Chiefs’, chaps, 1 and 3Google Scholar; see also Howard, A. M., ‘Historical Centralities and Spatial Patterns in Northern Sierra Leone’Google Scholar, Conference on Spatial Hierarchies in African Interurban Systems, Joint Committee on African Studies, American Council of Learned Sciences/Social Science Research Council, New York, 1970. Among the clusters were those on the lower Rokel River (including Magbeli, Rokon, Rokel, Forodugu, and lesser points), in the Karina-Bumban area, and in the Kambia-Rowula complex. While few plains centres exceeded a thousand residents, clusters of hamlets, villages and towns resulted in both much larger populations and greater functional complexity around the key nodes. The lower Rokel cluster, for example, served a large collecting area for bulky foods and exports and also a north-south traffic in indigenous commerce.

13 Christaller, W., The Central Places of Southern Germany, trans, by Baskin, C. (Englewood Cliffs, N.J., 1966)Google Scholar; Brush, J. E. and Bracey, H. E., ‘Rural Service Centers in Southwestern Wisconsin and Southern England’, in H. M. Mayer and C. F. Kohn, Readings in Urban Geography (Chicago, 1959), 210–17Google Scholar; and Berry, B. J. L., Geography of Market Centers and Retail Distribution (Englewood Cliffs, N.J., 1967).Google Scholar

14 Berry, and Horton, , Geographic Perspectives, 106–228Google Scholar; and Hurst, , Geography of Economic Behavior, 194–222Google Scholar, summarize other classical theories and the recent literature.

15 By sorting out centres according to functions and, where useful, making functional typologies, the researcher can better explain relationships among places and the characteristics of regions and economic and political systems. See Howard, A. M., ‘Pre-Colonial Centres and Regional Systems in Africa’, Pan-African Journal, viii (Fall, 1975).Google Scholar

16 Hill, P., ‘Landlords and Brokers: A West African Trading System’, Cahters d'Etudes Africaines, vi (1966), 349–66CrossRefGoogle Scholar; and ‘Two Types of West African House Trade’, in Meillassoux, C., ed., The Development of Indigenous Trade and Markets in West Africa (London, 1971), 303–18Google Scholar. For big men and landlords in the coastal plain, see Howard, , ‘Big Men, Traders, and Chiefs’, chaps. 5 and 6Google Scholar; and Howard, A. M., ‘Commercial Innovation Associated with Temne Compounds’Google Scholar, Annual Meeting, African Studies Association, Philadelphia, 1972, which examine Temne and other entrepreneurs in Magbeli, Port Loko, and other river towns, 1870–1920.

17 Functions associated with chieftaincy were important but not the only critical indicators of political centrality. Power and authority were diffused, and there were a large number of titled and untitled chiefs, plus warriors and notables who headed villages and small towns, often while acting as councillors. Some heads of ethnic enclaves within chiefdoms ruled over wider areas and more people than did ‘paramounts’ of tiny chiefdoms.

18 Techniques developed by social anthropologists, such as the construction of social networks, can be applied spatially. See Mitchell, J. C., ed., Social Networks in Urban Situations (Manchester, 1969).Google Scholar

19 I am presently engaged in a study of statistics collected in the Sierra Leone National Archives and Senegal National Archives which pertain to the movement of traders and carriers and of foodstuffs, African manufactures, exports, and imports among towns and regions from the 1850s to 1920.

20 Thinking spatially, the question of diffusion of food and export crops involves examining how rapidly innovations spread within a region and among regions, the mechanisms by which they spread, and the adaptability of existing commercial networks. See the valuable study of Brooks, G. E., ‘Peanuts and Colonialism: Consequences of the Commercialization of Peanuts in West Africa, 1830–1870’, J. Afr. Hist., xvi (1975), 29–54.CrossRefGoogle Scholar

21 Christaller, , Central Places, 17–18, 58–72, 152–68Google Scholar. Christaller did caution that size and population of a city were not in themselves indices of complexity. For a discussion of central place hierarchies and non-hierarchical approaches, see Berry, , Geographic Perspectives, 150–228Google Scholar; and Geography of Market Centers, 26–42 ff.

22 Mabogunje, , Urbanization in Nigeria, 63–71, 90–103.Google Scholar

23 Russell, J. C., Medieval Regions and Their Cities (n.p., but Bloomington, Ind., 1972).Google Scholar

24 Good, C. M., ‘Salt, Trade, and Disease: Aspects of Development in Africa's Northern Great Lakes Region’, Int. J. Afr. Hist. St., v (1972), 543–86CrossRefGoogle Scholar, which also illustrates the way disease spreads along trade routes.

25 For a discussion of various patterns of centrality, see Howard, , ‘Pre-Colonial Centers’Google Scholar. Similar conclusions have been reached for East Africa by Taylor, D. R. F., ‘Invisible City in East Africa’, Spring Regional Social Science Research Council Regional Seminar, New York, May 1975.Google Scholar

26 Two basic assumptions operate here: first, that a certain minimum level of demand (‘threshold’) is needed for a particular good or service to be offered at a given place; second, that a consumer will travel up to a maximum distance to obtain a particular good or service, and thus that each good has a certain market area. Aggregates can be constructed and such problems as multiple purchase of high and low threshold goods on a single trip can be analysed. Berry, , Geographic Perspectives, 169–228.Google Scholar

27 Smith, R. H. T., ‘West African Marketplaces: Temporal Periodicity and Locational Spacing’, in Meillassoux, ed., The Development of Indigenous Trade and Markets, 319–46Google Scholar. Important studies of twentieth-century hierarchies which go beyond marketing include Abiodun, J. O., ‘Urban Hierarchy in a Developing Country’, Economic Geography, xliii (1967), 347–63CrossRefGoogle Scholar; and McNulty, M. L., ‘Urban Structure and Development: The Urban Systems of Ghana’, J. of Developing Areas, iii (1969), 159–76.Google Scholar

28 Howard, , ‘Historical Centralities’Google Scholar; and ‘Big Men, Traders, and Chiefs’, chap. 6.

29 Non-hierarchical performance of such functions could occur, however, as when Jula, landlords, and authorities all worked to govern currency movement; see discussion below.

30 Curtin, P. D., Economic Change in Precolonial Africa. Senegambia in the Era of the Slave Trade (Madison, 1975), 62–6 ff.Google Scholar

31 Christaller posited that arrangement of centres was governed by certain ‘principles’. In the case of the ‘marketing principle’, location was such that ease of consumer movement to retailing points would be maximized; with the ‘transport principle’, ease of movement along major routes would be maximized and transport costs reduced; with the ‘administrative principle’, ease of access from governing centres to all lesser places would be maximized. Geographers have found that in certain areas one of the ‘principles’ may be operating very strongly, so that the arrangement of centres and connecting roads approaches the ‘ideal’ of Christaller's model. The best examples are where the ‘marketing principle’ prevails. Christaller hypothesized that because of competition among centres, hexagonal ‘market areas’ would be generated, with different ‘nesting’ of hexagons for each principle. A. Losch, arguing from contrasting premises, also concluded that a hexagonal spatial geometry would result from market forces, but later scholars have posited rectangular patterns of spatial organization. Berry, , Geography of Market Centers, 62–73Google Scholar; and Hurst, , Geography of Economic Behavior, 194–206Google Scholar. For historians, however, the critical point is that each ‘principle’ can be operating to varying degrees in a given area and that actual spatial configurations represent a combination of varying forces which may be exerting different influences upon location.

32 For Futa Jallon, see Diallo, T., Les Institutions Politiques du Fouta Dyalon au XIXe Siicle (Dakar, 1972), 31–52Google Scholar, and research now being undertaken by Diallo and B. Barry. Relationships of the Samorian Almamate with the coastal plain have been explored by Y. Person, see Samori, tome I (Dakar, , 1968), 98–101, passimGoogle Scholar. C. M. Fyle and R. L. Ganga recently have carried out important research on the foundation and growth of Falaba town and the Solima state.

33 Development of roads and towns and relationships of commerce and spatial change are discussed in Howard, , ‘Big Men, Traders, and Chiefs’, chaps. 3 and 7Google Scholar. For an interpretation of Rokel River history which differs somewhat from my own, Ijagbemi, E. A., ‘The Rokel River and the Development of Inland Trade in Sierra Leone’, Odu, n.s., in (1970), 4S-70.Google Scholar

34 Vansina, J., ‘The Use of Process-Models in African History’, in Vansina, J., Mauny, R., Thomas, L. V., eds., The Historian in Tropical Africa (London, 1964), 375–89.Google Scholar

35 In the Northern Sierra Leone plain during the eighteenth century, the predominant spatial pattern was that of settlements clustered around chiefdom capitals and other political and military centres. The expansion of inter-regional and regional trade en couraged the growth of points along main roads, spaced at fairly regular intervals like beads on a string (more in keeping with the ‘transport principle’), while in the rivers and nearby hinterland, configurations resembling a marketing hierarchy came into existence. Howard, ‘Historical Centralities and Spatial Patterns’. For changes in settlement pattern since the nineteenth century, see Siddle, D. J., ‘War-towns in Sierra Leone: A Study in Social Change’, Africa, xxxviii (1968), 47–56CrossRefGoogle Scholar, which, however, overemphasizes the role of war in town formation and ignores commercial and certain political centres.

36 In many instances, ‘indigenous’ and ‘stranger’ are relative terms because of the processes of accommodation through which newcomers acquired land, entered local networks, and gained a political voice. Often, such conflicts were channelled into limited forms of competition, being institutionalized by such mechanisms as shared offices, guaranteed roles in decision making, and recognized semi-autonomous spheres. Howard, ‘Big Men, Traders, and Chiefs’, chaps. 3, 4, and 7. D. E. Skinner is now carrying out a detailed study of Moria (Kissi Kissi) which had problems in finding a satisfactory political structure.

37 For a comparison, see Akintoye, S. A., ‘The Ondo Road Eastwards of Lagos, c. 1870–95’, J. Afr. Hist., x (1969), 581–98CrossRefGoogle Scholar. When struggles take place among middle-or lower-level places, they tend not to disrupt a system severely. Functions can usually be relocated with much greater ease than those performed at higher-ranking places. Specialists migrate and traders re-route goods.

38 This process went furthest in the Sierra Leone River hinterland. From the 1870s, politics increasingly split along those lines, and, by the mid-1880s, trading alliances were prevailing. By the early 1890s, however, direct European intervention was causing political upheaval which deeply affected alliance patterns.

39 Asante provides another example of how indigenous trade, primarily in kola, affected spatial arrangements and a case where the premier centre—Kumasi—was able to limit spatially disruptive effects of expanding commerce. See Wilks, I., Asante in the Nineteenth Century (London, 1975), 113–20.Google Scholar

40 In commenting on an earlier version of this article, B. A. Agiri suggested that the term ‘growth’ be avoided because of the overtones associated with it; here it is used to mean increased output and exchange, with no further implications unless stated.

41 Regional analysts have set out at least two broad meanings for polarization: first, the creation of extremes—a rich core, an impoverished periphery; or an overly wealthy West, an exploited Third World—second, ‘a succession of different fields of forces which generate a changing sequence of different vectors over functional and geographical space’. Polarization in the latter sense was employed by F. Perroux in his original writings on ‘growth poles’. The notion of a ‘growth pole’ is primarily sectoral; it pertains to ‘leading’ firms and industries from which growth impulses radiate. Some people, therefore, have preferred to use the term ‘growth centre’ when dealing with industrial location and the geographic aspects of regional growth. Lasue'n, J. R., ‘On Growth Poles’, in Hansen, N. M., ed., Growth Centers in Regional Economic Development (New York, 1972), 47Google Scholar. See also Hermansen, T., ‘Development Poles and Developmental Centres: National and Regional Development’, in Kuklinski, A., ed., Growth Poles and Growth Centres in Regional Planning (The Hague, 1972), 1–67.Google Scholar

42 Friedmann, J., ‘A General Theory of Polarized Development’, in Hansen, ed., Growth Centers, 82–107Google Scholar; and The Spatial Organization of Power in the Development of Urban Systems (School of Architecture and Urban Planning, University of California, Los Angeles, July 1972). For historians, a critical issue is the correlation of degree of regional inequality with ‘stage of development’, see Williamson, J. G., ‘Regional In equality and the Process of National Development: A Description of Patterns’, Economic Development and Cultural Change, xiii 4 pt. II (July 1965), 1–84.CrossRefGoogle Scholar

43 Compounds were, in effect, firms directed from above by the big men and other ranking people. They could match talent with jobs and use labour and capital on an efficient basis by concentrating on planting, tree tending, harvesting, and varied transport and commercial activities in a scheduled way. Howard, , ‘Commercial Innovations … Temne Compounds’Google Scholar. Interview with Pa Bai Kamara, Pa Lenisa Kabbia, and Pa Kapr, Magbeli, 8 Apr. 1968; and interviews with Alhaji Dumbuya, Port Loko, 8 Mar. 1968, and 20 Oct. 1968.

44 Writers assessing the potentialities of pre-colonial economies have focused on transport costs, emphasizing that they were high and thus a barrier to economic expansion. Gray, R. and Birmingham, D., eds., Pre-Colonial African Trade (London, 1970), 1–23Google Scholar. The most comprehensive assessment has been made by Hopkins, A. G., An Economic History of West Africa, 78–166Google Scholar. Generally, though, the question of spatial efficiencies has been overlooked, as have certain related issues. Much more work is needed, for example, in measuring the ‘external economies’ provided by rulers, traders, landlords, and others in towns and cities. Modern marketing studies have demonstrated that production can be increased and marketing improved by better information flow, collection, transport, and storage, and also that producers raise their output when they feel certain about getting crops to market. See Jones, W. O., Marketing of Staple Food Crops in Tropical Africa (Ithaca, N.Y., 1972).Google Scholar

45 Beginning in the 1870s, French and British officials sought to narrow the range of African alternatives by enlarging their customs spheres. What they considered smuggling across customs lines was for African farmers an effort to get a better price and for traders an attempt to profit from price differentials. In the era of formal colonialism, taxation, the development of a more dendritic marketing system, and other factors brought about great producer indebtedness, with crops being mortgaged against the future. Some large-scale African traders engaged in this practice, but it became much more pronounced as Syrians and Lebanese gained greater control over produce buying. For such practices by Temne and other entrepreneurs in the pre-colonial and colonial periods, see Howard, ‘Commercial Innovations … Temne Compounds’.

46 See note 19 above. Preliminary figures comparing the value of indigenous and overseas trade were presented at the African Economic Workshop, University of Wisconsin, Madison, July 1974. Barry, M. D. quantified Guinean trade in his stimulating study, Les Formations Sociales de la Guinie: Les Formations Sociales Pre-Coloniales (United Nations, Institut Africain de Development Isconomique et de Planification, Dakar, Aug. 1973).Google Scholar

47 Lovejoy, , ‘Interregional Monetary Flows’Google Scholar; and Lovejoy, P. E., ‘The Kambarin Beriberi: The Formation of a Specialized Group of Hausa Kola Traders in the Nineteenth Century’, J. Afr. Hut., xiv (1973), 633–51.CrossRefGoogle Scholar

48 Kiwanuka, M. S. M. S., A History of Buganda (New York, 1972), 148–52Google Scholar; and Tosh, J., ’The Northern Interlacustrine Region’, in Gray and Birmingham, Pre-Colonial African Trade, 103–18.Google Scholar

49 Awe, B., ‘Militarism and Economic Development, in Nineteenth Century Yoruba Country: The Ibadan Example’, J. Afr. Hist., xiv (1973), 65–77;CrossRefGoogle ScholarArhin, K., ‘Aspects of the Ashanti Northern Trade in the Nineteenth Century’, Africa, xl (1970), 363–73CrossRefGoogle Scholar; and Wilks, , Asante in the Nineteenth Century, 666–724.Google Scholar

50 Weigend, G. G., ‘Some Elements in the Study of Port Geography’, reprinted in Mayer and Cohn, Readings in Urban Geography, 365–75Google Scholar; and Bird, J. H., ‘Of Central Places, Cities and Seaports’, Geography, lviii, 259 (04. 1973), 105–18Google Scholar. Towns formed by immigrant traders, such as Mande Jula, often served as gateway towns and could be analysed as such with respect to their surrounding area.

51 Howard, A. M., ‘The Role of Freetown in the Commercial Life of Sierra Leone’, in Fyfe, C. and Jones, E., eds., Freetown: A Symposium (London and Freetown, 1968), 38–64.Google Scholar

52 Historical and contemporary relationships between primacy and development are discussed in El-Shakhs, S., ‘Development, Primacy, and Systems of Cities’, J. of Developing Areas, vii (1972), 11–35Google Scholar. It should also be possible to study correlations between primacy and regional inequalities in colonies and nations, on the one hand, and various historical economic relationships with the larger world economy on the other.

53 For interpretations of the relationships between Freetown and its northern hinterland which contrast somewhat with my own, see Ijagbemi, E. A., ‘The Freetown Colony and the Development of “legitimate” Commerce in the Adjoining Territories’, J. Hist. Soc. Nigeria, v, 2 (1970), 243–57Google Scholar; and Deveneaux, G. K., ‘The Political and Social Impact of the Colony in Northern Sierra Leone, 1821–1896’ (Ph.D. dissertation, Boston University, 1973)Google Scholar. Fyfe, C., A History of Sierra Leone (London, 1962)Google Scholar, surveys political relationships both north and south of the Sierra Leone peninsula.

54 For a discussion of dendritic networks, see Johnson, E. A. J., The Organization of Space in Developing Countries (Cambridge, Mass., 1970), 83–92CrossRefGoogle Scholar. Prior to the 1840s, dendritic, extractive networks existed in a few parts of the plain. As in many areas of Africa, slave trading for export promoted dendritic organization.

55 John, C. St., ‘Kazembe and the Tanganyika-Nyasa Corridor, 1800–1890’, in Gray and Birmingham, Pre-Colonial African Trade, 202–30Google Scholar. Other essays in the same book confirm the point that trade in wasting resources, including slaves, tends to be dendritic in organization. The complex relationship between trade, social change, and growth of centres in one part of East Africa is examined by Alpers, E. A., Ivory and Slaves in East Central Africa (Berkeley, 1975).Google Scholar

56 Personal communication, Patrick Manning, 24 June 1975.

57 Awe, B., ‘Militarism and Economic Development’, 68–74.Google Scholar

58 Afigbo, A. E., ‘Trade and Trade Routes in Nineteenth Century Nsukka’, J. Hist. Soc. Nigeria, vli, i (1973),Google Scholar

59 For spatial patterns in Sierra Leone during the twentieth century, see Howard, , ‘Big Men, Traders, and Chiefs’, chap. 9Google Scholar; Clarke, , Sierra Leone in Maps;Google ScholarHarvey, M. E. E., ‘A Geographical Study of the Pattern, Processes and Consequences of Urban Growth in Sierra Leone in the Twentieth Century” (Ph.D. dissertation, University of Durham, 1966)Google Scholar; and Riddell, J. B., The Spatial Dynamics of Modernization in Sierra Leone (Evanston, Ill., 1970).Google Scholar