Hostname: page-component-586b7cd67f-gb8f7 Total loading time: 0 Render date: 2024-11-27T15:40:38.723Z Has data issue: false hasContentIssue false
  • English
  • Français

Notes on Traditional Market Authority and Market Periodicity in West Africa1

Published online by Cambridge University Press:  22 January 2009

Polly Hill
Get access Rights & Permissions [Opens in a new window]

Extract

These notes on traditional market authority and market periodicity in West Africa are intended to draw attention to the neglect of a fascinating historico-anthropological field. Given the lack of any satisfactory typology of West African markets, generalizations about all types of market must be avoided—and certain erroneous assumptions relating to the traditional relationship of chief and market are called in question. Nor should it be assumed that Christaller's ‘central-place theory’ has any application in such regions as Yorubaland and Iboland, where rural periodic markets cannot be classified hierarchically as, for instance, they can be in rural China. Although daily (or continuous) markets which, both traditionally and today, occur only in the very largest centres of population, are typologically distinct from rural periodic markets, this is rather because of their high proportion of full-time vendors than because they are necessarily apices of an hierarchical market system.

Everywhere in West Africa there is a standard ‘market-week’ such that all the periodic markets in the locality are based on the same cycle. Although this has been well-known for centuries, few writers in English have been interested in mapping these weeks, or examining the relationship between them. After summarizing existing geographical material, which clearly establishes an association between 4- and 8-day market weeks (which are extraordinarily widespread in the forest zone), as well as some connexion between the 3- and 6-day market weeks, it is concluded that there are effectively only four common market weeks, viz. 3- (or 6-) day, 4- (or 8-) day, 5-day and 7-day. The 7-day market week prevails throughout Hausaland and many other Islamized savanna districts. The fact that it has existed for centuries throughout the Akan areas of southern Ghana and Ashanti (though hardly anywhere else in the forest zone) has not attracted the attention it deserves—the same applying to other questions, here posed though not answered, such as why the 4-day market week cuts right across some ethnic boundaries and not others.

The article concludes with brief notes on the two traditional types of market taxation in West Africa, tolls and stall-rents, emphasis being laid on the traditional lack of market taxation in Eastern Nigeria.

Type
Research Article
Information
The Journal of African History , Volume 7 , Issue 2 , July 1966 , pp. 295 - 311
Copyright
Copyright © Cambridge University Press 1966

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

2 The monumental article on African markets by Willy Fröhlich, on which I lean so heavily in my discussion of market periodicity (see p. 301), is ethnographic bibliography not social anthropology.Google Scholar

3 I am currently working on many other aspects of West African market organization. It is a pleasure to note that it is the geographers of Nigeria who are rapidly becoming the real pioneers on the study of markets, especially in the field of market periodicity and location.Google Scholar

4 Final Report (H.M.S.O., 1891), 2.Google Scholar

5 Report on Markets and Fairs in England and Wales, part I, General Review, Economic Series no. 13, Ministry of Agriculture and Fisheries (H.M.S.O., 1927), 5. The following quotation is drawn from p. 71 of this scholarly report: ‘There is a tendency in some quarters to think that because markets are institutions of great antiquity…they do not fit in with modern business and have little to offer to modern trade. Nothing could be further from the truth.’Google Scholar

6 Wayside gatherings are especially common in some regions: Henry Barth commented on their absence when he left Hausaland travelling eastwards (Travels and Discoveries in North and Central Africa (1849–55), II, 164).Google Scholar The law ‘which prohibits men from buying and selling elsewhere than in a duly constituted market’ (Maitland, F. W., Domesday Book and Beyond, Library, Fontana ed., 1960, 236) usually applies only to the vicinity of markets on market days.Google Scholar

7 Like the gatherings of women, these concourses may happen to be incipient markets, though specialist wholesale markets, other than livestock markets, are rather rare in West Africa.Google Scholar

8 See ‘Landlords and Brokers: A West African Trading System’, by the present author, in Cahiers d'Etudes Africaines (1966), forthcoming. The institution of mai gida may either complement or be independent of any market in the locality, or it may exist in marketless places such as the Sarakole country in Mali where, although the people were great traders, organized in strong chiefdoms, there were formerly no markets (personal communication from Dr C. Meillassoux).Google Scholar

9 ‘Marketing and social structure in rural China’, parts I, II and III, by G. William Skinner, The Journal of Asian Studies (Nov. 1964, Feb. 1965, May 1965).Google Scholar

10 Battuta, Ibn, Travels in Asia and Africa, 1325–1354, ed. Gibb, H. A. R. (1929), ch. XIV, provides evidence for the existence of considerable markets over a large area of the western Sudan in the mid-fourteenth century.Google Scholar In the mid-seventeenth century Jenné was accounted (see The Northern Factor in Ashanti History, by Wilks, Ivor (University of Ghana, 1961), 8)Google Scholar ‘one of the great markets of the Moslem world’. Pieter de Marees, who was on the Gold Coast 1600–1, reported that ‘every town hath market days especially appointed’ (quoted from Pageant of Ghana, by Freda Wolfson, 55). In the late seventeenth century Barbot especially noted the ‘public markets of Benin’. Thus the early history should be seen in terms both of the development of trade routes and their associated markets (where they existed) and of the establishment of markets to serve the local community. In East Africa, by contrast, long-distance trade was rare before the nineteenth century and local markets hardly existed before the colonial period.

11 I am not here disagreeing with Professor Skinner's suggestion (part 1,3) that the marketing structures he describes are ‘characteristic of the whole class of civilizations known as “peasant” or “traditional agrarian” societies’. He regards the China case as ‘strategic’ for comparative study because (i) ‘the integrative task accomplished there was uniquely large’, (ii) ‘the exceptional longevity and stability of Chinese society have allowed the marketing system in many regions to reach full maturity prior to the beginnings of modernization’, (iii) documentation of marketing extends over several centuries. West Africa may be ‘different’mainly because the state of ‘full maturity’ has not been reached.Google Scholar

12 The literature abounds with references to extra-settlement markets, two of the most notable sources being Meillassoux, C., Anthropologie Economique des Gouro de Côte d'Ivoire, 286,Google Scholar and Hodder, B. W., a well-known authority on Yoruba markets, who in ‘Distribution of markets in Yorubaland’, Scottish Geographical Magazine (April 1965), 51, asserts that ‘the distribution of periodic markets shows little correspondence either with the distribution of population or with the distribution, size or hierarchy of rural or urban settlements…; many of the more important settlements have no periodic markets’.Google Scholar

13 This assertion is partly based on my own, as yet unpublished, field observations, mainly in Ghana. It may not apply to Hausaland. I think it is much more likely to apply in the forest zone than in the savanna, owing to the extreme bulkiness of the basic starchy foods of the south (cassava, plantain, yams), which means that every trader is desirous of cutting down handling to the minimum. The increased consumption of less bulky processed foodstuffs, such as gari, may lead to the development of more trade between markets (see also p. 301 below). I am, admittedly, ignoring trade in imported goods.Google Scholar

14 ‘The primitive market was not necessarily in or near a town or village, in fact there is much to be said for the view that it was originally always remote from human habitations’ (‘The market in African law and custom’, by Thomas, N. W., Journal of the Society of Comparative Legislation (1908), 91). Wilks, op. cit. 7, notes that in 1679 trade between the Mande and the pagans of the forest was conducted in the bush outside Begho, ‘which is declared to be the market’.Google Scholar

12 Natives, also, may establish markets outside the walls, to prevent strangers entering (see Travels in Western Africa in 1845 and 1846, by John Duncan, 11, 30). Duncan also noted that when a town or city has two walls, the public market, which was often attended by strangers from a great distance, was held within the outer walls (11, 99).Google Scholar

16 See ‘Nineteenth-century trade in the Bamenda grassfields, Southern Cameroons’, by Chilver, E. M., Afrika und Übersee, XLV (1962), Heft 4, 240.Google Scholar

17 See the chapter by Smith, M. G. in Markets in Africa, ed. Bohannan, et al. , 305–6.Google Scholar

18 According to Skinner, E. P. (The Mossi of the Upper Volta (1964), 42), most Mossi market officials held hereditary posts: the same may be true of Hausaland, though M. G. Smith, virtually the only authority, is nowhere explicit on the point. See also E. P. Skinner's article in Markets in Africa, 265–6.Google Scholar

19 Even as recently as 1962, the market in the largest stranger-suburb of Accra (Nima) was ‘owned’ and controlled by the man on whose land it was situated, his employees collecting the tolls.Google Scholar

20 At the local-authority-controlled market at Wechiau, near Wa, in north-west Ghana, the tendaana was still exacting tolls, in kind, in 1965.Google Scholar

21 Markets in Africa, 255.Google Scholar

22 Although markets are such ancient institutions in many regions, marketless areas also exist. Hodder (‘Distribution of markets in Yorubaland’, 50, states that in Yorubaland there is a critical density of population of about fifty persons to the square mile, above which ‘there is a regular pattern of periodic markets and below which there are very few indeed’. To the present writer it seems highly unlikely that Meillassoux's challenging generalization about the origin of markets, which follows, can possibly be of universal validity in West Africa, so variable are the circumstances: ‘local needs of exchange for foodstuffs and craft products do not seem sufficient to promote marketing activities…’; ‘markets are primarily induced by external exchanges of complementary products with an alien population’ (from the chapter on Gouro markets by Meillassoux, C., in Markets in Africa, 297).Google Scholar But, as Bovill noted in 1922 (Bovill, E. W., ‘Jega Market’, Journal of the African Society, 10 1922, 54), seven of ‘the most noted markets of the Western Sudan’, viz. Bamako, Ouagadougou, Fada-n-Gourma, Gaya, Jega, Kano and Dikwa, ‘were all situated within a day's march of the twelfth parallel of north latitude’, in the intermediate zone separating ‘the oil palms of the forests from the borassus and hyphoene palms of the Sudan’ and all of them, save Kano, which also and largely served the local population, were basically entrepêts linking savanna and forest.Google Scholar

23 Sources include ‘Afikpo markets: 1900–1960’, by S. and P. Ottenberg, in Markets in Africa; Anthropological Report on Ibo-Speaking Peoples of Nigeria, part IV, Law and Custom of the Ibo of the Asaba District, by N. W. Thomas (1914); and Ibo Village Affairs, by M. M. Green (1947).Google Scholar To these must be added ‘Markets in Iboland, Eastern Nigeria’, by Ukwu, U. U., Cambridge Ph.D. thesis (1965), which is an extremely interesting unpublished work by a geographer. See also n. 72 below.Google Scholar

24 Night markets, such as occur in Yorubaland, may perhaps be regarded as a species of periodic market, the point about the daily market being that it is in constant session during the working day; daily (or twice-daily) cattle markets may also be regarded as periodic, for they are always of brief duration.Google Scholar

25 In Ghana, for instance, there may have been no more than about half a dozen such markets in the nineteenth century, compared with perhaps about a dozen today.Google Scholar

26 I do not, however, think that daily markets are nearly common enough in Hausaland to justify using their existence as a criterion distinguishing town from village—as suggested by M. G. Smith, op. cit. 310.Google Scholar

27 See Duncan, op. cit., which is a remarkable source on markets. (I do not know whether the numerous daily markets he observed have since been converted, in so far as they still exist, into periodic markets.)Google Scholar

28 Skinner, op. cit. part I, II.Google Scholar

29 Ibid. part 1, 10.

30 The final preparation of the food for marketing, or eating, is often the end of a long process which begins with the harvesting of the crop (if not earlicr), so that the activities of farming and food processing are not necessarily distinguishable. (The women traders of rural Yorubaland are perhaps more apt to be trading specialists than any other rurally based women in West Africa.)Google Scholar

31 Journal of the Royal Anthropological Institute (1924).Google Scholar

32 Zeitschrift für Ethnologie (1940). Of 104 pages in length, it includes a bibliography of 406 works and other reference material of historical interest. It is not listed in the bibliography of Markets in Africa.Google Scholar

33 Thomas, ibid. 193. Thomas was concerned to dispute A. B. Ellis's theory, in relation to the Akan and the Yoruba, that the West African week was based on the rest-day, which was itself a transformed lunar festival.

34 The list of market weeks that follows is mainly based on those of Thomas and Fröhlich: although I have made some additions, most of them based on post-war sources, I am well aware that it remains very incomplete, especially in its omission of the market weeks of Guinée and Sierra Leone.Google Scholar

35 Du Niger au Golfe de Guinée, by L. C. Binger (1892); see, for instance, 1,496. For E. P. Skinner's reference to the 3-day week see Markets in Africa, 257.Google Scholar

36 Thomas, ibid. 191, points out that the strong ethnographic evidence of the former existence of this week is confirmed linguistically.

37 I am grateful to Professor J. Vansina for informing me that the 4-day market week prevails solidly in a huge area south and west of the central basin of the Congo River. This area extends very roughly from the latitude of Brazzaville and Lusambo (in the north) to just north of the source of the river Cuanza in Angola (in the south). See, also, fn. 60 below.Google Scholar

38 Possibly Ardener's reference to markets being held ‘every nine days’ in rural areas near Duala implies an 8-day week. (Coastal Bantu of tile Cameroons, Ethnographic Survey of Africa, Int. Af. Inst., 1956, 48.)Google Scholar

39 All markets in Yorubaland are on 4-day (or 8-day) schedules—Thomas was probably misled by the Yoruba practice of referring to 4-day as s-day markets.Google Scholar

40 See Dahomey, by M. J. Herskovits (1938).Google Scholar

41 It is the most common market-week in Eweland, but there are 5- and 7-day weeks in some districts.Google Scholar

42 See the following volumes of the Ethnographic Survey of the International African Institute: Peoples of tile Niger-Benue Confluence, sections on the Igala and the Idoma by R. C. Armstrong, 1955; The Benin Kingdom and the Edo-speaking Peoples…, by R. E. Bradbury, 1957; and Peoples of the Central Cameroons, section on the Bamileke by M. Littlewood, 1954.Google Scholar

43 Two-day marketing schedules, also, are found only in the ‘4-day zone’—in southern Dahomey (see ‘Traditional market economy in South Dahomey’, by C. and C. Tardits, Markets in Africa, 97) and in neighbouring Yorubuland. In the former district large and small markets are said to alternate, suggesting an association with a 4-day week. In Yorubaland (see fn. 71 below) the 2-day markets may represent mergers of two 4-day markets. So, rather arbitrarily, I have chosen not to regard the 2-day schedule as a market week proper. It is to be noted, with reference to n. 74 below, that nearly all the 2-day markets lie very far south.Google Scholar

44 See The Tiv of Central Nigeria, by L. and P. Bohannan (Ethnographic Survey of Africa, International African Institute, 1953).Google Scholar

45 I assume that Nadel's statement (A Black Byzantium, by S. F. Nadel, 321) that markets are held ‘every fifth day’ means a 5-day, not a 4-day week.Google Scholar

46 Armstrong, ibid. 93.

47 See a map showing the distribution of 4- (or 8-) day, 5-day and 6-day market schedules in Eastern Nigeria in The Peoples of Southern Nigeria by Talbot, P. A. (1926), III, 871.Google Scholar

48 See Die Glidyi-Ewe in Toga, bs D. Westermann (1935), 95 ff.Google Scholar

49 See Binger, op. cit. 1, 318, 362, 370.Google Scholar

50 See Les Tribus du Rameau Lobi, by Labouret, H. (1931), 353.Google Scholar

51 Though see An Expedition into the Interior of Africa by the River Niger, by Laird, M. and Oldfield, R. A. K. (1837), I, 165, 409, for references to a 10-day market near the Benue confluence.Google Scholar

52 See the map referred to in fn. 47 above.Google Scholar

53 See The Konkomba of Northern Ghana, by Tait, D. (1961), 18,Google Scholar and Les Populations du Nord-Togo, by Froelich, J. C., Alexandre, P. and Cornevin, R. (Ethnographic Survey, Int. Af. Inst., 1963), 24, 83, 133.Google Scholar

54 Cardinal regarded this smaller market as a ‘woman's market’ (The Natives of the Northern Territories of the Gold Coast (1920), 97). (Such an interpolation is wholly absent in some places, for instance in the large town of Wa in north-west Ghana.)Google Scholar

55 Ethnographic material is still inadequate for this purpose.Google Scholar

56 The 7-day market week is peculiar in that it is quite common, perhaps increasingly so, for markets to be held on two fixed days of the 7-day cycle, such as Mondays and Thursdays.Google Scholar

57 Barth provides a splendid description of Kukuwa market, near lake Chad, which was held every Monday and attracted some 12,000 to 15,000 people (Travels and Discoveries in North and Central Africa, 1849–55, by Barth, H., II, 306 ff.).Google Scholar East and south of Bornu there is also a 7-day week—see Les Populations païennes du Nord-Cameroun et de l'Adamawa, by Lambezat, B. (1961).Google Scholar

58 See Meillassoux, op. cit.Google Scholar

59 Fröhlich (264) was doubtful about this.Google Scholar

60 Fröhlich stated categorically (265) that there is no evidence for the existence of a ‘pre-Islamic’ 7-day week in Africa. It is for historians to comment on this, in relation both to the Akan 7-day week and to the existence of a 7-day market week throughout a huge area of the central basin of the Congo which, according to Professor J. Vansina (personal communication), lies south of the Congo river bend between Stanleyville and the Kasai confluence, extending as far south as Lusambo.Google Scholar

61 Such as the 16-day cloth market at Oje in Ibadan (personal communication from Dr B. W. Hodder). (But other markets are held at this market place on other days.)Google Scholar

62 In considering the Aro trade fairs, reported on by Dr Ukwu(op. cit.), which were on a 24-day schedule, and certain Niger river-fairs, one comes up against the question of whether markets and fairs should be distinguished in Africa. Perhaps any market on a long schedule which lasts for several days is appropriately regarded as a fair? Barbot (A Descriplion of the Coasts of North and South Guinea (1746), 78–9, 269) refers to up-country fairs.Google Scholar

63 Migrants who have established markets in new places sometimes carry their old day-names with them, which may assist the study of migration.Google Scholar

64 In Iboland ‘the markets are called after the day of the week, whereas in nearly the whole of the rest of the country, except Yorubaland, the contrary is the case’ (Talbot, op. cit. III, 869).Google Scholar

65 He was, in fact, himself doubtful of its application to what he called ‘the Yoruba– Twi area’, op. cit. 193. (The 4-day and 7-day weeks are, anyway, the only West African weeks which might be lunar in origin.)Google Scholar

66 Though see Meillassoux, op. cit., and E. P. Skinner, Markets in Africa, 225 ff.Google Scholar

67 But see Green, M. M. (op. cit. 134 ff.) and S. and P. Ottenberg (op. cit. 121 ff.) on consultations and disputes relating to the establishment of Ibo markets.Google Scholar

68 Nadel, op. cit. 323, provides an example. P. C. Lloyd (Yoruba Land Law, 72) states that in Yorubaland public markets ‘may only be created by the command of the Oba’.Google Scholar

69 The most famous of these circuits are those in Yorubaland described by B. W. Hodder—see his chapter in Markets in Africa, and ‘Rural periodic day markets in part of Yorubaland, Western Nigeria’, Transactions and Papers of the Institute of British Geographers (1961). However, as Dr Hodder has himself pointed Out on p. 56 of the source cited in fn. 12, ‘adjacent market rings often impinge on one another, resulting in a loose chain-mail pattern of rings stretching over most of Yorubaland’. This means that any market may belong to more than one ring, in the sense both that there is no fixed market community corresponding to the area served by a ring, and that the full-time women traders who move round the markets may frequent more than one ring. Dr Meillassoux, also, has revealed (op. cit. 287) the existence of overlapping market areas stretching across northern Gouro country, nearly all of them with a different calendar based on the names of markets. See also Tait (op. cit. 18), who regarded certain groups of Konkomba markets as providing ‘a space and time dimension’. Many observers, including Barth (op. cit. II, 157), have commented on the fact that in densely populated country there is apt to be a market within reach of a town every day of the week. But why not? ‘If at the outset, when markets were beginning to arise, it was already the practice to fix the day and place of markets by agreement between different towns, it is no long stretch of the imagination to suppose that the order of the markets once fixed, by mutual agreement, the unit we now call aweek came, as it were, automatically into existence…’ (Thomas, op. cit. 194).Google Scholar

70 Skinner, op. cit. part II, 209.Google Scholar

71 Dr B. W. Hodder has mapped all the periodic markets in Yorubaland in 1962–63, distinguishing ‘8-day’, ‘4-day’ and ‘2-day’ markets (see ‘Distribution of markets in Yorubaland’, loc. cit., map 2. His map shows nearly 500 4-day markets (21 of them in Dahomey), compared with about fifty 8-day markets and seven 2-day markets. The 8-day markets occur in definite groupings (north of Ibadan, in Ondo, in the Ijebu-Remo area and strung Out along the lagoons from Badagri in the west to Epe in the east) and Dr Hodder, following Meillassoux (see fn. 22), presents, p. 55, the fascinating hypothesis that they represent ‘the original foci of the market network in their location at [ethnic] junction zones’ —the lagoon-side markets also being very old. The 2-day markets probably represent mergers of two 4-day markets. Dr Hodder informs me that he has cartographic evidence of a very few markets having changed from 8-day to 4-day schedules, but none of the converse process.Google Scholar

72 The Ibo have a 4-day (calendrical) week. Although nowadays nearly all Ibo (as well as Ijaw and Ibibio) periodic markets are on an 8-day schedule (personal communication from Mr G. I. Jones, see also Markets in Iboland, by U. I. Ukwu), the general evidence seems to be that this was not always so, and there are some markets (e.g. those of the Afikpo village-group described by Ottenberg, op. cit.) on a 4-day schedule. It is usual for the first 4 days of the market week to have the Ibo word ‘big’ attached to the day-name, the second 4 days having the word ‘small’; M. M. Green (op. cit. 37) is quite explicit in seeing this as signifying a former lengthening of the market week. Then Talbot (Tribes of the Niger Delta by P. A. Talbot (1932), 281) records that ‘in each town of importance throughout southern Ibo country small markets… and big ones…are held alternately on every fourth day’, though he notes a few exceptions. Although I am told that there is some doubt about the reliability of this material, I note that Basden also states (Niger Ibos, 1938, 33), that markets occur every 8 days ‘in certain places’ only. Finally, many writers have generally referred to the 4-day market week of Eastern Nigeria, e.g. the following, in reference to the lower reaches of the Niger: ‘Every town has a market, generally once in four days’Google Scholar (A Narrative of the Expedition to the River Niger in 841 under the Command of Capt. Trotter, H. D., by Allen, W. and Thomson, T. R. H. (1848), I, 399).Google Scholar

73 There is evidence from northern Ghana of the interpolation of a second market in a 6-day schedule, but none of a lengthening of the 3-day schedule. However, it is recorded (see Skinner's, E. P. contribution to Markets in Africa, 257) that Mossi 3-day markets are also on a 21-day cycle, in that there is an especially large market every 21 days—and 21 is not divisible by 6.Google Scholar

74 Certainly in tropical forest conditions, a market week of longer than 4 days is apt to be very inconvenient (more so than in the savanna) to the consumer of marketed perishable foodstuffs, unless there is more than one market within convenient walking distance. The other general point to make is that Skinner's hypothesis is more likely to apply in the savanna (than in the south)—where, for all we know, an hierarchical system of periodic markets may sometimes exist.Google Scholar

75 There are, also, very few oflicial statistics relating to numbers of markets, which makes it worth drawing attention to such material in Ghana's 1961 Statistical Year Book (Central Bureau of Statistics, 1962), table x. Relating these figures to 1960 population statistics, surprisingly little variation is found, as between regions, in the average number of persons per rural market, which is some 8,000–10,000, a figure which might compare roughly with that for Yorubaland (see fn. 71).Google Scholar

76 In general, as urban populations grow, the largest periodic markets are converted into daily markets.Google Scholar

77 Sources include Duncan, op. cit., and explorers' writings on Hausaland. Bowdich (Mission from Cape Coast Castle to Ashantee, by Bowdich, T. E., 1819 ) found a daily market in Kumasi. Professor Jan Vansina informs me that the distinction between daily and periodic markets applied in the Congo/Angola region in pre-colonial times.Google Scholar

78 See p. 300 above.Google Scholar

79 See p. 296 above.Google Scholar

80 The counterparts of Kumasi Central Market today.Google Scholar

81 As it is today in the daily market of Koforidua in the eastern region of Ghana.Google Scholar

82 Which must account for the fact that their existence was entirely ignored by Fröhlich, who identified daily markets with the small daily gatherings referred to on p. 296 above.Google Scholar

83 In his unpublished thesis, op. cit.Google Scholar

84 They are, in particular, unclear as to the rights they wish to accord to the permanent vendors. While traders are never expressly prohibited from entering a daily market to sell their produce directly to the consumer, this traditional right of entry is often effectively lacking owing to the attitude and numerousness of permanent vendors. It is interesting to note that a few market authorities, including Kumasi, are now preserving this tradition by establishing special market annexes for those denoted as ‘villagers’.Google Scholar

85 The traditional organization of lorry parks is briefly discussed in the article mentioned in fn. 8 above.Google Scholar

86 This being reflected, for instance, in the general lack of material on market taxation in Markets in Africa, though see Skinner's, E. P. article which, pp. 266–7, gives much detail on Mossi market taxes.Google Scholar

87 Thus Clapperton noted that in Kano ‘A small duty is also levied on every article sold in the market; or, in lieu thereof, a certain rent is paid for the stall or shed’ (cited from West African Explorers, ed. Howard, C., 283).Google Scholar

88 See Appendix III of the Northern Nigeria Lands Committee (H.M.S.O., 1910).Google Scholar

89 Duncan, op. cit. I, 124.Google Scholar

90 Personal communication from Dr P. C. Lloyd.Google Scholar

91 Elaborate lists of rates of toll chargeable on different quantities of different types of produce have been compiled by some local authorities, but they usually mean very little and are never displayed in the market. There are few references in the literature to rates of toll varying with the value of goods—but see ‘Der Markt in Süd-Togo’, by Wucherer, A., Zeitschrift für Ethnologie (1935).Google Scholar

92 Traders commonly erect their own stalls, even in city markets, and if lock-up sheds are erected in an unplanned way, as in Makola no. a market in central Accra, the chaos is indescribable. Skinner, E. P. (Markets in Africa, 257) notes that the ‘solid stalls’ in Mossi markets were ‘built and tenanted by merchants’.Google Scholar

93 The estimated revenue from stall rents for the newly built main market at Onitsha in 1963–64 was £75,000—no revenue was raised from tolls. The total revenue raised in Kumasi Central Market in 1960–61 was £110,000, of which only £30,000 was from stall rents.Google Scholar

94 This may not be true of a modern market like Onitsha, where most of the stallholders handle non-food (mostly imported) merchandise. (In this respect Onitsha market seems rather to resemble an entrepêt than a West African market.)Google Scholar