¹ The volumes of the Ethnographic Survey of Africa issued by the International African Institute do not yet cover more than half the coast. Relevant volumes are cited below, but it should be noted that some of the earlier volumes display an anti-historical bias and pointedly ignore sources of the period with which this paper is concerned.Google Scholar

² Major sources are Westermann, D. and Bryan, M. A., Languages of West Africa: Handbook of African Languages, Part II (International African Institute, 1952),Google Scholar cited as Westermann, and Bryan, ; de Lavergne De Tressan, M., ‘Inventaire linguistique d'Afrique Occidentale Française et du Togo’, Mém. I.F.A.N. no. 30 (Dakar, 1953), cited as De Tressan. Sources for individual regions are cited below.Google Scholar

³ That is, substantial groups of people (‘tribes’) traditionally defined by their use of a distinct language as a mother-tongue.Google Scholar

⁴ Our method is therefore more rigorous than that of the classic study in European ethnohistory, Chadwick, H. M., The Origin of the English Nation (1907), where Germanic tribes are pursued across Europe and through centuries of fragmentary references, almost solely by name-comparison. Doubts about the validity of this method of inquiry have been expressed by contemporary European Iron Age scholars.Google Scholar

⁵ The languages of the Guinea coast can generally be distinguished by the comparison of a few lexical items with more ease than the current discussion of their ultimate genetic relationship may have led some to suppose. It has to be remembered that even between ‘closely related’ languages there are normally more lexical dissimilarities than similarities.Google Scholar

⁶ It would be valuable if all language references in historical studies carried a date indicator, e.g. distinguishing Temne¹⁷ from Temne²⁰, that is, Temne of the seventeenth century from Temne of the present day. However, in this study the time-factor is a limited one of not more than five centuries, and it is unlikely that the languages concerned have shown much basic development in this period.Google Scholar

⁷ It is being presented in a paper, ‘An ethnolinguistic inventory of the Guinea coast before 1700’, in the Sierra Leone Language Review, VI (1967).Google Scholar

⁸ The major sources for the coast from Senegal to Sierra Leone are the following (cited in the editions used): Gomes, D., [De prima inventione Gujnee] De la première découverte de la Guinée, ed. Monod, T., Mauny, R. and Duval, G. (Bissau, 1959);Google ScholarCadamosto, , Viagens de Luis de Cadamosto e de Pedro de Sintra, ed. Peres, D. (Lisbon, 1948);Google ScholarPereira, Duarte Pacheco, Esmeraldo de situ orbis (Côte occidentale d'Afrique), ed. Mauny, R. (Bissau, 1956);Google ScholarFernandes, V., Description de la Côte Occidentale d'Afrique (Sénégal au Cap de Monte), ed. Monod, T., da Mota, A. Teixeira and Mauny, R. (Bissau, 1951);Google Scholard'Almada, A. Alvares, Tratado breve dos rios de Guiné, ed. Silveira, L. (Lisbon, 1946),Google ScholarDe Lemos Coelho, F., Duas descrições seiscentistas da Guiné, ed. Peres, D. (Lisbon, 1953).Google Scholar

⁹ On the Wolof and Serer, see Gamble, D. P., The Wolof of Senegambia (Ethnographic Survey, 1957).Google Scholar

¹⁰ Identification in Mauny, R., Tableau géographique de l'Ouest Africain au Moyen Age (1961), 458–2. On the Wolof and Serer languages, see Westermann and Bryan, and map 8 in De Tressan.Google Scholar

¹¹ Gamble, op. cit. 17.Google Scholar

¹² Mauny, op. cit. 450, which also refers to Mandingo terms found in earlier Arabic sources.Google Scholar

¹³ Gamble, op. cit., map; De Tressan, map 9.Google Scholar

¹⁴ Westermann and Bryan; De Tressan, map 8;Google ScholarThomas, L. V., Les Diola (1958–1959), map;Google ScholarSapir, J. D., A Grammar of Diola-Fogny (1965), map.Google Scholar

¹⁵ Add to the references in the previous footnote, the map (‘Carta étnica da Guiné Portuguesa’) by Da Mota, A. Teixeira, in Da Mota, A. Teixeira and Neves, M. G. Ventim, A habitação indigena na Guiné Portuguesa (1948). This map, cited as ‘Carta étnica’, is a major source of information for Portuguese Guinea, but must be modified in detail in terms of maps based on a more recent census in A. Carreira, ‘População autóctone segundo os recenseamentos para fins fiscais’, Boletim Cultural da Guiné Portuguesa (a series of articles appearing in), xv–xviii, 1960–1962. In this case, see the map in XVII (1962), 415.Google Scholar

¹⁶ Dalby, D. and Hair, P. E. H., ‘A word from a West African language of 1456’, Journal of West African Languages, IV (1967, to appear).Google Scholar

¹⁷ Westermann and Bryan; De Tresaman, map 8; Carta étnica.Google Scholar

¹⁸ On the Bidyogo and Biafada, see the references in the previous footnote, adding the map in Boletim Culturalda Guiné Portueugsa, XVI (1961), 105. On the Balanta advance, see the map in XVII (1962), 225,Google Scholar and Wilson, W. A. A., ‘Outline of the Balanta language’, African Language Studies, 11 (1961), 139.Google Scholar

¹⁹ Two Venetian portolans in B.M. MS. Egerton 73, reproduced in Kamal, Y., Monumenta cartographica Africae, tome, fasc. 1 (1951), 1508, 1511.Google Scholar

²⁰ Add to the references in footnote 10, the map in Boletim Cultural da Guiné Portuguesa, XVII (1962), 225.Google Scholar

²¹ Westermann and Bryan; De Tressan, map 8;Google ScholarDalby, D., ‘The Mel languages: a reclassification of Southern “West Atlantic”’, African Language Studies VI (1965) 1–17 especially the map.Google Scholar

²² Koelle, S. W., Polygotta Africana (1854), 16;Google Scholar cf. the note by da Mota, A. Teixeira in Fernandes, p. 168, n. 156.Google Scholar

²³ Westermann and Bryan p. 14;Google ScholarDe Tressan p. 163;Google ScholarWilson, W. A. A., An Outline of the Temne Language (1961) 1.Google Scholar

²⁴ Wilson, W. A. A. ‘Temne, Landuma and the Baga Languages’, Sierra Leone Language Review, I (1962), 27–38; D. Dalby, op. cit.Google Scholar

²⁵ Almada stated, certainly incorrectly, that all the Sape peoples ‘understood each other’.Google Scholar

²⁶ de Sandoval, Alonso, Naturaleza … de todas Etiopes (Sevila, 1627), lib. 1, cap. 1, p. 7.Google Scholar

²⁷ Westermann and Bryan; De Tressan, p. 163 and map 8.Google Scholar

²⁸ On the Bullom and Temne, see Westermann and Bryan; Dc Tressan, map 8; the map ‘Language distribution in Sierra Leone 1961–2’, by Dalby, T. D. P. in Sierra Leone Language Review, 1 (1962), at end.Google Scholar

²⁹ On the Susu, see Westermann and Bryan; De Tressan; map ‘Language distribution…’, cited in previous footnote; Houis, M., Étude descriptive de la langue Susu (Dakar, 1963), map.Google Scholar

³⁰ Guerreiro, F., Relaçam anual das cousas que fizeram os Padres da Companhia de Jesus…, ed. Viegas, A., III (Lisbon, 1930–1942), 243–4.Google Scholar

³¹ Fyfe, C., A History of Sierra Leone (1962), 2;Google Scholar the account in Kup, P., A History of Sierra Leone 1400–1787, chapter 4, is unreliable.Google Scholar

³² Fyfe, C., Sierra Leone Inheritance (1964), 226 (referring to 1795);Google Scholar for a possible reference to the Mende before 2700, see Hair, P. E. H., ‘A bibliography of Mende’, Sierra Leone Language Review, I (1962), 39–6,, on p. 39.Google Scholar

³³ Hair, P. E. H., ‘An early seventeenth century vocabulary of Vai’, African Studies, XXIII (1964), 529–39;Google Scholar this paper discusses the ‘Kquoja’ material in Dapper, O., Naukeurige Beschryvinge der Afrikaensche Gewesten… (Amsterdam, 1668).Google Scholar

³⁴ Dalby, D. and Hair, P. E. H., ‘“Le laugage de Guynee”: an early sixteenth century vocabulary from the Pepper Coast’, African Language Studies V (1964), 174–91;Google Scholar this paper discusses all the early material on the Kru peoples and includes references to modern sources. On the peoples of the Liberian coast and hinterland, see Cartes, Ethnodémographiques de l'Afrique occidentale, no. 2 (Dakar, 1960).Google Scholar

³⁵ In footnote 5 on page 190 of the paper cited in the previous footnote it was suggested that Zeguebo might correspond to Gereabo: we now think that it more probably corresponds to Sikrekpo, given in Herskovits, M. J. and Tagbwe, S., ‘Kru proverbs’, Journal of American Folk-lore XLIII (1930), pp. 225–93, on p. 225.CrossRefGoogle Scholar

³⁶ The main sources for the eastern part of the Guinea coast before 1700 are as follows: Pacheco Pereira, op. cit.; Dapper, cited in footnote 33: Barbot, J., A Description of the Coasts of North and South Guinea, (1732) (such original material as this work contains was collected in the early 1680s or in 1699–1700): W. Bosman, A New and Accurate Description of the Coast of Guinea (Dutch original 1704), (1705). The last three sources are sometimes referred to jointly as ‘late sources’; Barbot plagiarized heavily from Dapper and Bosman.Google Scholar

³⁷ On the languages between Cape Palmas and Assinie, see Westermann and Bryan; Dc Tressan, maps 6 and 8. For Towerson, see Dalby and Hair, op. cit. 187, n. 3. On Quaqua, see S. Brun, Schiffahrten (1624), ed. S. P. I'Honoré Naber (1913), 26 (a voyage of 1614), and D. Ruiters, Toortse der Zee-vaert (1623), by the same editor in the same volume, p. 71. The quotation is from T. Phillips, ‘Journal of a voyage…’ (1698), Churchill's Voyages, VI, 213. Identifications in J. Clarke, Specimens… (1848), 94: M. Delafosse, Essai de manuel de la langue Agni (1901), 196.Google Scholar

³⁸ On the Alaws/Alares, see Figueiredo, M., Hidrographia… (1608), 54 (copied in Ruiters, op. cit. p. 71), and the map cited below in footnote 40. The Adaows appear on the map ‘L'Afrique et Lybie’, by N. Sanson, 1656. The modern Adyukru have Kru as their neighbours—does this have something to do with the last syllable of their name? Note that Dapper and later writers extended the term Quaqua to include the peoples eastwards up to Assinie (Dapper, 1668, second pagination, p. 61).Google Scholar

³⁹ The Veterez were reported by Loyer in 1701, see Roussier, P., L'éstablissement de l'Issiny (Paris, 1935), 178.Google Scholar

⁴⁰ Particularly valuable sources for the Gold Coast are the following: a map of Guinea by Luis Teixeira with much detail on this district, based on a Portuguese mission to the interior of 1573, printed in 1602—see Cortesāo, A. Z. and Da Mota, A. Teixeira, Portugaliae Monumenta Cartographica, III (1962), plate 362, pp. 67–9, and v (1962), 185: P. de. M [arees], Beschryvinghe ende historische verhael van het Gout Koninckrijck van Gunea (1962), ed. S. P. I'Honoré Naber (1912).Google Scholar

⁴¹ On the 1480 vocabulary, see Hair, P. E. H., ‘A note on de la Fosse's “Mina” vocabulary of 1479–80’, Journal of West African Languages, III (1966), 55–7. On Tower- son's vocabulary, see Dalby and Hair, op. cit. (1964), 188, n. I. On De Marees' vocabulary, see I. Wilks, ‘An early Twi word-list: a lexico-statistical analysis’, a paper presented at the West African Languages Congress of 1964—this argues interestingly that the vocabulary collected at Elmina is of a dialect closely related both to modern Ashanti and to modern Fante.Google Scholar

⁴² Variant forms are mainly from documents in Brasio, A., Monumenta mssionaria Africana: Africa Ocidental, vols. I-V, and in K. Ratelband, Vijf dagregisters van het kasteel São Jorge da Mina 1645–7 (1953), and from Dapper. We do not accept the identification of the Ati with the modern Attié of Ivory Coast suggested by R. Mauny (in Pacheco Pereira, p. 189, n. 253). The Cacres are rather puzzling: following the 1602 map (cited footnote 40), where the spelling is Caceros, we believe the pronunciation to be Kasere (not Kakere): on the map they are credited with marauding activities around Axim, but ‘Cacers Anguines’ are also shown across the upper course of a river, probably the Camoe, in which case apparently near modern Asikasso, where the languages spoken today are Anyi/Akan and Baule/Akan. The map at least makes it unlikely that the Kasere were one of the interior non-Akan peoples with a similar name, e.g. Kasele/Gurma of Togo, Kasene/Grusi of North Ghana.Google Scholar

⁴³ Brasio, op. cit. III, p. 110.Google Scholar

⁴⁴ Dapper, 1668, second pagination, p. 107. The whole of Dapper's statement is diflicult to follow, since it begins by asserting that between Koromantijn and Akara four separate languages were spoken, an assertion contradicted by the rest of the statement. The French translator found the same difficulty and omitted the reference to four languages.Google Scholar

⁴⁵ On Akan and Ga-Adangme (and Ewe) see Westermann and Bryan; Manoukian, M., Akan and Gã-Adangme peoples of the Gold Coast (Ethnographic Survey, 1950); note that De Tressan, on map 6, wrongly includes Afutu/Akan in Gã-Adangme.Google Scholar

⁴⁶ Roussier, op. cit. p. 187;Google ScholarDelafosse, op. cit. p. 198, identifies Loyer's vocabulary as Agni.Google Scholar

⁴⁷ Ryder, A. F. C., Materials for West African History in Portuguese Archives (1965), 15 (document of 1517), and Brasio, op. cit. 111, p. 92.Google Scholar

⁴⁸ Wilks, I., ‘The rise of the Akwamu empire 1650–1710’, Trans. Hist. Soc. Ghana, III (1957), 99–136, on p. 134, n. 64. On p. 112 of the same article the Adangme–Ewe frontier in the seventeenth century is put slightly west of the Volta.Google Scholar

⁴⁹ Newbury, C. W., The Western Slave Coast and Its Rulers (1961), 2–3;Google ScholarCornevin, R., Histoire du Dahomey (1962), 241; many early—and mid-sixteenth-century maps show ‘Alhandra’ or ‘Alhadia’ on the Coast east of the Volta—it is uncertain whether this is a Corruption of a Portuguese term (? ‘aldea da’) or a form of ‘Ardra’. Poupous appears as the name of a country or people between Mina and Benin on a map of 1561, Cortesão and Teixeira Da Mota, op. cit. 11, 100.Google Scholar

⁵⁰ See Labouret, H. and Rivet, P., Le Royaume d'Arda et son évangélisation au XVIIe siècle (Paris, 1929).Google Scholar Westermann said of this work that it showed ‘that the language has scarcely changed in the last three centuries’—quoted in Armstrong, R. G., The Study of West African Languages (Ibadan, 1964), 12 (as part of the argument that ‘there has been a tendency to seriously underestimate the time-depths that underlie the present arrangement of languages and groups of languages in West Africa’).Google Scholar

⁵¹ Sandoval, op. cit. lib. 1, cap. 1, p. 7. On Locumi, see a review of a ‘Vocabulario Locumi (el Yoruba que se habla en Cuba)’ in Bull. I.F.A.N., B, XXI (1959), 577–9.Google Scholar On Yoruba, see Forde, D., The Yoruba -speaking Peoples of South Western Nigeria (Ethnographic Survey, 1950), map.Google Scholar

⁵² See Bradbury, R. E. and Lloyd, P. C., The Benin Kingdom and the Edo-speaking Peoples… The Itsekiri (Ethnographic Survey, 1957), maps.Google Scholar

⁵³ On the Ijaw, see the maps in Bradbury and Lloyd, op. cit.; in Forde, D. and Jones, G. I., The Ibo and Ibo-speaking peoples of South-Eastern Nigeria (Ethnographic Survey, 1950);Google Scholar and in Jones, G. I., The Trading States of the Oil Rivers (1963). Note that when Pacheco Pereira referred to a country called Opuu ‘one hundred leagues’ up the Forcados River, this probably meant, not the Nupe as Mauny suggests, but the Ijaw, opu-forming part of many Ijaw toponyms, e.g. Opobo, and Ijaw country being approachable from the west via the Forcados.Google Scholar

⁵⁴ Sandoval, op. cit. lib. 1, cap. 1, p. 59. The list is as follows: Ambo (?), Abolamo (Abolama, near Bonny), Bila (Belli in Dapper, Bile), Cubai (? Kugbo, near Nembe), Coco (?), Cola (Kula), Dembe (? Nembe), Dare (?), Evo (Evo/Sobo), Ibo (see next paragraph), Ido (Ido, near Buguma), Mana (?), Moco (see next sentence), Oquema (Egwema, near Akassa), Ormapri (? Orupiri, near Bonny), Quereca (Krike in Dapper, Okrika), Tebo (? Obiatobo), Tequo (?).Google Scholar

⁵⁵ Oldendorp, G. C. A., Geschichte der Mission der Evangelischen Brüder auf den Caraibischen Inseln S. Thomas, S. Croix and S. Jan (Barby, 1777), table of vocabularies; [H. Kilham], Specimens… (1828);Google ScholarClarke, J., Specimens… (1848);Google ScholarKoelle, S. W., Polyglotta Africana (1854), 11.Google Scholar A source of 1798, cited by Debien, G. et al. , ‘Les origines des esclaves des Antilles’, Bull. I.F.A.N., B, XXV (1963), 5–38, 215–65, inaccurately, or at least very loosely, included Mokos among ‘les nègres du Benin’.Google Scholar

⁵⁶ On the Ibibio, and the Ibo, see the map in Forde and Jones, op. cit. ‘Moko’ may derive from ‘Mboko’, a name attributed by a section of the Ibo north of Bonny to a Section of Ibibio, allegedly their part ancestors—Forde and Jones, op. cit. A less likely derivation is suggested in Jones, op. cit. 32. A vocabulary of the ‘Old Calabar language’ was supplied by Barbot, but unfortunately turns out to be mainly pidgin Portuguese; traces of Ibibio can, however, be found in it—Jeffreys, M. D. W., Old Calabar and Notes on the Ibibio Language (Calabar, 1935), 34 (which perhaps underestimates the amount of Ibibio).Google Scholar

⁵⁷ Cortesão and Teixeira Da Mota, op. cit. v, plate 557.Google Scholar

⁵⁸ Da Mota, A. Teixeira, Topónimos de origem Portuguesa na Costa ocidental de Africa (Bissau, 1950), 306–7. The text of Pacheco Pereira's account is corrupt and reads as if the mountain was on the island.Google Scholar

⁵⁹ On the modern peoples and languages of this region, see Westermann and Bryan;Google ScholarBryan, M. A., The Bantu Languages of Africa (1959);Google ScholarRichardson, I., Linguistic Survey of the Northern Bantu Borderland, 11 (1957), map;Google ScholarArdener, E., Coastal Bantu of the Cameroons, (Ethnographic Survey), 1956.Google Scholar

⁶⁰ In the original ‘hooger de reviere op’ (p. 137), whereas the French translation totally misleads with ‘loin de la côte, près de la source de cette rivière’ (p. 306). We suspect that by Rio del Rey the source means not the river normally known by this name, the Ndiang which flows from due north, but the River Masaka which flows from the east (i.e. from behind Cameroons Mountain) and emerges into the same general estuary.Google Scholar

⁶¹ The prefix ka- is not found in north-western Bantu, according to Johnston, H. H., A Comparative Study of the Bantu and Semi-Bantu Languages, 1 (1919).Google Scholar

⁶² For example, Embours—Jean Alfonce Sainctongeois, Les voyages avantureux, (Poitiers, 1559), f. 54; Ambous—Cortesão and Teixeira, op. cit. 11, 100, map of 1561; Zambus—Figueiredo, op. cit. 59; Cambo/Ambo/Amboises—Dapper.Google Scholar

⁶³ Ardener, op. cit. 22.Google Scholar

⁶⁴ It is perhaps worth pointing Out that lists of Guinea coast peoples appeared in well-known seventeenth-century works. It is true that the fullest list is in Sandoval (1627), a work not generally available and seldom noted in bibliographies of Africa (we owe our own knowledge of the work to references in C. R. Boxer's writings on Brazil). But a list covering the western part of Guinea appeared in the Jesuit letters edited by Guerreiro, whose French version by Du Jarric was extensively quoted in English by Purchas and in Dutch by Dapper—for instance, the peoples are named in sequence in a few paragraphs in Purchas, Book 5, pp. 710–12. It is curious that this list has not previously been compared with the modern inventory. However, we understand that the ethnolinguistic continuity in Portuguese Guinea has been noted in a work we have not been able to consult— A. Teixeira Da Mota, Guiné Portuguesa (1954).Google Scholar

⁶⁵ The only names unidentified are a handful in Sandoval's list (where no geographical locations are given); they are most probably either the names of constituent units (e.g. of settlements of Caravali) or of interior peoples.Google Scholar

⁶⁶ The term is Dr David Dalby's, in an oral communication. Examples of works to which the writer would apply the term—Cartes ethno-démographiques de l'Afrique occidentale, 11 (1960),Google Scholar historical map: Person, Y., ‘Les Kissi et leurs statuettes de pierre dans le cadre de l'histoire ouest-africaine’, Bull. I.F.A.N., B, XXIII (1961), 1–59:Google ScholarKup, P., A History of Sierra Leone 1400–1787 (1961);Google ScholarCornevin, R., Histoire des peuples de l'Afrique Noire, 2nd ed. (1962).Google Scholar