OFHUNA DURING THE NINBTEBNOH CWEOHF Li«S. Xropp Dakubu* In the following notes I attempt to address two questions that a^# basic to a con s^L deration of the conduct of verbal communi cation within and 1?et ween communities* The first one i s, what languages were spoken,, and what were their home territories? Or, more specifically, have the boundaries betweai language comnrunities shifted, before during or after the nineteenth century? The second question i s, what language or languages were used for communica- In approaching these ticn between linguistic communities, and at what level? questions I shall take the term "coast" very l i t e r a l l y, considering boundaries only where thqy meet the sea* 1 • Linguistic boundaries Today, the languages spokea along the ooast from west to east are: Nzema, Ahsnta, jseite, Awutu, Ga, Danigme? Jlwe. Thi-s ba«? beer! the g^ieral picture at least since the eignteeith century• Prottai (i764) listed the languages of the coast west of the Volta as nPantew, «Haata«, "Obutu", "Acra" aad " A d p p^ abd beyond the Volta, "Crepe"* Bowcich (181-9) gives the same distinguishes Fuema fxom iiiianta. His l i st is "Amsnahea" (Nzema), '»Fantee5', nAffoctoo'% Accra, locations of -';e boundaries between these languages do not seem to have changed radically curing the past tw^ hundred years, at least en the coast» but there has be.kh a certain amount oi drifting, -which' seems to reflect a. gradual expansion of ibe t e r r i t o r i es of the languages of the major centres of commerce and p o l i t i c s! i-e- of Pante and Ga, a& the languages of Elmina and Gape Coast and of Ac-Wa respectively • It would seei that t h is expansion Associate Professorf Ins Vitute of African Studies* 193 proceeded very slowly if at all during the f i r st part of the nineteenth century* but picked up considerably during the second half, and is perhaps s t i ll proceed ding today. Although this is t think a fair picture of the general trend, it is difficult to be precise about where the boundaries actually were (or even are) at any givda time* £here are several reasons for this* Most writers* whether their main interest was trade or evangelization, were most interested in the languages that were most generally useful* So when Christalier says (1881, Introduction) that Akan is spoken from the Tano to the Volta he does not mean that it is the only language, or even necessarily that it is the f i r st lan- guage of the majori-ty* sit any given spot between those boundaries* .Another reason ishy it is difficult to be precise about boundaries is that they do not generally occur at the major centres, about which everybody wrote, but at obscure villages between major centres* Even today, I find* although the general picture is common knowledge, precision requires field work* ' There is also a problem of time lag in observations* Ttien I speak here of shift of a language boundary, or expansion of the territory of a language, I mean a change in the language that children leaaii f i r st in l i fe and oontinue to use as their primary language within their native conmuni-ty, in the same geographical location* Such a change would not likely be noted by observers •until the children in whose generation it occurred were fully adult, and possibly not evai then if their reports were second hand, as they oftoi were in tlie case of more isolated places* There is also the fact that the name of a state may also be the name of a language, but the territories of the language and of the state do not invariably coincide* With these problems in mind I now remark on what is known of particular boundaries shortly before, during and after the nineteenth century* 194 Bowdich (18^9) said that the people of Axim speak a dialect of Ahanta. I gather from He definitely distinguished Ahaata from the Kzema or "Amanahea" language, which he implied was spoken on the other side of the iaakobra. Van Daatzig (1977s64) that in the eighteenth century 5 Prince's Tom was regarded as Ahanta. According to Mock (i9^9) the traditional ethnic or political "boundary is the Ankobra, but the language of Axim? knowi as Bv.aLues is quite definitely a dialect of Bzema. She also motions Ajemra ag a transitional dialect between fizema and Ahanta, but more like Nzema. I am told that Prince's Tom is Nzema-speaking today (p.A. Dolphyne, personal comnujiication)s but I haye no information on places betweaa there and Dixeove8 which is s t i ll Ahaaatar-speaking- eastward at the expense of Ahanta since the beginning of the nineteenth cm tuny. It would seem thai that Nzema has expanded Die eastern boundary of Ahanta, with Ptote, has probably moved west- wards so that the portion of the coast *«here Ahaata is spoken has contracted on both sides* In his chapter on Languages Bowdich maationed Sekaidi as the f i r st torn in Ahanta country, moving east to west, and also mgationed Takoradi, Butri, Busua and Dixco>se» According to Mock the territory of Ahanta extends as far east as the Pra, and she mentions Dixcove, Sekoadi and Eg/am but not Shama, but does not commit herself as to whether lihanta is s t i ll spoken there * I am informed (J»B^ Abban, personal cottinunication) that although Ahanta traditional territory extaids as far as the Pra, the most easterly Ahsnta- speaking towi is now Apowa, four or fire miles west of Takoradi. This has apparently beaa the case for at least fifty yearss but it is possible that Sekondi was Ahanta-speaking th-iKJughout the nineteenth century. Migeod C 1911s38) spoke of "Ahaita, extending from between-Sekondi and Ghama as far as and iaacluding Aximw. Migeod is not always entirely reliable? hut in he was working in Sekqadi aid may have had f i r st hand information. 1-95 Ttie eastern boundary of Faites with Awutu, seems to have been stationary since the eighteenth century ° Prottai moi timed Obutu without saying where it was spoketts and listed "Tiriba" (Winneba) and "Afutu breku" among the states where Fante was the mother loigue* Since it is hardly likely that Winneba and Senya shifted from Fante to Awutu during the following two centuries, I interpret this to mean that in the mid-eighteenth century those tovais were already strongly bilingual. Perhaps Winneba had a Fante-speaking section. Bowdich specified that the people of Simpah or Winnebah were "Fantees", but that t h e ir language was called "Affootoo*. He then mentioned Senya, but not i ts language, ffliere are intimations, which I find convincing, that a Guang language, possibly Awutu, was spoken all along the Fante-speaking, coast in earlier times* but plainly before the nineteenth century it had disappeared west of Winneba, Winneba and Senya are of course Awutu-speaking today, as is the village of Amitu or Obutu just inland about eight miles to the north- east, but Fete aid Nyanyano, m the coast immediately east of Senya, are Fante-speaking (G«P. Hagan, personal communication). Eradit ion ally, Ga extends westwards to the Sakumofio lagom, at the moutii of the Ban su, but today it apparently extends to the My any a lagocn. The eastern boundaiy of Ga» with Dangme, is relatively well documented, and has certainly shifted eastwards. Writers up to and including Zimmermann (1858) regarded fema as Dangme-speaking. P. de M (arees) (16O2) gave numbers from 1 to -jO in what is evidently an early form of Dangme collected from "Ohiaacke", probably Kink a, a vanished settlement near Tema. Qn the other hand, Protten (1764) claimed that Ga was spoken as far as "Tema, and perhaps Ningo or Ls|jrw. Since the l a t t er two are in country that is now Dangme- speakingr, and since Protten does not seem to be speaking from first-hand observation, it is likely that he meant that that vra-s as far as Ga was known as either first or second language, in contrast to Fante9 which was known much further. Bowdich1 s information is a ll quoted from I s e rt ("1788), and does not mmtim Teffla or any other towi west of Ningo, vrtiich I s e rt said was 196 Dangme speakings but I s e rt himself .gave cne word, the name of a fish "hardis" i»e» _a&i which appears to "be Dangmes not Ga, and apparently came from Kpane (see Dakubu 1969). Ziinmermann claimed that Dannie was the language as far west as Mungua, •which of course was originally founded by Ningo people« He f i r st stated (-1858; VIII) that Ga was spokm in Ga? Kinka i.e« Dutch Accra, Osu, Laa and Teshie, and that Adangme was the language of Nungua, Tema, Kpoae, Gbugbla, Ningo axxd Ada- Later however (385) he omitted Nungua, which -nay alrea<%r have become Ga-speaking, since it is much closer to Teshie -*Jhai to Tema, but he is quite definite about Tema* He goes en to say (?86) that "the difference betweai the Adanme of Tema t il Nurio en the cne and Krobo an the other hand is nearly as great as that between the former and Ga proper, the limits of both being uncertain....M But Field (1936) stated unequivocally that Kpane was Ga-speaking in daily l i f e, aad regarded Tema as Ga without questioning it* The way in which this shift from Dangme to Ga in Nungua aad Tema came about can be deduced from the situation in Kpcne todqy» Kpcne has two main section Sj me (Jo Shi) that is traditicn ally of Dangme origin, • aad. s t i ll uses Daigme for some customary observances, and another, much larger secticn (Alata) which has no Dangme traditions but is said to have been founded later than the first, by Gas and other nm-Dangmes? and has always beaa Ga-spaaking. The two sections have intermarried a great deal. Evidently there must have been a time whan Dangme was the f i r st language in J_o Shi, Ga the f i r st language in Alata, and communicaticn between the two took place more often in Ga thai in Daagme^ Under ttie combined pressure of the greater number of Grv speakers in Alata and the economic and cultural dynamism of the Accra orbit, the Dangme-speakers of the torn became totally bilingual in Ga, -aid eventually came to use it as the home language, so that now it is Daagme that is the second language. Something like this must have happened earlier in Bttngua end 197 Tema.. It may even already have been far advanced in Zimmermann' s time, althou^a he and his informants were not aware of i t. Field (1936; -jO) refers to fema as "Ihis isolated l i t t le town»«..»» In Kpane however it seems that the shift was not complete until sometime in the second half of the nine- teenth century, or evei early in the twentieth. Pield (1936:77) spoke of Adaagme having been "spoken generally in Kpoig t i ll within fairly recent years11* '" I s e rt said that Mangme was spoken as far as Ida, and gav-@,a phrase he collected in Ada that is indeed in Daigme (1788;44> see also Dakubu op«cit«)» It is thought that Ewe-speaking people preceded the .Datigme en the western side of the Volta, but throughout the nineteenth century they were apparently not pexmanaitly ox the coast i t s e lf in any numbers, or if they were they were soon linguistically assimilated to the Dangme* Sources are unaaimous that the language east of the Volta was Ewe, usually referred to by some versicn of the name •Krepe**' I s e rt noted that much Dangpie was also to be heard in Aflao, due to much Dangme migration eastwards* There are s t i ll linguistic traces of these migrations farther inland., for example in Sjg-Zogbadji in Togo, but they seem to have ceased during the nineteenth century and any Dangmes that settled along tie i»e- speaklng coast have been assimilated* One or more Guang languages have also been spoken in the past en the Ga aid Baagsne sectiois of the coast* Daere is very strong evidence for this* She people of Larteh, according to tradition, were earlier established along with thoir language at labadi. Pield (1962) reported an Awutu traditiaa of earlier Residence near Ningo, and mgntioaed good place-name evidence in support of i t. Today, any southern Guang-speaking communities have either moved out Cr beaa assimilated linguistically, and it would seem that in general this had already happened before the nineteenth century. However, Bowdich (1819) published a short "Adainpe*1 word l i s t, which I have shown 198 elsewhere (Dikubu 1969) to be at least partly Guang, and which apparently came from- somewhere between Ningo and Ada* Oils is evidence that early in the century some Guang speakers were s t i ll to be found there, but the fact that Bowdich's nAdampe" numbers ?/ere indeed the modem Dgigme ones is evidence that the shift to Dangme was in progress. There is no doubt that in the nineteenth century Fante was the largest coastal language? both as a f i r st and as a second language, as it is today• Zimmermann (1858s VIII) estimated the numbers of speakers of Ga as 4° to 50,000, and of Dangme as 5O to 60,000. The l a t t er figure was probably based primarily on the inland Krobo and Sh&i areas, with which he was more familiar thai the coast* A few years later, Ghristaller and others (quoted in T-rutaaau 1973) estimated Ga at somewhat less, thirty to forty thousand, Daigrae at rather more, over eighty thousand, and Akan, including Asante, Akyem and Akwapem as well as Fjfite, at three or four million. The relatively high figure for Dangme was surely based on Krobo. It may be noted that the coastal towns where the shift from Dangme to Ga occurred are (or were) quite isolated geographically from the bulk of the Dan gme~ speaking population* Today, Ga and Dangme are both estimated at around half a million, and Akan at five or six, a proportional increase for Ga that is probably due to accelerated development of the area after the close of the nineteenth century. Jrorn the seventeenth caitury onwardsj the coast of Ghana displ^red the trilingual pattern that has come to dominate modem Africa, whereby cue finds the local .language used at home and within the community; an African, language of wider communication used with neighbouring peoples who do not speak the local language; and a "world" language for communication over an evsi wider radius but in this case specifically for commercial and diplomatic dealings with visiting Europeans- The African language of more than local application may be used locally for special purposes, and where the wider African language 199 is also the local language, the two functions are filled by me language* In the preceding section I sought to establish the territorial'extent along the coast of the sevsi locally used laa,guages» There is abundant evidence that well before the nineteenth century, Fante was established in the function of the language of external communication at an intermediate range- Prottea, for example, stated that Fante was known from Axim to the Tolta and MCrepe or Popo" and Christaller said essentially the same thing. Todqy, Fante is apparently universally knom as a second language in the Awutu area, and in Ahaata aid most of Hzema too, but in the Ga-Daagme . speaking .-areas* although Akaa is widely knowi as a second language it is not so nearly universal as in the areas that immediately border the local Fante area, and there appears to have bean a shift in the predominant dialect, to those of Akyem aid jSkwapenu *Ehis shift probably occurred in the l a t t er half of the nineteenth omtury, with the development of trade between the Ga coast aid the country iimiediately north of i t, and the decline Of -tine seaborne trade based ax forts dotted along the coast• 51x8 lower degree of acquaintance with Faite or other varieties of Akax Is undoubtedly also related to the fact that the language of Accra gave it some competition. Pxott-gn stressed the greater spread of Fante, but admitted that Ga was knom as far as Ningo or even Lay. The fact that Tetta aad Kpone switched to Ga as f i r st language is also evidaice of a wide knowledge of Ga as a second language in those places during the ninetemth century- Todsgr it is the major second language at least as far as Ada. I ts spread was no doubt givga additional impetus by the decision of the Basel Mission in the middle of the nineteenth caatury to use Ga for evangelism and education in Dgngme- speaking territory. major The language of overseas trade today of oourse is EngLish, but in the eighteaiifa century it -was Pidgin Portuguese, as it had bem since the beginnings of this trade- Protten refers to it as »Heger-Portugisiskn, knovoi 200 l i ke Fante a ll the way to popo, and does not mention Baglish at a l l, althou^i that language mist have been in use to some degree where there were English forts* Bowdich, on the other hand, does not mention pidgin Portuguese* This does not of course mean t h at it was not spoken, but it may "be an indication that in the Gape Coast area at l e a st it was by thai n ot important. During the f i r st half of the century the Basel missionaries plainly regarded i h g l i sh as the language of the future? and by the end of the century Pidgin Portu- {rioce had completely disappeared. References Bowdich, T»E. Ghri st all er, J. G • Dakubu, M«E»K» Field, M. Field, M. Isert, P.E. P. de M(arees) Migeod, P-T/.H. Mock, C«C« 1819 1881 1969 1936 1962 1788 16O2 1911 1969 A Mission from Gape Goast Gastle to Ashanteeo London. Dictionary of the Asaate 9nd Fante Laagua^ called Tshi. Bowdich1s "Adampe" Word L i s t. Hesearch Review, Vol.% Mo. 3 PP- 45-9 Religion and Medicine of the Ga r .'c;,:^ G • Lai dca . Awutu-Bereku. Bawjiase* Eeise nach Guinea. Copaahagen. Beschryvin^he ende Historische Terhael van het Gout Koninckrijck van Gunea.. Edition of Naber 1912» The Lsn^uages of West Africa Vol. 1* London. The grammatical u n i ts of the Nzema language; a qystendc analysis* Ph*Do t h e s i s, liiiversiiy College London • 201 1764 1973 1977 Pro tt en 9 C • Trutgnau, H»M.J. Dantzig, A» Van Zimmerman nf J» 1858. Eta n y t t ig gramnati.calsk. indledelse t il tymde h i d i n d t il ubekigidte Sp.ro g? Pan teiak og Aci^-isk... A h i s t o r i c al note en the size of the Akan, Ga and Daigme speaking popula- tions in Ghana. Mitteilxxngen der Bagler Afrika Bibliographigi $. pp. 34-7. South-westea» Ghana and the Akan- speaking areas of the Ivory Coasts a survey of the h i s t o r i c al evidence^ .Pro oeedin gs of ..the Seain arri ioa ; Ghan^ai an. Historiography. and^Historical Beseardh, ed» J»0» Htmwick» legen- A Grammatical^.Sketch of the Akra- or ^Ga-