Raids and Refuge: The Bulsa in Babatu’s Slave Wars
Research Review (Institute of African Studies, Legon), New Series, vol 24, 2 (2008), p. 25-38
.
RAIDS AND REFUGE THE BULSA IN BABATU’S SLAVE WARS
[endnote 1]
.
Franz Kröger
.
.
Abstract
From the 18th century the Grunshi, a group of peoples in the north of modern Ghana, suffered from the slave raids of centralised states, their southern neighbours, especially the Dagomba. Under Babatu (d.1807), the leader of Zabarima warriors, slave raiding reached its peak. This paper attempts to throw light on the defensive strategies of the Bulsa (Builsa), an ethnic group living southwest of Navrongo (U.E.R.), in battles and also on their habit of retreating into caverns when attacked. An examination of these caverns proved that, at least in the Bulsa area, a particular type is predominant. This consists of a secluded natural cavity, which can be entered only through a shaft and a horizontal tunnel. It could be demonstrated that several caverns of the Bulsa and the Koma, their immediate neighbours, are today either earth shrines themselves or closely associated to such.
.
Resumé
Depuis le 18ième siècle, les Grunshi, une ethnie du nord de l’actuel Ghana, souffraient des chasses aux esclaves effectuées par des Etats centralisés, et tout particulièrement par les Dagomba, leurs voisins du Sud.
Cette étude examine les stratégies des Bulsa, un groupe ethnique au sud-ouest de Navrongo (U.E.R.), aussie bien leurs batailles que leurs habitudes de se retirer dans des cavernes lors d’une attaque. Un examen de telles cavernes a eu pour résultat qu’un certain type de cavernes prédomine au moins dans la région des Bulsa. On ne peut entrer de l’extérieur dans ce type de cavernes naturelles que par un puit et un tunnel horizontal. On a pu démontrer que plusieurs cavernes des Bulsa et des Koma, leurs voisins directs, sont elles-mêmes soit des sanctuaires, soit des endroits étroitement liés à un sanctuaire.
Eighteenth and nineteenth century reports on the conquest of African villages in the course of slave raids and by colonial powers are usually seen through the eyes of the colonialists. Even in recent publications by African authors, a certain European influence cannot always be avoided, since the written primary sources on specific issues such as historical conflicts are often of European origin.
Assailed by superior enemies equipped with fire arms and horses, the indigenous population often fled so early that the aggressors never even saw them. In default of adequate sources about the reaction of the people attacked by superior powers from outside, the situation of 1902 (i.e. five years after the end of the slave raids, when a British punitive expedition invaded the Bulsa area) [endnote 2] is described in this paper. We may surmise that the local population used strategies of flight at the hint of danger rather than confrontation in the decades before this date [endnote 3]. The reasons for the expedition are summed up by the British Governor, Major Nathan, in a letter to Mr Chamberlain, the Colonial Minister:
Lieutenant-Colonel Morris acted judiciously in confining his operations to what was necessary to secure the main purpose of the expedition, viz.: the punishment of those tribes who made a practice of raiding and ill-treating their more loyal and peaceful neighbours…
When, in March 1902, Lt. Col. Morris and his troops entered Sandema, perhaps the most important Bulsa settlement, this village was completely empty of people.
(Diary, March 21st, 1902) …At 7.35 a.m. the outskirts of Sinlieh [Sandema] were reached (6 miles), and the compounds were found abandoned…
(22nd March 1902) …I again went out to Sinlieh from Tiana [Chana] with the same force as yesterday… with the intention… of destroying as many compounds as possible. No signs of the enemy were seen, the whole district, about twenty square miles, being abandoned (endnote 4).
As the British troops were keener on demonstrating their military superiority and breaking any potential armed resistance than tracing out unarmed people in their hiding-places, the Bulsa refuge caverns were apparently of no significant interest to the British in these colonial encounters. The situation was completely different at the time of the slave-raids, when Zabarima slave raiders haunted those inhabitants of the Northern Territories of the Gold Coast, who, as stateless societies, were generally classified as ‘Gurunsi’ or ‘Grunshi’.
Although slave raids among the Bulsa were also documented before those of the Zabarima [endnote 5], the following historical survey is confined to the activities of Babatu, leader of the Zabarima warriors between 1885 and 1897.
The Zabarima (Zambarima, Zarma, Djerma), a branch of the Songhai people, initially inhabited the region of Niamey (modern Niger). According to Holden (1965, p. 60), “in the early 1860s a small group of Muslim Zabarima horsemen came to Dagomba as traders, mercenaries or malams or perhaps all the three”. As mercenaries, they raided Grunshi settlements to provide slaves for their Dagomba overlords, before they decided to carry out slave raids on their own behalf and establish themselves first in Kassana (Kasena land), then in Seti or Sati (24 km south-west of Léo, today in Burkina Faso). Although there had been raids under Babatu=s predecessors, he “brought the Zabarima state to its peak of power and prestige” (Holden 1965, p. 61) and “by the late 1880’s, they… had conquered and were probably continuously controlling an area stretching from Ouagadougou to Wa” (p. 60).
Using Malam Abu’s report [endnote 6] and the results of modern research, we can distinguish two main campaigns against the Bulsa in the 19th century: In the first, Babatu set out from Seti (his main residence) in about 1890/1891 and, according to Holden and Malam Abu, started attacks on Bulsa villages. The second campaign began perhaps in 1896 with the conquest of Kanjaga. Babatu stayed there for quite a long time, according to Tamakloe (1931, p. 54) “for some months”, and used this village as the launching stage for raids on other villages. It was also at Kanjaga that at last he was beaten decisively. Among the Bulsa of today the battle of Sandema, in which a united force of Bulsa defeated Babatu, seems to play a greater role in their historical memory than the battle of Kanjaga.
In August 1973 the late Sandemnaab Azantilow (d. 2006) gave me an account of the most important events of the Bulsa-Zabarima confrontation.
For about three months the Bulsa had suffered the attacks of Babatu. Then, in a battle near Wiaga [endnote 7], many warriors of that village were taken prisoners. Others rushed to Sandema and Babatu followed them. In the battle near Azagsuk [a hill in Sandema Fiisa] Babatu was beaten. In this battle all the Bulsa were united; warriors from Fumbisi, Kunkwa and even Navrongo fought together. The people of Navrongo had fled to Kolugu, but when the chief of that village could not help them, they came to Sandema.
The battle of Sandema is mentioned in some published and unpublished historical accounts. Cardinall (1920, p. 10-11), who presumably interviewed Babatu’s contemporaries, comments:
Babatu meanwhile had continued the annual raiding, sustaining a severe defeat at Sandema from the temporarily united Builsa [Bulsa]. He was forced to recross the Sissili River [for Seti? F.K.] to recuperate, and on his return met the French forces…
Ollivant (1933, p. 6), District Commissioner of the Navrongo District, collected an account told by Mamadi Kantosi, Babatu’s horse boy:
He [Babatu] went back to Seti and recruited more men, and came back on several occasions and for many years invading Nakon, Chiana, Paga, Navoro [Navrongo] and then into Builsa country where he was again beaten at Sandema and again returned to Seti. At this time there was a quarrel between Babatu and his two generals, Seti and Amarea.
We owe an interesting and detailed account of the battle of Sandema to P. Perrault (1954, p. 94), based on “tales of the battles against Babatu” collected in several villages:
…it seems that the Builsas were the only ones to inflict on him a serious defeat.
Babatu had raided many Builsa villages in the South (around 1896). The people were running in front of his army. They finally decided to meet in Sandema and there to defeat the enemy or die. Sure of the outcome Babatu attacked Sandema. The local people were spread between the market and the Primary School. They had learned that the Dane guns were dangerous but that Babatu’s soldiers were powerless while they were reloading…
In the above quoted passages about the battle of Sandema there are some remarkable discrepancies, especially concerning its location. While in the Sandemnaab’s account, the battle took place in the north of Sandema (Fiisa), Perrault mentions a battlefield between the market and the [Old] Primary School, i.e. in the south of Sandema. If Babatu’s warriors entered the village after some encounters with Southern Bulsa, and if, according to a Sandema tradition, the chief of Sandema had even been leading a “Builsa army against a Zabarima invasion at Giadema, Uwasi, Fumbisi and Kanjaga” (Holden 1965, p. 75, fn 70), a battlefield south of Sandema seems to be more probable.
This is in agreement with a Wiaga tradition as rendered by Apajie from Wiaga-Tangkansa [endnote 8]:
After the conquest of Kanjaga, the Zabarima wanted to proceed to other villages of the Southern Bulsa but were afraid because of the very big rivers [?]. Then they made for Wiaga but, before they arrived there, the people had fled to Sandema for refuge. The people of Wiaga were armed before they reached their brothers at Sandema.
Babatu and his large army felt strong enough to attack the Sandema people. When the message arrived that Babatu and his large army were coming to attack Sandema, all the people who had come there for refuge waited no longer, but, with the help of the white men, attacked them on their way from Wiaga.
This great war between Babatu and his army and the people of Sandema, with the help of the white men, took place at “Akumcham”, an area within [Sandema-] Suwarinsa [endnote 9].
In 1973 I interviewed Anyenangdu, then the oldest man of his clan Wiaga-Sinyangsa Badomsa. As a small child he had experienced the raids of ‘Anaatu’, as he called Babatu. According to him, Babatu reached Sandema by way of Bachonsa, Doninga, Kanjaga and Wiaga.
The tradition of a battle to the north of Sandema is still very much alive in the memories of Fiisa people, and Zabarima rifles, booty of that battle, are still shown in one Fiisa compound. Also Ollivant states that, before coming to Sandema, Babatu had invaded villages to the north and east of Sandema (Nakon, Chiana, Paga, Navoro [Navrongo]) and “… then [he moved] into the Builsa country where he was beaten… at Sandema” (Ollivant 1933:6). In this case, Fiisa would have been on Babatu’s route to the centre of Sandema.
The fact that people from Navrongo sought refuge in Sandema (cf. Azantilow) may be explained by their defeat in Babatu’s campaign against these villages before he entered Sandema. In the course of many years the place and time of possibly two battles at Sandema might have become mixed up in the memories of the Bulsa.
The exact year of the Sandema battle is also disputed. Most notes on this battle do not give an exact date. The terminus ante quem is March 14th, 1897, the day when Babatu was beaten at Kanjaga. According to Azantilow’s account, there was no great time gap between the Sandema and Kanjaga battles, for immediately after the Zabarima had been beaten in Sandema, French troops approached Kanjaga from the north via Leo, Preta, Basisa, Nakong, Bachonsa and Doninga. Thus, for the Sandema battle, we might surmise a date around 1896. This is also the year mentioned by Perrault (p. 94).
On the other hand, a battle of Sandema might have taken place during Babatu’s first campaign against the Bulsa. We are informed by Malam Abu that Babatu attacked and conquered Kajagâ (Kanjaga), Hubisi (Fumbisi) and Kukuwa (Kunkwa) and then fought against the people of Nafaro (Navrongo), Faga-Vuru (Paga-Buru) and Zûzûlo (Chuchuliga) [endnote 10]. Holden, too, was told in Sandema “that the Zabarimas came to Builsa after conquering Paga” (p. 75, fn. 71). As for the time of this campaign, he was informed in Wiasi that Babatu conquered Wiasi and Kunkwa “four or five years before the whites came to fight the Zabarimas, i.e. about 1891” (ibid.). This does not mean that Perrault’s surmise of a battle south of Sandema in 1896 is wrong, for at that time Babatu had started his second campaign in Bulsaland, including perhaps a second unsuccessful encounter with Sandema.
Our knowledge, so scanty with regard to Babatu’s earlier raids and battles in Bulsaland, is more extensive concerning the final battle of Kanjaga [endnote 11] on March 14th, 1897, in which Babatu and his warriors were beaten by the united forces of the French under Lt. Chanoine and Amariya (Amaria,
Amarea, Hamaria), a renegade from Babatu’s army who called himself “King of the Gurunsi”. This battle and further defeats in the same year ushered in the decline of Babatu’s power and influence in Northern Ghana. Finally, he fled to Yendi, then a town in the German Colony of Togo, where he died of a spider bite (cf. Holden, p. 85). Meanwhile the slave raids among the Bulsa and other people had come to an end and a few years later they all became British subjects in the Northern Territories of the Gold Coast.
Reviewing our present knowledge about the Zabarima campaigns in the Grunshi area, we have found some data and accounts in written and oral sources about Babatu’s raids on mostly unfortified villages, about the sale of slaves on the slave markets [endnote 13] and about the afflicted people’s reactions to the continuously threatening dangers. Although the Europeans as well as the slave-raiders knew quite well that the local population had established an effective network of hiding-places and several sources tell us that they fled into the bush or hid in caverns and hollow trees [endnote 14], we have little detailed knowledge about these places of refuge. In historical and ethnological literature we find only occasional references to such places in Northern Ghana.
Benedict Der (1998: 21) describes the reaction of the local people on the Zabarima raids as follows:
At the approach of the raiders, the villagers fled their homes and sought safety in the bush, in caves, where these existed, or in the hills where the Zabarima cavalry could not operate; they returned to their homes only when the raiders had departed.
With regard to the Sisala and Dagara, Carola Lentz (2003: 137) mentions hills, caves and rivers as “safe havens for those who had fled from the surrounding area”. Bako Alhassan (1991, p. 12) describes an episode of Babatu=s campaign in Dagara territory when “all the people in the Sekai district climbed up the hill of Wutoma with all their belongings… and hid in the caves of that hill”. The UNESCO World Heritage Tentative List (2008) mentions that, near Mole National Park, “there is a cave in the Konkori escarpment that was used as refuge from slave raiders by the local indigenes”.
From these quotations we may conclude that caves of refuge were widespread in Northern Ghana and probably also in other parts of West Africa.
Again and again Bulsa informants mention caverns as hiding places when they tell about their ancestors’ encounters with the slave-raiders. One of these caverns of refuge (10° 31′ N; 1° 31′ W) is situated between Wiaga-Zamsa and Uwasi, another one (10° 38′ N; 1°18′ W) between Wiaga and Gbedema. Although the linear distance between the two caverns is more than 15 km, they display striking similarities in their outward appearance, construction and in the combination of a natural rock cavity and extensions dug by man.
I visited a third cavern in Wiaga-Yimonsa, similar to those mentioned above. This cavern (10° 37.97′ N; 1° 18.46′ W), situated in uninhabited bushland, cannot be entered today. A steel measuring tape inserted into the hole, indicated a depth of over two metres. I did not find any traces of an exterior shaft or tunnel as they are described below for the other two caverns. Close to the Yimonsa cavern is an allegedly very old settlement mound (10° 38.18′ N; 1° 18.35′ W)-[endnote15].
I was able to examine the so-called Posuk [endnote 16] Cavern between Wiaga and Gbedema carefully and carry out some measurements, because there is now comfortable access to the main “room” due to partial collapse of the roof. The old entrance, which consists of a perpendicular shaft with a diameter of 0.47 m and currently a depth of 1.10 m, has survived without damage. This shaft was certainly dug by human beings, probably Bulsa, into the upper layer of lateritic rock and displays some resemblance to the shaft of a modern Bulsa grave, which also extends into a cavity at its bottom to accommodate the corpse [endnote 17].
The tunnel branching from the bottom of the shaft of the Posuk Cavern is about 30 cm high at its entrance increasing to 60 cm as you move on. It is directed towards the main cavity, the room for the fugitives. After 60 cm, however, it runs into another shaft, which, in view of its irregular and larger diameter may be regarded as a secondary shaft, possibly resulting from a caving in of the tunnel roof. Contrary to my expectations, the tunnel starting from this secondary shaft does not continue towards the main cavity, but branches off at an angle of about 90° into a very broad passage, which might be a natural cavity. As it was only 30 cm high and inhabited by porcupines, I could not enter it. My steel tape, passed through a hole, indicated a depth of at least four metres without the tape reaching the end of the cavity. A direct connection of the two shafts with the main cavity could not be found, but may have existed before it was filled up with gravel and sand.
My main informant in Zamsa, Agbandem, who also holds the office of an earth priest (teng nyono), asserted that the main cavity, which today only allows an unsatisfactory insight through some small clefts, could be reached through the tunnel until both this and part of the shaft were filled up with sand and gravel.
The main cavity of the Posuk Cavern was possibly caused or enlarged by a rivulet flowing through this part of the cavern in the rainy season. Today, all parts of this cavity can be reached through the above-mentioned large entrance. The maximum height of the roof is 1.10 m, but the average height is considerably less than this. The width is approximately 2.20 m and the length approximately 8 m. There is only a small hole at the opposite side of the large new entrance: the entry of the rivulet. It is big enough to be looked out of and, with some inconvenience, to be used for leaving or entering the cavern.
The situation of the fugitives in the time of the slave raids can well be imagined. The Posuk Cavern might at most have sheltered about twenty people, who crowded together and, in fear for their lives, were listening intently to any sounds from outside, for although they could not easily be seen, the existence of small holes and clefts permitted them both to hear and be heard. When slave raiders approached the cavern, the persecuted people had to maintain absolute silence. A crucial problem would certainly have been that of preventing small babies from crying. Kwame Jua’s account regarding his own ethnic group, the Sisaala, would probably also have applied for the Bulsa, their immediate neighbours (Howell 1998, p. 15):
While people were fleeing from slave raiders, if the babies cried ceaselessly, they were taken from their mother’s backs and killed so that the crying would not betray them to the slave raiders.
Although the main purpose of the two caverns described above may have been their use as shelters and places of refuge in times of danger, both are also associated with religious ideas and institutions.
The Posuk Cavern is an earth shrine (tanggbain), although sacrifices to it ceased some decades ago. The shrine was originally in the possession of a Gbedema family, but was later seized by people from Wiaga-Longsa, who also sacrificed to it for a time.
The Zamsa cavern is not an earth shrine itself, but is situated close to and associated with the tanggbain of the Zamsa clan-section. The earth shrine consists of a large heap of stones in which, under some stones, two terracotta heads are hidden [endnote ]. They receive their libations together with sacrifices to the earth shrine and represent Anaanateng and his wife, the first ancestors of the autochthonous Bulsa or even of mankind, for Agbandem, the earth priest (teng-nyono), half humorously, called them Adam and Eve. There is also a tradition that in primeval times the first ancestor ascended from the cavern to live henceforth on earth.
The belief that the first ancestors lived in caverns does not appear to be limited to Zamsa. When, in the relevant literature, authors use the term ‘holes’ rather than ‘caverns’, this may refer to the type of cavern described above which was entered through a hole (shaft) or to the type of “well” described by Graf Zech.
Ollivant (1933, p. 1) was told the following about the Bulsa, “…when the settlers [Atuga’s family]… came from Mamprussi they found a few wild and uncouth people who are said to have lived in holes in the ground”. According to the same author a similar situation existed among the Kasena when the first Nankani immigrated into their territory, “…they found Kasena here who… are said to have lived in holes in the ground” (p. 4).
Cardinall (1920, p. 13) states that
… the sections named Biung [endnote 19] … nearly always lay claim to the fact that they themselves were from the earth and that their ancestors dwelt in holes in the ground. I have never been able to persuade one of these people to show me these caves or holes, although, both at Navarro and at Zuarungu, Biung sections exist.
I surmise that nearly all Northern Ghanaian caverns, including those which are not of the shaft-and-tunnel type, are associated with the apical ancestors, the early history of the tribal group and/or with an earth shrine. Apart from the Bulsa caverns, I could prove this only for the two most well-known Koma caverns: Buyori (10° 16.4′ N, 1° 33.56′ W; h= 216 m) and Gubong (10° 19.22′ N; 1° 38.2′ W; h= 347m), both of which were also places of refuge during the Zabarima wars [endnote 20]. Buyori is a shrine, which still receives sacrifices from a family living in Yizesi, and Gubong Hill (347 m), including the cavern, is a sanctuary of the Koma village of Tantuosi.However, all these caverns have probably been multifunctional: permanent residences of the early ancestors, shelters for shepherds and hunters and, above all, places of refuge during all types of conflicts. To my knowledge most of these caves have not yet been explored thoroughly [endnote 21]. Oral traditions, archaeological excavations and comparative studies will certainly provide new and fascinating knowledge and thus contribute a new chapter to Ghana’s eventful history.
The old entrance of Posuk Cavern
The interior of Posuk Cavern with the new entrance (on the left of the person)
Within the main cavity of Posuk Cavern
.
Agbandem in the entrance of Zamsa Cavern
.
Endnotes
[endnote 1] I wish to thank Prof. Dr. Albert Awedoba and Dr. Barbara Meier for reading my manuscript and for their useful comments
[endnote 2] The modern Bulsa live in the Bulsa District in the Upper East Region of Ghana. They number about 100,000. Even today they live mainly on agriculture by growing chiefly millet, guinea corn, rice and groundnuts.
[endnote 3] Cf. Lentz 1998, p. 174 (translation F. Kröger): “Seen from a local view, the first inspection and punitive expeditions were presumably hardly to be distinguished from the raids of the Zabarima and Sofa troops”. – Cf. also the episode when Sandema people attacked a French delegation mistaking them for Zabarima.
[endnote 4] Public Record Office, London, Co 879, 78, 05939, No. 25352, No. 5: Governor Major Nathan to Mr Chamberlain (1902); enclosure: Report on the expedition into the Tiansi [Bulsa] country.
[endnote 5] About 1850 S. Koelle interviewed two released Bulsa, who had been kidnapped five and six years before in their native villages: “Atim [Atiim]… born in the district Kándsâre [Kanjaga], where he lived till about his eighteenth year… and Adsúmâno… born in Nyása…” (Koelle 1854, p. 6). The language of a “Gurunsi” vocabulary, collected by Nina Rodrigues, presumably at the end of the 19th century in Bahia (Brazil), was identified as Buli, the language of the Bulsa, by J. Zwernemann (1968, p. 147-156). This probably means that the Bulsa of Brazil were deported as slaves before the Zabarima slave raids of 1891 and 1897, because natives captured at that time were no longer transported to America.
[endnote 6] Piłaszewicz (ed.) 1992.
[endnote 7] I could not find this battle of Wiaga mentioned in any written source, but, because many people lost relatives in it, it is still vivid in the memories of present-day Wiaga people. It took place near the present Anabeli Dam, 5 km south of Wiaga’s centre.
[endnote 8] Interview and translation by Danlardy Leander Amoak in 2001. Before I (F.K.) could verify the data and ask additional questions, Apajie had died (2002).
[endnote 9] Akumcham (literally‘death-sheabutter tree’) is situated near the Sandema Boarding School in the south of Sandema, i.e. not far from the place mentioned by Perrault. According to Sebastian Adanur (Sandema, 1978), the tree was called Akumcham because, after the battle, the Bulsa captured one of Babatu’s wives, killed her, and hung her body on that tree. Although the tree no longer exists, its exact former place is still known today. Any involvement of the ‘white men’ (British soldiers) in this battle is not very probable in the light of other sources.
[endnote 10] Interpretation of Abu’s spelling by Piłaszewicz 1992: 83-84. fn. 68-75.
[endnote 11] For a more detailed description of this battle, its prehistory and its political consequences, see Duperray1984, p. 100-103 and Piłaszewicz 1992, p. 92-93 (Malam Abu, p. 45). Malam Abu does not admit that Babatu was beaten at Kanjaga. Holden (p. 83) and Perrault (p. 95) mention the battle of Kanjaga. Tamakloe (1931, p. 54-55) and Ahmed Bako Alhassan (p. 17), who relies on Tamakloe, do not refer to Amariya’s share in the victory.
[endnote 12] Benedict Der (1998, p. 23): “Amariya is said to be a Builsa who was captured as a boy in the raids of Gazari and was brought up in the Zabarima army where he rose to be commander”. Ollivant (p. 6): “Another was called Amarea, and came from Kanjaga”. Details about the mutiny can be found in Bako Alhassan 1991, p. 16-17 and Holden 1965, p. 78. The name Amariya might be Buli. All Bulsa individual names of persons start on A-; ma(a)ri means ‘help’ (verb), while -ya is a suffix denoting incompleteness; maari ya = ‘has helped’. Personal names in the form of short sentences, usually without a subject, are quite common among the Bulsa. – In the form Ameria the name might, however, be a Hausa adoption.
[endnote 13] According to Bulsa informants the local slave markets were Doninga and Yagaba.
[endnote 14] big hollow baobab tree in Wiaga Sinyangsa could have been used for up to five people as a hiding place. The use of such trees is not unusual; at Tengzug (U.E.R., Bolgatanga District) tourists are shown such a tree which once served as a place of hiding.
[endnote 15] I could not collect any information on the cavern or the abandoned settlement (guuk).
[endnote 16] Posuk is the Buli term for ‘house with a flat roof’.
[endnote 17] 1904-10 Governor of the German colony of Togo, observed similar phenomena in Konkombaland (1904, p. 130, translation by F. Kröger): “Striking are the many… cylindrical wells with a diameter of 0.5 m whose shafts have been hollowed out quite smoothly and clearly. They lead vertically through the upper rock layer. These holes must be of very old origin… Under the penetrated rock layer the earth has been dug out like a cavern… In Eastern Konkombaland such a well has been found, in which, at one side, small cavities [i.e. steps, F.K.] have been dug to give hold to the feet when climbing down”. From Zech’s account it is not clear, whether these shafts had always served only as wells or whether human beings stayed in them for short periods. As wells are entered only extremely rarely, it seems unlikely that the steps were cut for this purpose. A. Awedoba commented, however, in a letter to the author that “steps facilitate the digging, since people rest on them to dig deeper”. The extension under the rock layers would be unnecessary if the wells were used for drawing water only. Tamakloe (1931, p. 58) mentions “water holes” which he found in the Dagomba province of Karaga: “In this province, again, we find many water holes excavated through rock strata, these being attributed to one of those Giants whose name was Zodebi”.
[endnote 18] Figures in this style are named ‘Komaland terracottas’ nowadays, although such artefacts are found in the southern part of Bulsaland, too. They are between 400 and 800 years old. Further information about these terracottas in general can be found in Anquandah 1998 and Kröger 1988, and about the two Zamsa-terracottas in Kröger 1982, p. 16-20 and Plate 16-17
[endnote 19] One of the two villages called Biung (today also Biuk or Biu) with a Bulsa majority among its inhabitants is situated 15 km south of Navrongo.
[endnote 20] As regards the Gubong cavern, Koma history demonstrates that caverns of refuge may turn out to be real traps, since in the battle of Gubong, when the cavern was planned to be a refuge for women, children and old people, Koma warriors were driven into this cavern by Babatu’s forces and died there. – Today, the two entrances to the cavern (or entrances to two caverns?) are blocked with big stones.
[endnote 21] One rock “shelter”, mentioned in a paper by P.L. and P.J. Carter (1964), is situated near the foot of the Gambaga Escarpment on the Ghana-Togo border. The main concern of the two authors was not the shelter itself but the description of the site’s rock paintings
.
References
Alhassan, Ahmed Bako, 1991. Babatu. Tamale (typescript).
Anquandah, James, 1998. Koma – Bulsa. Its Art and Archaeology. Rome: Istituto Italiano per l’Africa e l’Oriente.
Cardinall, A.W., 1920. The Natives of the Northern Territories of the Gold Coast. Their Customs, Religion and Folklore. New York: Negro Universities Press.
Carter, P.L. and P.J., 1964. Rock Paintings from Northern Ghana. Transactions of the Historical Society of Ghana, vol. VII, p. 1-3.
Der, Benedict G., 1998. The Slave Trade in Northern Ghana. Accra: Woeli Publishing Services.
Duperray, Anne-Marie, 1984. Les Gourounsi de Haute-Volta. Conquête et colonisation 1896-1933. Stuttgart: Franz Steiner Verlag Wiesbaden.
Holden, J.J., 1965. The Zabarima Conquest of North-West Ghana, Part I. Transactions of the Historical Society of Ghana, vol. VIII, p. 60-86.
Howell, Allison (ed.), 1998. The Slave Trade and Reconciliation: A Northern Ghanaian Perspective. Accra: Bible Church of Africa and SIM Ghana.
Jua, Kwame, 1998. The Impact of Slave Trade on the Sisaala. In: Howell 1998, p. 12-17.
Kröger, Franz 1982. Ancestor Worship among the Bulsa of Northern Ghana. Hohenschäftlarn: Klaus Renner Verlag.
—–, 1988. Die Terrakotta-Funde des Koma-Gebietes (Nordghana). Paideuma, vol. 34, p. 129-142.
—–, 2001. Materielle Kultur und traditionelles Handwerk bei den Bulsa (Nordghana). 2 vols., Münster, Hamburg and London.
Lentz, Carola, 1998. Die Konstruktion von Ethnizität. Eine politische Geschichte Nord-West Ghanas, 1870-1990. Studien zur Kulturkunde 112, Köln: Rüdiger Köppe Verlag.
—–, 1003. Stateless Societies or Chiefdoms: A Debate among Dagara Intellectuals. In: F. Kröger and B. Meier: Ghana’s North. Research on Culture, Religion, and Politics of Societies in Transition. Frankfurt am Main, Berlin, Bern, Bruxelles, New York, Oxford: Peter Lang Verlag.
Morris, A., 1902. In: Report on the late expedition into the Tiansi country, Gambaga, April 26, 1902; Enclosure in a letter of Governor Major Nathan to Mr. Chamberlain (Accra, May 31, 1902), Public Record Office, London, Co 879, 78, 05939, no. 25352 (Enclosure no. G/105/N.T./02).
Ollivant, District Commissioner of Navrongo District, 1933. A Short History of the Buli, Nankani and Kassena speaking People in the Navrongo Area of the Mamprusi District (typescript).
Perrault, Rev. Fr., 1954. History of the Tribes of the Northern Territories of the Gold Coast. St. John Bosco’s Press, Navrongo (typescript).
Piłaszewicz, Stanisław (ed.), 1992. The Zabarma Conquest of North-West Ghana and Upper Volta – A Hausa Narrative “Histories of Samory and Babatu and Others” by Mallam Abu. Warszawa: Polish Academy of Sciences.
Schott, Rüdiger, 1977. Sources for a History of the Bulsa in Northern Ghana. Paideuma 23, p. 141-168.
Tamakloe, Emmanuel Forster, 1931. A Brief History of the Dagbamba People. Government Printing Office Accra.
UNESCO World Heritage, Tentative List Ghana, 2008. http://whc.unesco.org/en/tentivelists
Zech, Julius, Graf von, 1904. Aus dem Schutzgebiete Togo. Land und Leute an der Nordwestgrenze von Togo. Mitteilungen aus den deutschen Schutzgebieten, 17, p. 107-135.
Zwernemann, Jürgen, 1968. Ein “Gurunsi”-Vokabular aus Bahia. Ein Beitrag zur Afro Amerikanistik, Tribus, 17, p. 147-156.
