INTRODUCTION
Gold mining and trade were at the heart of the great West African empires such as Ghana and Mali. Until the Americas were discovered, West Africa was the major provider of gold for the European market. The topic of gold mining has always fascinated explorers and researchers, partly because so little was actually known about it:
The source of West African gold was a remarkably well kept secret. It puzzled almost all outsiders for over two thousand years, and for half that time their efforts to discover where the gold came from were as unremitting as they were unsuccessful.Footnote 2
In this article I will give an overview of the current state of knowledge about precolonial gold mining in the Mouhoun (Black Volta) region of what is today Burkina Faso.
Both because the topic of gold generally seems to inspire myth-making and because of the dearth of reliable sources for precolonial history in West Africa, gold mining and trade have been contentious issues among archaeologists and historians. In her article ‘A reconsideration of Wangara/Palolus, Island of Gold’,Footnote 3 Susan Keech McIntosh unraveled many previous assumptions and hypotheses about the ‘islands of gold’ mentioned in Arabic sources. She convincingly argued that, contrary to prevailing ideas, the ‘islands of gold’ were not identical with the gold fields at Bambuk (at the confluence of Senegal and Baleme) or at Bure (on the Upper Niger), but with the Inland Niger Delta.Footnote 4 Two major assumptions had previously led to the rejection of this hypothesis: (1) that the ‘islands’ were identical with the gold deposits (and not with trade centers); and (2) that the middle Niger was not a major trade axis until the fourteenth century. On the basis of archaeological research and a re-reading of the Arabic sources Keech McIntosh was able to contradict these assumptions.
One myth that till today is repeated in popular accounts about historical gold mining in West Africa is the ‘silent trade’, whereby two trading parties exchanged goods without speaking to or even seeing each other. Moraes FariasFootnote 5 showed that the ‘silent trade’ was not a fact but a topos that kept re-surfacing in Arabic, Portuguese and other sources after the ninth century, without ever having been backed up by eye-witness accounts. He concluded that the ‘silent trade’ was most probably not silent in a literal sense but conducted by middlemen who acted as intermediaries between producers and traders. Accounts of the ‘silent trade’ have to be considered as a misconception of an ancient pattern of ‘trade-through-broker’. The myth of the ‘silent trade’ was spread by those middlemen who were interested in keeping rivals away.
In a similar attempt to verify some prevailing assumptions about gold mining and trade in West Africa, this article re-visits received knowledge about gold mining in a region called ‘the Lobi’. The Lobi gold fields are often referenced in secondary works as operating by at least the late medieval period, but based on very little substantive historical evidence. What Keech McIntosh said for Bambuk and Bure also holds true for ‘the Lobi’: ‘The identification of the mediaeval Island of Gold with Bambuk/Bure is so commonplace in the literature that the authority for it is rarely cited’.Footnote 6 In most cases Lobi is used as the name for a region which is vaguely located, for instance, ‘on the upper Volta’,Footnote 7 ‘along the banks of the Black Volta river’,Footnote 8 in the ‘Volta basin’Footnote 9 or not at all. But our knowledge of these Lobi gold fields – their exact geographical location, who mined the gold, how it entered the regional trade networks and when the name ‘Lobi’ first came into use – has always been imprecise, though scholars have recycled old information as given over the decades. I will show that the usage of ‘Lobi’ as the name for an ancient gold-producing zone is misleading, both because it is an ethnic label for groups that have only recently moved into the region and because very little is actually known about precolonial gold production and trade in that area. What has come to be accepted for a fact needs to be reconsidered in the light of new research in southern Burkina Faso.
Map 1. Southern Burkina Faso, place names mentioned in text are framed. Based on a map by Richard Kuba.
PRECOLONIAL GOLD MINING AND POLITICAL ORGANIZATION
In her article ‘The political organization of traditional gold mining: the western Loby, c. 1850 to c. 1910’,Footnote 10 Marie B. Perinbam presented her hypothesis about the precolonial political organization of a region in present-day Burkina Faso. She believed that gold mining and trading activities in this region were directly controlled by the ‘Mande-Jula’ from Kong. In what follows I will show that her conclusions are flawed and that there is in fact no evidence for them in the archival sources on which she based them.
Perinbam claimed that there were 118 ‘mining towns’ in the region she termed ‘the westerly part of the Loby’.Footnote 11 In this region, Jula and Hausa trade routes between Ouagadougou and Bobo-Dioulasso and between Bondoukou and Bouna intersected. Gold production was in the hands of ‘animist, non-centralized polities’ who held land rights over gold-bearing territories, but these polities were under the political domination of the ‘Mande-Jula of Kong’. According to Perinbam, each of these 118 ‘mining towns’ was subject to the ‘Wattara’Footnote 12 of Kong and had a Jula ‘chief in residence’.Footnote 13 Perinbam presented a hypothetical model of the ‘urban political organization’ of the ‘Loby’ that consists of three distinct levels.
Level I: ‘Corporate semi-autonomous towns’
On this level there were localized kin-groups who settled in different quarters of multi-ethnic ‘towns’, a ‘town’ being defined as a ‘corporate agglomeration of different ethnic wards’. These towns were subdivided into sukulas (clusters of adjoining houses) and governed by a ‘resident chief’. The different ethnic groups represented were ‘Dagari’, ‘Dyan’, ‘Buguli’, ‘Birifor’, ‘Wule’ and ‘Nienigue’. Mande-speaking Muslim traders lived in separate quarters. Since the seventeenth century Bobo-Jula were part of the population and contributed to the emergence of trade routes from Kong to Jenne, along which gold was traded. According to Perinbam, it was these Bobo-Jula who acted as ‘resident chiefs’ and ‘advisors’.Footnote 14
Level II: ‘Chiefly towns’
Lineage heads (‘ancestral’ or lineage chiefs) lived in larger settlements that were surrounded by ‘satellite’ settlements. These larger towns also had markets and were located at a distance of several days’ journey from smaller settlements. The relations between the smaller, largely autonomous settlements and the ‘chiefly towns’ were both political and economic. According to Perinbam, Jebugu (Diébougou) was for its population of ‘Dyan’, ‘Dagari’ and ‘Fulbe’ a ‘corporate town’, ‘central place’, ‘chiefly seat’, ‘main town for jural community’ as well as a ‘market town for gold producers’ that attracted Jula traders.Footnote 15
Level III: Kong
Both level I and level II were superseded by level III, the domination by the ‘Mande-Jula’ of Kong in present-day Côte d'Ivoire. Warrior clans of Mande origin bearing the patronym ‘Watara’ had introduced certain forms of political organization into the ‘Loby’ since the sixteenth and seventeenth centuries. Around the middle of the nineteenth century, Watara had settled along the trade routes between Kong and Bobo-Dioulasso and into the region west of the Black Volta. They delegated ‘representatives’ to allied settlements who were supposed to protect the trade interests and the security of trade routes. According to Perinbam, the Jula patronyms of village heads prove that these were in fact Jula ‘representatives’.
Perinbam concluded:
political organizations introduced by Mande-Jula into Loby towns were designed on the one hand to control gold mining populations, and on the other to maximize the region's commercial potential. In practice, this meant the consolidation of Mande-organized autonomous and semi-autonomous urban quarters, corporate towns and urban agglomerations, in which Mande-Jula representatives resided, into a larger organization.Footnote 16
Now, what is the evidence for the existence of 118 ‘mining towns’ controlled by Jula ‘resident chiefs’? A check-up on the sources she referred to in the National Archives of Mali in Bamako-Koulouba revealed that these sources contain in fact very little information about gold in that region and no evidence whatsoever for the existence of ‘mining towns’.Footnote 17 Neither is there any information about the precise location of gold mines or the activity of gold mining.
The document ‘Renseignements sur les villages de la circonscription de Diébougou’ consists of a list of 120 villages, none of them a ‘Lobi’ village in an ethnic sense (speakers of the Gur language Lobiri), but 63 ‘Dagari’, 19 ‘Dian’, 14 ‘Birifon’, 10 ‘Oulé’, 12 ‘Pougouli’ and 2 ‘Niénégué’ villages. The entry for each village is subdivided into: name of village head, number of compounds, number of inhabitants, number of cattle, goats and sheep; special crops (for instance, rice or sweet potatoes), mineral resources and special resources (e.g. fish). The entries for ‘mineral resources’ almost always say: ‘néant’, ‘0’, or ‘?’. Only in one case, the village of Moa (Ouan), does the entry say: ‘peut être de l'or’. In another village the entry consists of question marks, and a note scribbled on the side of the page says: ‘Il y a peut être dans les environs des gisements miniers’. A table attached to this document lists all the products that were offered on the markets of the Cercle du Lobi, including cattle, sheep, chicken and gold, but the latter without any specification concerning amount, price, origin or any other information.
Likewise, the information about the political organization of these villages is limited to the names of village heads and some remarks about the general economic state (e.g. ‘assez bon village riche’, ‘hameau pauvre’) or the degree of cooperation with the colonial administration (‘totalement insoumis’). One only finds some casual references to itinerant Jula traders, and absolutely nothing about Jula chiefs.Footnote 18 Another source cited by Perinbam as evidence for the existence of ‘mining towns’ contains in fact one single sentence about gold: ‘Enfin de l'or était autrefois recherché à Oronkua, à Naro, à Golibatégué’.Footnote 19
Apparently, Perinbam never did research in Burkina Faso herself. In her article, she cited three oral sources from Bamako, Mali.Footnote 20 Neither these interviews nor the archival sources confirm her theories about the precolonial political organization of the ‘Loby’ or the organization of gold mining and trading. The article rather represents a ‘Mande-centric’ view of a region that was in fact settled by segmentary societies that were only loosely connected with Jula trade activities and never under direct control of Kong.
The assumption that village heads who bore Jula patronyms were ‘representatives’ directly installed by the Watara is completely unfounded. People adopted Jula patronyms at different times and for a variety of reasons: to show allegiance; to adopt their master's name when being a slave; to become traders and thus Muslims; as a ‘nom de guerre’; under colonial pressure; or simply as a nickname after having traveled to the Mande-speaking regions.Footnote 21
To conclude: Perinbam's hypotheses about precolonial political organization and gold mining in southern Burkina are not confirmed by any of the sources she herself cited as evidence. Now, what do we actually know about the precolonial role of the Watara in what is today southern Burkina Faso?
THE WATARA IN THE HISTORY OF SOUTHERN BURKINA FASO
One of the problems with the existing literature is that ‘Lobi’ is used both for an (imprecisely defined) region and for the ‘ethnic’ Lobi in present-day Burkina Faso, Côte d'Ivoire and Ghana as if it were an ancient name. This usage, however, only became common in the period that preceded the colonial domination: ‘Pour le cas de la Haute-Volta/Burkina Faso, il en est ainsi des locutions “le Mossi”, “le Lobi”, “le Gourmantché”, etc. utilisés à longueur de publications et de rapports administratifs. Cette terminologie fut très certainement puisée (pour ne pas remonter plus loin) dans la bible des coloniaux que constitua le livre de Binger’.Footnote 22
During the British and French colonial period in the Black Volta region, ‘Lobi’ was used for different territorial, ethnic and cultural entities.Footnote 23 Labouret applied the term ‘rameau Lobi’ to a zone that also included populations such as Birifor, Dagara, Dyan (Dian, Jãane), Gan and Pougouli (Phuo).Footnote 24 The ‘ethnic’ Lobi or ‘lobi proprement dit’ in Burkina Faso speak variations of the Gur language Lobiri.Footnote 25 However, inhabitants of the northern and central regions of Burkina Faso may refer to every southerner as a ‘Lobi’. Since colonial times, ‘Lobi’ was a name applied to the allegedly ‘uncivilized’ and belligerent segmentary populations in the Mouhoun region. A present-day stereotype depicts the Lobi region as particularly backward (‘le Burkina profond’), and the population as the ‘Indians’ of Burkina Faso, who since the colonial period have been known for resisting modernization (in the form of missions, schools or development projects).Footnote 26
Up to the 1890s, there are basically no sources other than oral traditions about the precolonial settlement history of the ‘Lobi’ region: ‘There seem to be no Arabic manuscripts from the nearest Islamic centers, Kong, Sya (Bobo-Dioulasso), Ouahabou or Wa. No major trade route had crossed it’.Footnote 27 Some of the oral sources were collected by colonial officials or Christian missionaries and have to be read carefully against the backdrop of Eurocentric stereotypes about African and especially segmentary societies, which were probably even reinforced by the Muslim intermediaries and interpreters on whom the colonial administrators depended.Footnote 28 According to the available sources, the present-day populations of this region, today known as the Birifor, Dagara, Dyan, Lobi and Phuo,Footnote 29 settled there after having arrived from the other side of the Mouhoun from the middle of the eighteenth century, possibly as a result of the wars and raids connected with the expansion of Gonja and Dagomba in what is now northern Ghana.
Recent historical and anthropological research into this region could not confirm the existence of a Kong hegemony and Jula ‘resident chiefs’.Footnote 30 OuattaraFootnote 31 wrote that the Jula warlord Barakatou raided the settlements along a route that included Gaoua, Nako, Hemkoa, Dapola, Nako and Diébougou between c. 1715 and 1720 (a date that seems to be much too early) and levied tributes in gold. Although Ouattara claims that Barakatou had in fact thus ‘subjugated’ the area, he later states:
les Watara seraient-ils venus dans le but de conquérir cette vaste région pour implanter un pouvoir nouveau et d'organiser la région sur une nouvelle base? Il semblerait que rien n’était codifié au préalable comme ce fut le cas des conquêtes Européennes. Cette absence de planification expliquerait à notre avis, le caractère relativement sommaire de leur structure administrative.Footnote 32
The ‘ethnic’ Lobi became especially known as a population that resisted the Watara warriors by way of fortified villages and open battle. According to the informants of Madeleine PèreFootnote 33 it was only after the arrival of the French in 1897 that Jula traders started settling in some larger Lobi settlements like Kampti, Gaoua, Doudou and Loropéni. Watara warriors had probably hunted for slaves in that region since the end of the eighteenth century but were not interested in farming and thus in becoming ‘chefs de terre’.Footnote 34 However, some Jula from the Kong area settled in Dyan villages such as Diébougou, Loto and Sorondigui. They married local women, and some Dyan families in those villages adopted Jula patronyms like Watara and Coulibaly. Jula traders and weavers also settled in other Dyan and Lobi villages. They sold hoes and salt to the local people, and some of them even farmed.Footnote 35
Although they repeatedly raided the area, the Watara warriors were never able to sustain domination: ‘Les populations n'obéissaient pas d'une façon régulière à ces étrangers qui n’étaient que des pillards et des chasseurs d'esclaves sans organisation’.Footnote 36 An exception was Loto where Watara warriors had settled since the end of the eighteenth century. From there they attacked villages in the vicinity. Although Dyan and Phuo villages were victims of these raids, there were some Dyan and Phuo lineages in Diébougou and Loto who later allied with the Watara and fought with them against the Dagara-Wiile.Footnote 37
In his account of the region's history, the present-day Imam of LotoFootnote 38 downplayed the role of the Watara as warriors and stressed the peaceful integration of the Jula in Loto. In his version, Diébougou's Dyan founder had asked a Jula trader and Muslim scholar for spiritual help against his enemies, the Dagara and Birifor. The Jula remained in his house for several weeks and prayed for an amicable cohabitation of the different populations. Then he spent some days in Loto. On his way back to Sya (Bobo-Dioulasso), while staying in the village of Mougé, he learned that the Dagara and Dyan in Diébougou had in fact made peace with each other. His Dyan host in Loto asked him to join them again, and so he settled in Loto with his family and some friends. It is highly probable that there were both Watara warriors and peaceful Muslim traders in the area, as elsewhere.Footnote 39 In any case, it is interesting that the most recent account of Loto's history stresses the non-violent integration of Jula into the local population. This account reproduces a common theme of the ‘dispersion of Muslims’Footnote 40 in the Volta region, who often acted as mediators and providers of spiritual services.
Around 1880, the Muslim warlord Moktar Karantao from Ouahabou reached Diébougou. He was accompanied by some Dagara-JulaFootnote 41 who remained in Diébougou and formed the core of a group of wealthy traders who also engaged in the slave trade. With the arrival of these and other Muslim traders, Islam came to the region, but did not make great inroads. A report from 1904 remarks:
Les Ouattaras venus de Kong sont très peu nombreux, et malgré qu'ils soient à la tête des Dians, ils n'ont pas fait de prosélytes parmi eux. Ils sont de la secte des ‘Tauri’ [?] de Kong et le marabout le plus important est un Touré nommé M'Faleban Touré qui réside au village de Loto … Tous les autres étrangers musulmanes se réclament du Marabout de Ouahabou. Il y a à Diébougou un délégué du Marabout de Ouahabou, il n'est pas chef et personne ne le reconnaît pour tel. Son rôle consiste à crier l'heure de la prière du haut de l'arigamas de sa soukala car il n'y a pas de mosquée pas plus à Diébougou que dans la Circonscription.Footnote 42
This latter citation makes it especially clear that there was obviously no political domination over the area by either the Watara from Kong or the Karantao from Ouahabou prior to the arrival of the French. Likewise, Kambou-FerrandFootnote 43 pointed out that the Watara from Kong never managed to dominate the segmentary populations of the ‘Lobi’ area, and that even attempts by Samori Touré to conquer the area had failed.
When the French arrived in 1897, the Watara warriors had already lost considerable force in battles with the local segmentary populations, and the Watara warlord Baratou (Barkatou) had retreated from Loto to Kobogho (Kubo), about 30 km from Diébougou. Nevertheless, he was able to present himself to the French as ‘Chef des Gans et d'une partie des Zians [Dyan] et des Pougoulis’, and he became a major auxiliary for the French in the first phase of colonial domination. On 11 September 1897, Caudrelier and Baratou signed a treaty in which Baratou was called ‘roi de Koubo’.Footnote 44 The French made the region around Diébougou a protectorate, the treaty for which was signed by Olé Gbona, the grandson of Diébougou's founder Bè Syo. Another grandson of Bè Syo (a son of one of his daughters and presumably a former slave from Mali), Aldiouma Ouattara, had become a local Big Man. The French made him Chef de groupement Dagara, but in 1903 he was arrested and jailed for ten years on the grounds of continued slave trading.Footnote 45
It was only the support of the French that brought some Jula and Dyan into positions of political authority. Bakary Traoré notes the late arrival of Jula in the ‘Lobi’ region:
Dans le pays lobi, où l'absence des Jula anciennement installés est notoire, on rencontre également des colonies de ces Jula récemment arrivés. Attirés par les rumeurs sur la richesse en or du pays, et encouragés par l'administration coloniale qui voyait en eux d'utiles ‘instruments de pacification’, ils s'installent à Gaoua, à Kampti et à Batié.Footnote 46
Through the influence of Jula traders, Diébougou developed into an important market town. Traders from Kong, Bobo-Dioulasso and further away frequented the five-day market.
As I have shown, there is no sufficient evidence for a ‘Mande-Jula’ – or rather Kong-Watara – hegemony over this region. Different parts of the segmentarily organized populations had allegiances to powerful ‘war houses’, as elsewhere.Footnote 47 Even within ‘ethnic’ entities such as the Dyan there were different clans and lineages, some of whom sided with the Watara and some of whom did not. The adoption of the patronym ‘Watara’ was in no case an indication for a Watara ‘resident chief’ delegated by the rulers of Kong. Perinbam's model of the precolonial political organization of the region thus has to be considered as invalid.
PRECOLONIAL GOLD MINING IN THE ‘LOBI’ REGION
According to the available sources and present-day evidence, precolonial gold mining in Africa was predominantly a temporary activity exercised during the dry season and done by groups of relatives or friends. In contrast with iron ore smelters and smiths, there were apparently no professional gold miners. Slave labor may have been involved as in other economic activities. The mining itself was organized in analogy to agricultural work, including gender differences: it was normally the men who did deep-shaft mining and women and children who did the washing, panning and transporting. Women also did shallow-pit mining.Footnote 48 Gold mining, especially deep-shaft mining, required ritual preparation and certain taboos had to be observed.
McIntosh assumed that one source of the gold that was traded in Jenne-Jeno was the ‘Lobi’ along the Black Volta.Footnote 49 The name ‘Lobi’, however, was not used in any written source until the end of the nineteenth century. Although the Mouhoun (Black Volta) region in present-day Burkina Faso was probably known as a gold-bearing zone for a long time, archaeological and historical evidence for gold mining and trading remains scanty at best. McIntosh hypothesized that Soninke traders from the Western Sudan may have reached the ‘Lobi’ and Akan gold mining zones as early as around 500 CE, but there is no proof for this hypothesis.Footnote 50 Schneider thinks that gold from this region may have entered the trans-Saharan trade in the fourteenth century.Footnote 51 He assumes that the large buildings at Loropéni, the ruins of which remain, were probably built as fortified outposts of Jula/Wangara trade routes in an attempt to control the gold and slave trade in the sixteenth century. Madeleine Père's Gan informants said that it was their forefathers who constructed these buildings, but this claim has not yet been substantiated with other kinds of information.Footnote 52 To date, no archaeological excavations have been undertaken to determine the age, origin and function of the ‘Lobi’ ruins.
The Lobi region is named after its present-day inhabitants, but the identity of the people who exploited gold deposits in ancient times remains unknown. ‘Lobi’ as the name for a region and its inhabitants was first mentioned by Louis Gustave Binger who between 1887 and 1889, traveled through parts of what is today Burkina Faso, but did not reach the ‘Lobi’ region itself.Footnote 53 In his description of the ‘Gouin(g)’ he mentions that their language is related to the Lobi language, and that there were Lobi warriors in the army of Kong.Footnote 54 In Kong he learned that most of the gold traded there originated from ‘Lobi’ and was exchanged against copper bars and slaves.Footnote 55 On the way from Boromo to Ouagadougou, he passed the gold fields of Baporo near Poura: ‘Au dire des gens que j'ai interrogés, le rendement est, paraît-il, plus grand que dans le Lobi; malheureusement, les gisements sont assez éloignés du fleuve pour que les indigènes renoncent à s'y pourvoir d'eau pour effectuer le lavage’.Footnote 56 According to Binger, the gold miners at Baporo were ‘Bobo-Niéniégué’ and ‘Nounouma’, but Lobi gold miners were supposed to be more refined in this craft.
From Binger's account it is clear that what is often referred to loosely as ‘Lobi’ in the literature about gold mining in West Africa was in fact two different regions along the Mouhoun (Black Volta), around Poura and Gaoua, respectively, about 200 km apart from each other.Footnote 57 Precolonial gold mining in Poura has been studied by the archaeologist Jean-Baptiste Kiéthéga.Footnote 58 According to him, the region was settled by Gurunsi, Dyan and Dagara around the end of the eighteenth century. These populations, however, did not mine gold, and neither did the Bwaba and Moose (Mossi) who arrived later. Gold mining was done by immigrant Bobo-Jula (Zara) from further west. During the first half of the nineteenth century, these Bobo-Jula immigrants were active in gold exploitation and trade in the region around Poura. In the second half of the nineteenth century, gold production declined due to the increasing raids by the Watara warriors from Kong, the Karantao jihadists from Ouahabou and the Zerma slave raiders who devastated the region up to the colonial ‘pacification’ in 1896. Kiéthéga assumes that there was a period of gold mining prior to the fifteenth century, but here again the archaeological evidence is virtually non-existent.
Likewise, there is no conclusive evidence concerning the identity of the gold miners who left hundreds of circular mining shafts in an area around Guéguéré, north-east of Diébougou.Footnote 59 Those pits – which until today have not been subject to archaeological research – were attributed to the YeriFootnote 60 in Oronkua by SavonnetFootnote 61 and to the Bwaba by Somé.Footnote 62 Savonnet tentatively dates them to between the fifteenth and eighteenth centuries, referring to some oral sources collected by Hébert,Footnote 63 and based on estimates about the erosion of overburden left over from mining activities. Somé claims that these gold miners must have been Bwaba. His sources are five interviews in seven villages and the similarity of these mining shafts with remnants of gold mining at Bangassi, Dossi and Poura, where, according to Binger, it was the Bobo-Niéniégué (Bwaba) who were mining the gold (see above). Somé contradicts Savonnet on the grounds that Oronkua was too far away from the mining pits (about 25 km), so that the Yeri who lived in Oronkua could not have been doing the mining.
Evidence based on a very small number of oral sources which are taken at face value, the attribution of ‘ethnic’ identities or precise locations to populations in precolonial times in a region that is even today characterized by a pronounced mobility of kin-based or socioprofessional groups, and the tentative dating of mining pits according to vague estimates concerning erosion must remain highly speculative. Nevertheless, as a hypothesis, it may be that both Savonnet and Somé are right. It may have been more than one population who engaged in mining activities during the dry season, and these populations may easily have traveled to and camped at locations that were at a distance of 25 km or more from their settlements – even today many people cover distances like that on foot or on bicycle in order to attend important markets or funerals.
Many of the populations in present-day southern Burkina Faso seem not to have practiced gold mining or panning except as forced labor during colonial rule. According to Lobi informants, their forefathers only started panning for gold after Jula traders told them that gold could be exchanged for salt.Footnote 64 Because panning was done with calabashes which are typically associated with female domestic tasks, gold panning became women's work. According to the Imam of Loto, the Sanogo bought gold from Birifor women and sold it to ‘Maninga-mori’ traders from Guinée. Labouret wrote in 1931 that gold panning done by women was actually in decline, which he attributed to women's numerous other tasks.Footnote 65 He assumed that one woman could obtain an average of 8·6 g of gold per season but admitted that this was a vague estimate. The gold was sold to traders from Bouna against cowries, cloth or hoes. These traders then sold the gold in Bobo-Dioulasso, Bondoukou, Ouagadougou and Mopti.Footnote 66 Labouret thought that gold mining had been introduced into the Lobi by the Kulango from Bouna,Footnote 67 who in turn had adopted gold mining from the Asante via Bondoukou.Footnote 68 This hypothesis rests upon Labouret's assumption that the ‘Lobi ruins’ were erected by the Kulango.
The historian Bakary Traoré confirms that Jula traders from Bouna were involved in the gold trade: ‘À en croire les renseignements reçus, le métal précieux exploité dans les pays lobi était surtout commercialisé par les Jula de Gbona [Bouna], car ceux-ci étaient en relation privilégiée avec les producteurs, notamment les Lokhon’. Traoré also points out that:
ces marchands de Gbona ne semblent pas avoir joui du même privilège dans la partie du bassin aurifère située au Nord de Diébougou où les Marka, apparaissant sous des désignations aussi diverses que Zara, Zéri, Yèri, Bobo-Jula et Dagari-Jula, connaissent une présence fort ancienne. De ces centres, l'or suivait des circuits d’échange qui échappent cruellement à notre connaissance.Footnote 69
These passages highlight once more that present-day ‘ethnic’ names are notoriously unreliable for identifying specific groups at specific points in history. The label ‘Marka’ may refer to Muslim groups with Mande origins (like ‘Jula’), but it can also mean different things in different regions.Footnote 70 People who refer to themselves as Zara or Dagara-Jula today may be surprised to find themselves labeled as ‘Marka’.
With the exception of the Gan around Obiré who said their forefathers mined gold,Footnote 71 the present-day populations of southern Burkina Faso, such as the Birifor, Dagara, Dyan, Lobi and Phuo, do not know who was mining gold prior to their arrival in this region.Footnote 72 In any case, gold mining and trade in that area were certainly stimulated by existing trade routes and relations, by, among others, itinerant Jula and Marka traders. But there is still no evidence that gold mining and trading were politically controlled by the Watara from Kong.
COLONIAL PRESSURE TO MINE
After the arrival of the French in Diébougou in 1897, a part of what is today southern Burkina and northern Côte d'Ivoire was subdivided into three Circonscriptions: Diébougou, Lokosso and Bouna. In 1898, the Cercle de Lobi was established, but Diébougou remained the administrative center. In 1902, Gaoua became the administrative center and Diébougou was made ‘Chef-lieu de la Subdivision de Diébougou’. The colonial border at the Mouhoun between the French and the British territories separated the settlement zones of populations who were found on both sides of the river.
The French administrators who tried to force the Lobi population to mine gold on their behalf were faced with outright resistance. In fact, gold production declined rather than increased, because the Lobi (according to French archival sources due to their ‘savage’ and ‘distrustful’ character) simply refused to cooperate.Footnote 73 In 1901, Capitaine Ruby wrote:
Chez les Lobis des régions aurifères les caractères de cette race, amour de l'indépendance, méfiance excessive, sont accusés d'une façon spéciale, l'on pourrait même dire poussés à l'excès. Sitôt que l'on aborde la question de l'or le Lobi devient intraitable et il refusera de vous faire voir un gramme d'or, qu'il aura en sa possession, même si la vie de son père est en danger (l'exemple l'a prouvé). Aussi sera-t-il très long et très difficile de leur faire accepter notre voisinage.Footnote 74
One year later he seemed more hopeful. The relations with the local populations had improved, and Ruby found that there were more gold deposits than he had initially assumed. Professional prospecting had been undertaken by three mining engineers. They thought the alluvial deposits to be negligible but quartz mining promised to be profitable. In Ruby's comment on local mining prospects, a Eurocentric stereotype about African economic activities was brought forward:
Laissant de côté la question des filons de quartz, qui ne peut être traitée que par des techniciens, le Capitaine Commandant estime qu'il y a au sujet de la richesse des alluvions une contradiction évidente entre l'opinion des ingénieurs de la société du Haut Niger et le travail considérable auquel les habitants ne craignent pas de se livrer pour rechercher l'or. Or les noirs en général et les Lobis en particulier ne sont pas de grands travailleurs et pour qu'ils creusent tant de trous et remuent tant de terre, il faut réellement qu'ils aient de grandes chances de rencontrer des alluvions qui ne sont pas pauvres.Footnote 75
Until 1909, the French saw no possibility of prospecting in the Lobi region without provoking hostilities: ‘Les Lobis considèrent les placers d'où ils extraient l'or comme des fétiches et il est encore dangereux à l'heure actuelle de s'en approcher’.Footnote 76 It was only the interdiction to pay taxes in kind or in cowries in 1909Footnote 77 that eventually forced the local people to mine and sell gold in order to obtain cash.
In the region around Diébougou, gold mining was part of the forced labor imposed by the French administrators: ‘Le chef de canton dans le temps délégait des femmes par village pour exploiter l'or. Ces dernières creusaient dans les marigots. L'or était remis au chef de canton, pour être remi ensuite au commandant. Les hommes n’étaient pas concernés’.Footnote 78
Since the end of the colonial period, Lobi and Birifor women have continued to perform seasonal small-scale shallow-pit mining and panning.Footnote 79 It was only when the present-day mining boom began from the end of the 1990s that larger parts of the populations of southern Burkina Faso became increasingly involved in gold mining.Footnote 80 It is interesting to note that labor migration to the industrial gold mines in GhanaFootnote 81 was not infrequent for men after the colonial period, but those very same men may oppose gold mining on the territory of their own villages.
DANGEROUS GOLD
Even today, some local populations (or some lineages among them) in the region around Diébougou and Gaoua are opposed to gold mining. One reason for this is that gold is considered as a ‘supernatural resource’. Many people believe that gold is a living entity that moves and can suddenly appear or disappear, accompanied by strange lights or sounds. Gold can render a person blind, lame or crazy and can even kill. Some groups or certain lineages are forbidden even to go near gold. If someone discovers gold in the bush or in the fields, one must first ‘kill’ it by sprinkling it with blood or urine before it can be taken, but the corresponding body part will ‘die’ too.Footnote 82 It is not the gold as such that brings bad luck, but rather the hill, tree or water spirits to whom the gold ‘belongs’. Just to give a few examples from the area around Diébougou:
Gold is a bad thing. In the past, nobody hoped to come across gold. If a certain place was known to be rich in gold, nobody would dare build a house there. Ordinary Dagara people knew gold only by name, not in nature.Footnote 83
The Dagara did not know gold as such, because it did not concern them. If someone came across it and recognized it, he would not be allowed to touch it immediately. It was said that you had to sprinkle blood on it before taking it.Footnote 84
I know our ancestors were afraid of gold. They said that if someone built a house on a gold vein, his house would always attract sickness and grief. My own mother is very old. She told me that she was digging for gold during the colonial period. But she stressed that every year, my grandfather would purify her by way of killing a black chicken and a guinea-hen [on the ancestral shrine].Footnote 85
For the Dyan, gold is wealth. Everybody likes wealth. So, all the Dyan should like gold. However, this is not the case; there are Dyan families who have no right to touch gold.Footnote 86
For us, the Dyan, gold is a taboo. It is forbidden to touch gold. One does not even want the money made with gold. One should not pay for things like food with this kind of money. If somebody does, he will die.Footnote 87
Until today, this interdiction is valid. If one of our children goes to a gold mining camp [in order to work in the mine], he is finished; the ancestors will kill him.Footnote 88
Even groups who do not consider gold as a taboo, such as the Birifor, did not engage in gold mining but added gold they had found incidentally to house shrines.Footnote 89 If gold is sold at all, the proceeds from the sale are ‘bitter money’ that must not be used for certain social or ritual payments. Among the Dagara, for instance, the cowries or cash that are exchanged for gold must under no circumstances constitute part of a bride price.Footnote 90
These beliefs explain why gold mining among the Lobi was and still is an activity which causes discomfort and is predominantly considered a task of post-menopausal women. Today, the arrival of gold diggers from other regions sometimes causes violent confrontations between them and the Lobi villagers. For instance, when gold was discovered in a sacred site near a Lobi village in the province of Noumbiel in the year 2000, a fight between villagers and gold diggers ensued that left ten gold diggers injured, two of them seriously. According to the villagers, they had twice asked the gold diggers to leave, because they feared that gold mining on sacred ground could bring death and drought. The fight occurred when the gold diggers returned for a third time in spite of these requests.Footnote 91
CONCLUSION
Though gold mining obviously predated the settlement of the ‘ethnic’ Lobi in the Mouhoun region from the eighteenth century, as numerous remains of old mining shafts prove, there is little evidence from Arabic writings, archaeological excavations or oral traditions to support this observation. There is no doubt that there were Jula warriors and traders in the region from at least the eighteenth century, and that the region was connected to the existing trade networks. There is also no doubt that gold was mined and traded. But there was almost as certainly no hierarchical Mande-Jula political system in this area overseen by the Watara from Kong during the precolonial period. What was there was not a highly organized structure, but segmentary societies who were involved in trade networks, matrimonial and political alliances or temporary allegiances to shifting centers of power in a region which was, according to all available evidence, not dominated by any one polity prior to colonial rule. The Watara neither dominated the whole region, nor controlled gold mining and trade in its entirety.
The situation in this region was similar to the adjacent regions discussed by Mahir Şaul:
In summary, the strong military leaders of the Mouhoun region came from a variety of backgrounds, and the impression of homogeneity created in the literature by cursory reference to descent ties or to the widespread nature of the Watara patronymic is very misleading.Footnote 92
As Şaul has convincingly shown, the Watara warriors and slave raiders who founded powerful ‘war houses’ did not necessarily depend on Kong, although there might have been genealogical relations to its founder Sekou Ouattara. They were independent actors who aimed at maximizing ‘wealth and fighting power’, but not territorial control:Footnote 93 ‘The relations between agricultural communities and war houses, or among the war houses themselves, can be reviewed under the rubrics of diplomacy and war, but not under political dominion, administration, or revolt’.Footnote 94 Warriors or warlords were not ‘chiefs’ but a special kind of entrepreneur. The local populations had ways of dealing with those warlords, either by combating them, or by hiding, or by forging alliances that lasted only as long as the respective warlord was alive. In this case, the villagers paid tribute (probably including gold) to those warlords and were in turn protected by them against the raids of other warlords.
As elsewhere in Africa, gold mining was a seasonal activity, generating small yields which were primarily exchanged for non-local goods like salt, iron tools, or textiles. It was probably also paid as tribute at some time to someone, but since our sources for the history of that region before the middle of the nineteenth century are virtually non-existent, there is no way to say more about it than that. One may speculate that the beliefs about ‘dangerous gold’ originated in a time when gold mining was exercised under conditions of brutal domination by some political entity, but we will probably never know.
