9. Class Struggle and Cannibalism

Storytelling and History Writing on the Copperbelts of Colonial Northern Rhodesia and the Belgian Congo

This chapter, like several others in this book, posits two distinctive sets of historical materials, one expressed in the vocabulary of blood and abduction and one that we might call conventional, the standard narrative and the facts and figures with which that narrative is proved. This chapter is intended to take the notion of two kinds of histories a step further. It argues that the two sets of materials, which are, as we have seen, different in different places, can be compared across political and cultural boundaries to produce a more comparative history, which can in turn reveal a broader range of African workers’ strategies than other sources have done.

Jigsaws, Holograms, and Labor History

This chapter compares vampire stories on the two central African copperbelts, that in the Belgian Congo, where mining was controlled by one company, Union Minière d’Haute Katanga (UMHK), and that on the Northern Rhodesian side of the border, where mining companies—the Anglo-American Corporation and the Rhodesian Selection Trust—competed for labor and profits during the 1930s. Although African laborers, artisans, and hangers-on crisscrossed this border well into the 1940s, working men told different stories of abduction and extraction in both places. The banyama who sought their victims around Northern Rhodesia’s copper mines took men’s blood, but they also captured men’s wills. Banyama stories from the Northern Rhodesian Copperbelt produced some terrifying descriptions of how men became workers:

This story was going around when I arrived at Luanshya in 1935.…The Banyama…snatch live men and sell their bodies to people who need them.…Another story is that the European has a special license from the Government, and he employs natives armed with charms to do his work in secret. These secret agents are never known and they go about at night. Having picked their victim they lure him to a secret spot where they hit him on the head with a “mupila.” A “mupila” is said to have a rubber tube full of medicine which steals the man’s mind and he even forgets his name and can be taken anywhere.^[1]

In colonial Katanga, batumbula captured Africans with the same methods found in banyama and wazimamoto stories elsewhere, but ate their flesh and did not take their blood. Africans were hired by white men to capture other Africans and give them an injection that made them “dumb.” Finally, victims became fat, white or pinkish, and hairy like pigs. Sometimes a special diet transformed the victims into cows. White men then ate them on special occasions like Christmas and New Year.^[2] Sometimes these white men worked for Americans. The revolts of the early 1940s—Luluabourg and Katanga—were fueled by rumors of white cannibalism.^[3] A Belgian priest, Dom Grégoire Coussement, was said to be a batumbula in Elisabethville. After he was transferred to Kasenga, on the Luapula, in 1943, he was said to kidnap Africans on both sides of the river, imprison them in the belfry of the mission church, and drive them to Elisabethville, where they were killed and eaten.^[4]

The next section shows how laborers from Northern Rhodesia worked in Katanga’s mining industry well into the 1930s. This raises the question, why, if the men telling the stories were the same, are these stories so different on the two copperbelts? Finding out is not easy. The source materials for both copperbelts have little in common. Even before the 1935 strike on the Northern Rhodesian Copperbelt, officials and missionaries had been concerned about the impact of migrant labor on African society. As we have seen in chapters 6 and 7, officials worried about the moral fiber of migrants—already compromised in their eyes by how they imagined matrilineality—and the vulnerability and privation of women left alone to farm. The Depression had shocked these officials even more, showing them the grim realities of rural poverty as nothing had before: when the world price of copper dropped by almost half, and all but two of Northern Rhodesia’s mines closed, many African laborers did not return home. By 1933, the International Missionary Council’s Department of Social and Industrial Research, founded in 1930, published the results of its research on Northern Rhodesia’s mines, Modern Industry and the African, which offered a wealth of evidence with which to reconstruct the lives of copperbelt workers, and how they managed urban life. Of the African workforce at Nkana Mine, 10 to 15 percent had Barclays Bank accounts, for example, and a large number of miners at Roan Antelope Mine ordered blankets and other goods by mail rather than patronize local shopkeepers.^[5]

The 1935 strike generated another set of concerns. Typical of the labor protests of the 1930s, African miners had actively and collectively left work, apparently to protest a tax increase, and in the subsequent panic the police shot and killed six Africans and wounded many more at Roan Antelope Mine. Although Africans soon returned to work without incident, officials launched a commission of inquiry that served to inscribe the strike with a level of leadership and organization that their findings disputed. The commission also produced page after page of miners’ and managers’ testimony about work, remuneration, and relations underground and in the compound.^[6] More than dockworkers’ strikes elsewhere in Africa, the 1935 Copperbelt strike had a profound impact on British imperialism: the anthropology of urbanization was born, as were studies of the conditions of mine labor and urban life. Indeed, the banyama story quoted above comes from a government-run newspaper sold—without much success—on the Northern Rhodesian Copperbelt in the wake of industrial strife. Its editor sought to make it the vehicle by which educated Africans could offer information to their less educated fellows.^[7] This particular tension of the reporting of a story by those attempting to prove it wrong informs many of the banyama stories in this chapter. P. K. Kanosa, the author of the quotation above, cautioned that Africans “who have some sort of education” had an obligation to “kill this story,” but noted that as long as “ignorant natives continue to arrive on the Copperbelt there will always be people to believe stories like this.” ^[8]

No such sources, let alone uses of storytelling, exist for UMHK in the 1930s. The details of consumption and pleasure that percolate in and out of the Northern Rhodesian material are absent, as far as I can tell, from the Congolese material. Missionaries in Katanga, as we shall see, did not see themselves as opponents of state labor policies. Visitors who had just come from South Africa, such as Margery Perham, were usually impressed by the differences between the two countries. The Belgian government in exile conducted no large-scale investigation of Congolese labor protests of the early 1940s; it was assumed that Belgium’s status as an occupied country encouraged African extremism. Northern Rhodesia and the Congo do not have equivalent data; the material presented here cannot, in and of itself, yield a reliable historical comparison: data on African banking on the Rhodesian Copperbelt cannot usefully be compared with travelers’ anecdotes about the Congo.

So I ask readers to bear with me and think of this chapter as two kinds of histories at once—the history of the puzzle, in which each and every piece clarifies and completes the picture, and the history of the hologram, in which the image is constructed from the light shed on a fragment. This is a history-writing strategy, a way to get at two different kinds of knowledge at once. The material on consumption and savings on the Northern Rhodesian Copperbelt allows for a detailed examination of how workers regarded their wages and all they contained. There are no comparable data for the Belgian Congo. But there is a level and commonality of description from the Congo that is unlike anything for Northern Rhodesia—descriptions that are so similar that, seen in the same light, they make a hologram, a single image that becomes vivid and three-dimensional when seen through a single, consistent illumination. African miners were not alone in describing the physical transformation that accompanied their sojourns in Katanga. Margery Perham used the metaphor of castration to describe the hospitalized Congolese mineworkers she visited in February 1930. Listening to them “rasping out” responses to a nun, she reflected: “There is something almost ludicrous about it. Ten, five years ago (now, if they had the chance) these men were indulging in tribal warfare, perhaps in torture and cannibalism; now they look as docile and as lifeless as bullocks in a fat-stock show.” ^[9] The barely fictionalized autobiography of a willful Belgian nun in Elisabethville in the late 1930s includes the powerful image of a hospital orderly, sent to sleep off a binge on local brew, eaten by ants: “. . . on the dirt floor was a man-shaped mound of white ants that had eaten Banza clean to the skeleton. Not even a tuft of hair was left on the skull.” ^[10]

Let me argue that the Northern Rhodesian copperbelt is the jigsaw. Between missionaries’ studies, testimonies about the 1935 strike, a generation of anthropologists and another generation of labor historians, the data about the Copperbelt in 1930–45 provide a number of “pieces” with which I can reconstruct a picture. The Congolese side is the hologram—whether I recreate an image from Dame Margery, African miners, or former nuns, the image is always the same: Africans are being eaten alive, their shapes transformed, and emasculated. Each of these images is blurred and incomplete. Taken together they do not add up to the kind of picture one gets from diverse pieces; they add up to something somewhat different, made three-dimensional by the way it is looked at.

Two Copperbelts, Two Histories

The history of both copperbelts may have been made into separate colonial and national histories because few historians worked in both French and English. The history of UMHK has been seen as separate and distinct from that of the mines on the British side of the border. In part this has to do with Africanists disinclination to mix the history of francophone and anglophone regions, and in part it is because the two histories do not provide a good chronological or comparative fit. Although copper mining in Katanga had begun well before World War I, copper production on the Northern Rhodesian Copperbelt only began in the late 1920s, and a few mines were only ready for production when they had to close due to the Depression. More important, perhaps, was the fact that labor policies differed in the two colonies. By 1926, before the first copper mine in colonial Northern Rhodesia was fully operational, the UMHK had begun to stabilize its labor force. Officially, copper mines in Northern Rhodesia did not encourage a family presence, but most mines allowed women to settle in compound housing.^[11]

Both copperbelts had separate and distinct histories with deep play between them. They also had separate and distinct historiographies, and the different sources for production and reproduction on both sides of the border have meant that, with few exceptions, the threads that link the two copperbelts have been overlooked, and the fluidity with which African labor penetrated colonial and cultural borders has not been the source with which history was written. It has also meant—incidentally, I think—that employers’ categories have dominated the analysis more than was necessary, and that colonial mythologies have survived longer than they might otherwise have done. The idea that labor was stabilized in Katanga, while the copper mines of Northern Rhodesia relied on a system of migrancy has gone unchallenged; there has been a disinclination to look for sources that might suggest what these categories might have meant to the men engaged in mine labor in either place.

Belgian and British copper mines in Central Africa shared labor and managerial expertise from a very early date. In its early years, copper mining in Katanga was dependent on Northern Rhodesian labor and British expertise: until World War I, in fact, the main language in Elisabethville was English.^[12] As late as 1929, there were almost 11,000 miners from Northern Rhodesia working for the UMHK, and the most common language spoken in the mines of Katanga was Bemba. In the early 1930s, Northern Rhodesian Africans also seem to have dominated petty trade and artisan labor outside the mining sector in Katanga.^[13] The wages paid to Northern Rhodesian workers were the same on both sides of the border, but had a greater purchasing power on the Belgian side. The preference for working in Katanga may have had to do with commodities, not wages.^[14] The Depression served to make the border more porous than it had been before. Workers of all races migrated in search of better wages. In 1930, white miners crossed the Congolese border in numbers great enough to alarm Belgian officials, and within a few years, skilled, literate African workers, most of them from Nyasaland, crossed the border from the Congo into Northern Rhodesia after the UMHK reduced the wages for clerks and capitães.^[15] But it was not only mine workers who crossed the border between Belgian and British Africa. European peddlers and hawkers and a few Indian traders carried goods across the border and sold them for less than they could be purchased for at Katanga’s newly opened shops. After 1930, African and Greek traders sold dried fish to migrants on both sides of the border.^[16] Northern Rhodesian women also crossed the border with ease. In 1938, district officers complained that several hundred women lived in temporary unions in Katanga with men from Kasai: “There is no intention that it should be a proper marriage—it is only a means of getting money and clothing.” Parents routinely crossed the border to visit their daughters; “they all admitted that they were going to see what they could get from their ‘sons-in-law.’” ^[17] Some sources suggest that one reason for the 1935 strike was that the Belgians had closed the border to Northern Rhodesian trade and produce the year before, thus making Africans more dependent on income from wage labor than they had hitherto been.^[18] Even when the border was officially closed, religious movements, witch-finding movements, and new dance fads crossed the border regularly.^[19] These movements must be added to the picture of African labor—the jigsaw puzzle again—that has emerged over the past twenty years, in which African workers’ gossip passed on the latest information about the safest working conditions, the best living conditions, the best wages. Such talk sent African workers across cultural and colonial boundaries to work and to make money, where they found themselves in worlds and seams and stopes that had different safety records, different standards of living, and different rates of remuneration.^[20] Moreover, such talk placed Africans in worlds and seams and stopes that required different descriptions and different imaginings.

But given the intensity of travel and association, why are the region’s stories about whites who, some way or other, consumed Africans imagined so differently? Is it a matter of storytelling conventions, that a genre of story shaped and set in one place becomes the framework that all future storytellers use? Is there a standard plot in which characters and place-names are inserted? Or are these differences a matter of experience? Is the reality of work and life so different in each place that it can only be described with different narratives? There may not be an either/or answer, of course—stories and imaginings and lives are not such separate domains that they can be rigidly segregated—but the question of whether these differences of detail are about the social construction of narrative or the social construction of experience may reveal far more than is usually discerned from regional studies of the Central African copper industry. I do not argue that banyama and batumbula stories are more important than any other aspect of this region’s history, however. My point is that by looking at these stories, historians can examine workers’ strategies and experiences more closely than they could with other sources.

The writing strategy of the jigsaw and the hologram notwithstanding, how solid a wall is the divide between storytelling in colonial Katanga and the Northern Rhodesian Copperbelt? Images and clichés did not stay put on one side of a border, unable to cross colonial boundaries. They slipped across, where, based on the evidence I have, some took hold and others did not. Two separate accusations involving priests from the first half of 1932 show how stories might travel without settling. In March 1932—in the midst of intense banyama accusations, many directed at two priests at this mission—the monsignor of Chilubula, the White Fathers’ Mission in Kasama District, a major recruitment area for mine labor of both copperbelts, received a letter written in English containing “gross insults not to be repeated” in the mission’s daily diary. It was signed “your good roast mutton captivity, imprisonment, and bandages.” ^[21] It hardly matters whether this letter can be demonstrated to be a banyama accusation, although I think it is. The word “bandages” echoes the medical idiom of contemporary banyama accusations. But “roast mutton captivity” does not correspond to blood accusations in Northern Rhodesia either of this time or later. It recalls Congolese stories of captured Africans being made into animals to be consumed. Two months later, a young priest in the Belgian Congo asked to be sent home. For six months he had been accosted wherever he went by Africans who accused him “of imprisoning women and mistreating children,” he said.^[22] Imprisoning women was a standard theme of batumbula accusations but harming children was not. Indeed, the Congolese painter Tshibumba recalled that when he was a toddler, in about 1948, his mother was accosted by two batumbula, but they said to each other, in Swahili, “Shouldn’t we let her go? She’s with a child. We’ll let her go.” ^[23] But by the early 1940s, however, kidnapping children was a standard theme of banyama stories in the Kasama District of Northern Rhodesia. One official reported that these kidnapped children were killed on Christmas Day, a theme of Congolese stories absent from Northern Rhodesian ones.^[24]

But if the images and clichés of banyama and batumbula crossed boundaries as regularly as they went back, what about individuals, the stock characters of these stories? Did they travel across borders as well? Dom Grégoire appeared in many banyama accusations in colonial Northern Rhodesia. By the time he was transferred to Kasenga, stories about him circulated on both sides of the Luapula River: in Northern Rhodesia, where he was said to be a White Father, Africans claimed that an African man offered the oldest of his wives to Grégoire for the banyama; the priest refused and asked the man to kill the younger one for him.^[25] By 1949, stories about “Gregory” involved the crisscrossing of boundaries, conceptual and political. He was said to pay men on behalf of the Belgian government to cross the river and capture Africans in Northern Rhodesia, who would be sold to his mission, said to be Chibondo, where there was a strange building connected to the mission station in which captured Africans were killed and their brains eaten. The few victims who were not killed were unable to speak when they were found. Images of dumb captives seem to have stuck in Northern Rhodesian banyama stories, whether bundled with those about Dom Grégoire or not. A headman was forced to leave Northern Rhodesia for the Congo; people insisted that he was banyama, charging that he had captured a man from his own village and when they found the man he could not speak.^[26] But not every banyama—or every priest—crossed boundaries. The European called Yengwe on the Copperbelt was said by Africans on the Luapula to have been working with Greek traders in Kasenga and with Belgian plantation owners in the Congo, but only his victims were said to cross the border.^[27] In 1944, an administrator in Northern Rhodesia wrote that Africans in the Luapula River valley believed that an Italian priest was the “head” of banyama, “and if he should cross to our side of the river they intended to kill him.” ^[28] Does this mean that stories and rumors and ideas do not diffuse, but that some individuals and objects carry with them bundled traits and associations that do cross borders and rivers? If so, it may be that these images and clichés, the traveling priests and silenced Africans, may reveal regional rather than local concerns.^[29] The individuals who became the subject of transnational accusations may have been those who embodied local ideas about colonial policies in the wider regional economy. In other words, Dom Grégoire and the headman were associated with speechlessness because he and speechlessness were associated with Belgian rule. The Italian priest who was considered a banyama who did not cross the Luapula had no such associations. I return to these issues at the end of this chapter, but for now my point is that while parts of stories cross borders, they do not do so as storytelling idioms. Whatever their narrative strengths and appeals, they do not survive in local storytelling unless they reflect local thinking and local experiences.

Work and Talk

My question is simple: what is the relationship between the jobs men do and the stories they tell about them? The answer, however, may be complicated: I have no oral material that I collected so as to let Africans reconstruct their workplaces of the past, no “voices.” But even with voices, with Africans “speaking for themselves,” what kind of picture would I have? The idea that a pure voice can be distilled and disembedded from the struggles of colonial experiences is itself problematic. It argues that colonial African language and thought and imagination were not sullied by the categories and constructs of the oppressors. A clear, pure African voice may be an impractical vehicle for the ambiguities that rested in the relation between underground work and the drinking of African blood. Had I, or anyone else in the 1990s, interviewed African miners about banyama in the 1930s, would the interviews be shaped by the increased control and the intensified supervision of the post-1935 Copperbelt? The complexities and contradictions of a past workplace might best be reconstructed from sources officials did not take seriously, stories considered the domain of “ignorant natives,” stories that were in fact handed over to Africans like P. K. Kanosa to debunk.

Let me return to the story Kanosa heard at Luanshya in 1935. Who are these men with licenses and tubes of drugs that can steal another man’s mind? I suggest that this is a discussion (as opposed to a description) of the relations of production at Roan Antelope Mine at Luanshya. At Roan Antelope, there were far more flat, scraping stopes than at other mines, and these stopes required more semi-skilled hoist drivers than other mines. More important, because of these flat stopes Roan Antelope required more underground blasting, and therefore more mineworkers with blasting licenses than any other mine in the 1930s. Indeed, in 1934, 34 percent of the underground workforce at Roan Antelope was classified as skilled.^[30] At Roan Antelope and Nkana mines, Africans were required to take a training course to get blasting licenses, and 112 obtained licenses by examination in 1934.^[31] At Roan Antelope, such a high percentage of skilled workers must have decreased the level of supervision underground, an ideal version of which was set at Nkana at 1 white miner:1 boss boy:16 underground workers.^[32]

But what of the term mupila and its place in mine labor in Central Africa? Mupila is a word for rubber, ball, or football that probably entered Central African languages from Portuguese or possibly French. In the languages of colonial industrialization, Shaba Swahili and Town Bemba, mupila came to refer to the qualities of expansion, a quality stressed in usage more than anything to do with resilience. Thus, while throughout the region kapila means ball, in northeastern Zambia the same term can be used for the mechanism that inflates balls. On the Copperbelt, kapila can refer to miners’ boots; in Shaba Swahili, it can mean a pullover sweater. What then do we make of an early 1930s gloss of mupila as a white tube of drugs?^[33] The use of pila means this cannot refer to tablets of medication, whatever their size or shape. It is conceivable that it refers to a syringe, the contents of which do not expand but contract, although no Bemba-speaker I asked had ever heard this usage. More likely, the meaning of mupila as a tube of drugs rests on another translation, that of drugs: it probably glosses the Bemba umuti, meaning medications or drugs in the sense of any substance taken internally or applied externally for healing purposes. It can thus include bandages—such as those in the salutation of the letter to the monsignor of Chilubula—and splints and poultices. Tubes filled with drugs may well gloss rolls of bandages, which were white and which expanded, the standard first aid equipment carried by boss boys throughout the region. Indeed, the possession of bandages identified an African miner underground as the boss boy.^[34] Men enthralled by mupila may have been captured by boss boys at the very time that blasting licenses fixed the hierarchies of underground work teams. Where most underground mine labor was unskilled, boss boys maintained their position by continually negotiating coercion and consensus. Starting in the mid 1930s, however, blasting licenses may have given them a degree of permanence they otherwise would not have had.

But if the banyama who capture men have many of the attributes of boss boys, what are they doing above ground, and why must they snatch a man’s mind to get him to follow them? Such a story seems to have been specific to the Copperbelt. It had little in common with other 1930s versions reported in rural Northern Rhodesia.^[35] But an earlier version of Copperbelt banyama stories told of European banyama. “In 1930 thousands of Africans…left the Copper-belt and fled home to their villages because they believed that white men were going into the compounds and capturing Africans. The method used was to strike an African with a stick of rubber—mupila—which paralysed him, and then throw him into a lorry and drive him off,” Thomas Fox-Pitt recalled.^[36] The changes between 1930 and 1934 and 1935 might not have had anything to do with the extractive power of mupila or even the uses to which they were put; it had to do with the race of supervisory personnel and skilled labor in the copper mines that operated in Northern Rhodesia during the Depression.

In copper mining, the Depression years were 1929 to 1936, with copper prices starting to fall in mid 1930 and hitting bottom in 1932. But throughout the 1930s, there were substantial improvements in underground conditions, so that in many mines the conditions in which men worked improved and remuneration for skilled labor increased. As late as 1930, the proportion of skilled workers on both sides of the border was high; at the Roan Antelope Mine, the percentage of labor classified as skilled more than doubled in 1931. This meant that even though starting wages were reduced in most of the mines in the region, skilled wages remained high enough that the average wage bill per shift increased. In Northern Rhodesia, the increase in skilled labor coincided with an improvement in underground efficiency, so that there was a reduction in accidents and mortality, while—as would seem probable during the Depression—turnover was at an all-time low, with workers staying on the job for about sixteen months.^[37] By 1935, an underground worker engaged at 22/- could, with overtime and bonuses, earn almost 40/- a month.^[38]

But was the man working overtime the same man engaged at 22/-? Did more than one man work on the same ticket and share the month’s earnings?^[39] Copperbelt workers testifying before the Russell Commission spoke of unemployed Africans living in the compound, men on whom some were eager to blame the 1935 strike. Did these men work a few shifts each month in exchange for accommodation or a portion of a wage? There is probably not enough evidence available to answer this question, but it does provide an additional context for the banyama who could make a man forget his own name. This might refer to the name under which a man worked a few shifts a month.^[40] But the man made to forget his name may have been a fleeting trope from Katanga, a part of the story best understood by men who had spent some time working in and around Elisabethville. Men with the same name—or even men who chose to be called the same name—claimed a name relationship, a practice that in urban Katanga developed into a bond that was often considered stronger than kinship.^[41] There is no reason to have an either/or interpretation of part of a story about banyama, of course. What is important is to note how part of a story might have different and overlapping meanings to different men, depending on where they had worked.

In Katanga, UMHK’s African workforce increased dramatically during the Depression years. In general, each mine had a large core of semi-skilled or skilled workers on contracts supplemented by the number of short-term unskilled employees production processes required. At the same time, the 1920s policy of replacing white skilled labor with black skilled labor was reversed, and the policies of migrancy were applied to white workers. In an attempt to lower skilled wages overall and to take advantage of the decreased mechanization of 1930s mining, a great number of white workers were laid off between 1929 and 1933. But it was usually local white workers who were let go—thus saving UMHK the family wages these men required—and white contract workers from Brussels or London were hired, the cost of whose repatriation was to raise the wage bill per shift.^[42] In 1932, more than 80 percent of UMHK’s African workforce were on three-year contracts. This percentage declined throughout the 1930s. By 1936, contract laborers were graded according to qualities that had little to do with skills. Each laborer was evaluated on a point system according to personal qualities such as intelligence, dexterity, endurance, and education and according to the requirements of the job, its hazards, the training it required, and how many others wanted that job. A wage scale was devised according to the personal qualities with which a worker performed his job. Thus, an “indifferent” surface worker with a rating of four on the scale and three years’ experience would earn a quarter of what a “very good” worker in the same job with a rating of fourteen on the scale and many more years’ experience would. A man stripped of his labor classification could be transferred to an unskilled job in another mine or factory.^[43] This rating system disrupted all the fluidity and all the violence by which men advanced through the ranks of mining workforces elsewhere. Not only did it put African advancement more firmly under European control, it undermined stabilization and the camaraderie of a skilled workforce.^[44]

Although the number of UMHK workers on long-term contracts decreased slightly during the Depression, the company’s involvement in the private lives of those workers increased dramatically. In 1931, UMHK executives asked Dom Grégoire to organize leisure-time activities for their workers who lived in the camp, UMHK’s compound: “It would be quite dangerous if this time were to be organized by Africans themselves, but it is likely that organized activities within the camp will be more successful if there is an intermediary between UMHK and the Africans.” Dom Grégoire was only too eager to be that intermediary, and within a few years he had introduced scouting, TB clinics, and camp schools that banned the children of Protestant and customary unions.^[45] Catholic missions, urban and rural, began to medicalize African childbirth as early as the 1920s. Bottle-feeding regimes to shorten birth spacing and thus end polygyny were introduced in UMHK camps in the early 1930s, supported by Dom Grégoire and coerced by Leopold Mottoulle of the UMHK. The company policy of paying bridewealth in some regions for some workers began in the 1920s as well.^[46] It is not possible to prove that these embodied interventions were translated into batumbula accusations—nor do I think it is necessary to my argument—but both Catholic priests and white mine supervisors were accused of being batumbula.^[47] In several oral batumbula accounts, people describe embodied sensations—inebriation, full stomachs, full bladders—along with cheerful recollections of Thursday vaccinations or mine dancing, while they describe white supervisors eating Africans. A woman complained that men who lived in the camp often went to the Cité—the African settlement outside the camp, and the control of UMHK—to drink and returned drunk and violent. “One group of batumbula would beat men up when they found them drunk at night. These men would lead these men to the house where white men ate human flesh. They never managed to eat all the flesh so they saved the rest in tins, like corned beef,” said Thèrese Mwadi.^[48] According to Kasongo Ngoiy, bodily discipline could protect Africans from batumbula. “People did not go out at night, at least not after 9 o’clock. If you wanted to go to the W.C., you would have to take every precaution to go to the W.C. before that hour. Whoever dared to go out after 9 o’clock was at the mercy of batumbula.” ^[49]

And then there is the story of Kanka Jean, the “official leader” of a pick-and-shovel team at Kolwezi, who disappeared just as UMHK began to dismantle the ticket system early in 1943. His team had been evaluated by a time-and-motion expert, and he disappeared the night before his team was scheduled to compete with a mechanical shovel driver. His fellows refused to believe that he had gone home because of marital problems, as they were told; they went to the compound manager’s home to insist that Jean had been taken by batumbula and demand his release.^[50] Although John Higginson argues that this incident revisited the issues of the 1941 strike, I suggest something more mundane: Kanka Jean was a boss boy, a man whose considerable authority would vanish if the ticket system was replaced by mechanical tools. Although work stoppages and threats thereof were common in batumbula stories in 1940s Katanga, this disappearance may not be of one man only but of a way of organizing labor and extraction underground. According to Kasongo Ngoiy in the 1940s, “Things changed a little when workers got together and demanded that the compound manager help them find their kidnapped brothers.” ^[51] Without a ticket system, the rankings of semi-skilled workers became infused with new meanings. Moreover, these rankings were temporary and not based on seniority. The new manpower regulations that replaced the ticket system favored some men, but such favoritism had only temporary advantages. Some men were treated well, but they were firmly under European control, described in terms of imprisonment and migrancy. “The captives of batumbula did not eat maize meal. They drank sugar water or they ate sugar cane. The captives who were favored this way became fat and hairy, and were taken to the Hotel Biano, where they were killed and eaten. When there were enough captives, one group was transported to Belgium and another to America.” ^[52]

Migrancy, Stabilization, and Clothes

This section addresses the differences between the jigsaw and the hologram, the history reconstructed from many different pieces and the history reconstructed from one piece illuminated over and over again. Through banyama and batumbula stories, it reexamines workers’ strategies on both copperbelts to problematize some of the ways in which historians have dealt with stabilization and migrancy. The batumbula accounts from the early 1940s preface the strikes of 1941 and 1944 and thus provide a point of comparison with the Northern Rhodesian Copperbelt in which the vampire stories of the early 1930s are read through the details provided by missionary writings, the literature of urban anthropology, and the commission of inquiry into the 1935 strike.

In recent years there has been much debate about the nature of migration to and on the Northern Rhodesian—and, indeed, the Zambian—Copperbelt.^[53] Despite a colonial and developmentalist description of rural Northern Rhodesia as a classic labor reserve, forty years of scholarship have shown that migrants to the mines of the Copperbelt tended to migrate from mine to mine, or from job to job on the Copperbelt, rather than alternate periods of work with long periods not working at home. I want to shift the terms of this debate somewhat and look at migrancy in and out of Northern Rhodesia in the 1930s and stabilization in the Belgian Congo, not to establish which copper regime was truly migrant and which was truly stabilized, but to examine the reasons why one or the other might have provided workers with a reliable accumulation strategy in the Depression. My argument is that both migrancy and stabilization were employers’ categories that when taken up by historians have tended to obscure the various strategies by which workers enhanced the value of their wages. For example, the clothing issued to African workers was not simply symbolic or “smart.” Clothing was part of the wage, easily converted to cash or goods valued in and of themselves, and I want to suggest that the wages Africans sought—either as migrants or long-term workers with a family presence—can best be understood in terms of the commodities embedded in those wages.

The wages paid to African miners on the Northern Rhodesian Copperbelt during the Depression are by no means a straightforward issue. Indeed, after recording 800 pages of testimony, the Russell Commission could not decide whether miners had gone on strike in 1935 for higher wages or to protest a tax increase. This was not because workers and mine owners did not know which was which, but because they could not tell which caused the greater hardship. More to the point, “wage” was a pervious category, covering not only the money paid to workers as their regular stipend but bonuses, deductions, tickets, commodities, and clothing. During the Depression, at the latest, Africans in Northern Rhodesia had begun to problematize money both as a medium of exchange and as a token of political authority. As copper mines closed in 1932, there was a shortage of money in the colony, along with widespread rumors that unemployed miners had been promised exemptions for their 1932 taxes.^[54] In the Isoka District, returning miners told the district officer that they could not work in Tanganyika Territory “because the white ants had eaten all the money.” ^[55] But on the Copperbelt there were rumors of devalued English coins and the king of England being jailed for taxing his subjects too heavily.^[56] The next year, there were rumors of an end to British rule, government by black Americans—bringing American currency with them—and, as South African silver coins were withdrawn from circulation, at least one rural administrative headquarters was said to be closed.^[57] Tickets, on the other hand, were things that muddled ideas about time and about money in both Northern Rhodesia and the Congo. Tickets were the method by which the length of service by African laborers was measured: a ticket was a booklet containing thirty slips of paper, each representing a shift, not a day. Wages were paid when each slip or ticket in the booklet had been signed by the ganger in charge; “a contract worked out in months was in practice worked out in tickets,” Charles Perrings notes. Margery Perham, admiring the torn-off tickets workers’ placed on nails in the Congolese mine office each day, wrote: “At the end of six months the whole book goes on. Then there are a whole series of other labels…a blue one for bad work and so a reduction in pay; a yellow one for absence in hospital; a green one for prison, etc. It is possible by looking at a man’s nail to see his whole industrial history.” As late as the 1960s, Africans in Elisabethville saw these tickets as synonymous with time: they worked for UMHK for “6 or 10 tickets” before returning home.^[58] In the 1930s, many mines in Northern Rhodesia paid a bonus to Africans who completed their ticket books in a month, and Nkana Mine paid a bonus of 2/6 for every six months’ continuous work.^[59]

On both copperbelts, commodities, especially clothes, were the subject of uneasy negotiation. Whether boots, leggings, and hard hats were issued free or charged for, they seem to have been regarded quite differently by managers and African workers, for whom they seem to have constituted part of the wage. Certainly, they were converted to cash as often as they were used. In Elisabethville in 1930, Margery Perham heard of a Belgian businessman ruined by the liberal regulations of the Congo. He had provided each of his African workers with a blanket of good quality as required by law. The workers promptly sold them, and when the inspector came, the businessman received a heavy fine for not providing his workers with blankets.^[60] In Northern Rhodesia, however, which items of clothing were free and which were deducted from wages, and from which kind of wages, varied from mine to mine. According to the Russell Commission, at Nkana Mine, where well over 80 percent of the workforce earned under 30/- for thirty days’ work, there was no bonus, but one hard hat, one blanket, and one pair of boots were issued free. After that, the cost of these items was deducted from wages. Two candles per shift were issued free to underground workers in the early 1930s, but the uses to which they were put and Nkana’s solutions to free issues and deductions reveals the extent to which Africans sought to make homes under the conditions their employers called “migrant.” Africans tended not to use their candles underground, saving them to light their rooms at night. Enforcing candle use proved difficult, and in 1932, Africans asked if they could have lamps, offering to buy them. “The same light that is supplied to Europeans” was deducted from African wages at 8/6, “voluntarily, for the reason that they have not only got good light underground but can also use them at their houses at night.” ^[61]

At Mufulira, most of the workforce was unskilled and worked above ground in construction. Ninety-five percent of Africans earned under 30/- for thirty shifts, and one hard hat and one waterproof coat were issued free; boots were compulsory, but miners were charged 10/- for a pair (the Commission noted that they cost Mufulira Mine 15/9) and lamps could be purchased for 8/-. The cost of these items was deducted from thirty days’ pay at the rate of no more than 10/-. At Roan Antelope Mine, the system was almost arcane. Just over half the African workforce earned under 30/- per thirty shifts, and 85 percent earned under 40/- per thirty shifts; all workers were issued hard hats free, but they were charged 20/- for boots, 13/- for coats, and 2/6 for leggings. Lamps could be purchased for 7/9. Workers received a bonus of 11/3 for working thirty days, from which the cost of clothing was deducted at the rate of no more than 11/3 per thirty days.^[62] Workers could recoup the cost of boots, leggings, and hard hats by thirty days of work for wages. African workers did not necessarily think of this bonus the way their employers did, however. In 1936, miners complained to the visiting Ngoni paramount chief that they were charged for their first issue of clothing.^[63] It seemed that African workers considered sequential issues of clothing to be part of their wages. By 1935, Nkana’s safety officers ruefully noted that the boots issued African miners were almost at once sent home or sold. The second pair was deducted from their pay.^[64] These deductions—20/ for boots and 8/- for lamps at Nkana—were contested; a mineworker from Mufulira said this was a bigger problem than taxes, and the compound manager from Roan Antelope said that Africans frequently said they would prefer the money.^[65] But at Roan Antelope Mine, many men seemed to have stayed on the job for the commodities provided there. When output restrictions decreased production at Roan Antelope by one-third in early 1935 and retrenchments threatened semi-skilled African workers, desertions were at an all-time low.^[66] But what about men who worked short periods at different mines, each time getting a new pair of boots? What about the man who had worked at two mines in eleven years, and then, during the Depression, worked at Mufulira for eight months, then went back to Nchanga for five months, then to Bwana Mkubwa for six months, and then back to Nchanga?^[67] Were such migrations within the Copperbelt—the source of academic debates about how stabilized the Copperbelt actually was—to do with wages, retrenchment, or the opportunities of a free issue of boots and candles?

Different mines’ amalgamation of free issues, compulsory clothing, and wage labor muddle any apparent difference between the three categories on the Northern Rhodesian Copperbelt. From the evidence of banyama stories, it would seem that these policies naturalized clothing, making it much more a part of African bodies than compulsory dress underground. Thus, in 1931, the Copperbelt rumor was that banyama had white balls, called mupila, “and the balls were thrown into the path of the victim, and as the victim came up the banyama spoke to him. If he answered then all his power left him, his clothes fell off, and he no longer had any memory or will.” ^[68] Clothing was not a free issue or a compulsory purchase; it was part of a person, like will or speech or potency, that could be stripped away by the dreadful power of the boss boy. This suggests that commodities do not become commoditized when the waged context in which they are acquired blurs the line between money and commodity.

With substantially less secondary evidence, African ideas about clothing in Katanga can be read in batumbula stories. Vampire stories from elsewhere in East Africa are less concerned with clothing, whether that of the victims or the abductors. At least one Ugandan said it was impossible to say what the bazimamoto were wearing, because they did their work at night. Others said they wore black, or white coats and black trousers. Clothing figures frequently in twentieth-century industrial urban legends and rumors, however. Folklorists have written extensively about the stories of snakes found in the sleeves of department-store coats, of young women abducted from dress shops in France, and of racist slogans sewn into the linings of jackets sold to American youths.^[69] All of these are stories about clothing or consumption rather than stories in which something happens to clothing. These narratives animate clothing with relationships, dangers, and risks. They valorize and criticize commodity fetishism with stories of clothing’s hidden affiliations. The clothing described below has no such hidden meanings; the relations of production and reproduction are present in it and in what happens to it. The clothing in the batumbula stories below does not depict fetishized commodities that conceal the relations of production; on the contrary, it depicts the processes of wage labor, leisure, and capture. According to the artist Tshibumba, batumbula “wore long black coats and miners hats with lamps”; when they caught someone they took his clothes.^[70] A former mineworker recalled that in the 1940s, batumbula “wore hats so they could not be recognized.” ^[71] In Leopoldville in 1960, Africans spoke of muntu wa mudele, “the men with the lamp,” Europeans who captured Africans and ate them.^[72] Again, in the 1960s, batumbula in Katanga were said to wear dirty, slovenly clothes,^[73] a wary comment on stabilization and independence. Indeed, if Northern Rhodesian migrancy allowed for several pairs of boots and hats—and the money obtained from reselling them or the loyalties accrued from sending them home—stabilization seems to have left Congolese workers feeling naked and vulnerable, their hats, lamps, and uniforms turned against ordinary miners. Recalling how batumbula captured people in Elisabethville in the 1940s, men and women interviewed in 1991 said “they undressed them and tossed their clothes far away from the place of capture. When people found these clothes the next morning, they knew batumbula had captured someone.” ^[74] “When batumbula kidnap people, they do not come back to pick up the traces. You see clothes thrown on the road,” Kasongo Ngoiy claimed.^[75]

Why were clothes thrown out on the road such a powerful image that they informed popular paintings and were talked about fifty years later? Karen Hansen has noted that there was a trade in second-hand clothes from the Belgian Congo to Northern Rhodesia beginning in the 1920s.^[76] The existence of such a trade would have made stories of abandoned mounds of clothing all the more chilling, and make the purchasers of such garments seem cynical and monstrous. At the same time there seems to have been a demand for luxury clothing. As early as 1929, White Fathers complained that villages in Northern Rhodesia nearest Katanga had emptied as men sought clothing: “The fashion is hats and white shorts, shoes and stockings.” ^[77] In Elisabethville itself, men needed dress suits for malinga—ballroom—dancing. In the domestic workers’ history of Elisabethville, written in 1965, men recalled that even in the midst of the Depression, when “everyone suffered a lot from hunger.…The big thing was to go to the malinga dance.” When a man had work, his friends came to his place “to eat and to clothe themselves.” ^[78] The layers of relationships and layers of clothing may have had a straightforward meaning in the regional economy of both copperbelts, however. A man who had lived in Elisabethville for many years recalled that in shops, clothing was purchased in sterling.^[79] Ready-made clothing may have been more than a scarce commodity; it may only have been available to the few who with access to the currencies of British colonies. But what about the clothes available to Congolese workers? Well into the 1960s, domestic servants complained of Belgians’ unwillingness to allow Africans to wear shoes on the job.^[80] For miners, the clothing issued in the late 1930s and early 1940s was thought by miners to be insufficient for the work required of them. By 1940, underground and factory workers complained bitterly that they were not issued shoes, jackets, or shirts that were sturdy enough to protect them from serious injury. Furnacemen demanded heavy woolen shirts to protect them from burns from flying ash. In November 1941, a few weeks before the first of many wartime strikes, a company official complained that African peddlers, selling mainly clothing, invaded the camps every payday. In strike-torn war years, commodities became disaggregated from miners’ wages. John Higginson has noted that the connection between the workers and the peddlers, with its inevitable relations of credit, informality, and multiple currencies, was strengthened by the weakness of the company’s relationship to its workers.^[81] Africans, however, demanded a return to the bundling of clothing, wages, and stabilization. On 4 December 1941, when wives joined strikers at the central workshop at Panda, they demanded that the company reinstate the incentives it had formerly given women to remain in the camps, mainly sewing machines and cloth.^[82]

In the three tense years between the strikes of 1941 and 1944, a number of batumbula accusations terrified European personnel in UMHK camps. But the African men armed with machetes who surrounded the houses of mine supervisors and compound managers did not demand a return of their clothing, but of their wives. Their grievances stemmed from the tensions of a family presence. In January and February 1943—around the time of Kanka Jean’s disappearance—European supervisors and camp directors were threatened. In the camp at Sofwe, men armed with axes and machetes surrounded the manager’s house shouting, “You have eaten the woman, you took her by force.” ^[83] In other camps, Europeans required police protection when workers attacked their houses, accusing them of eating missing women.^[84] The director of Mwale camp was trapped in his house by workers shouting, “You have taken the wife of our brother, you are not looking for her, but we will finish the whites. You, you think you are allowed to eat this woman?…whites will not eat sugar after tomorrow, Monsieur Donnay is finished, he is batumbula, a lion of Europe.” ^[85] Tensions around clothing, embedded in other versions of these stories, were part of the nervous relations around matrimony, property, and the place of UMHK in allocating them. Even the most cautionary tales had clothing in them. Joseph Kabila Kiomba Alona recalled “that in many arguments between husband and wife, wives are in the habit of packing a suitcase and returning to their parents’ home. For batumbula this was an opportunity to capture women who abandoned their husbands late at night.” ^[86]

Stories and Migrants

Many of the qualities of batumbula stories appeared in accounts of banyama in the late 1940s in eastern Northern Rhodesia, on the border with colonial Nyasaland. These were, according to the young researcher John Barnes, “people (i.e., Africans) who are employed by Europeans to capture people…by touching the captive with a wand, which made the captive invisible and helpless.” Sometimes they used a lorry at night: “All the children from the villages run out to see and the amnyama who have come in the lorry touch the children with their wands.” The children “cannot help getting on the lorry” and are taken away, across the border. Captives were taken to Lilongwe “where they were fed on special foods that make them very fat. Then they are in some cases killed and their blood drunk by the European employers of the amnyama, in other instances their blood is just drained, and eventually they get back to their villages very emaciated.” Barnes’ informants, many of whom had worked on the Copperbelt, assured him that amnyama also operated there, but they used a “flexible piece of rubber” instead of a wand.^[87]

Kamupila or kampira (ka- is a Bemba prefix meaning diminutive) became synonymous with banyama in various places: in Lusaka in the early 1950s, in Broken Hill a few years later, and in Southern Rhodesia in 1960. In Lusaka, kamupila was a synonym for banyama, and stories similar to the Congolese ones were told using specifically Northern Rhodesian idioms and images. The capture of children that had become a feature of Northern Rhodesian stories in the early 1940s was refined in Lusaka, where kamupila were said to capture children, give them injections that made them docile and dumb, and lead them to a faraway place, where their captors drank their blood. Sometimes, adults were captured, and a rubber ball was forced into their mouths as they were marched off to the Belgian Congo for slave labor.^[88] In the ferment of anti-Federation activity in Northern Rhodesia in the early 1950s, rumors circulated that whites had poisoned the sugar sold to Africans, that whites gave away soap that would sap the will of Africans, that tinned meat marked for African consumption was made from human flesh.^[89] Officials took extreme action: two clerks who printed a pamphlet warning Africans about the poisoned sugar were tried and sentenced to three years each, while on the Copperbelt the district officer in Kitwe ate a can of meat at a public meeting to show how harmless it was.^[90] But stories still circulated: the pro-Federation Capricorn African Society was banyama, and government newscasters—two of whom played guitar and sang in the Central African Broadcast Service’s most popular show, the Alec Nkhata Quartet—were banyama as well, reading news that made Africans lose their will.

In 1957, the African Welfare Department in Broken Hill had to confine its youth and boys’ club activities to daylight hours because no parents would let their children out at night for fear they would be taken by banyama to their headquarters in the Belgian Congo.^[91] A year later, in Fort Jameson, it was said that the White Fathers had chosen their victims in advance and had marked their clothes with “the Sign of the Cross,” which was invisible to all but the priests and their African henchmen. When enough victims had been marked, lorries were sent out to the villages, whistles blew, and “victims of the Cross” were “collected.” ^[92] In 1960, in what was then Southern Rhodesia, stories resembling those specific to Katanga far more than those specific to the Copperbelt of Northern Rhodesia were told. Kampira in “a typical district”—according to Brelsford, Fort Victoria—was something that was “alleged to mark miraculously the clothes of people for future slaughter by the cannibals known to exist in the northern territories and particularly the Belgian Congo.” People in the district “believe that a big lorry comes across the country and when it hoots all those who have a mark on their garments will rush to the lorry to be carried away to ‘Burumatara’—a mythical land. On arrival there they are injected with a chloroform solution which changes them from human beings to pigs. They are then fattened up prior to slaughter.” In Salisbury, the capital, Africans were captured, given injections, loaded onto a Sabena airplane while they were unconscious, and changed into pigs during the flight.^[93]

What is migrating here? The stories, the details and the words, the storytellers, or the circumstances that made these good and credible stories? Why do idioms and images that do not take hold fifty miles away from where they are first told take hold a thousand miles away fifteen years later? All of these stories were most likely carried by returning migrants. Why did they tell these stories, in all their variants and power? To entertain or to articulate grievances and animate political analyses? Did a part of the story that was not believed in Mufulira in 1934, for example, find a credible audience in Fort Jameson in 1948?

Such questions may isolate these stories from other parts of speech. Men and women may not repeat rumors so much as they report the hodge-podge of events, ideas, and images they have heard circulating in and out of where they worked. If rumors are narrative attempts to figure out the meanings behind events, postwar Africa was a fertile ground for rumors as the cost of living rose amid stagnant wages and increasing political demands, made both by Africans and colonial powers. In Central Africa, this tension was heightened by the Central African Federation, which was to unite Northern and Southern Rhodesia and Nyasaland into one political unit, making settler control greater in Nyasaland and Northern Rhodesia than it had been before. There is no clearer indication of the distrust felt by Africans during this period than the widespread stories of drugged and ingested Africans, kidnapped children, poisoned sugar, and dangerous soap. Thus, in this section, I shall not discuss what these rumors are about—they seem to describe how the last gasp of colonial control was imagined—but how they come to contain so many elements of 1930s and 1940s Congolese stories that had not been in widespread use in stories on the Northern Rhodesian Copperbelt during that time.

The question is, who knew the bits and pieces of these stories, whom did they tell them to, and how did they tell them—which details did they put into a story? All the evidence above comes from written sources, so it is impossible to tell who said what, let alone who said what to whom. A. L. Epstein’s study of gossip on the Northern Rhodesian Copperbelt in the late 1950s shows how people conversed across a number of domains. These conversations did not distinguish between migrant and countryman, between stabilized worker and new arrival, between wife, mother, and young man in the context of the urban centers to which people have come and gone and come again. Epstein heard of an unremarkable (but nonetheless amusing) case of adultery in Ndola from one of his research assistants, who heard the story from a neighbor. The assistant knew the adulterer because both men played on the same football team. The neighbor had gone to school with the wronged husband, and they were still close friends. The neighbor had told Epstein’s assistant the story when he visited his house while he (the assistant) was entertaining a close friend. Epstein also heard the story from a schoolmate of the wife’s, whom he knew because her husband was a senior clerk and political activist on the Copperbelt. She, however, had not heard it from the wife herself but from another woman who had been to the same school. This woman then told several woman friends and a younger man from her home area, who had quite a good job in town. Epstein then heard the story from another research assistant, who had grown up in the mine compound at Mufulira, where his parents were close friends of the adulterous wife’s parents. This assistant had met the adulterous man at a boxing match at a boy scouts’ jamboree in 1947. But he had not heard the story from the wife or the adulterer but from the wife’s mother, who lived on the outskirts of Ndola, whom he visited often. This assistant soon heard the story again from the young man who had heard it from the wife’s schoolmate. The young man had told Epstein’s assistant one night in the beer hall because he himself was involved with a close friend of the wife’s, who sometimes let the adulterous couple use her house but feared that her husband would think that her sanctioning such actions would cast doubt on her own fidelity. The young man was quite concerned about the extent of gossip in Ndola, since his activities might easily be disclosed, and complained bitterly about the woman who told him this story and the impact her gossip about adultery had had on his wife.^[94]

The issue then is not who tells stories but the number of overlapping ways in which people hear stories in urban situations of great mobility and even more affiliations and loyalties. Childhood, school, sport, home area, new friends, old friends, and relations all provided occasions where news was exchanged; untangling who heard a story first and who passed it on may not be as important as noting that someone who hears news might not be the person to whom it was originally told and may pass it on for different reasons than it was first repeated in their presence. And stories may not be heard the same way by all who listen to them: some may hear about the threat of banyama to their children, others may hear a story about the danger of injections, and still another may hear an explanation for the increased availability of canned meat.

The capture of children and the draining of their blood that emerges so strongly in banyama accusations in Northern Rhodesia’s Northern Province in the early 1940s may have been a way to talk about the status of children in households in which some husbands demanded bridewealth and increased control over their children, while others wanted to be able to pawn those children.^[95] On the Copperbelt, children had been considered sources of industrial disorder for years. Between 1935 and 1940, officials proposed a variety of plans to send children over ten back to rural areas. These were all unworkable, and officials noted that however much they wanted children gone, parents wanted them residing with them.^[96] But in Fort Jameson, the situation might have been different for the patrilineal Ngoni. There the kidnapping of children may have reflected a wider change in the regional economy, and the involvement of children in agricultural labor in neighboring Nyasaland.^[97] But what of the secretly marked clothing of the late 1950s? Is this a version of the clothing of batumbula stories? I think not. Even though many Ngoni men worked on the mines, and many of them and perhaps a few of their fathers worked in Katanga for a time, stories about cast-off clothing in the workplace do not necessarily diffuse into stories about marked clothing hundreds of miles away. The power of supervisory staff in Katanga, with their hats and dark glasses, to strip their victims of the very clothes that wages and employers provided reveals the anxious place of dress in wage labor. The power of banyama in Fort Jameson fifteen years later to mark clothes and persons with a sign only they could see reveals something both broader and more menacing, the subtle power of the colonial state and the variety of operatives that did its work. These stories made the most sense in both places when they included trucks and rich foods rationed to Africans for the most horrific of motives, just as stories of kidnapping and consumption in 1940s Katanga and late 1950s Southern Rhodesia made the most sense when told with details of injections and canned meat. How stories make sense and how they become credible returns us to the question with which I began this essay: do banyama and batumbula stories describe experiences or do they conform to rules of storytelling? Do they sound like reasonable stories because of how they are told or because of what they are told about?

Stories and experiences, performances and topics are not separate and discrete domains; a story sounded reasonable because it described what Congolese mine workers or Ngoni fathers considered the likely motives of Europeans; rumors contained other proofs—hazardous injections and stolen clothes—that both added credibility to the story and structured it in credible ways.^[98] But the issue here is not that people must believe stories to retell them or act on them. They probably do, but how is that belief established—and for how long and in which contexts—by the stories of kidnapping and injections themselves or by how those stories are told?^[99] But such a distinction may not be necessary and may not explain why a story becomes widespread and powerful. Instead, banyama stories function both as rumor and gossip, simultaneously containing contradictions that underscore the wider contradictions in the late colonial project. The details of storytelling reveal the details of experience—the hidden motives of the state, the dreadful importance of clothes, the true meaning of colonial medicine, and the dramatic and contradictory tools with which it is carried out.

Northern Rhodesian rumors about government newscasters in the early 1950s may fuse some of these issues in particularly useful ways. The proposed Central African Federation was opposed by African political organizations in Northern Rhodesia, particularly the African National Congress, led by Harry Nkumbula. It was also opposed with popular distrust that was probably not manipulated by the ANC. If Africans in Northern Rhodesia actually supported Federation, it was generally through membership in the liberal, multiracial, elite Capricorn African Society, originally founded in Southern Rhodesia. This group’s members tried to recruit educated Africans in Northern Rhodesia and other British colonies.^[100] Although it never had serious status as a political party either for whites or Africans, it did actively support Federation. Africans in the Capricorn Society, or Africans suspected of such membership, were accused of being banyama and police informers as early as 1950.^[101]

The association of the Capricorn Society with banyama reached its peak in Lusaka in 1952. This was in part fueled by the play of rumors of drugged food and commodities between the Copperbelt and Lusaka, and in part by the tactics of Capricorn Society members, whom European liberals opposed to Federation accused of terrorizing women and children in the capital with “drunkenness and hooliganism.” ^[102] By the end of the year, fears were such that an ANC meeting in Lusaka passed a resolution condemning the government for “failing to deal with the Vampire men threatening the peace and order…of the country.” Two months later, Harry Nkumbula wrote to the member of Northern Rhodesia’s Legislative Council who represented African interests “with regard to the vampiremen incidents it is high time that Government took action.” ^[103]

At the same time, radio announcers were accused of being banyama. According to Peter Fraenkel, the director of the Central African Broadcast Service, the reasoning behind these new accusations was logical: “How could the announcers broadcast ‘bad news,’ news which displeased Africans, unless they had lost their will-power? How else could they be made to read pro-Federation propaganda on the air?” ^[104] These accusations have their own histories: one broadcaster was so frightened of being poisoned that he cycled miles to his home village once a week to buy food; another, Edward Kateka, was accused of kidnapping a child and took refuge in a police station; Alick Nkhata, arguably the most influential of Central Africa’s influential guitarists, wrote a song about banyama stories, which was well received as a dance tune but made little impact on the rumors.^[105] These however are beyond the scope of this chapter. What I want to discuss now is how men whose job it was to tell stories came to be accused of being banyama—of being, in short, something made real by storytelling.

Did every accuser believe these broadcasters were banyama? Did they think they actually sold Africans to whites for their bodies and their blood? Some did. An anonymous letter threatening Nkhata and Kateka contained many of the details and hierarchies of Copperbelt banyama stories: “You people Capricornists,” it began, “you are selling your people to Yengwe in Ndola.” Although Yengwe/Arthur Davison had died in 1951, the court interpreter explained to the judge that he was a European who lived in Ndola who supposedly bought Africans and sold them to the Congo to be eaten. “[Y]ou wanted to kill Nkumbula you even received revolvers from your Minister the General President of the Capricornists…you are all civil servants…you are the people pretending to become maneaters, kamupila, ” the letter claimed. The colonial state’s response was swift, but perhaps not what Nkumbula had in mind: handwriting samples were sent to police in Livingstone and South Africa, who examined 200 of these before they found the culprit, a twenty-year-old office worker from Western Province. He claimed that he had only wanted to warn the broadcasters. The magistrate observed, “This is not a letter of kindly warning” and sentenced the man to five years’ imprisonment.^[106]

But others believed the story, rather than the literal existence of banyama and Yengwe. According to Nkhata, “My best friends are afraid to come see me…I am quite alone. They don’t even greet me in the street. They cross over to the other side to avoid me. It is not that they believe this nonsense, but they’re afraid. They think they too will be accused if they are seen with me.” Another announcer complained that he could not fathom the reason for this accusation: “Why us?…They’ve always trusted us. I know they’ve mistrusted other departments, but then people like the Forestry officers do bully them to stop cutting wood. But we, we’ve always been popular. We bring them education and entertainment and we don’t even ask a license-fee. Why pick on us now?” ^[107]

What did cause these accusations? Was it the general panic, the men themselves, the stories newscasters told on the air or the way they told these stories? Would Federation described, however favorably, with reference to injections or lorries have been as suspicious, while Federation described as a negotiated political process aroused fear? Given the evidence available—banyama accusations filtered through so many layers of European translation and summary that the narrative is lost—I cannot say what kind of stories were told about banyama, much less about the kind of stories newscasters told. But the experience of Alick Nkhata in concert, performing a song that joked about banyama, supports the view that it was the work that was suspicious, not the man. Nkhata and his quartet performed in the midst of the most aggressive banyama accusations: “A large audience turned up, paid their entrance fee and applauded his amusing songs about modern town-life.…The new song about vampire-men went down fairly well and caused some laughter.” But during a break Nkhata went outside, and members of the audience standing there shrieked and ran away.^[108]

Nkhata the broadcaster might have been thought banyama, but Nkhata the guitarist was not. The song itself named Kateka, named the Information Department, but included none of the details of head banyama, injections, and fattening captivities that made banyama stories powerful and immediate. Nkhata’s song discredited banyama stories because they were false, not because of what they were about. “Who is Nyama? / Nobody knows him / Some say it is Kateka / Who works for the Information Department / There are no Vampiremen / Do not scare people.” ^[109] The experiences, real or imagined, of marked clothes, stolen children, dangerous injections, and fattened captivities were absent from a song that only reprimanded African ideas. It may not have been frightening because it contained no evidence of banyama.

A banyama or batumbula story that travels, that carries words and ideas hundreds of miles across borders for many years, is told with experiences. Without those experiences, however compressed and cryptic they might be, it is just another song—another example of educated Africans’ condescending advice, perhaps the reason for the audience’s laughter. These experiences did not need to have the same meaning to all who heard the story and repeated it, but their presence in a narrative, told with injections and marked clothes, rather than any specific storytelling conventions, made it a banyama story.

Notes