Truth in Vampires, Truth in Oral History

A simple premise undergirds my interpretation of vampire stories in this book: people do not speak with truth, with a concept of the accurate description of what they saw, to say what they mean, but they construct and repeat stories that carry the values and meanings that most forcibly get their points across. People do not always speak from experience—even when that is considered the most accurate kind of information—but speak with stories that circulate to explain what happened.

This is not to say that people deliberately tell false stories. The distinction between true and false stories may be an important one for historians, but for people engaged in contentious arguments, explanations, and descriptions, sometimes presenting themselves as experts, or just in the best possible light, it may not matter: people want to tell stories that work, stories that convey ideas and points. When Gregory Sseluwagi in Uganda became exasperated with my assistant and I hectoring him to admit that vampires did not exist, he said, “They existed as stories,” and it was that existence with which he and his fellows were confronted daily.^[74] For this man—and for historians—true and false are historical and cultural constructions. They are not absolutes but the product of lived experience, of thought and reflection, of hard evidence. “During the colonial period, I could not believe there were some people who could abduct people. I would ask myself, how could someone go missing? Could somebody disappear like a goat? But when I learned of my brother-in-law…taken by the Amin regime…then I understood. But for some of us, who did not know anybody captured by bazimamoto, it was impossible to understand it.” ^[75]

For most of the people quoted in this book, experience was true, but not as reliable as hearsay, the circulating stories that helped a person understand what had previously been incomprehensible. There was a widespread belief that talk was rigorously grounded in fact. Its opposite was the “loose talk” that characterized the Swahili-speaking people of the East African coast.^[76] Children were brought up not to speculate idly.^[77] The way to prove that vampires were real was to say so: “This is not just a tale, nor something you gossip about,” the Congolese painter Tshibumba told Johannes Fabian.^[78] Experience shaped narratives insofar as it was assumed that everyone spoke the truth. “If I am stealing bananas and they talk about me, they say I always steal bananas. But can they talk about somebody they don’t know, and say he is stealing?…Now I have seen this recording machine. If I had not seen it, I wouldn’t be able to talk about it, but because I have seen it I can talk about it.” ^[79] Put simply, “people were not crazy just to start talking about something that was not already there.” ^[80] The issue was not how well argued a story was—what Paul Veyne has called “rhetorical truth,” established by eloquence and elegance^[81]—but how readily and commonly a story was told. “It was a true story because it was known by many people and many people talked about it. Therefore it is a true story and it is wrong to say that it is not because they would not talk about it if it was not true.” ^[82]

But how well can oral historians trust informants to talk about what’s true, especially if, as I argue, what is true is so historically constructed as to be beyond generalization. Some believe that, like trial lawyers, oral historians should not ask leading questions to elicit facts that can be evaluated on their own terms to arrive at a single truth explaining one version of events. Interviewers must be neutral; otherwise they risk people telling them the stories they think the interviewers want to hear. Jan Vansina has cautioned against leading questions with a calculus of participation and exclusion: “Any interview has two authors: the performer and the researcher. The input of the latter should be minimal.…Indeed, if the questions are leading questions, such as ‘Is it not true that . . .’ the performer’s input tends to be zero.” ^[83] This book argues something very different, that absolute notions of true and false, of interviewing technique and legalistic practices, are simply overwhelmed by local ideas about evidence, ideas that are continually negotiated and renegotiated by talking.^[84] In the following exchange, who is leading whom, the way informant and interviewer toss concerns about expertise and knowledge back and forth, indicate the ways in which evidence, especially oral evidence, is produced in contentious dialogue:

q:

Some people have told us that wazimamoto kept their victims in pits. Did you ever hear this?

a:

No, I never heard anything like that.

q:

Some people have said that wazimamoto used prostitutes to help them get victims. Did you hear that also?

a:

Yes, I heard that wazimamoto used prostitutes for such purposes.

q:

That means these stories were true?

a:

Of course they were. Who told you they weren’t?

q:

Nobody told me, it was just my personal feeling that these stories were false.

a:

These stories were very much true. Those stories started in Nairobi when racial segregation was there. Whites never shared anything with other races and whites were also eating in their own hotels like Muthiaga.^[85]

The slippage between confirming facts, hearsay, and geographical knowledge bordering on political economy is typical of wazimamoto stories told by former migrants in western Kenya. But the slippage also poses a disjuncture between academic historians’ and the speakers’ notions of truth. While historians might be most concerned with which parts of the account are true and are thus useful in historical reconstruction, the speakers seem engaged in problematizing what is true, and establishing how and with what evidence a story becomes true. It is not that truth is fluid, but that it has to be established by continually listening to and evaluating new evidence. The material basis of historical truth is not eroded in such accounts, and the mediation of language is no stronger than the events it describes. Something much more subtle is going on, something oral historians may be better placed than other historians to appreciate, that the use of language is the analysis by which people ascertain what is true and what is false, what they should fear and what they can profit from. It is through talking that people learn about cause and intention. Language and event—even language and événement—are not opposites, but in constant dialogue and interrogation. Accounts of the past are documented with words, with descriptions of social relations and of material objects, even as the relationships between the men and women narrating these accounts are negotiated as they speak.^[86] Old words, new terms and neologisms, circulating stories and eyewitness accounts, and the insights of the odd interviewer all add up to make a bedrock not of experience but of the ideas on which experience can be based. Turning those words and stories into the tools with which a historian reconstructs the past is not a matter of transforming them into something else, but of giving the words and stories the play of contradiction, of leading question, of innuendo and hearsay that they have in practice. Oral historians have not always done this well. In an early, important critique of the use of oral tradition, T. O. Beidelman complained that historians tended to make African culture static to make traditions into historical facts; finding out what really happened obscured how traditions were used on the ground, how they held “social ‘truths’ independent of historical facts.” ^[87] But the line between different kinds of truth is flexible. Historical facts—like knowledge of segregation and the elite settlers’ club in colonial Nairobi—emerge from social truths, just as social truths develop from readings of historical facts. Hearsay is a kind of fact when people believe it. It is impossible to say that wazimamoto stories, told and retold in East African cities, are independent of historical, or social, or sociological fact. The 1947 riot at the Mombasa fire station is but one example. In October 1958, Nusula Bua was arrested at the Kampala fire station for offering to sell them a man for 1,500/-. He told the fireman he spoke to that he had “about 100 people to sell.” Bua was sentenced to three years, because it was his first offense. According to the magistrate, “People must know that the Fire Brigade is not buying people, but is intended to extinguish fires in burning buildings and vehicles.” ^[88] It took more than officials’ statements to get people to believe that firemen just put out fires, however. In 1972, the Dar es Salaam section of the Tanzania Standard published a half-page article, “Firemen are not ‘chinja-chinja.’” ^[89]

Part of what made hearsay so reliable to those who repeated it was that it could resolve some of the confusions that experience actually contained. What happened to people was not always so clear and explicable that they would immediately appreciate its full import, or always have the right words to describe it at the time. Joan Scott’s essay “The Evidence of Experience” (1991) and its critics have noted the limits of the project of social history. The goal of widening the range of experiences that could constitute a national, occupational, or sexual narrative simultaneously reinforced a notion of experience in which individuals are the foundation of evidence, the ultimate authorities on what they lived through. Whatever fractures and fissures in individuals’ senses of themselves and their worlds that shape first-person accounts are lost: instead, Scott argues, “raw events” produce raw analyses, visual and visceral, outside language, and thus beyond the reach of historians who seek diverse experiences in order to relocate subjects in the historical records.^[90] Scott’s critics argue that she has gone too far, that even the most counterhegemonic of experiences are described with words borrowed for the purpose: no words are free from the materialism that generates them, and words are often densely packed with historical meanings. Meanings change—like that of benandanti—and terms can lose one historical specificity and take on another. The use and, as chapter 4 argues, misuse, of words carries material histories of work, objects, and places. It is only when these new words are taken up and transformed into personal narratives—when circulating stories are refashioned into personal experiences and the knowledge such experiences contain—that people participate in shaping the language with which they describe the world.^[91] When these new words are spoken unproblematically, as hearsay, they offer a contextualization that older terms do not provide. The repetition of hearsay provides a glimpse of the everyday talk and gossip that is a thick description of what otherwise remains as confusing as distinguishing between a wink and a twitch.^[92] For example, a woman who thought she was almost captured by wazimamoto could be reassured by hearsay. Mwajuma Alexander was going to her farm late one night in 1959 after an evening’s drinking with her husband and co-wife. Near a neighbor’s farm, she saw a group of men, one of them white, standing around a parked vehicle. One man threw her to the ground. She ran away and hid while they searched for her. Finally, she heard one of them say, “Oh, oh, oh, the time is over,” and they drove off. She fled home. The following day, her husband, on his way to a market in a nearby village, heard that the wazimamoto had caught a woman in the area; this confirmed what everyone suspected.^[93] If someone told someone, who told someone, who told Mwajuma’s husband, that wazimamoto were capturing women in the area, then they were.

But what constituted hearsay and circulating stories? Was the common knowledge that wazimamoto had caught a woman here or there made up from bits and pieces of the diverse experiences of many people, or was something else at play, a notion of experience that was not necessarily personal, a notion of experience that incorporated that which was heard about? If historians have worried that experience may be our own rubric for unifying diverse elements into a narrative that subsumes differences, men and women in western Kenya, at least, have suggested that diverse experiences, taken over and told as personal narratives, can reveal the power of difference and the speakers’ knowledge thereof. Such domestication of circulating stories was not boastful exaggeration, or at least it was not only that. Circulating stories were told with convention and constraint. The act of making a wazimamoto story personal—adding names and places and work relations—had nothing to do with making it a better, more detailed story that explained the intricacies of bloodsucking.

The idea that a story may be true although its details are unknown to its tellers is at odds with most of the methodologies used to assess the reliability of testimony or an informant.^[94] Vampire stories are neither true nor false, in the sense that they do not have to be proven beyond their being talked about; but as they are told, they contain different empirical elements that carry different weights: stories are told with truths, commentaries, and statements of ignorance. These do not make wazimamoto stories seem unlikely; it is a true story and no one would make a compositional effort to change it to make it more credible. Anyango Mahondo of Siaya, for example, explained that the police were actually the bloodsuckers, something he could neither tell his wife nor his brothers. It was “ordinary people” who could not distinguish between police and firemen. In Kampala

When a man joined the police, he had to undergo the initial training of bloodsucking.…When one qualified there, he was absorbed into the police force as a constable.…At night we did the job of manhunting…from the station, we used to leave in a group of four, with one white man in charge.…Once in town, we would hide the vehicle somewhere that no one could see it. We would leave the vehicle and walk around in pairs. When we saw a person, we would catch him and take him to the vehicle.…Whites are a really bad race.…They used to keep victims in big pits.…blood would be sucked from those people until they were considered useless.…Inside the pits, lights were on whether it was day or night. The victims were fed really good food to make them produce more blood.…The job of the police recruit was to get victims and nothing else. Occasionally, we could go down in the pits, and if we are lucky, we can see the bloodsucking, but nothing else.^[95]

This is presumably the account Mahondo could not tell his wife. Chapter 4 argues that this particular chunk of narrative describes on-the-job experience, supervision, promotion, and the place of race and rank therein. Now I want to examine this account as testimony, as a narrative told with different kinds of truths and frank admissions of ignorance. I am not interested in why he told this story, but in how. Mahondo has made hearsay into a narrative of personal experience: the vehicles, the nighttime abductions, the pits, the feeding of victims were all commonplace in the region’s wazimamoto stories. Mahondo does not seem to be talking simply to enhance his dubious prestige;^[96] instead, he seems to be establishing truth about wazimamoto—the role of the police, the evils of white people—by telling the story as personal experience and by describing his own role as a participant and a bystander. Indeed, what is important here is the way that Mahondo informs his own storytelling; the process of making a personal narrative was constrained by hearsay: if Mahondo was not speaking the truth, or claiming that he as an eyewitness knew more truth, why did he not make up a better, more elaborate story about what happened in the pits?

It is possible that it was only by conforming to the standards and conventions of hearsay that Mahondo could have been thought credible. Had he stated what actually went on in the alleged secret pits under the Kampala police station, or if he said that he knew what whites did with the blood, he might have revealed himself to be a fraud, rather than a man with insider knowledge. Performance is part of every interview, not the work of specific practitioners in specific places.^[97] Speakers use a genre by giving a good example of its use shaped to meet their needs at the moment.^[98] Mahondo’s eyewitness account was told the way hearsay wazimamoto stories were told. How the story was performed, and the elements with which it was performed, made it credible. Where it stood on some imaginary line between hearsay and experience had nothing to do with how accurate it was.

Zebede Oyoyo had been captured by Nairobi’s fire brigade in the early 1920s. All his neighbors knew his story, which was how I came to be sent to him early on during my stay in Yimbo in 1986. My research assistant and I interviewed him twice. The first interview was a barely disguised account of his strength—“My fists were like sledgehammers.” “Nobody could come near me.” “When I saw the chance, I dashed out of the room…I outpaced them.” “Those kachinjas really chased me, and when I had completely beaten them, one of them told me, ‘Eh, eh you! You were really very lucky. You will stay in this world and really multiply.” ^[99] The second interview, ten days later, provided a much more detailed and subtle account of his encounter in a urinal with an African man.

I was caught near River Road. It was near the police station. I had gone for a short call in one of those town toilets. The time was before noon.…When I finished urinating, someone came from nowhere and grabbed my shirt collar. He started asking me funny questions, like “What are you doing here?” I told him I was urinating in a public toilet. On hearing that, the man started beating me. He slapped me several times and pulled me toward a certain room. On reaching that room, I realized that something was wrong. It was then that I started to become wild, and since I was still young…that man could not hold me.…I fought with the man until I got the chance to open the door. I shot out at terrific speed.…When they realized they could not catch me, one of them told me, “You, you are really lucky. You will really give birth to many children and will only die of old age. You were lucky and pray to God for that luck.” ^[100]

I am not the first to notice that people often revise the answers they have given in a first interview when they are interviewed for a second time. Neither am I the first to find this unremarkable. Historians routinely mediate between different accounts of the same event; why should this mediation be methodologically any different when the different accounts are provided by one person? It is only when a voice is conceived of as a single, spoken rendition of experience that contradictions become extraordinary rather than ordinary. To argue that an informant is mistaken because he or she says different things at different times, or even to argue that one account is wrong, makes linear demands on speech and self: lives and experiences are not such simple, straightforward things that they lend themselves to easy representation; people do not give testimony that fits neatly into chronological or cosmological accounts. Instead, they talk about different things in personal terms; they talk both about what happened to them and about what they did about it, but they also use themselves as a context in which to talk about other things as well.

The idea that a voice, however produced, would not change its mind or its words serves historians, not the speaker’s own complicated interests. What, after all, constitutes the authority of the voice? That historians use what it says? But what happens when voices willingly speak untruths, telling stories the veracity of which they might learn, but that they do not always believe? This raises another question entirely: what makes oral evidence reliable? That it can be made to be verified just like documents, or that it is taken as a kind of evidence produced in circumstances unlike the ones in which people write diaries, reports, and memoranda? What would make oral material true: that truth is spoken during an interview or the repeated social facts and hearsay with which people talk that give us insight into local knowledge beyond one man or woman’s experience? Mwajuma Alexander or Zebede Oyoyo or even Anyango Mahondo were not telling “the truth” but misrepresenting and misconstruing something that happened into vampire stories; they were constructing experience out of widespread hearsay.

Indeed, Oyoyo’s second story seems to have been circulating throughout East Africa in the early 1920s. In 1923, a “Believer” wrote to the Tanganyikan Swahili-language newspaper Mambo Leo saying that he was now convinced that “mumiani are cruel and merciless and kill people to get their blood.” He had seen this himself in Nairobi. Near the new mosque in River Road, there was a long, narrow building and a “government toilet but no permission was given for people to use these toilets. Inside the long narrow house, people stay, wear black clothes and are called Zima Moto, but the thing that is astonishing is that somebody isn’t in this group and they go inside this building, they never come out again.” A Luo man who worked there would not allow his brother to come near the building, not even to greet him.^[101] Did Oyoyo bring this story home and craft it to depict his own strength, his own talents, and his own memories?

Zebede Oyoyo may not have been what North Americans would call henpecked, but his wives seemed dubious of his bravado. Once, visiting his compound, my research assistant overheard his senior wife asking him why he “always” spoke English to her but not with the visiting white woman. I would argue that the first version of his near-abduction was the one he wanted his wives to appreciate: it was the story of his strength and his fame. It may not have been a story Oyoyo told with any success anywhere else; we may have heard it precisely because it was received so badly at home. The narrated bravado of the first interview may have been Oyoyo’s chance to get that story taken seriously. He told my assistant and me that this was, after all, a men’s story: “None of my wives could realize the seriousness of these stories, but [he turned to my research assistant, a man] a man like you can realize the value and seriousness of any story.” ^[102] The story he told us ten days later is what I like to think was the result of his reflection: having thought about the incident, he may have recalled more, and he was able to tell me this version when we returned to interview him again. While this interpretation “explains” the second interview, it is one that puts my questions at the center, just as Oyoyo’s first story puts him at the center. In a provocative article, Justin Willis notes that informants may change key parts of their lives in different interviews, not because of anything the interviewer says or does, but because of other people in the room: the audience for which lives are negotiated and re-presented (as opposed to represented) is not even the interviewer.^[103] Such an insight problematizes concerns about the politics of interviewing—the interview and his or her questionnaire may have little to do with what’s being said or why.

My point is most emphatically not that Africans saw things in urinals and police stations that they did not understand and then told stories to explain them. My point is the opposite: that what went on in the government’s strange toilets or police stations was so well known that, despite attempts to conceal pits or forbid people to use toilets, it could best be described in the commonplace terms everyone used in talking about it. Telling a more unique and detailed story—describing the happenings in the pits, for example, or boasting of one’s youthful strength—risked disbelief and derision.

These stories, even when told with all the conventions and constraints of hearsay, were not all received and heard the same way. Not everyone believed these stories, or believed them all the time, or believed every version a neighbor or acquaintance repeated. Nevertheless, each repetition, each repudiation, each amendment and refinement did not make a story more true or more false, but made it a more immediate way to talk about other things. Every argument or discussion a vampire story generated created a debate—stories could be evaluated on the merits of their contents, not their performance, and men and women argued over the importance of cars or men who worked only at night. Vampire stories could be refashioned and made personal or local by a few names and examples. Indeed, as part 2 argues, vampire stories are matters not so much of belief as of details: the stories are false, but the names and places and tools in them are true, and the stories are about the real fears those places and tools aroused. When men and women in Uganda recalled that bazimamoto captured people with chloroform, they were not literally describing bloodsucking or hospital practices: they were, as chapter 3 argues, talking about a European drug that had intense meaning for them because its application was similar to those of medicines used by Ugandan healers.

But did Africans believe these stories? The answer, which may not be that important to my purposes, is probably both yes and no. Or, to put it another way, Africans’ understanding of these stories went beyond assessing their truth, or even the motives of those investigating these stories. In Uganda, I wrote a questionnaire that I had my research assistants give; on it, I asked about the embalasassa, a speckled lizard said to be poisonous and to have been sent by Prime Minister Milton Obote to kill Baganda in the late 1960s. It is not poisonous and was no more common in the 1960s than it had been in previous decades, as Makerere University science professors announced on the radio and stated in print.^[104] But I was curious about embalasassa stories, and in one of those errors that oral historians are never supposed to admit in print, I wrote the question, What is the difference between bazimamoto and embalasassa? Anyone who knows anything about a Bantu language—myself included—would know the answer was contained in the question: humans and reptiles are different living things and belong to different noun classes. I had not asked a stupid question to see how informants might respond, but because I was thinking with English rather than Bantu-language categories when I wrote it. A few of my informants corrected my ignorance: “There is a big difference between them, bazimamoto are people and embalasassa are lizards,” ^[105] but many, many more ignored the translation in my question and moved beyond it to address the history of the constructs of firemen and poisonous lizards without the slightest hesitation. They disregarded language to engage in a discussion of events. “Bazimamoto finished by the time of embalasassa; that was during independence.” ^[106] “Embalasassa came after independence, I think in 1974, it had never happened in Uganda before.” ^[107] In fact, when people expressed confusion in answer to my question, it was not about the differences between species, but between the policies of the late colonial era and the first decade of independence: while most people said that embalasassa came during Obote’s first regime, a few said it was “sent” by Governor Andrew Cohen in the 1950s or by Idi Amin in his first years in power.^[108] My point is not about the truth of the embalasassa story, or even the inadvertent good sense of my questionnaire, but rather that the labeling of one thing as “true” and the other as “fictive” or “metaphorical”—all the usual polite academic terms for false—may eclipse all the intricate ways in which people use social truths to talk about the past. Moreover, chronological contradictions may foreground the fuzziness of certain ideas and policies, and that fuzziness may be more accurate than any exact historical reconstruction.^[109] Stories about poisonous lizards, spoken to men who only came to believe in the truth of bazimamoto stories because of the violence of Amin’s Uganda, raised questions about colonial and postcolonial states and the differences between them with each telling and retelling. Whether the story of the poisonous embalasassa was real was hardly the issue; there was a real, harmless lizard and there was a real time when people in and around Kampala feared the embalasassa. They feared it in part because of beliefs about lizards, but mainly what frightened people was their fear of their government and the lengths to which it would go to harm them. The confusions and the misunderstandings show what is important; knowledge about the actual lizard would not. Vampire stories are, then, confusions and misunderstanding of the best kind: they reveal the world of power and uncertainty in which Africans have lived in this century. Their very falseness is what gives them meaning; they are a way of talking that encourages a reassessment of everyday experience to address the workings of power and knowledge and how regimes use them.