Kenyatta’s lament

This work is licensed under the Creative Commons | © Robert W. Blunt. Attribution-NonCommercial-NoDerivs 3.0 Unported. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau3.3.007

Kenyatta’s lament

Oaths and the transformation of ritual ideologies in colonial Kenya

Robert W. BLUNT, Lafayette College

This article provides an ethnographically and historically grounded critique of Giorgio Agamben’s claim that oaths, like religion and law, reflect a universal experience of the failure of language to obligate people. I show how Agamben’s critique of the oath only becomes relevant in the colonial Kenyan context after missionaries and administrators had facilitated the “unmaking” of Kikuyu ideologies of ritual efficacy. Specifically, it tracks how colonial administrators and Kikuyu elders came to understand the customary nature of elder authority through the representational and regulatory capacity that objects known as ĩthathi were held to have. In the end, administrators and elders came to share an ideology linking ritual efficacy and elder authority, jointly participating in the naturalization of that relationship while failing to anticipate where it would lead.

Keywords: oath, ritual, sovereignty, Agamben, semiotic ideology, Mau Mau, colonialism

Jumbled and obscene rites

In 1954, Max Gluckman published a short article in The listener entitled the “The magic of despair” (Gluckman 1963: 137).1 The article was based on a BBC radio broadcast of the same name and delivered during the height of the Mau Mau uprising in colonial Kenya; in it, Gluckman attempted to convince a skeptical British public that the ritual oaths Kikuyu “Mau Mau” guerillas were using to cement loyalty and secrecy within their ranks, “with their high obscenity, were not a return to African pagan religion, as was commonly alleged, but were quite a different kind of phenomenon” (Gluckman 1963: 137). Instead, Gluckman argued that Mau Mau had been “produced by the colonization of Africa, and not by indigenous Africa itself” (Gluckman 1963: 139). As Peter Worsley would later point out, the “obscenities and sacrilege” at work in Mau Mau oathing rituals had prevented the uprising from obtaining full support from Britain’s left. Mau Mau’s aura of violent religiosity, an aura that was highly embellished by prominent settler voices, did not sit well with a British working class, which Worsley argued had “been so influenced by our imperialist history that it is perhaps the most chauvinist working class in the world, and one of the least active on colonial issues” (Worsley 1957: 16).

Worsley’s essay, “The anatomy of Mau Mau,” published toward the end of the conflict in 1957, appeared in the anti-Stalinist journal, The New Reasoner, and was a scathing indictment of both conservative and liberal settler representations of Mau Mau. Like “The magic of despair,” it also argued cogently about what Mau Mau was not, namely a nationalist cult movement like the one that blossomed early in the Melanesian colonial encounter, which he analyzed in The trumpet shall sound. Worsley noted that Kikuyus had already developed secular political associations and independent Christian churches as early as the 1920s. He argued that Kikuyus had turned to violence, both inside and outside of the ritual frame, because the colonial government had banned their political organizations and forced the majority of the male Kikuyu population into concentration villages. Mau Mau’s oaths were thus rituals of last resort. They were, to quote Worsley, “quite unlike any traditional rituals even though they use elements of various traditional rituals” (Worsley 1957: 20).

In this article, I build on Gluckman’s and Worsley’s observations about the colonial inspiration of Mau Mau ritual to examine historically and ethnographically the nature of this transformation. Specifically, I argue that the mandates of indirect rule in Kenya sparked a process of “unmaking” (in Weiss’ [1996] words) of Kikuyu ideologies of ritual efficacy. If “traditional” oaths were once grounded in a distinctly acephalous semiotic ideology that refused the ontological centering of sovereignty, they came to be increasingly mediated, for Kikuyus and colonials alike, by a Christian capitalist understanding of how oaths were supposed to work, one that foregrounds sovereignty as an ontological rooting / fixation of power. This article thus also speaks to broader theoretical concerns pertaining to the genealogy of oathing.

Specifically, I use the lens of Kikuyu oathing to engage with Giorgio Agamben’s provocative ruminations on the topic. Kikuyu materials here expand and complicate Agamben’s idea that oathing works to both create and instantiate political authority. My analysis cautions against Agamben’s assumption that oathing (and its failures) rest on a faith in language’s ability to definitively bind words to things and deeds. Kikuyu materials suggest that this connection is founded upon a specifically (Judeo-Christian) ideology of mastery, indexing a divine sovereign whose speech acts are true (like the Hebrew god Yahweh naming things in the garden of Eden before the Fall). I argue that oaths and their efficacy may be implicated in other semiotic logics. In the case of the precolonial Kikuyu, ritual efficacy entailed a semiotics of management where the power to join word to deed was seen as excessive and dangerous, requiring careful corralling and channeling. It is not until missionaries and colonial administrators sparked the unraveling of Kikuyu ideologies of ritual efficacy that Agamben’s critique of the oath becomes relevant for Kenya. In this context, Agamben’s lament converges with Kenya’s iconic first president, Jomo Kenyatta’s own lament about sovereignty’s slipperiness and the erosion of the oath’s power to ground authority. Both capture something of the limitations of Mau Mau oathing, yet both fail to account for the complex semiotic and historical mediations underwriting the aporias of Mau Mau ritual signification.

Writing ritual out of Mau Mau studies

The claim that Mau Mau and its opaque and confusing ritual complex was a byproduct of colonialism may sound obvious in hindsight. As an object of inquiry examined by historians and anthropologists alike, Mau Mau has been exhaustively taken up as a way of arguing about the relationship between ethnicity and violence and a dense site for the production of different analytic frames for understanding dynamics of continuity and change. However, Gluckman’s and Worsley’s emphasis on violence and signification in the ritual frame, as well as the apparently jumbled nature of oathing rituals in the 1950s, was never really taken up by historians. As John Lonsdale and subsequent generations of historians of Kenya have persuasively argued, the “Mau Mau rebellion,” a Kikuyu anticolonial uprising and civil war that took place during the 1950s in the Kenya colony, was a violent moral debate among Kikuyus about the proper avenues for achieving cultural notions of civic honor and adulthood (Lonsdale 1992). Daniel Branch has nicely shown that this was equally true for most people who fell between the poles of “loyalism,” those who sided with the colonial state’s crackdown on Mau Mau, and those who tended more toward militant support of, or active participation in, the violent aspects of the uprising (Branch 2009). Against the backdrop of the massive appropriation of land by white settlers and the forced introduction of private property and capitalism more generally, other avenues for voicing Kikuyu grievance had been cut off. The Kikuyu poor increasingly understood themselves as unable to achieve ithaka na wiathi, or “land and freedom,” the mantra of the “Kenya Land and Freedom Army,” what Mau Mau forest fighters actually called themselves.

Regarding oaths specifically, much of the Mau Mau literature deals with the various content and typological classifications of oaths as indices of differing levels of militancy and commitment to the movement (cf. Kershaw 1997). While much of this literature traces the historical origin and evolution of Kikuyu oaths,2 none of it really examines the historicity of the Mau Mau oath at the level of form. There are very good descriptions of Mau Mau oathing rituals from participants (e.g., Barnett and Njama 1966; Clough 1998). Yet there is a tendency among scholars to write off the oath’s more perplexing details—as settler fantasy about dark rites by firelight, epitomized in the ethnopsychiatry of psychologist Dr. John Colin Carothers (1954). Historians with a more nationalist bent have emphasized how the oaths were a continuation of a deeply engrained tradition of warrior oaths (Githige 1978; Hudson-Koster 2010) but they fail to address the almost inflationary expansion of oathing symbols (cf. Githige 1978: 109) in each new iteration of the oath as the war dragged on. They also do not attend to the reason why forest fighters found it acceptable to kill oath breakers rather than wait for the deleterious effects of the oath’s curse to take effect (Githige 1978: 69).3 Some explain away the more blatantly symbolic dimensions of Mau Mau ritual such that oaths become an expression of practical reason, albeit a violent one (cf. Green 1990). Thus, regarding oaths, it has become more or less commonplace to assume, as Caroline Elkins does, that they were “the rational response of a rural people seeking to understand the enormous socioeconomic and political changes taking place around them while attempting to respond collectively to new and unjust realities” (Elkins 2005: 27–28). Yet, so were Kenyan experiments with trade unionism. Here ritual loses any specificity.

Whether those who administered oaths—or those who “ate” them—knew they were violating “sacred” taboos, as Worsley claimed, is debatable. However, Worsley was right: the rituals contained jumbled collections of significata mined from all over the colonial social field (e.g., the banana fronds of circumcision thrown together with overtly Christian symbols). This jumbled (and violent) quality of Mau Mau ritual still requires explanation.4 Worsley’s and Gluckman’s focus on the oathing ritual itself, from its allegedly sacrilegious and obscene qualities to its violent character, the ritual’s apparently baroque jumble of ritual significata, has thus given way to efforts to recuperate Mau Mau’s rationality from colonial era depictions of its savagery.

This exclusion is perhaps understandable. Gluckman argued, channeling Freud, given the incapacities of traditional Kikuyu religion to make sense of the rapidly transforming colonial world, the ritual deployment of “blood, sex, excreta, bestiality,” or “the threat to murder near kin” expressed universal instinctual drives that can bubble up when societal controls are weakened (Gluckman 1963: 145). At the level of description alone, Gluckman came close to reifying what many scholars have since written off as settler salaciousness. But concerns about the degree to which the descriptions of the oath’s details are dependable or not perhaps blind us to the value of Gluckman’s analytic clustering of violence with what he called “symbols” at work in the oaths. Both he and Worsley captured the end of a process, again, following Brad Weiss (1996), which we might think of as the unmaking of Kikuyu cosmology, evidenced in the uprising’s wartime oathing procedures.

Ritual and sovereignty

Webb Keane’s notion of “semiotic ideology” is helpful for thinking about the often vaguely articulated and constantly changing assumptions about Kikuyu ritual efficacy at play in Kenya’s late colonial encounter. Semiotic ideologies—the beliefs people articulate and hold about how sign systems (which encompass both words and things) are structured and operate—rationalize and justify particular forms of action (Keane 2007: 16). The act of swearing an oath implicitly indexes some kind of semiotic authority that enforces the oath’s statutes. As will become clear, in the Kenya colony, colonial administrators and Kikuyus increasingly shared a single semiotic ideology, characterized by a perspective that views sovereignty as “an ontological ground of power and order expressed in law or in enduring ideas of legitimate rule,” in Thomas Blom Hansen and Finn Stepputat’s words (2006: 297).

Readers will note the parallels between this notion of sovereignty and the one examined and critiqued in the influential works of Giorgio Agamben (1998, 2005, 2011). As the title suggests, The sacrament of language: An archaeology of the oath (2011) deals specifically with oaths. More than any other form of action, Agamben argues, the oath expresses the problematic linkage of the political and the theological. In oaths, he suggests, semiotic efficacy and political efficacy amount to the same thing since oaths simultaneously create and instantiate political authority. According to him this is an error, but one linked to a larger problem, namely, that humans assume that language has the capacity “to tie speech to things and to bind, by means of curses and anathemas, speaking subjects to the veritative power of their speech” (Agamben 2011: 58), which, for Agamben, is a mistaken belief. The problem is people keep trying to make them work, often through violence. As Richard Fitch notes, Agamben’s archaeology of the oath, like the larger Homo sacer project, “unfolds a brilliantly provocative thesis as to why ideological apparatuses, developed to deal with the problem of grounding social life, have led to unnecessary political horror” (Fitch 2012). Like religion and the law, oaths don’t really work. Thus, the oath, according to Agamben, is the sacrament of language because in performing it, those who swear oaths are not really swearing to God but are attempting to partake in the divine power of language itself; think the Hebrew god Yahweh in the garden of Eden whose speech acts, Agamben argues, are “pure being” (whose very words are oaths).

However, Agamben does not seem to consider how non-Western societies might have very different semiotic ideologies that shape understandings of language’s promissory capacities that do not model the oath on the speech acts of a divine sovereign. Here, I seek to extend Blom Hansen and Stepputat’s insight that Agamben’s philological vision of sovereignty needs to be replaced by an anthropological one that examines sovereignty as a form of authority that is always emergent and often plural in colonial social fields (Blom Hansen and Stepputat 2006: 297). I argue that prior to colonial intrusion, the semiotics of Kikuyu oathing were based less on mastery than an elaborate logic of gambling (Peterson 2002), in which participants tried to manage and channel forces that were inherently “wild” and excessive. Success in that regard might bring limited resolution to highly localized problems, but it did not ground authority per se. The colonial state’s assumption that sovereignty ontologically grounds order eventually generated a crisis of authority in the Kikuyu ritual frame, a crisis that was particularly distilled in Mau Mau’s ritual practices.

Unmaking Kikuyu cosmology

The process of unmaking Kikuyu cosmology was one in which the colonial state and certain Kikuyu elders attempted to naturalize gerontocratic authority as the font of ritual power, but turned on the contradictory social trajectory of fetishes that Kikuyu called ĩthathi (gĩthathi sing.). ĩthathi were small, sacred objects that functioned as both oracle and judge; the gĩthathi “decided” which litigant was “lying” in customary legal disputes, cursing the guilty party’s entire lineage with a condition of bodily affliction5 that could ultimately lead to death. As I will show, under indirect rule, ĩthathi, and the ritual system in which they were embedded, were utterly transformed as they were incorporated into processes of native administration. While ĩthathi may have existed before the formal establishment of indirect rule, it was under colonialism that they were ideologically invested with a peculiar gerontocratic potency. Part of this novelty was due to the fact that the colonial state invested ĩthathi not only with unprecedented regulatory capacities but also with the power to commensurate native custom with their own “universal” law. Of course, they were unaware of the latter category’s parochial character (cf. Sheik 2011). “Oathing stones,” as British district officers often called ĩthathi, were thus fetishes par excellence (cf. Pietz 1985, 1987, 1988; Spyer 1998). Ĩthathi sublimated actual cultural differences between completely different ritual ideologies into a shared enthusiasm for the oath as a technique of governance.

This is not to say that this process of calibration and commensuration was in any measure a success for those involved. The formal similarity of oaths masked the underlying dissimilarity of their latent ideologies of efficacy. Previously, Kikuyu oaths were characterized by a sense of certainty with respect to the force that backed oaths, but uncertainty regarding its management and effects. This force, generated by nature, known as thahu, was understood to be dangerous in Kikuyu society. Its polluting, contagious quality created abject beings, both human and animal, and always potentially overwhelmed the elders who speculatively channeled it to solve specific social problems. In this respect, the force that backed oaths was not intrinsic to the oath itself or even to Kikuyu society; rather, it was the constitutive outside of Kikuyu society, an aspect of nature really, which was only speculatively and fearfully invoked.

The imposition of Protestant Christian and capitalist notions of sovereignty onto the Kikuyu ritual system thus conflated elder authority and ĩthathi with the force they previously could only hope to direct efficaciously. By attempting to bring thahu into society, the colonial situation in the Kikuyu reserves compelled new questions about thahu’s origins, qualities, and effects. More specifically, available sources provide evidence for a fundamental shift in attitude regarding the certainty of oathing procedures. Whereas before Kikuyus regarded the force that backed the efficacy of oaths as frightfully certain, its outcomes were not viewed as overdetermined or even necessarily manageable due to thahu’s overwhelming and excessive character. Colonial administrative retooling of oaths produced what was arguably a new sense of uncertainty about what exactly backed oaths, combined with a strange sense of the oath’s definitive capacity to produce social order (if performed correctly). As we shall see, however, this newfound ritual certainty that oaths had to work was plagued by persistent doubts as to the authenticity of the institutions (elders) and media (ĩthathi) through which they were performed. What is most peculiar about the eventual addition of violence to oaths by Mau Mau insurgents was its complete reversal of what was supposed to have been the unique nature of oaths: they produced death without blame, blocking the spirals of retaliatory violence characteristic of acephalous societies, as E. E. Evans-Pritchard long ago demonstrated in his discussion of the Nuer feud (Evans-Pritchard 1967).

At the center of this story is the figure of Jomo Kenyatta, who was both the first president of independent Kenya and a student of Bronislaw Malinowski at the London School of Economics. Kenyatta is often described as having presented Kikuyu society as an overly ordered “Arcadian republic of elders” (Berman 1996: 333). To him, oaths insured the largely mythical mechanical solidarity of a Kikuyu community that had, in truth, always been politically fractious (Berman 1996: 313). According to Kenyatta, societies, and Kikuyu society specifically, were held together by what he refers to several times in his ethnography of the Kikuyu, Facing Mount Kenya, as “symbols of truth” or “symbols of the oath” (Kenyatta 1965: 214–17). Kenyatta’s function-alist understanding of oaths was also expressive of a widespread sense among Africans that something was actively undermining the structures and processes of traditional authority, which both the British and many Kikuyu were invested in maintaining. For Kenyatta, this corrosive “something” was money (Kenyatta 1965: 216). But in isolating money as an independent force undermining customary institutions, Kenyatta misrecognized the colonial cash economy and colonial power structures as separable things. In this respect, his lament about how money destroyed oaths paralleled attempts by Mau Mau insurgents to keep money’s allegedly deleterious effects on ritual efficacy and sincerity outside of the ritual frame; during the war, oath-takers had to remove European coins from their person before uttering the oath’s statutes (Barnett and Njama 1966: 57). Colonial officials as well as Kenyatta’s political rival Louis Leakey were subject to the same misrecognition. In truth, money and elders were both integral to larger forces of abstraction, whose symbolic logics already thoroughly mediated the different elements of the Kikuyu ritual economy (cf. Comaroff and Comaroff 1991: 18), a problem we will return to. For now, Kenyatta’s experience with the Kiambu Local Native Tribunal in 1949 frames our entry into the oath’s longer history.

Kenyatta’s lament

By the late 1940s in the British Kenya colony, settler capitalism was a mode of production that had begun to implode, given the increased mechanization of labor and the postwar decline in demand for Kenyan wheat (Anderson 2000: 484; Sorrenson 1967: 79). Newly redundant Kikuyu laborers flooded the native reserves6 after 1945, but when they returned home many found themselves in a situation that threatened the very possibility of achieving cultural notions of male public respectability, which demanded control over land, livestock, and women (Lonsdale 1992; Meillassoux 1972). Lands that migrant laborers had entrusted to others or had left behind were enmeshed in complicated processes of determining what the basis for land ownership was. Claims were largely based on stories of clans having obtained land from Ndorobo hunter-gatherers over fifty years earlier, a messy basis for establishing individual private ownership, the goal of the colonial administration. Often multiple parties would claim the same property. In accordance with indirect rule, claims over land in the native reserves were largely left to colonially reconstructed native authorities, in this case, Local Native Councils (LNCs) and Local Native Tribunals (LNTs), known more colloquially as the kiama, or “elder’s council.”

In 1949, Jomo Kenyatta wrote to the colonial administration to complain about a crisis unfolding in the Kikuyu reserves. At the heart of Kenyatta’s complaint was the method by which LNTs were determining land ownership, namely through a divinatory practice called kuringa gĩthathi (“to hit or activate the generation stone”). The sacred objects known as ĩthathi were patrilineally inherited by Kikuyu elder males and fashioned out of stone or the atlas vertebra of an elephant. Their ethnographic descriptions generally paint them as conduits used to channel thahu, the ever present, dangerous energies Kikuyus associated with nature, in order to enforce the swearing of oaths. In the case of a land dispute, the oath would have run something like, “If I am not the owner of this land, may this muma (oath) kill me,” implying that one would contract thahu, which was also the name of the condition of bodily wasting that would ensue in the man’s clan if he were deemed guilty. Bodily affliction served as evidence of the gĩthathis verdict. But according to Kenyatta, unscrupulous elders were using fake ĩthathi to adjudicate land tenure and ownership rights.

Kenyatta worried incessantly about the durability of customary institutions of conflict settlement and their ritual media: He believed them to be the only guarantors of due process. This was because not only were litigants swearing oaths against the stone’s curse, local native tribunal members were also oathed to insure honesty in judging the stone’s effects on the “guilty” party. This fact, however, presented a problem: if ĩthathi were extensions of elder power, as Kenyatta tacitly believed, the fact that elder judges as well as litigants were oathed made the exact source of a gĩthathis imprimatur something of a puzzle. The following was in Kenyatta’s letter:

My father Ngengi brought and won a case against Gitao Nruki for land. He won the case. Gitau appealed and when LNT Elders John, Mhugwa, Kahucho and Samuel with Registrar, Bewes, came on to the land yesterday. They ordered the Gĩthathi Oath (Thenge) be taken by the parties concerned. The kiama told the parties to collect the stone from Nganga Kabitu, in his/chief Waruhius location at the hire cost of Shillings 40/, and it was brought by the two young boys, representing both parties, on a bicycle. It was taken to Gatundu tribunal and yesterday morning it was brought to the land.

My suspicions as to the authenticity of the Gĩthathi stone were aroused by the fact that boys went for it (children normally would not handle it) and it was carried in a tin container. So I asked that I might inspect it, and in doing so, I considered it to be a false one and it was made of ordinary stone and had clearly been manufactured. I hand it to you as an exhibit. I inspected it after the elders had come to a decision in the case to the effect that as my father was not prepared to take the oath on the whole land as he considered that his brother should take it on his share. My father must lose the case, pay shillings 200/ cash and 40/ shillings for the hire of the gĩthathi.

On inspecting the stone, namely for anthropological interests, I was amazed to find it was a fake and that it must have been used wrongly and without meaning in many cases throughout Kikuyu country. So I took possession of it. This morning I showed it to the LNT Elders John and Kabucho and the registrar, and also in the presence of the LNC councilors and they agreed that it was a faked gĩthathi.

So I hand it on to you for investigation as the wrongful use of this stone has been and is a serious matter affecting many people, and at the same time encouraging bribery and corruption among Elders who hear cases who might agree at the request of one party to the Oath being taken knowing full well that it would have no effect—but at the same time causing the land to be decided in order that the person giving the bribe might get his share of the land to which he was not entitled.7

Although Kenyatta assumed that the deployment of a fake gĩthathi was a deception being purposely perpetrated by unscrupulous and greedy elders, his own inability to locate the definitive source of its authenticity raises serious doubts about the degree to which this was true. While Kenyatta claims anthropological authority to question the stone’s authenticity, this authority is not sufficient to appeal the stone’s verdict. In handing it over to the District Commissioner (DC), Kenyatta rather unconsciously races through several theories about what the specific authority might be that the gĩthathi’s power is allegedly rooted in. No one authority emerges definitively sovereign over any other. The force of custom, at least in the version that Kenyatta is hinting at, does not seem to be operating here. Boys, rather than elders, brought the gĩthathi to the tribunal, and it is this violation of taboo that causes him to question the material properties of the gĩthathi itself. If the stone was substantively “real,” he reasons, uncircumcised boys would not have been brave enough to touch it. In turning the gĩthathi over to the colonial authorities, Kenyatta is, in effect, asking the colonial state to guarantee the relationship between elder authority and the gĩthathi. In stating that the fake was manufactured from “ordinary stone” he seems to be asserting the natural sovereignty of such objects, at least when allegedly real ones were present. In other words, at this moment, ĩthathi seem to require some kind of supplemental authority to authenticate not only themselves, but also the ritual procedures that they are supposed to authenticate. How are we to understand Kenyatta’s appeal to so many certifying authorities? Why such epistemological confusion regarding the stone’s criteria of authenticity, the source of its regulatory force, and the qualities it was supposed to have?

The oath

Colonial era ethnographic descriptions of Kikuyu ritual like those of Louis Leakey’s must be treated with a certain degree of suspicion. If Kenyatta’s description of Kikuyu society was that of an Arcadian republic of elders, Leakey’s was no less so. Both understood Mau Mau oaths to be “illegal” by standards of legitimate ritual somewhere in the past. But Leakey’s ethnography presents other problems as well. His flat ethnographic present in which correct Kikuyu custom and ritual is read back into time more accurately depicts a society that had undergone massive transformation. Leakey describes ĩthathi as Kikuyu sacred objects in the Durk-heimian sense of being set apart or taboo. Kikuyus considered them rare, both in terms of their availability and because of elder strategies to keep them out of general circulation. Leakey notes that when disputants agreed to kuringga gĩthathi (to hit or activate the generation stone) the elder male representatives of the plaintiff and defendant families would go to the stone’s hiding place and carefully wrap it in banana tree bark for its journey to the oathing location, keeping it hidden until the last possible moment before its activation (Leakey 1976: 1011).

According to Leakey, under no circumstances were the objects to be sold. Moreover, if women or, even worse, junior males, touched them, they would immediately be desacralized, necessitating the ritual production of a new one—again, a strictly elder male activity.8 Cutting against the grain of these descriptions however, Leakey recalls that in the early twentieth century when he was collecting data, people claimed that the stones had mysterious origins and may not have been made by the Kikuyu at all (Leakey 1976: 1010). Sometimes they were assigned a Maasai origin. In this regard, it is interesting to note that the Kikuyu word for oath, muuma, is derived from the Maa word for oath, ol mumai. This maintenance of a lexical Nilotic foreignness—they could have just as easily adopted the more familiar Bantu Kiswahili word for oath, kiapo—suggests an acephalous theory of political power analogous to that of the stranger king and the power of the foreign (cf. Sahlins 2008; Rutherford 2003). However, Leakey interprets this recollection as further proof that the objects were unique, incomparable, and rare, reinforcing the proscription on their circulation as common items.

He continues. Gĩthathi oaths entailed the insertion of the representative’s finger into each of the stone’s seven symmetrical holes, followed by the insertion of wands of various “good” and “bad” woods (thought to amplify the effects of thahu) while the recitation of the oath’s statutes were uttered by the disputing parties in front of a group of elders. Here is Leakey’s description of the verbal oathing format for a homicide case: “If our family is in fact responsible for the death we are accused of, may we die, but if we are not guilty, may the family that is guilty and that has kept quiet in this matter and caused us to be accused all die” (Leakey 1977: 1010).

Elders warned the disputing parties that they had seven agricultural cycles to clear themselves—the statute of limitations for how long the stone’s power was active (Leakey 1977: 1010). Whichever family suffered misfortune first, evidence of the gĩthathi’s verdict, lost. Essentially, people served as the functional equivalent of Zande chickens (Evans-Pritchard 1963); detecting insincere speech, the stones could kill the “lying” disputant’s entire clan and descending generations.

Throughout these seven growing seasons, a host of activities and requirements were taboo:

During the whole of the first planting season, no member of either family might have sexual intercourse, nor might they cause their children to go through the second birth or initiation ceremonies. Moreover, during the first planting season all he-goats of both families had to be castrated and kept from serving the she-goats, and all rams and bulls had to be kept away from the females…. If the oath had been taken in connection with a boundary dispute, the land lying between the two alleged boundaries might not be cultivated till the matter had been settled by the effects of the oath. (Leakey 1977: 1010)

It is unclear whether or not halting all reproductive processes was actually done or achieved. Contrary to Leakey’s tendency to take Kikuyu elder male claims about how things really worked in the precolonial past at face value, it is more likely that the impossibility of such prescriptions indexes the magnitude of the crises that kũringa gĩthathi was supposed to resolve. The presumed rarity of such an occasion was homologous to the alleged rarity of the gĩthathi itself. Singularity of truth was the aim. However, as we shall see below, the allegedly customary ritual emphasis on producing this singularity cannot simply be read as customary, at least in the sense of being an unproblematic continuation of past ritual ideology and practice as Leakey seems to suggest.

If the activation of a gĩthathi’s power seemed to reinforce the equivalence of elders and symbols in Leakey account, its deactivation involved a degree of excess that threatened the terms of this relationship. In Leakey’s account, the problem shows up as one of control over simulacra—namely, the possibility that some ĩthathi were “fake.” Kũhohoria gĩthathi (“pacifying or quieting the generation stone”), Leakey says, was necessary to spare the innocents in the “guilty” party’s family from certain death or misfortune (1977: 1011). The deactivation of the curse entailed the manufacturing of a clay facsimile, upon which the verbal statutes of the original oath were again sworn but were cut off halfway through their utterance (Leakey 1977: 1011). The threading of sticks through the stone’s holes, part of the original oathing ceremony, was similarly prevented. (Leakey 1977: 1012). At the end of the ceremony, what Leakey calls the “imitation gĩthathi” was brought out into the bush, broken, and its pieces were hidden. Clearly, the ritual production and destruction of the copy was supposed to make the semantic value of words as substantive and singular as their solid stone backer of truth, the original gĩthathi. But in doing so—and Leakey misses this—it indexes doubt about the efficacy of the stones themselves. While the uniqueness and potency of the gĩthathi was supposed to be ensured by their limited circulation, the rules that were supposed to keep ĩthathi as relative singularities (cf. Kopytoff 1986) appeared insufficient to the task of truly keeping them that way. I will explore why this was the case below.

For Kenyatta, the problem of excess shows up differently. In recalling the Kikuyu past, he argues that the stones were excessive in their power, but he is silent on the capacity of elders to deactivate them through kũhohoria gĩthathi. This is clear in his narration of the activation ritual. “The elders stood at a little distance facing the spot where the oath ceremony was being prepared. The place had to be a barren ground not likely to be cultivated, for no one would allow the ceremony to be performed on or near his cultivation. It was feared that the evil of the oath symbols might spread to a cultivated crop and destroy it (Kenyatta 1965: 216). Clearly thahu, what Kenyatta refers to as the oath’s “evil,” could exceed the power of elders to direct it toward socially productive ends. In fact, it threatens to overwhelm the elders handling the gĩthathi, eviscerating the very forms of wealth—land and crops—that back their status as elders in the first place. And, since elders are described as the only ones capable of activating such forces, the exact origin of their imprimatur becomes an issue. Kenyatta seems at times to assume that it originates in elders themselves. He presents elderhood as equivalent to certain representational media that he calls “symbols of the oath,” which are supposed to embody the sovereignty of the social vested in the authority of elders. Ritual efficacy is thus unambiguous: Representational media stand for elder authority, in accordance with how much authority elders wished to exercise in any oathing context.

However, if as Keane argues, ritual brings to light the possibility that “authority, legitimate agency, and the various sources of social and political power do not necessarily cohere” and thus may “hint at the nature and sources of potential failure” (1997: 9), then in much the same way, kũhohoria gĩthathi, Leakey’s deactivation ritual, anticipated, signified, and enacted the anxieties about the potential failure of elderhood to be able to do just that. The important question to ask of Leakey is whether or not kũhohoria gĩthathi was “traditional” at all. In a 1918 account of ĩthathi there is no mention of kũhohoria gĩthathi and the ritualized production and destruction of a clay copy (Tate 1918). In Kenyatta’s ethnography there is no mention of it either. Yet, in both Leakey’s ethnography and Kenyatta’s letter, the elimination of copies is essential to the future efficacy of ĩthathi. In Leakey’s case, the production and destruction of the fake is contained within the ritual frame and thus depicted as traditional. In Kenyatta’s letter, the removal of “fake” ĩthathi from circulation must take place outside of the ritual frame, a process to be aided by the colonial state. Arguably, the anticipation of the corrupting power of the counterfeit in both cases are symptoms of the same problem, regardless of whether or not the solution to this problem is assigned to the domain of traditional ritual or modern political entities. The ratcheting up of the possibility of ritual failure is intriguing and suggests that other forces are at work.

Kenyatta’s equation of elders and symbols also becomes the ground of second order equivalences outside the ritual frame, which again, are indicative of other forces at work. In Facing Mount Kenya, this is evident in the logical and narrative ordering of his claims about the fixity of the rates of compensation through which legal torts could be righted. Not coincidentally, he discusses these equivalences in the section immediately following his discussion of oaths. For example, in cases of homicide, Kenyatta states that “one hundred sheep or goats or ten cows was assessed on the assumption that the man would have been able, had he lived, to bring property to his group equal in value to that fixed for compensation for the loss of his life” (Kenyatta 1965: 220). This durable exchange rate in turn guarantees a whole host of other equivalences (1 man = 10 cows, 10 cows = 100 sheep, 10 cows = 100 goats, 100 goats = 100 sheep, etc.). Here is a particularly salient example of what Jean-Joseph Goux would argue is a more cultural manifestation of the Marx’s notion of the general equivalent (Goux 1990). Like theories of value, which vest authority in the supposed universal equivalence of something like gold (cf. Caffentzis 1989), elder authority and their symbolic equivalents become the mutually reinforcing embodied grounds for all exchange. Instructive here is Marx’s critique of British political economists who understood money as that which rendered commodities exchangeable, rather than as the expression of their exchangeability (Marx 1976: 125–240). Oaths, being grounded in the equivalence of elders and their symbols, are posited as that which facilitates the alleged stability of larger systems of meaning, value, and exchange. In reality, Kenyatta’s description of stable and traditional compensation rates was indicative of an emerging principle of universal exchangeability epitomized in wage labor, hut and poll taxes, and Christian notions of a universal sovereign that could guarantee the stability of signifier and signified.

Unstable sovereigns

As stated earlier, what is so peculiar about the Kikuyu concept of thahu is that in the precolonial Kikuyu world its invocation and channeling to hold parties accountable to what they had sworn (and thus lethally detecting the “liar”), represented the extreme opposite of an elaborate complex of quotidian avoidance practices designed to evade the excessive and contaminating aspects of thahu, a constant threat lying just beyond the hedgerows of the domestic compound. This acephalous theory of power as emanating from outside the social world, amorphously manifest in thahu, created an extremely unstable metonymic equivalence between elder authority in the profane world and the magical objects they used to channel thahu within the ritual frame.

Kikuyu practices for avoiding thahu struck one of the Kikuyu’s earliest ethnographers as baroque. In his 1922 tome about Kikuyu and Kamba “religion,” Bantu beliefs and magic, C. W. Hobley (the onetime coastal Transport Superintendent for the British Imperial East Africa Company and eventual Provincial Commissioner of Kavirondo district) lists no fewer than sixty-eight prescriptions for avoiding thahu, dealing with matters as diverse as livestock theft, incest, adultery, sodomy, domestic violence, contact with warriors, handling blood, menstrual blood, corpses, the proper methods for transporting children, handling the corpse of a goat after a hyena kills it inside the homestead, certain forms of interspecies contact (if a baby goat suckles a lactating woman, for example)—and these are just prescriptions for humans (Hobley 1922: 103–23). Cattle and small stock could contract thahu as well.

In a review of the book from the same year, published in the International review of missions, the reviewer, South African Methodist missionary anthropologist Edwin W. Smith, praised the volume for its “practical and scientific value” (Smith 1922: 600). Understanding the regulatory institutions and proscriptions of Kikuyu society could greatly aid administrators and settlers in their management of labor (Smith 1922: 600), if only Hobley could compensate for his poor anthropological theorizing. Hobley would be the first in a long line of thinkers to mistake the Kikuyu for a rule-governed society rather than a rule-producing one. At the root of this confusion was a fixation on the alleged regulatory capacities of thahu as a form of “curse” administered by definitive agents:

We think he (Hobley) is wrong in giving “curse” (in the medieval sense) as the equivalent of thahu. A curse, as Sir James Frazer points out, implies a personal; agent, human or divine, who has called down evil on the sufferer. It is true that natives say that the thahu condition is caused by the spirits of departed ancestors, the ngoma, but this clearly is a rationalization of the belief; an examination of Mr. Hobley’s cases shows that there is, for the most part; no suggestion of such an agency, and the thahu is removed not by prayers and offerings to the spirits but by devices of the medicine-men. Sir J. Frazer thinks that “ceremonial uncleanliness” is a better rendering of the word, but even this is not satisfactory. Both renderings obscure the fact that people or animals that are thahu are a danger to the community. By their condition, or by their acts, they betray the presence in them of an evil-working principle, a mystic force, whose action may be fatal if steps are not taken. (Smith 1922: 600–601)

Thahu was thus a constant force in daily life. Similar to Georges Bataille’s “general economy,” in which a plenitude of cosmic energy generated by the sun must be expended with no return or profit (2007: 25–26), thahu required discharge or else would spread contagiously, wreaking social havoc. Contrary to British ideas that gĩthathi and elder were metonymic equivalents, thahu actually functioned more like a mediating—but not necessarily commensurating—third term. However, as a mediating third term, its excessive qualities could not effectively guarantee the equivalence between elder and stone as a dependable basis for producing social order in times of conflict. It could build up, leak, and infect people and animals when utilized for oaths. The main method for expelling thahu was purgative vomiting, a practice known as tahikio, requiring the prosthetic assistance of cool leaves. As Derek Peterson writes:

Wise men induced this restorative vomiting by holding the hoof of a goat up to the lips of the victim, enjoining him or her to vomit out the thahu. The wise men then brushed the sufferer with the leaves of the mukenia tree, dipped in a collection of powders. Ngondu, one of the powders wise men used in this process, was derived from the stomach contents of the tree hyrax, sugar cane, sweet potatoes, the stomach contests of a ram, and fig tree leaves. Each of these substances was in some way “cool”—the ram for instance, ate only the soft leaves of non-thorny plants. The contents of its stomach, which distilled the cooling leaves, were useful in cleansing the sufferers of heated contaminants. (Peterson 2002: 40)

Nature was thus both the eternal combatant of Kikuyu civilization and also that upon which Kikuyu social reproduction was dependent in certain situations. It seemed capable of endlessly replenishing hot and wild energies, which in turn always impinged upon the coolness of the homestead (Peterson 2002: 39). Such hot forces and their opposite were omnipresent, but overall, they defied exact signification and routinization. Of course, there was a logic to dealing with the un-predictability of nature and thahu. People made attempts literally to corral it: Peterson speaks of magongona—Kikuyu attempts to “protect the living from the uncharitable dead” (Peterson 2002: 41)—in which lines were drawn separating the dangerous wilderness, where ancestors ended up, and the safer confines of the home from which they had departed. Ultimately, though, these attempts, which were ongoing, remained just that: attempts. As Peterson observes, the substances used to treat thahu “were never simply symbols, standing in a fixed relationship to [its] heat.” Rather, “they were unstable, unpredictable, probably dangerous substances whose effects were not mechanically determined” (Peterson 2002: 40). Put simply, tahikio and magongona was speculative, what Peterson rightly refers to as “gambling” (Peterson 2002: 40). The ultimately excessive nature of thahu thus prevented the emergence of a fixed repertoire of avoidance techniques, or exact methods for handling it. In this sense, it was an engine for the proliferation of what appeared to colonial observers as rules.

This fact was difficult to absorb for anthropologists, administrators, and missionaries alike because nowhere was the ambiguity more evident than in efforts to cram thahu and other ritual concepts into the conceptual categories of Christianity and its implicit semiotics of mastery. For instance, Scots Presbyterian missionaries at Tumutumu were disappointed that the Kikuyu lacked abstract thought about their “high god” Ngai who lived on Mount Kenya. In fact, they were making a category error. As Peterson explains, natural and social disasters like drought often involved drawing boundaries, to speculatively contain “an unfamiliar deity” Ngai, whose name (like the term for oath, recall) was etymologically Maasai (Peterson 2002: 41). For Kikuyu, it was unwise to attract Ngai’s attention at all. Still, the seeming absence of a divine sovereign authority interrupted the missionary hope for interreligious dialogue about abstract theological precepts. It destabilized the assumption that religions were always “fully formed systems” (Peterson 2002: 48), an assumption that was smuggled in via the very concepts of “religion” and “ritual” and the specific Christian genealogy that undergirded these categories (Asad in Peterson 2002: 38). This aspiration to a fixed set of rules was perhaps most potently epitomized in the Tumutumu missionaries teaching converts to “reenact the story of Adam, when animate objects were identified in the hierarchy of Creation” (Peterson 2002: 43). As Peterson succinctly puts it, “naming and mastering, in Protestant thought, were intimately related” (Peterson 2002: 43). Just like in the Garden of Eden before the Fall, the missionary compulsion to name evinces the same semiotic ideology Agamben argues undergirds oaths. Missionaries encouraged potential converts to imitate God’s perfect speech, whose words were true the instant they left the deity’s lips.

Thahu, however, represented a particularly difficult translational challenge. It was the Kikuyu concept most similar to the Christian notion of sin or so missionaries thought: just as sin involved infractions against God’s law, so thahu involved straightforward consequences for trespasses—or at least, what missionaries assumed to be trespasses (Peterson 2002: 44). In contrast, as Peterson notes, “Ngai left no commandments with which to judge human action,” a problem that often led missionaries to accuse Kikuyu of having no moral conscience at all (Peterson 2002: 44). No matter how they tried, thahu would not lend itself to translation, and sin clearly could not convey the sense of pollution without sovereign judgment that thahu originally conveyed. The missionaries were eventually obliged to invent a new word for sin, “mehia,” and in so doing abandoned efforts to recuperate thahu, which hampered rather than helped to establish the notion of the “immaterial transcendence” of God (Peterson 2002: 47).

While missionaries may have given up on the regulatory capacities of thahu, the colonial state did not. Instead, together with Kikuyu elders, it attempted to distill the force of thahu into ĩthathi, to develop a “native” political resource without excess. Early Kikuyu understandings of kũringa gĩthathi had put the practice on par with other speculative practices. Again, the relationship between elder and gĩthathi was mediated by an unstable third term whose excess could destroy biological and social reproduction altogether. By limiting the gĩthathi oath’s use, as Leakey argues, elders came to imagine themselves able to maintain relative control over the full destructive power of thahu. This strategy of managing the circulation and deployment of the gĩthathi oath, however, was arguably an attempt by elders to carve out a narrowly circumscribed independence from the colonial state. This would be overturned by the state’s usurpation of the ownership of the stones themselves. After a certain point, they could only be housed in colonial administrative offices.

The limits of tradition

The earliest mention of the gĩthathi oath in the colonial administrative archive, however, is a 1933 request from the Forest Office in Londiani to Kiambu DC J. E. H. Lambert asking if the DC had a stone “to spare.” Apparently, the stone the DC had borrowed from the Resident Magistrate in Nakuru had arrived bro-ken.9 Clearly, “oathing stones,” as the forest officer in Londiani called them, were already well out of the hands of specific elders and their patrilineages and were now a portable and generic form of authority appropriable by colonial governors. The fact that by 1950 the President of the Nyeri Local Native Council would need four more stones in addition to the ones the DC’s office possessed would seem to indicate a sudden increase in gĩthathi oaths, a situation we might describe as “supply side oathing.” The more murky and numerous the claims to historical land ownership, the more the stone’s powers were in demand. Rumors of corruption, however, in the form of massive bribes to court elders, had emerged in district reports as early as 1948, echoing Kenyatta’s concern that money had rendered purchasable the elder imprimatur that he assumed were embodied in the stones, turning them into just another commodity.

This contradiction between universality and particularism struck at the core of the colonial project in Kenya. Administrators preoccupied with a perceived lack of universal durable abstractions like law or religion were equally intent on ensuring that the instruments of universal social regulation be locally recognizable enough to be effective. Their solution was a generalized traditional oath, which actually failed to quell, and may have actually exacerbated, the anxieties of district officials that traditional authorities could be effective agents of social order. Eventually, elder attempts to limit the supply of ĩthathi by keeping them strictly taboo and out of circulation gave way to administrative demands for their increased usage. The colonial state’s attempt to fashion elders into the embodiment of an increasingly abstract Kikuyu law thus launched ĩthathi on a contradictory trajectory. Ithathi began circulating as an all-purpose ritual object in colonially reconstituted oathing procedures, but simultaneously challenged the local particularities upon which indirect rule was allegedly anchored. This contradiction rendered completely ambiguous the question of whether ĩthathi backed the authority of elders or elders backed the authority of ĩthathi. At the same time, it intensified the logic of proliferation that was once attributable to thahu, an explanatory concept that was now vague if not wholly absent.

Administrators would eventually attempt to generalize the new “customary” oath across all administrative districts. In 1929 the Kiambu DC wrote a memo to his peers, who had expressed the need for consensus on proper oathing procedures:

It was suggested at the last District Commissioners’ meeting that uniformity in the above matter is desirable and that I should begin the discussion by circulating for comments the form of the oath for the Native Council which is in use in this office. It is attached. Will you please pass it on with your comments to the District Commissioner, Fort Hall, asking him to send it with his comments back to me.10

Actual administrative practices were thus pragmatic experiments in commen-suration within the terms of indirect rule.11 While the DC for South Nyeri asserted the correctness of swearing to a remote sovereign authority like God, he saw no point in trying to fill oaths with explicit administrative agendas. Clauses designed to prevent Native Council members from making critical commentary about government orders, he was quick to point out, would create a situation in which elders would be violating their own oaths much of the time.12 To attempt the ritual micromanagement of Kikuyu elders would be to risk accusations of trying to facilitate something of a ritual coup, a sleight of hand through which Kikuyu and British authority could be transubstantiated. In other words, if colonial administrators wanted oaths to be effective in securing elder loyalty in councils and elder honesty in tribunals, they could not be used to secure everything on the colonial state’s agenda. Like Kikuyu elders, colonial administrators had to keep them relatively singular in purpose and restricted in circulation. In attempting to use customary law to enforce the contractual basis of new institutions like individualized land ownership, elders themselves would often ask the colonial state to back the oaths’ oracular verdicts, while at other times native administration would continue without any such appeals.

Aspirations for the standardization and generalization of oaths across districts were also undermined by the revolving door of district commissioners cycling in and out of administrative districts. Every few years, new district officials would circulate telegrams requesting information from other districts about proper oathing procedures so that they could be effectively replicated elsewhere. By the late 1940s, despite the fact that the specifically Kiambu LNC oath had been generalized all the way to Machakos, Elgeyo-Marakwet, and Pokot districts, commissioners were still anxious about what backed sincerity in native tribunals. In other words, colonial administrative turnover, while hamstringing standardization at one level, kept producing further desire for standardization across districts and ethnicities, as seen in this brief 1949 memo from the DC of Elgeyo-Marakwet:

I am anxious to introduce a form of oath to be sworn by all LNC Members when taking Office, and I would be grateful if you would let me have a copy of the form of oath used in the Kiambu LNC.13

The infinite regress in the search for truth-backing authority in moments of conflict led administrators to fixate on the right words for “natives” to recite during oaths. In attempting to find the right words, administrators hoped to strike a balance between universal aspirations of law on the one hand and native capacity to recognize their own local legal procedures on the other. In focusing on the oath’s statutes, colonial administrators produced a version of oathing in which elder authority, words, and things were not only solidly “within a single representational economy” (Keane 2007: 20) but also thought to be unproblematically equivalent. But as ĩthathi entered general circulation, just like the Kiambu LNC oath, “changes in some domains” of the gĩthathi oath’s representational economy had “consequences in others” (Keane 2007: 20).

To their credit, some administrators saw the dangers of generalized oathing. At a District Commissioners’ meeting in 1947, one Central Province official noted that generalization of the gĩthathi oath had gone far in destroying the regulatory aspirations that district officials had for it in the 1920s:

The question of formulating a procedure generally applicable to the whole Province was fully discussed. It was considered doubtful if agreement was either desirable or possible even within the three Kikuyu districts of Nyeri, Fort Hall and Kiambu.

The nature of the original Gĩthathi oath was considered…. It was formerly employed only in important cases like stock theft etc. The ritual was extremely complicated and took a long time. It was pointed out that there is now a Gĩthathi stone in every Native Tribunal in Nyeri District and that it is regularly taken. The complicated ritual is, in many cases, no longer followed, and the oath has therefore lost much of its original sanction….

The Meeting … further considered that generalization is one of the surest ways of making the oath ineffective.14

Some administrators simply washed their hands of trying to figure out proper oathing procedures, leaving such matters to the very institutions whose basis for ritual authority had been rendered ambiguous by colonial tampering. This provided already powerful elders ample opportunity for yet more ritual and legal improvisation. The following letter is from the Registrar of the Native Tribunal in Chura, Kikuyu district during the late 1940s, regarding the way oaths were used to decide whose competing claims to land ownership were “true”:

I beg to inform you that when both parties are ordered by the court to take the customary oath “muma” or “thenge” sometimes one party refuses to take it. We obviously award the whole land or anything in dispute to the party which does not refuse to take the oath. Later we see the party which had refused to take the oath applying for an appeal.

Kindly inform whether such party can be allowed to appeal the same case anymore unless it agrees to take the oath as ordered. In this court’s opinion, such party cannot be allowed to appeal the same case. Once it had already refused to take the oath that is the final judgment of the case unless it agrees to take it later on.15

The Kiambu DC simply referred the matter back to the Local Native Council. Clearly the colonial administration wanted oaths to be backed by elders, but the ambiguous hierarchy of institutions and objects led to a problematic surplus of authority, in turn generating the problem of competing legal temporalities. By the 1940s, for example, the generalization of the oath and the increased circulation of ĩthathi entailed a noticeable compression of the curse’s statute of limitation of seven agricultural seasons. Many who asked for the oath, especially to settle land ownership claims, had the expectation that its verdict would be more or less instantaneous. Some complained that litigants in land cases were jumping to appellate procedures prematurely, since their opponents had not yet experienced the oath’s effects.

In 1945, minutes from the office of the Kiambu DC reported that Chief Magugu “said that these days people, after taking the oath in a case, did not wait for the prescribed time before filing an appeal,” and that “this is not Kikuyu custom.” The DC and LNC were in agreement that “people who have taken an oath should not be allowed to appeal before the prescribed time of seven seasons.”16 Yet this assertion of the primacy of the temporality of the curse conflicted with other emerging understandings of tradition, in which the gĩthathi was understood to be the equivalent of swearing on the Bible, the practice of English courts, where a verdict potentially could be rendered the same day. Indeed, Christians who did not want to take “pagan oaths” introduced another competing authority into the framework of oathing: mission authorities could authenticate practicing Christians and certify whether a convert could be relieved from the duress of the oath altogether.

These competing legal temporalities thus also undergirded conflicting understandings of evidentiary procedure. One understood the gĩthathi as oracular, able to divine wrongdoing. Another framework suggested the gĩthathi was merely the first step toward allegedly more robust evidentiary procedures. What seems to have happened over the twenty year period examined here is that, despite administrative aspirations for oathing to help commensurate legal regimes, the actual process produced a cacophony of “enunciative procedures” (Bayart 1993: 240–41) that both Kikuyus and colonial administrators imagined to be backed by different authorities within the same ritual/legal form. The problematic assumption that elders and symbols were equal sovereigns became an engine for divergent fantasies of authoritative backing. The question of what drove the desire for the standardization of traditional oaths and the increased circulation of Tthathi requires further explanation, and returns us to the figure of Jomo Kenyatta, with whom we began.

The problem of proliferation

Whereas earlier accounts emphasized the stones’ exotic and mysterious origins and narrowly controlled circulation, administrative correspondence reveals the relative ease by which they could be obtained by 1950. While administrative mandates were put into place to prevent elders from charging “traditional” fees for a gĩthathi oath in 1950, administrators in Kiambu had no problem with the DC’s office in Nyeri paying fees to the “traditional supplier of gĩthathi stones”:

  1. No charges are being levied on the parties wishing to or required to swear the gĩthathi oath in the Tribunals of this district.
  2. Hon. Senior Chief Wambugu is the traditional supplier of gĩthathi stones. I have informed him of your requirements and he is willing and able to supply two stones immediately. He can produce two more in due course.
  3. The traditional price of a gĩthathi stone is thirty sheep and three rams which on present day valuation amounts to Shs. 720/4.
  4. Chief Wambugu is willing to supply the Kiambu council with stones at a cost of Shs. 280/- each.

Will you please inform me if we are prepared to pay this amount.

I personally consider this price to be very reasonable.17

Colonialism on the cheap indeed! What the DC reveals is that despite attempts to cordon off the ritual from the persuasive power of money, money is nonetheless revealed to be what was thoroughly mediating the equivalence between elderhood and the symbolic authority distilled in Tthathi. Stones were being replicated and reproduced when they were supposed to be singular and unique.

In this light, we might read Leakey’s description of the ritual creation and destruction of the “imitation gĩthathi” as symptom of, rather than the antidote to, the generalization of the gĩthathi oath. If the identity of a stone was associated with specific patrilineages and traveled to and from oathing locations with its “owner,” it seems strange that such an emphatic destruction of the counterfeit would be necessary at all. Clearly by 1950, the distinct patrilineal association and history of an individual gĩthathi had been replaced with a “traditional supplier” willing to offer them to colonial administrators at a “reasonable price.” But with the mass production of ĩthathi came increased paranoia about counterfeit stones and the sincerity of elders, an intensification of the contradictions we can now see were latent within the form of the oath itself from an earlier period. Nowhere in the colonial archive is this condition of proliferating ritual authority and doubt more distilled than in the letter from Kenyatta with which we began, in which he ultimately asks the colonial state to guarantee the relationship between elder authority and stone as an allegedly stable, mediating third term. Like all accusations of corruption, he suggests that the right way of doing things is still achievable. Not only can there be an authoritative backer of truth, he suggests, but the specific agents of dissimulation are also discoverable and controllable. Again, these are assumptions Kenyatta shared with colonial administrators.

Hence Kenyatta’s lament: if nothing could prevent money from rendering elders dubious backers of truth in general, the state should intervene. And it did. But at this point the state’s response to his letter should not surprise us: The DC simply asked the owner of the suspect gĩthathi, one Nganga Kabitu, to swear an oath to its authenticity:

The following are the words that Nganga Kibitu is going to swear to prove that his gĩthathi is the true one.

  1. This gĩthathi which I hold in my hand is the gĩthathi which belonged to my grandfather; if I do not speak the truth I must die for this Thenge.
  2. The gĩthathi which I hold in my hand was left to my father by my grandfather; if this is not true I must die for this Thenge.
  3. This is the original gĩthathi which was left to me by my father; if this not true I must die for this Thenge.
  4. If I myself have made or if I obtained from some other person Thenge besides the original one which was my grandfather’s and which was left to me by my father I must die for this Thenge.
  5. If I purify myself after taking this Oath before three and a half years have expired I must die for this Thenge.
  6. If I by cunning means take any steps whatsoever to attempt to cause the death of Kenyata [sic] I must die for this Thenge.
  7. If any persons take any steps whatsoever to kill me or my family so that it will appear that the cause of my or my family’s death was the taking of this Thenge Oath he must die for this Thenge.18

The problem of the counterfeit is proliferating here in multiple registers. Since Kabitu’s conflict is with Kenyatta’s claim regarding the stone’s authenticity, even the oracular power of the stone can be potentially faked. Kabitu is asked to swear that he will not seek to cause the death of Kenyatta in a way that might appear to simulate the effect of the curse, thus counterfeiting the oath’s verdict. But in light of the already ambiguous nature of the stone, it is unclear what exactly is backing this new certifying oath beyond a reference to another certifying media used in oaths, a slaughtered goat, or thenge. Kenyatta’s desire for state intervention, compared with the state’s actual response by authenticating the stone through another oath, reveals the question of what backed ritual authority to be an aporia.


I have argued that the precolonial Kikuyu ideology of ritual efficacy entailed a semiotics of management and not one of mastery. Far from being a sovereign that performed a guaranteeing function, thahu was a constitutive outside to Kikuyu society, whose ultimately excessive character made it something of an antisover-eign. It mediated the relationship between elders and “oathing stones” but not in a way that really represented them as political/ritual equivalents. The colonial era incorporation of thahu as a property and extension of elder power and authority in effect, bringing it into the realm of the social and of politics (recall that Tthathi were kept in district officers’ offices and mass-produced by “traditional suppliers”), simultaneously denatured thahu as the force backing oaths, raising new questions about the truth of the oath’s constituent elements and agents: Tthathi and elders. This problem became especially acute when oaths were repurposed for assisting in the establishment of a wholly new regime of private property.

Kenyatta had thus tried to bypass the problematic mediation of local elders who, like their Tthathi, he considered counterfeit. Yet, while Kenyatta attempted to recharge and make singular the institution of the gĩthathi oath, to reauthenticate it with the power of the state, the fact that district officials had been delegating such certifying functions to elders only revealed the irresolvable nature of the problems of ritual authority and efficacy that oaths now epitomized. In the upshot, the oath’s most powerful sovereign, the gĩthathi, had become little more than a stone. The assumption of equivalence between elders and ritual symbols as the font of sovereign power, its “ontological ground,” as already stated, would eventually develop into an inflationary ritual economy in which young, dispossessed Kikuyu men attempted to make oaths efficacious through the manipulation of unmoored “traditional” symbols, augmented by a supplemental violence, in order to produce the sovereignty that they believed thahu to both found and maintain in the past (cf. Derrida 2002: 228–98).

Mau Mau thus became a war characterized by the type of incendiary and reciprocal violence that—in the precolonial world—a gĩthathi was supposed to prevent. A 1946 letter by one J. Karanja of Limuru foreshadows the confusing rites that Worsley and Gluckman tried to explain, but at the tail end of their development. In the letter, Mr. Karanja still held out hope that if the colonial state would bring the gĩthathi oath to his area, then problems of land distribution brought on by its commoditization could potentially be resolved:

I appeal to the Government of Kenya for help on account of the poor people who are in the Kikuyu Country, and the small sub clans, because they are always deprived of their property…. “Gĩthathi” is the only symbol of a truth.

It would be of great assistance to the poor Kikuyu people if the Government would allow “Gĩthathi” as the rich people would than [sic] have no chance of oppressing and exploiting the poor persons.19

In the face of predations by wealthy Kikuyu chiefs, council members, and big men, Karanja longed for a definitive oath that could protect the poor. Before colonialism however, it is arguable that Kikuyu would not have understood an oath as a sacrament at all. Inasmuch as swearing an oath entails an effort to sacramentally participate in God’s power to join word and deed (Agamben 2011: 71), Agamben is right: this semiotics of mastery is deeply problematic. And it is here that Agamben’s critique of the oath becomes relevant, but—and this is key—only after indirect rule made it so. Kikuyus struggled to keep thahu’s power beyond Kikuyu society’s boundaries and gerontocratic political structures. Missionaries and administrators had placed it at its normative center. With this in mind, Mr. Karanja’s request was predictably referred back to the Local Native Council.


I gratefully acknowledge the financial support for my fieldwork granted by Fulbright-Hays, the Social Science Research Council, and the Mellon-Sawyer seminar for the study of ethnicity in Africa at the university of Michigan during my 2010–2011 postdoctoral year. Besides the anonymous reviewers and editors of HAU, I would like to thank the following friends and colleagues for providing generous and critical comments while I was developing this article: Jeremy Jones, Zebulon Dingley, Derek Peterson, Paul Barclay, Courtney Handman, James Slotta, Sasha Newell, Rihan Yeh, Joseph Hankins, and last but certainly not least, Francois Richard.


Agamben, Giorgio. 1998. Homo sacer: Sovereign power and bare life. Stanford, CA: Stanford University Press.

———. 2005. State of exception. Chicago: University of Chicago Press.

———. 2011. The sacrament of language: An archaeology of the oath. Stanford, CA: Stanford University Press.

Anderson, David. 2000. “Master and servant in colonial Kenya.” Journal of African History 41 (3): 459–85.

Barnett, Donald L., and Karari Njama. 1966. Mau Mau from within: Autobiography and analysis of Kenya’s peasant revolt. New York and London: Monthly Review Press.

Bataille, Georges. 2007. The accursed share: An essay on general economy. New York: Zone Books.

Bayart, Jean-Francois. 1993. The state in Africa: The politics of the belly. London: Longman.

Berman, Bruce. 1996. “Ethnography as politics, politics as ethnography: Kenyatta, Malinowski, and the making of facing Mount Kenya.” Canadian Journal of African Studies 30 (3): 313–44.

Blom Hansen, Thomas, and Finn Stepputat. 2006. “Sovereignty revisited.” Annual Review of Anthropology 35: 295–315.

Branch, Daniel. 2009. Defeating Mau Mau, creating Kenya: Counterinsurgency, civil war, and decolonization. Cambridge: Cambridge University Press.

Caffentzis, Constantine. 1989. Clipped coins, abused words, civil government: John Locke’s philosophy of money. Brooklyn: Autonomedia.

Carothers, John Colin. 1954. The psychology of Mau Mau. Nairobi: Printed by the government printer.

Clough, Marshall. 1998. Mau Mau memoirs: History, memory, and politics. Boulder, CO: Lynne Reinner Publishers.

Comaroff, Jean, and John L. Comaroff. 1991. Of revelation and revolution: The dialectics of modernity on a South African frontier: Volume One. Chicago: University of Chicago Press.

Derrida, Jacque. 2002. “Force of law: The ‘mystical foundations of authority.’” In Acts of religion, 228–98. New York: Routledge.

Elkins, Caroline. 2005. Imperial reckoning: The untold story of Britain’s gulag in Kenya. Toledo, OH: Owl Books.

Evans-Pritchard, E. E. 1963. Witchcraft oracles and magic among the Azande. Oxford: Clarendon Press.

———. 1967. The Nuer: A description of the modes of livelihood and political institutions of a Nilotic people. Oxford: Clarendon Press.

Fitch, Richard. 2012. “Review of ’the sacrament of language‘; An archaeology of the oath by Giorgio Agamben.” Marx and philosophy review of books. www.marxandphilosophy.org.uk/reviewofbooks/reviews/2012/501.

Githige, Renison. 1978. “The religious factor in Mau Mau with particular reference to Mau Mau oaths.” Unpublished thesis, University of Nairobi.

Gluckman, Max. 1963. “The magic of despair.” In Order and rebellion in tribal Africa: Collected essays, with and autobiographical introduction. New York: Free Press of Glencoe.

Goux, Jean-Joseph. 1990. “Numismatics: An essay in theoretical numismatics.” In Symbolic economies: After Marx and Freud. Ithaca, NY: Cornell University Press.

Green, Maia. 1990. “Mau Mau oathing rituals and political ideology in Kenya: A re-analysis.” Africa 60 (1): 69–87.

Hobley, Charles William. 1922. Bantu beliefs and magic: With particular reference to the Kikuyu and Kamba tribes of Kenya colony; Together with some reflections on East Africa after the War. London: Witherby.

Hudson-Koster, Mickie. 2010. “The making of Mau Mau: The power of the oath.” PhD diss., Rice University.

Keane, Webb. 1997. Signs of recognition: Powers and hazards of representation in an Indonesian society. Berkeley: University of California Press.

———. 2007. Christian moderns: Freedom and fetish in the mission encounter. Berkeley: University of California Press.

Kenyatta, Jomo. 1965. Facing Mount Kenya: The tribal life of the Gikuyu. New York: Vintage Books.

Kershaw, Greet. 1997. Mau Mau from below. Oxford: James Currey.

Kopytoff, Igor. 1986. “The cultural biography of things: Commoditization as process.” In The social life of things: Commodities in cultural perspective, edited by Arjun Appadurai, 64–94. Cambridge: Cambridge University Press.

Leakey, L. S. B. 1954. Defeating Mau Mau. London: Methuen.

———. 1977. The southern Kikuyu before 1903. London: Academic Press.

Lonsdale, John. 1992. “The moral economy of Mau Mau: Wealth, poverty, and civic virtue in Kikuyu political thought.” In Unhappy valley: Conflict in Kenya and Africa, by Bruce Berman and John Lonsdale, 315–468. London: James Currey.

Marx, Karl. 1976. Capital, Volume 1: A critique of political economy. New York: Vintage.

Meillassoux, Claud. 1972. “From reproduction to production: A Marxist approach to economic anthropology.” Economy and Society 1 (1): 93–105.

Peterson, Derek R. 2002. “Gambling with God: Rethinking religion in colonial Central Kenya.” In The invention of religion: Rethinking belief in politics and history, by Derek R. Peterson and Darren R. Walhof, 37–58. New Brunswick, NJ: Rutgers University Press.

Pietz, William. 1985. “The Problem of the fetish, I.” Res 9: 5–17.

———. 1987. “The problem of the fetish, II.” Res 13: 23–45.

———. 1988. “The problem of the fetish, IIIa.” Res 16: 105–24.

Rutherford, Danilyn. 2003. Raiding the land of the foreigners. Princeton, NJ: Princeton University Press.

Sahlins, Marshall David. 2008. “The stranger king.” Indonesia and the Malay World 36 (105): 177–99.

Sheik, Nafisa Essop. 2011. “Colonial rites: Custom, marriage law and the making of difference in natal, 1830s-c.1910.” PhD diss., university of Michigan.

Sorrenson, M. P. K. 1967. Land reform in the Kikuyu country: A study in government policy. Nairobi: Oxford University Press.

Smith, Edwin W. 1922. “Religious beliefs in Kenya.” International Review of Missions 11: 599–601.

Smith, James H. 1998. “Njama’s supper: The consumption and use of literary potency by Mau Mau insurgents in colonial Kenya.” Comparative Studies of Society and History 40 (3): 524–48.

Spyer, Patricia. 1998. “Introduction.” In Border fetishisms: Material objects in unstable spaces, edited by Patricia Spyer, 1–12. New York: Routledge.

Tate, H. R. 1918. “Kikuyu ‘Ithathi.’” Journal of the East Africa and Uganda Natural History Society 6: 261–65.

Weiss, Brad. 1996. The making and unmaking of the Haya lived world. Durham, NC: Duke University Press Books.

La complainte de Kenyatta. Serments et transformation des idéologies rituelles dans le Kenya colonial

Résumé : Cet article fournit une critique ethnographiquement et historiquement fondée de l’affirmation de Giorgio Agamben que les serments, comme la religion et la loi, reflètent une expérience universelle de l‘échec de la langue d’obliger les gens. Je montre comment la critique d’Agamben du serment ne devient pertinente que dans le contexte du Kenya colonial après que les missionnaires et les administrateurs ont contribué à « défaire » les idéologies kikuyu de l’efficacité rituelle. L’article suit la façon dont les administrateurs coloniaux et les anciens Kikuyus sont venus à comprendre la nature coutumière de l’autorité des anciens à travers la capacité réglementaire et représentationnelle attribuée aux objets dits ĩthathi. En fin de compte, les administrateurs et les anciens partageaient une idéologie associant efficacité rituelle et autorité des anciens, et participaient conjointement à la naturalisation de cette relation sans pour autant reconnaitre où cela mènerait.

Robert W. BLUNT is Assistant Professor of Religious Studies and Africana Studies at Lafayette College in Easton, Pennsylvania. His book manuscript, Of money and elders, explores the colonial transformations of Kikuyu cosmology, society, and ritual and how these transformations were partially generative of Kenya’s postcolonial geronotocratic political order, and concomitant social anxieties around truth, value, and elders. He has also published work on the relationship between state-sponsored counterfeiting and Pentecostal discourses about Satanism in Kenya. More recently, he has published an article on the experience of being accused of witchcraft during fieldwork.

Robert W. Blunt
Department of Religious Studies
Lafayette College
Easton, PA 18042 USA


1. The article was reprinted in Gluckman’s Order and rebellion in tribal Africa.

2. Most scholars of Mau Mau understand the origin of Mau Mau oaths to be rooted in the membership oaths of the Kikuyu Central Association, an early Kikuyu political oragnization that reprsented Kikuyu grievances to the colonial government before the organization was banned in 1940.

3. Githige argues that forest fighters killed oath breakers before the curse could take effect because they knew the curse would do so anyway.

4. One notable exception is James H. Smith’s (1998) analysis of how forest fighters redeployed colonial literary-bureaucratic fetishes, like typewriters and stamps, to carve out a “counter-state” in the forests of Mount Kenya and the Aberdares.

5. The exact symptomology of thahu remains vague but C. W. Hobley, the colonial administrator turned comparative religionist described thahu as a condition of bodily wasting or erruption of boils that were apparently lethal (Hobley 1922: 103).

6. Kenya’s reserves were areas set aside for Africans along ethnic lines that colonial authorities understood to be relatively fixed in nature. The colonial state tightly controlled the movement of Africans between the reserves and settler dominated areas. The Native Administration Ammendement Ordinance passed in 1920 required all African males of working age to wear the kipande around their necks at all times (literally “a small piece” in Kiswahili), a small metal box that contained the name, fingerprints, and labor history of the wearer. The kipande was also known as the Native Administration Docket (Anderson 2000).

7. KNA MA/1/1, Jomo Kenyatta, Letter in Reference to Gatundu No. 116/48 and LNC. No. 65/49, 1949.

8. The few elder informants with whom I spoke who could even remember ĩthathi, claimed that in order to produce a new one, an elder would have to first hew a piece of rock, which would become the gĩthathi, from the earth behind his house where “no one had walked.”

9. KNA MA/1/1, Forest Office Londiani to DC Kiambu, “Oath Stone Kikuyu” 22 February 1933; DC Kiambu to Forester Londiani, 23 February 1933.

10. KNA MA/1/1, DC South Nyeri to DC Kiambu, “Form of Oath for Native Council and Native Tribunals,” 27 February 1929.

11. KNA MA/1/1, DC Fort Hall to DC South Nyeri, “Forms of Oath for Native Council, Native Tribunal,” 16 April 1929.

12. KNA MA/1/1, DC Fort Hall to DC South Nyeri, “Forms of Oath for Native Council, Native Tribunal,” 16 April 1929.

13. KNA MA/1/1, DC Elgeyo-Marakwet to DC Kiambu, “LNC Oath,” 17 January 1949; DC Kiambu to DC Elgeyo-Marakwet, “LNC Oath,” 25 January 1949.

14. KNA MA/1/1, DC’s Meeting, Minute 51/47, “Native Oaths,” 5–8 November 1947.

15. KNA MA/1/1, Registrar, Native Tribunal Kikuyu, “Re: Customary Oath ’thenge’ or ‘Muuma’ Procedure,” 21 July 1948, Kiambu LNC. Meeting Minute 25/48, 10–12 August 1948.

16. KNA MA/1/1, D.O. for DC Kiambu to all Tribunal Clerks Kiambu, “Re: Oaths,” 1 October 1945.

17. KNA MA/1/1, D.O. for President of Nyeri LNC to DC Kiambu, “Re: African Customary Oath,” 20 July 1950.

18. KNA MA/1/1, D.O. for DC Kiambu, “Thenge Oath to be Taken by Nganga Kabitu,” 7 October 1949.

19. KNA MA/1/1,J. Karanja to DC Kiambu 2 January 1946.