The author would like to thank the Samburu communities who permitted and supported this research, and the Kenyan Ministry of Education for permission to conduct research on which this paper is based (1992–1994, 2001–2007). Funding was provided by National Science Foundation Grant # (Jon Holtzman, co-PI), Western Michigan University, Fulbright IIE, University of Michigan, and Ford Foundation.
1In 1934, a well-publicized case in Kenya concerning the murder of Theodore Powys, a young British settler, resulted in a strange acquittal—strange because an entire Samburu pastoralist community underwent harsh state-sponsored collective punishment in its aftermath. In this instance, the acquittal exacerbated rather than eased colonial settler animosity towards the Samburu. Much was at stake: A small group of settlers had lobbied for over a decade to displace Samburu from the well-watered dry season grazing of north central Kenya’s Leroghi Plateau, bringing the usual accusations to bear on their claim for indigenous resources; according to settlers, the Maa-speaking Samburu—like the Maasai already displaced from neighboring Laikipia Plateau in 1904 and 1911—did not efficiently manage the land and, moreover, they engaged in ritualized violent practices. Unfortunately for the settlers involved, not only were the accused warriors acquitted in the case but, moreover, the Samburu maintained their rights to Leroghi Plateau. These outcomes notwithstanding, the punitive policies that were implemented against the Samburu for decades after the case derived directly from the cultural tropes invoked in the case that still haunt Kenyan pastoralists into the twenty-first century, as this paper will show.
2This paper gives careful attention to one discursive trope about the role of young feminine agency in reinforcing violent masculine bravado in order to highlight the ways in which “culture” can be strategically deployed in the globally significant milieus of land policy and conflict. Specifically, the paper describes Samburu girls’ varied political-economic and contested cultural role in Samburu society in ways not described sufficiently in the literature on East African pastoralists. The paper argues that a culturally particularized trope of feminine agency helped efface devastating political realities to the benefit of colonial settlers. Samburu elders perceived some truth in the cultural trope settlers and colonial officers invoked and were therefore complicit in reinforcing the stereotype of uncontrollable warriors and violence-inciting girlfriends. This internal cultural resonance notwithstanding, these gendered cultural practices were merely a framing device for historical and political factors that far exceeded them.
3Recounted here is a familiar story: Discourses about a marginalized community legitimize exploitative policies that reproduce and exacerbate marginalization. The particularities matter, however, especially when their resonance persists from one century into the next, permitting repeated misreading of the causes of land-based conflict. Stated in other terms, the paper highlights “culture” as distraction: The cultural discourses surrounding the case drew attention away from the fraught complexities surrounding colonialism and Samburu-settler conflict and contributed to the legitimization of policies detrimental to pastoralists. Moreover, they effectively criminalized the performative expressions of adolescent girls. What if girls did sing songs about martial prowess? (They did.) What if Samburu young men did kill a colonial settler? (They most likely didn’t but may have.) Would that tragic murder justify massive land loss and requital? Given the scale of colonial reprisals against an entire Samburu community, it appears that the price of a blood feud exacted by Europeans for the death of a young British settler was large indeed.
4Samburu are pastoralists inhabiting northern Kenya’s semiarid lands, where they keep goats, sheep, cattle, and, in some parts of Samburu, camels. Additionally, donkeys are common in the Samburu lowland areas for carrying water and other supplies, including for cattle camps, and for moving household goods during migration. The British colonial government enforced integration into the cash economy beginning in the early twentieth century, and by the 1930s, the government conscripted Samburu men to labor on roads and to serve in the Kenyan army. The first schools were opened during this period also. Wage labor and petty hawking of goods such as tobacco, tea, sugar, and beads became increasingly common by the 1980s, with wage labor more common among Samburu men but with some women also employed as nurses and teachers. In the twenty-first century, Samburu in the highland areas (on Leroghi Plateau) engage in a mixed economy that includes small-scale farming where rainfall is sufficient, wage labor employment typically of at least one member of an extended household, and education of both boys and girls, while nevertheless relying predominantly on the livestock economy. Away from Leroghi Plateau, Samburu in the lowlands practice little or no farming, relying almost entirely on livestock and livestock trade, but nevertheless educate an increasing number of both girls and boys, and encourage wage employment for individuals for whom it is possible. Mobile phone use is ubiquitous throughout Samburu, even if this means only turning on phones when in the range of cell phone towers concentrated on Leroghi and in the major lowland towns.
5In Samburu society, the economic contributions of girls and boys are largely undifferentiated until approximately eight years of age, when boys are more likely than girls to be sent to assist adolescent and young adult men at cattle camps. Even at this age and up until puberty, however, roles can be somewhat fluid particularly in the lowlands, where Samburu girls without brothers of suitable age may be sent to cattle camps under the watchful eye of older male relatives, both boys and girls venture far from home in their herding tasks during the day, and boys as well as girls may perform domestic tasks whether at cattle camps or to assist frail grandparents. On Leroghi in the highlands, however, gender roles are diverging earlier and more sharply, so that it is unusual for girls to herd far from home, which leaves girls closer to the cooking pot. Even on Leroghi, however, some gender fluidity persists, as boys may perform domestic tasks for grandparents. Irrespective of the degree of overlap in their tasks, adolescent and young adult girls and boys perform most of the labor in livestock economy while experiencing food and water scarcity in a difficult semiarid environment.
6Even if their labor overlaps prior to puberty, rituals for newborn girls differ from those for boys, adornment practices differ from early childhood when girls begin wearing multiple neck beads, and, by eight years of age, fathers do not physically discipline daughters due to avoidance rules (Figure 1). Additionally, caregivers instill the bodily movements of dance into babies and toddlers in gender-specific ways, holding and moving baby boys in the poses and moves of young men’s dances and baby girls in the poses and moves of girls’ and women’s dances. That is, from babyhood, girls learn to become the dancing singers who will praise and incite their boyfriends to bravery, beseech Divinity for rain and fertility, and soothe their cows into letting down their milk. Even before they reach puberty, girls ready themselves for these roles, adorning themselves with colorful beads at the neck, head, ankles, and waist. Boys, on the other hand, prepare to be excellent singers and defenders of the community and its livestock—with potential for offensive livestock-theft expeditions against neighbors as well (on men’s singing and ceremonial cycles see Marmone, 2017, present volume; see also Spencer, 1965).
figure 1. Lowland Samburu girls
a. In casual beads; b. Dressed for picture
© Bilinda Straight, 2017
- 1 For scholarship recognizing the importance of gender for understanding violence, see Elshtain, 1987 (...)
7East African pastoralists like the Samburu are well known for their participation in intercommunity cattle raiding (Blystad, 2005; Bollig, 1990; Bollig and Österle, 2008; Fleisher, 2000; Fukui and Turton, 1979; Hodgson, 2001; Holtzman, 2016; Hutchinson, 1996; Knighton, 2002; McCabe, 2004; Mirzeler and Young, 2000; Spencer, 1988). While the scholarly focus is typically on the varied actions and interests of male warriors, a few authors have noted the role of girls and young women in motivating men to fight (see, for example, Hutchinson, 1996; Knighton, 2002). In highlighting feminine agency in violent action, these authors forged a path away from treating women as passive victims or pawns and towards considering the gender and sexual politics of warfare.1
8Samburu have often been quintessentially depicted in postcards, coffee-table books, and the popular media as generously adorned girls or as glamorous bachelor warriors roaming the countryside engaging in livestock theft against their neighbors (Kasfir, 2007). While the portrayal is often oversimplified, the accuracy of certain aspects of this stereotype cannot be denied for contemporary Samburu in the lowlands particularly. After a prolonged initiation rite with circumcision at its center, the Samburu age system prescribes 7–14 years (ideally) of bachelor “warriorhood” during which Samburu lmurran (“warriors,” or, more accurately, “young initiated;” see Marmone, 2017) are expected to engage in hazardous long-distance cattle herding and protection of the community and its livestock, which can involve some theft or raiding of neighbors’ livestock. As in other East African contexts (e.g., Fleisher, 2000; Heald, 2000; Peatrik, 2013; Saitoti, 1986), raiding has historically been an important component of the construction of lmurran masculine identity, and the reputation earned as a lmurrani (sing.) can have a bearing on a man’s status in the community throughout his life. His age cohort is given its own name, and, by way of these generational names, Samburu tell and sing their histories of migration and warfare. Girls are key protagonists in these tales as sexy, well-ornamented lovers. Indeed, while Samburu women do not have formal age grades, they identify themselves according to the generation of lmurran with whom they danced, and thus the songs of lmurran and girls affirm their mutual importance to one another’s identities.
- 2 Many Samburu in the highlands have stopped training for martial roles, merely following the ceremon (...)
9In the twenty-first century, the proliferation of small arms, competition for political representation, and highly organized theft resulting in the movement of large numbers of livestock have considerably transformed Samburu men’s adolescence and young adulthood (see, for example, Bollig and Österle, 2008, 2013; Straight, 2009; Greiner, 2012, 2013). In the highlands, lmurran practices have altered considerably, with a trend toward decoupling the role of long-distance herding and defense from ornamented glamor.2 In the lowlands, in contrast, fashion and the military role still converge, and a strange amalgamation has resulted that merges glamor, community protection, and aggressive militarized campaigns fought with AK 47 rifles.
10The forging of Samburu lmurran masculinity begins early, when small boys begin to emulate their lmurran older brothers and neighbors. Boys watch lmurran very carefully, attempt to sing like them, and are eager to wear beaded ornaments like them. The importance of beaded ornaments to lmurran identity is evidenced by the fact that older boys who wear too many ornaments can be beaten by lmurran for transgressing the border between boy and initiated lmurran. Nevertheless, boys will try to wear as many ornaments as they can get away with in their attempts to imitate the living icons of their own future selves. Both fathers and mothers play a role in encouraging the behavior that will transition into lmurran status, so that, by initiation, parents have done a great deal to shape their sons’ masculinity as eventual protectors of the community. Once they have become these glamorous young men, lowland (and some highland) lmurran are indeed visually stunning with their long braids covered in red ochre and elaborate beadwork lovingly constructed by mothers and girlfriends. Lmurran are also allowed considerable sexual license, and their martial deeds (subsequently celebrated in song) are fundamental to their masculine identities in a context within which their prestige is, to an important extent, bound up with their relationships with their young female lovers, who are likewise accorded sexual freedom. Many young girls even build a singira, a girl’s house, in their parents’ compound.
11For their part, girls begin preparing for their role as lmurran girlfriends in early childhood, when their mothers begin to put multiple strands of beads at their necks. By the age of eight or nine, they will join older girls in dances at weddings and other festive occasions, and, as they approach puberty, their adornment will increase considerably. Once girls have reached puberty, their parents have the obligation to “make them a person” by purchasing their daughters enough beads to stretch from neck to breast (Figure 2). This practice has declined considerably in the highlands but is frequently still practiced in the lowlands, as of 2019. Girls’ adornment should consist of a neck full of beads (saen e murt), a beaded headband (saen e nkwe), a beaded waistband (ntore), copper bracelets (surutia), and metal ankle bracelets (nkupili) with beaded decorations above and below (saen e nkeju) (see Nakamura, 2005 on adornment terms). Until recently, and still in 2019 in many cases in the lowlands, lmurran have been obligated to buy these beads and metal ornaments for their girlfriends, alleviating parents of the expense. Sexual pleasure itself is metaphorically referred to in the idiom of giving beads, and thus when older women recollect their youth and assert that their lmurran boyfriends gave them “enough beads,” they are playfully acknowledging the entanglement of adornment, dance, and sexuality. Where the practice continues, parents want their daughters to attract boyfriends and, later, marriageable suitors (Figure 3). Yet, as with lmurran martial behavior, these are ambivalent practices because lmurran should defend the animals and perhaps steal some livestock but not engage in “too much” raiding, while girls should enjoy the sexual freedom of their unmarried days but not get pregnant.
figure 2. Lowland Samburu lmurran girlfriends
© Bilinda Straight, 2002
figure 3. Lowland Samburu teen bride with baby; another girl in background
© Bilinda Straight, 2017
- 3 Lmurran often address girls as “copper bracelets” for the arm bangles lmurran boyfriends gave their (...)
- 4 These lines are from Kakisha, a girls’ and young women’s song form. The particular performance draw (...)
12“Beaded” girlfriend–lmurran relationships are monogamous unions that provide the foundation for nighttime song competitions at which girls boast about their own boyfriends and ridicule entire groups of lmurran as cowards, to encourage them into bravely defending the herds against predators and enemies (see Spencer, 1965; Straight, 2005). During their youthful song performances, when songs of “now” are sung, girls wear their full collection of beads and smear themselves with animal fat and red ochre. They sing praises to their boyfriends and insults towards other girls and their boyfriends, engaging in verbal jousts: “Copper Bracelets [metonymic for girls],3 we don’t praise each other’s boyfriends. Everyone praises her own boyfriend only.”4 Lmurran likewise sing songs praising their own girlfriends and mocking others, and they also sing songs boasting their accomplishments against predators and enemies. These playful song competitions between lmurran and their girlfriends often break off into romantic trysts later.
13It is through the songs girls direct towards their lmurran boyfriends that lmurran purportedly are goaded into violence; this appears to make girls agents shaping Samburu warfare and cattle raiding. The complicating factor is how these gendered cultural practices and gendered agency can become not merely stereotypes but also larger-than-life distractions from politically motivated conflicts over land and resources. Anne-Marie Peatrik has noted the “double obstacle, ethical and empirical in nature, resulting from temporalities and the manipulation of the present,” once the media learn of any murders attributed to pastoralist young men. Continuing, Peatrik observes that “the anthropology of ritual murder has to navigate the tortuous route between alarmist accounts of human barbarity and sacrificial violence, or the pitfalls of exoticism and voyeurism” (Peatrik, 2013: 218). As this paper proceeds, it will cautiously affirm adolescent girls’ playful role vis-à-vis lmurran boyfriends, describe the implication of sexual play in the Powys murder case, and highlight the hazards attendant on granting girls’ songs (or lmurran desires for masculine bravado) substantial causal power in land-based conflicts.
14In the past two decades, a rich literature has developed that emphasizes the intersection of micro and macro political realities in inciting or exacerbating contemporary “ethnic” violence, as well as the importance of collective memory-making to territorial claims or to the construction of ethnicity itself (see for example, Braathen et al., 2000; Broch-Due, 2005; Gallagher, 1997; Malkki, 1995; Pottier, 2005; Ross, 2005; Turton, 1997). It is by now well established that ethnicity in Africa is, as Turton (1997: 3) describes, an “imagined” but not “imaginary” idea through which colonial agents circumscribed, territorialized, and fixed their colonial subjects into groups forced to compete for resources and favor distributed through proliferating centers of colonial administrative action and power. So-called modern African ethnicity thus eloquently expresses the force of the imagination in shaping and perpetuating group conflict that too often is accompanied by horrific acts of violence.
15The processes of colonialism were everywhere marked by contradiction, ambivalence, and intersecting agencies as colonial agents and subjects mutually shaped one another imaginatively and practically (Comaroff, 1985; Comaroff and Comaroff, 1991, 1997; Cooper and Stoler, 1997; Veer, 2001). Nevertheless, intergroup conflict necessarily entails a series of bald oversimplifications through an eliding of subtleties and contradictions. With regard to the forging of identity, Mudimbe points out that identities are always in the process of becoming:
To the basic question of what I am, the response is a temporalization. Forgetting that as projects the individual and the community of which (s)he is a part are future-oriented, both instead define themselves from behind, from a past, and read themselves in terms of fixed characteristics and values (Mudimbe, 2000: 30–31).
16During the colonial period in Kenya, Samburu feminine agency became an effective and enduring occluding trope, harnessed to a process through which Samburu responses to circumscription of their grazing rights came to be represented as unstoppable artefacts of “culture” in ways that have resonance for contemporary land conflicts affecting the Samburu. Tropes about ‘warrior’ masculinity remain launching points of contemporary Euro-American and Kenyan assumptions about Samburu “ethnic” conflicts in the present in ways that bear on broader issues concerning the fraught position of “culture” in analyses of violence. Indeed, the kinds of assumptions examined in this case study persist in media reports and the popular imagination.
- 5 Also see Hodgson, 2001 for a well-researched historical case study of representations affecting Maa (...)
17In the Introduction to their special issue on persistent images in pastoralist representations, Kratz and Gordon (2002) highlight the enduring character of such images and the ways in which they have become normalized in the Euro-American popular imagination. As the authors point out, visual images require narratives to “underwrite them” (ibid.: 251). This essay does not focus on the visual but rather on the stories contextualizing the visual, yet the visual impact of contemporary pastoralists packaged for tourist consumption and ubiquitous in public spaces throughout northern Kenya is a crucial element to the power of these discourses. Thus, germane to this paper’s purposes, Bollig and Heinemann, Galaty, and Kasfir in the same volume all emphasize the political stakes of stereotypic representations of pastoralists. Bollig and Heinemann and Kasfir document the play of gender relations in a number of the popular images associated with the pastoralist mythos: In Namibia, Himba women came to occupy the role of the iconic tribal representative (Bollig and Heinemann, 2002) while Kasfir’s description of Samburu featured in popular films notes an exclusive attention to lmurran—“warriors” (see also Bruner, 2001; Kasfir, 2007).5
18While girls and young women have become muted figures in contemporary representations of Samburu, they are ubiquitous in postcard displays and in performances Samburu young people are asked to enact for tourists, in which young women take their place opposite the lmurran whose masculine identities they reinforce (see Bruner and Kirshenblatt-Gimblett, 1994 for a description of girls in relation to lmurran in tourist display; see also Sobania, 2002). If girls’ role in inciting masculine lmurran bravery alternately protrudes and recedes from the surface in contemporary discourses, it surged provocatively at its framing moment in 1930s colonial Kenya.
19For many scholars working with pastoralists in northern Kenya, the death of Theodore Powys is a famous story, just as it became legendary for the Samburu and the colonial officials who worked with them in the 1930s and for decades thereafter. The story is appealing as an unsolved murder mystery, although were it not for the politics of colonialism as they affected the Samburu, it would have remained a simple tragic accident in a country abounding with accidents. The simplest facts of the matter are that Theodore Powys, nicknamed Dicky, disappeared on October 19, 1931, after riding his horse in the area of Lpingwuan in settler-occupied Laikipia. Later, his mutilated, partial remains pointed to a terrible accident: The coroner determined that he had been thrown from his horse, perhaps when startled by a lion or leopard, and his body either subsequently or immediately scavenged. However, by December of that year, a King’s African Rifles (KAR) soldier reported that Samburu lmurran had murdered Powys and thus any scavenging of Powys’ body had been a later development. While the colonial administration did not believe the story—in fact they had jailed the teller for perjury—a British settler named Gilbert Colville took the story up as a personal crusade on behalf of a settler community that wanted to expel the Samburu from the rich highlands of the Leroghi Plateau. Thus, Powys’ death became symbolic capital in a settler-driven public relations campaign about violent pastoralists who needed to be denied land rights.
- 6 Also see Lesorogol, 2008 on historical and contemporary contexts shaping Samburu land tenure.
20The political-economic aspects of this story and the way in which investigation of the murder played out in colonial politics and subsequent actions affecting the Samburu have been examined elsewhere (Duder and Simpson, 1997; Straight, 2005; Straight and Holtzman, 2004; Waweru, 2002; see also Fratkin, 1991, 2015).6 This paper’s focus is on the intriguing way in which gender and sexuality played into legal discovery in the case and in subsequent policies enacted against the Samburu. In short, while a famous Samburu prophet/ritual specialist named Leaduma was assumed to have sanctioned Samburu lmurran to murder a British settler, in a strange twist, Samburu girls were also causally implicated. More notable still, although the accused lmurran were eventually acquitted and Powys’ death reaffirmed legally as an accident (Duder and Simpson, 1997), Powys’ family and at least some Samburu were not so sure. Indeed, resentments concerning the alleged murder have persisted into the twenty-first century. In numerous conversations I have had with Samburu over the past two decades, contemporary Samburu voice their belief that the case continues to motivate Powys descendants still living on Laikipia Plateau to engage in unfavorable political, land-related actions against Samburu to the present day. This perception, and the persistence of culture as distraction that this paper examines, are at least partly supported by statements the Powys family have made to the press as recently as 2017, accusing Samburu of land-grabbing expansionism (The Telegraph, 2017). Just as in the 1930s, land pressures are all but forgotten.
21My understanding of conflicting Samburu views about girls’ role in warfare crystallized early in my 1992–1994 doctoral fieldwork, when I participated in, and also recorded, song performances with lmurran and their girlfriends. Research in the Kenya National Archives in 1994 revealed colonial attitudes towards Samburu girls. Additional archival research was conducted, and songs recorded in 2001–2002 seem to confirm colonial stereotypes of girls’ role in inciting organized attacks by lmurran against their neighbors. In one song, for example, a Samburu girl asks her lmurran lover to: “Go raid the white-legged cattle of Colson/And don’t return until your blood spills in the ambulance of the D.C. (District Commissioner).” And for their part, lmurran of the Lmooli generation sang song lines like this: “Now listen to what we did on our way/We went to a place where girls incite men to kill and bring back a lot of livestock/And with that, every lmurrani set off to kill and bring back a lot of livestock” (song recorded in 1993).
22That some Samburu continue to attribute to girls the power to incite their boyfriends into violent acts cannot be denied. In interviews recorded from 2003 to 2006, some Samburu mothers and grandmothers complained bitterly about girls’ singing:
As for us who are old, like me now, don’t I have lmurran [sons]? When I hear about a girl who praises the lmurran I really hate that. I say, “Aish! She has sent them.” This girl, this bloody one, when I find her somewhere I will tell her not to say that again when she sings to lmurran because I fear she will send them to raid [respondent F9-2003].
23On the other hand, women nearly universally claim that girls and women have no power over or even knowledge of lmurran warfare:
No one encourages lmurran to go to war. There is no one to encourage them. They just go on their own. There is no one who wants them to go but they just go on their own. They sneak […]. And it is just thought that they went to visit some neighboring settlements but they have gone on that raid [respondent F29-2004].
24In most cases, women claim that there is nothing that can stop lmurran from going to war, and that in fact even their girlfriends may not know they are going until afterward. Thus, girls may goad lmurran as a group while fearing for their own boyfriends.
[Do women and girls want lmurran to go on raids?] Women don’t want that and girls don’t want that very much. But when they go, lmurran don’t tell them. When they go, they don’t tell them […]. Because they will be prevented from going. Nobody wants them to go on a raid [respondent F29-2004].
25As this respondent (F29-2004) continues, her own ambivalence is clear:
[Even a mother doesn’t want her son to go?] Yes, they say no one wants fighting. They just hide when they go. [Even girls don’t want them to go?] Girls also don’t want, but if they tell the girls that it is okay, then girls bless them when they go and they know where they have gone.
26In the same conversation, this respondent (F29-2004) vacillated between whether girls do want lmurran to go, and whether they don’t. Another respondent (F32-2004) demonstrated this ambivalence more clearly. First, she confirmed the nearly universal claim that mothers and sisters do not want their sons and brothers fighting but that everyone is powerless to stop them. Then she claimed that girls want lmurran to go while simultaneously revealing that lmurran cannot bring themselves to tell their young lovers they are going:
[Does a mother want her son to go?] Even if she does not want that person, even if she does not want [him to go] he will not hear her. Who will hear her? [Nobody] Can lmurran hear the words of a woman? Even if she says, “Don’t go,” who hears? They go straight away. He cannot even tell you that he is going. He follows his ltau [heart, will], the way he feels. He does what he feels in his body […]. Some crazy ones [girls] want them to go. [Does a girl beaded by lmurran want them to go?] Yes, she wants [them to go], but nobody tells her, “I’m going.” Who can tell her, “I’m going there” [to war]? They just tell those children [their girlfriends], “Bye, we’re going somewhere. We are going over to those homes.” “We are going to a place like Baragoi.” “We are going near Baragoi.” Or, “We are going to Maralal.” So, they will think they [their boyfriends] have gone there. So, they go for raids. It’s only later that they hear where they have gone [respondent F32-2004].
27The full poignancy of the situation is revealed by another respondent (F25-2004) who said that if everyone returns unharmed bringing animals for their mothers to milk, then the mothers of lmurran who stayed behind feel bad.
28Do girls send their lmurran boyfriends on raids? There can be no conclusive response to this question because the girls’ role is complex, contradictory, and ambivalent. On one hand, a Samburu discourse persists about girls shaming lmurran into violent collective action—a discourse that has undoubtedly been shaped and reinforced by over seventy years of intersecting colonial, settler, and tourist discourses. On the other hand, tropes like that of girls goading lmurran into violence dangerously elide the complexities contributing to intercommunity warfare whether in the past or in its continuing transformations into the present. The cultural entailments of raiding should be read cautiously as one contextual dimension, not as monolithic cause. Yet the colonial tropes described in this paper ignore important factors such as historical changes in land tenure, the role of political elites vying for elections, and even the impact of misreading causal factors for conflict. Thus, it is not only that complexity is flattened, that Samburu sensibilities concerning war are ambivalent and women’s and girls’ power vis-à-vis war precarious; these questionable discourses serve questionable ends through the distraction their deceptive simplicity achieves so effectively.
29Nevertheless, girls’ purported agency surfaced significantly during the colonial period whenever settlers and/or colonial officials rightly or wrongly accused Samburu warriors of violent acts. When it did so, this stereotype of girls’ agency legitimized harsh retaliatory practices against the Samburu.
30As earlier described, Theodore Powys’ death was first determined to be an accident, then rumors circulated about a possible murder—but colonial authorities officially and summarily dismissed these rumors. However, when Gilbert Colville took the rumors up as a personal crusade against the Samburu community on behalf of his own settler community, murder charges were finally brought in 1933 in the wake of alleged Samburu murders of African laborers on European farms. A key aspect of the resulting court case was the implication of the Samburu ritual specialist/prophet (loiboni) Leaduma, who was accused of sending lmurran to war with ritual protection.
- 7 See Straight, 2005 concerning the song in relation to the case.
31Another factor; however, and the one that would have broad consequences for the Samburu community, was the role of girls and their songs about warfare,7 including the daughter of one of the principals named in testimonies collected for the case. Her alleged role had already been raised in 1931, when rumors were first brought to the District Commissioner. While the testimony was dismissed at the time, the transcript resurfaced when the case was revived in 1933. The December 1931 testimony claims that a Samburu “chief” had received six lmurran into his home carrying Powys’ head and genitals, whereupon he had them swear an “oath” upon these and their weapons. Next,
- 8 Kenya National Archives PC/RVP.6A/17/1: statement of December 12, 1931 by Kiberenge to H.H. Traffor (...)
they were instructed by the Chief to take the articles to the girls, one of whom was his own daughter. She: particularly, had to see to the burying of the articles after satisfying themselves [sic] that they had killed their man. The Muran whom they considered had done the best work had the privilege of sleeping with the daughter, the others had their girls.8
32It is worth noting that Powys’ head was found two months later in the general vicinity of his body (in a place where it was probably dragged by scavengers) and yet the rumor of the missing head continued to circulate and appears in later testimony, as well as in interviews I conducted in the early twenty-first century.
33Colonial implication of girls in the alleged murder was further strengthened by anonymous testimony suggesting that Powys’ death was the result of clan competition in the wake of violence between lmurran of the Samburu Lorokushu and Lpisikishu clans. This violence continues to be remembered as a tragic event when Samburu relatives killed relatives, even as it simultaneously highlights the significance of inter-clan politics. At this stage of the case, the unnamed witness claimed,
- 9 Kenya National Archives PC/RVP.6A/17/1: statement by unidentified witness to G.H. Griffiths.
Lorogishu [sic] girls had been annoyed with the moran because of the defeat by Pisingishu [sic] Moran. These eight moran had then told the girls they were going out to kill a European and would not return until they had killed one. They did not go to kill any particular European.9
34The alleged role of girls in lmurran violence, it turns out, is a consistent trope throughout colonial discourse that reached its apex in the Powys case, in which a colonial preoccupation with a form of sexuality at odds with conservative European sensibilities is quite palpable. That is, already by 1921, colonial officials were commenting on the sexuality of lmurran and their girlfriends, with one officer noting what he referred to as “free love” by “both sexes from age ten” along with the prevalence of abortion (KNA/DC/SAM/1/1//1921/6).
35In the Powys case, the apparent colonial fascination with youthful Samburu sexuality is graphically demonstrated in the testimonies of three deposed Samburu girls—one of whom was the niece of Leaduma, the famous loiboni/ritual specialist accused of encouraging Samburu lmurran to kill a European. The similarity in the beginning of each of their testimonies leaves little doubt as to the leading questions colonial authorities were asking these girls. Thus, Leaduma’s niece (referred to as Macharia d/o Ladoma [sic]) begins:
I have lived at Ole Kopen’s [sic] Manyatta as long as I can remember. I have had sexual intercourse with a Moran named Lesoiba; his own name is Langoi; Lesoiba is his father’s name. The last time I had intercourse with him was about 2 months ago […]. I have several times had intercourse with him as is the custom of our tribe for moran and girls to do.
- 10 KNA/PC/RVP/6A/17/1/Enquiry into Death of Mr. Powys by R.G. Darroch, District Officer, Samburu Distr (...)
36The second girl, referred to as Rabasi d/o Lewan, then begins her testimony with “I have lived for a long time at Ole Kopen’s [sic] Manyatta. I have had intercourse with a moran named Kiseker Legalgitele.” The third girl, referred to as Irimiti d/o Lekopen similarly begins her testimony with “I have had intercourse with a Moran named Lotomon Loipiala. He is the only one. He is now in the Manyatta. We often lie together and have done for many months as he has been at the Manyatta for many days without leaving it.”10 Preoccupied with the sexuality of lmurran and their girlfriends, colonial authorities seemed to presume its relevance to violence and murder.
- 11 KNA/PC/RBP/6A/17/1/Letter to Secretary of State for the Colonies, February 24, 1935.
- 12 KNA/DC/SAM/4/1/Samburu District Commissioner Report, 1948 (emphasis added).
37In the aftermath of the acquittal, the presumed power of Samburu girls to goad their boyfriends into violent deeds continued to be referred to both implicitly and explicitly. Thus, in a report summing the entire Powys case, one J. Byrne, Brigadier-General, notes the tendency for a retiring lmurran age set generation to engage in “spear-bloodying” before retiring. While he affirms the finding of Samburu innocence in the Powys affair, he attributes Samburu motivation for the murders of non-Samburu African laborers to this propensity: “It is more than probable that the numerous ‘spear-blooding’ murders of other natives were caused by the outgoing ‘age’ endeavoring to distinguish itself […].”11 Over a decade later, in a District Officer report discussing precautions being taken for the ongoing 1948 initiations of the Lkimaniki lmurran age set and comparing Lkimaniki circumcisions to those of the Lmekuri (1936 onset) and Lkileku (1923 onset) lmurran age set generations, the author repeats this interpretation almost verbatim in reflecting on the violence of the previous (Lkileku) generation, with an addition that makes girls’ culpability explicit: “[… T]his trouble arose largely through the tribal tradition which demands that the warriors must prove their manhood by some exploit before winning a wife. The taunts of the girls seem to have been behind most of their raids.”12 From the perspective of colonial officers working in Samburu District, girls’ agency in provoking violence was obvious and unquestionable—leaving no space for alternative analyses that might touch, for example, on land pressures or resistance to the negative material consequences of colonial rule.
38The Powys case caught Samburu in the competing agendas of British settlers who wanted to alienate the Leroghi Plateau from the Samburu and the British colonial government that resisted that aim (Duder and Simpson, 1997). Formally acquitted of Powys’ death, Samburu lmurran were hanged in connection with the murders of non-Samburu African laborers, and the entire community was heavily punished with disarmament and huge livestock fines enforced by a levy force of King’s African Rifles and police. Waweru (2002) reports that 1,800 spears, 1,300 cattle, 4,200 sheep and goats, 450 donkeys, and cash in the amount of 1,060 British pounds were confiscated. Waweru also states that his respondents claimed a number of atrocities at the hands of the levy force, including numerous rapes. Yet the colonial government was dissatisfied with the number of spears collected, and a second levy force “unleashed terror again on the Samburu. It remained in their midst for the better part of 1936” and, this time, 5,000 spears were taken (Waweru, 2002: 19). In interviews I have conducted, Samburu remember the crippling fines, the beatings, and the hangings. In both interviews and in story-telling and song, Samburu additionally remember the Lmekuri generation that followed as the lmurran age set that used wooden sticks instead of spears, killed locusts instead of raiding livestock, and relinquished treasured beads they had given their girlfriends.
- 13 Paul Spencer, completing his PhD fieldwork at the close of the colonial period, likewise made the c (...)
39Moreover, beginning with the age set charged with Powys’ alleged murder, girls and lmurran sang songs in a code their elders could not understand (as colonial officers also noted), a practice that continued with the Lmekuri and Lkishili generations in order to conceal their activities with their girlfriends from their Samburu elders and the colonial government alike (Spencer, 1965; Straight, 2005). Their need for subterfuge was real. As late as 1955, the author of the Samburu District Annual Report relates: “During June there were reports of a warrior dance called ‘NDO’ or ‘MAMAI’ which was designed to encourage stock theft, disobedience, and seduction of elders’ wives. The dance was proscribed, one conviction resulted, and no further word was heard of it” (KNA/DC/SAM/1/3/1955: 6). Again, these officers implicated young women—and it is important to note that many of these elders’ wives would be in their teens and only recently formally separated from lmurran boyfriends by arranged marriages.13 For this reason, elders colluded with the colonial government—man to man—in repressing the sexual flamboyance of teenage girls (who might include elders’ young wives) and their lmurran boyfriends. In a move reminiscent of mid-twentieth century United States’ censorship of rock lyrics and Elvis Presley’s pelvis, certain beads and songs—the material, sensuous language of Samburu youthfulness—were banned. Decades later, Samburu who had experienced the enactment of these bans reported that colonial police had literally cut beads from girls’ necks, an action tantamount to violent abduction during pastoralist warfare. The colonial government was raiding the Samburu.
*
* *
- 14 Parts of Laikipia were returned to indigenous ownership at different times beginning at Kenyan inde (...)
40This paper has applied a well-known theoretical paradigm about the material efficacy of discourse and hegemony to a case study whose historical depth baldly reveals the power of language. Thus, I have implicated “culture” as a hidden trope because girls’ agency, timeless “warriorhood,” and ethnicity are all the more compelling in the popular imagination as cultural “facts” that are “good to think with” for explaining violence in ways that reverberate in contemporary media and popular discourses. Thus, cultural tropes and political will converge to shore up popular opinion. In the first decade of the twentieth century, Maa-speaking pastoralists lost Laikipia Plateau14 through a process that included the claim that dry season pasture was “vacant land”—a familiar story on several continents. In the 1930s, Maa-speaking Samburu retained control over Leroghi Plateau adjacent to Laikipia but the struggle left a public relations legacy that helped legitimate subsequent punitive campaigns and losses. In 2017, as Kenyan soldiers burned Samburu homes in Laikipia and killed their cattle, The Telegraph (UK) published an article crediting British-descended settlers with positively transforming “unproductive wilderness” in Laikipia while depicting Samburu warriors as invaders from Kenya’s “northern frontier” bringing “death and destruction” and possibly performing “witchcraft rituals” with wild animals (The Telegraph, 2017).
41Cultural tropes of ritualized “warrior” bravery—whether or not influenced by girls—continue then, to shape media reports, popular discourses, and scholarship when pastoralist land rights are debated. Examples abound every time violence of sufficient magnitude to register in the media occurs on Kenya’s Laikipia or Leroghi Plateaus. The following excerpt from a dissertation on elephant conservation in Laikipia—the fraught setting for the Powys death—is another case in point.
Among the pastoralist groups living within Laikipia and the wider region, stealing cattle from neighbouring tribes represents a traditional means of reinforcing social identity among young men belonging to warrior age sets […]. Laikipia from the 1970s onwards were particularly vulnerable to livestock theft. The livestock keeping pastoralists in the Laikipia region have strong traditions of sharing, close community relations and communal land management and they were (and are) quickly able to mobilise groups of cattle raiding parties comprised of young men for acts of ritualised theft. This was and is in sharp contrast to the socially divided immigrant Kikuyu small-scale farmers with divergent origins and livelihoods based on socially and economically individualised resource use and land-tenure (Graham, 2006: 276).
42Graham rightly sees insecurity as a problem for elephant conservation but avoids an analysis of the ways in which ecotourism and conservation policy are implicated in conflicts between pastoralists and farmers (and political elites). Instead, he leaves the reader with the image of young men committing “ritual thefts,” who pose a threat to elephant conservation due to the insecurity they cause, even as his rigorous data analysis simultaneously and somewhat ironically reveals that pastoralists are more tolerant of and compatible with elephants than farmers are. The power of these tropes is illuminated rather starkly with Graham’s note that ethnic tensions (between pastoralist groups as well as between pastoralists and farmers)
underpinned the successful transition of Ilngwezi Group Ranch to an internationally renowned community wildlife conservancy and eco-tourist destination; use of Ilngwezi Group Ranch by its members was hindered by the presence of the neighbouring pastoralists and thus the opportunity cost of designating livestock pasture for wildlife conservation in exchange for security through a deal with the neighbouring large-scale private ranch (now known as Lewa Wildlife Conservancy), was relatively low […]. This trend suggests that insecure pastoralist land is probably the most economical for conservation investment, in terms of area of land designated for wildlife per dollar, in Kenya (Graham, 2006: 277, emphasis added).
- 15 For somewhat more balanced popular media accounts of the 2017 violence in Laikipa, see Newsweek and (...)
43Considering the pastoralist conflicts with Laikipia ranch owners that erupted in 2017 and the violent (government) reprisals against not just the Samburu and Pokot but their cows, the imaginative struggle and its material entailments recounted in this paper are not historical trivia.15 Cultural tropes persist about pastoralists as insidious distractions legitimizing land policies that favor farmers, settler descendants, and other Kenyan elites. The cultural tropes surrounding pastoralist violence that surfaced in the 1930s Powys case invite thoughtful consideration of how damning cultural discourses about East African pastoralists continue to shape land policies.
44It is remarkable that throughout 2016 and 2017—with a severe drought causing widespread food insecurity; a national election on the horizon; and conflict not coincidentally erupting between pastoralists in Laikipia, wealthy landowners with open pasture, and a government with its own interests in Laikipia land tenure—many Samburu elders in the highlands adjacent to Laikipia voiced eagerness—just as elders had in the 1930s—to retire the current lmurran generation and supported the reduction of the practice of “beading girls.” While the colonial government banned certain types of girls’ beads, elders’ attitudes in the twenty-first century coincide with a European donor-supported campaign to eliminate girl “beading,” directly connecting it to harmful sexuality and the martial role of warriors (see, for example, Samburu Women’s Trust, 2016). Yet contemporary land policies in Kenya have little, if anything, to do with gendered adolescent coming of age practices, but everything to do with political marginalization and negative public relations campaigns reminiscent of the one settlers enacted in the 1930s. At the beginning of the twentieth century, the presence of “vacant” land (dry season pasture) justified transferring Laikipia Plateau from Maa-speaking pastoralists to European settlers who would hunt wildlife, farm, and keep cattle and sheep. In the twenty-first century, the same “vacant” land is conserved for wildlife and the government authorizes the shooting of trespassing cows.
45Does the Samburu age and generation system produce girls with the power to incite young men to violence? Samburu ambivalence on this point is key to the classic monograph Paul Spencer produced on generational tensions between elders and youth. Moreover, the question points to a “culture” and youth-blaming trope that simultaneously oversimplifies the varied political, economic and cultural role of girls and lmurran described at the beginning of this paper and legitimizes essentialist media reports and politically-motivated land grabs. Thus, the more precise questions are whether the morally correct response to youthful violent acts is to violently displace entire communities and whether “insecure” pastoralist land ought to be cheaply repurposed for ecotourism.